The Engineering Management Handbook, 2nd Edition 0997519509, 9780997519501

Citation preview

2ND EDITION American Society of Engineering Management

Editors: John V. Farr, S. Jimmy Gandhi, and Donald N. Merino

Engineering Management Handbook, 2nd Edition ISBN: 978-0-9975195-0-1 Published by: The American Society of Engineering Management P.O. Box 820 614 Pine Street, Ste. 206B Rolla, MO 65401

©2016 ASEM. All rights reserved.

Printed in the United States of America. No part of this work may be reproduced or transmitted in any form or by any means, electronic, manual, photocopying, recording, or by any information storage and retrieval system, without prior written permission of the publisher.

ACKNOWLEDGMENTS The American Society of Engineering Management (ASEM) and the authors and editors of this handbook would specifically like to thank Dr. Donald N. Merino, Alexander Crombie Humphreys Chaired Professor of Economics of Engineering Emeritus at Stevens Institute of Technology for his financial contribution that made the writing of the 2nd edition of the Engineering Management Handbook possible. We would also like to thank the editors Dr. John V. Farr and Dr. S. Jimmy Gandhi for their time dedicated to the handbook. The following organization through their generosity supported the final editing and publishing: CIMCIL Moore Stephens Management Training & Consultancy, Belgium Dr. He Jishan Honorary Member of ASEM, Member of the Chinese Academy of Engineering, Honorary President for Life of Junefield College of CSU Old Dominion University Dept. of ​Engineering Management & Systems Engineering Norfolk,Virginia TchI Innovation Change & Innovation in Technical Environments Training & Consultancy, Tremelo, Belgium RS Project Management Consultancy & Training Services Management Training & Consultancy Philippines & UAE URS l CH2M Oak Ridge LLC (UCOR) Nuclear Services & Engineering​ Oak​ ​Ri​​dge, Tennessee University of Arkansas Department of Industrial Engineering Fayetteville, Arkansas University of Central Florida Engineering Leadership & Innovation Institute (eli2) Washington State University Engineering and Technology Management Voiland College of Engineering and Architecture Online master’s and certificate program - etm.wsu.edu

We would like to also acknowledge Mr. Nakul Sharma, Graduate Student at California State University, Northridge for helping organize and supporting the updating of the handbook. He did an outstanding job of managing the development of this handbook. Ms. Lisa Fisher and Kristine Gallo served as the editor and layout, and designed the covers, respectively. Most importantly the editors would like to thank every author who gave of their time to support ASEM through this handbook. Lastly, we would like to thank ASEM leadership for continuing to recognize the value of this handbook.

iii

PREFACE Welcome to the 2nd edition of the Engineering Management Handbook. The first edition was published in 2010. Six years later we have updated much of the original material and added seven new chapters. We hope to continue to update this handbook with plans to produce case studies as supplemental materials, include new material, update and expand the existing material, etc. Engineering managers have traditional been educated to work in the manufacturing sectors but now must succeed in a world where services based industries account for most economic activity. In today’s global business environment, engineer managers must use a wide variety of traditional engineering and leadership skills from the fields of operations research, statistics, management, systems engineering, business, traditional engineering, etc. There is value to having one source that can summarize many of the methods, processes, and tools (MPTs) for mainly the practicing engineering manager. Given this backdrop, we chose to organize this handbook into six sections: • Historical, Professional, and Academic Perspective, • Governance and Management of Engineering Core Competencies, • Quantitative Methods and Modeling, • Accounting, Financial and Economic Basis, • Project Management and Systems Engineering, and • Business Acumen. There are 23 chapters that have been divided into these areas. Most of the 16 chapters in the first edition were updated and in some cases totally rewritten. Seven new chapters were added to this edition to include material addressing patents, intellectual property, multi-generational workers, informatics, quality, innovation, entrepreneurship, and supply chain. The MPTs presented must be viewed as enablers of solutions and not just a collection of traditional academic stovepipes. Like this handbook, engineering management (EM) must evolve to remain relevant in our globally society. This handbook is intended to serve engineering managers and a wide range of professionals in related disciplines to include systems engineers, software engineers, technology management, traditional engineers, etc. In particular, this handbook should serve engineers at all levels in industry, government, and academia involved in the management of professionals and technology. Our goals in writing in this handbook include: • Defining the different MPTs needed by the 21st century engineering manager in support of lifelong learning, • Handbook for academics, • General reference into what is EM, • Reference for the practicing engineering manager, and • Advance the understanding of the complexity of method, processes, and tools needed by all engineers and technologists for our technology centric society. Hopefully, this 2nd edition will continue to serve as a catalyst for follow-on editions. No one source can in depth capture all of the MPTs to be a successful engineering manager. However, our goal was simply to provide one reference that can introduce the readers to certain MPTs. Lifelong learning is the key remaining relevant into today’s technology driven workforce. v

Unfortunately, like many books we had to “freeze” this edition to produce a product with many topics still not addressed. As we were developing the handbook we realized that there are still many MPT gaps in the handbook. For example, chapters on the following topics would greatly improve the handbook: • Performance Appraisals and Metrics, • Mentoring and Coaching, • Statistical Quality Control, • Business Operations of Technical Organization, • Team Dynamics and Management, and • Multinational and Multicultural Issues. John V. Farr United States Military Academy

vi

S. Jimmy Gandhi California State University, Northridge

Donald N. Merino Stevens Institute of Technology

TABLE OF CONTENTS ACKNOWLEDGMENTS................................................................................................................................................................ iii PREFACE...................................................................................................................................................................................... v Chapter 1: Engineering Management—Past, Present, and Future...................................................................................................... 1 1.1 Introduction..................................................................................................................................................................... 2 1.1.1 Overview of Engineering Management....................................................................................................................... 2 1.1.2 The History of the Engineering Management Discipline............................................................................................. 2 1.1.3 Definition of Engineering Management...................................................................................................................... 4 1.2 Present State of the Engineering and Technology Management................................................................................... 5 1.2.1 The Connection of the Engineering Management Discipline to Other Disciplines...................................................... 5 1.2.2 Engineering Management Related Professional Societies............................................................................................ 6 1.2.3 Engineering Management Related Journals................................................................................................................ 6 1.2.4 Engineering Management Related Conferences......................................................................................................... 9 1.2.5 The Future of the Engineering Management Discipline.............................................................................................. 9 1.3 Emerging Engineering Management Related Trends, Drivers, and Challenges............................................................... 9 1.4 Engineering Management Discipline’s Knowledge Roles............................................................................................. 11 1.5 Engineering Management Discipline Stakeholder Needs............................................................................................. 12 1.6 Conclusions and Summary........................................................................................................................................... 13 1.7 References.................................................................................................................................................................... 14 Chapter 2: Professional Responsibility, Ethics, and Legal Issues........................................................................................................ 17 2.1 Introduction.................................................................................................................................................................. 18 2.1.1 Relevance and Importance........................................................................................................................................ 18 2.1.2 What are Ethics?....................................................................................................................................................... 18 2.1.3 What Constitutes Intellectual Property?.................................................................................................................. 18 2.2 Engineering Code of Conduct...................................................................................................................................... 18 2.2.1 Introduction to the NSPE Ethical Canons................................................................................................................... 18 2.2.2 Safety, Health, and Welfare of the Public.................................................................................................................. 18 2.2.3 Professional Service Only in Qualified Areas............................................................................................................. 19 2.2.4 Objective and Truthful Public Statements................................................................................................................. 19 2.2.5 Faithful Agents for Employers or Clients................................................................................................................... 19 2.2.6 Avoidance of Deceptive Acts..................................................................................................................................... 19 2.2.7 Enhancing the Profession Through Ethical Behavior................................................................................................. 19 2.3 Ethical Decision-Making............................................................................................................................................... 20 2.3.1 Introduction............................................................................................................................................................... 20 2.3.2 Utilitarian Rule........................................................................................................................................................... 20 2.3.3 Moral Rights Rule...................................................................................................................................................... 20 2.3.4 Justice Rule................................................................................................................................................................ 21 2.3.5 Practical Rule............................................................................................................................................................. 21 2.3.6 Implementing the Ethical Principles.......................................................................................................................... 21 2.4 Global Considerations in Ethical Conduct .................................................................................................................... 21

vii

2.4.1 The Global Environment............................................................................................................................................ 21 2.4.2 Laws and Codes for International Business............................................................................................................... 22 2.4.3 Ethical Decision Making in the Global Environment.................................................................................................. 22 2.5 Protecting Employees Who Raise Ethical Issues........................................................................................................... 22 2.5.1 Introduction............................................................................................................................................................... 22 2.5.2 Creating the Right Culture......................................................................................................................................... 22 2.5.3 Sarbanes-Oxley Act.................................................................................................................................................... 23 2.6 Responsibilities for Intellectual Property...................................................................................................................... 23 2.6.1 Why Protect Intellectual Property?........................................................................................................................... 23 2.6.2 Invention Disclosure Processes.................................................................................................................................. 23 2.6.3 Patents....................................................................................................................................................................... 23 2.6.4 Copyrights.................................................................................................................................................................. 24 2.7 References.................................................................................................................................................................... 24 2.8 Other Sources of Information....................................................................................................................................... 24 Chapter 3: Management Theory and Concepts ................................................................................................................................. 25 3.1 Introduction.................................................................................................................................................................. 26 3.2 Historical Perspective................................................................................................................................................... 26 3.3 Scientific Management................................................................................................................................................. 27 3.4 The Bureaucracy.......................................................................................................................................................... 27 3.4.1 A Critique.................................................................................................................................................................. 28 3.5 Behavioral Approaches................................................................................................................................................. 29 3.6 Quantitative Methods ................................................................................................................................................. 29 3.7 Summary...................................................................................................................................................................... 29 3.8 Attempts at Integration................................................................................................................................................ 30 3.9 What Is Working?........................................................................................................................................................ 30 3.10 Conclusion ................................................................................................................................................................. 31 3.11 References ................................................................................................................................................................. 31 Chapter 4: Managing Knowledge Workers......................................................................................................................................... 33 4.1 Introduction.................................................................................................................................................................. 34 4.1.1 Attempts at Integration............................................................................................................................................. 34 4.1.2 Summary................................................................................................................................................................... 35 4.2 How It All Works Together............................................................................................................................................ 35 4.2.1 The “Integrated” Part................................................................................................................................................ 35 4.2.2 The External Environment ........................................................................................................................................ 36 4.2.3 The Internal Environment ......................................................................................................................................... 37 4.2.4 Management Systems............................................................................................................................................... 37 4.2.5 Organizational Structure............................................................................................................................................ 37 4.2.6 People (Orientation).................................................................................................................................................. 37 4.2.7 The Model................................................................................................................................................................. 38 4.3 The “People” Orientation............................................................................................................................................. 38 4.3.1 Background on Behavioral Approaches..................................................................................................................... 38 4.3.2 McGregor’s Theory X and Theory Y........................................................................................................................... 39 4.3.3 Maslow’s Hierarchy of Human Needs ....................................................................................................................... 39 4.3.4 Herzberg’s Motivators and Hygienes......................................................................................................................... 40 4.3.5 McClelland’s Need to Achieve................................................................................................................................... 41 4.3.6 Pink’s Motivational Concepts.................................................................................................................................... 42 4.3.7 B. F. Skinner’s Operant Conditioning Theory.............................................................................................................. 42 4.3.8 Multiple Generations in the Workplace.................................................................................................................... 43 4.3.9 Summary.................................................................................................................................................................... 44 4.4 People Orientation—Team Management ..................................................................................................................... 44

viii

4.4.1 Likert—An Integrating Principle.................................................................................................................................. 45 4.4.2 Characteristics of High Producing Organizations........................................................................................................45 4.4.3 Characteristics of Low Producing Organizations ........................................................................................................45 4.4.4 Team Management..................................................................................................................................................... 46 4.4.5 Likert’s System IV........................................................................................................................................................ 48 4.4.6 Blake and Mouton’s Managerial Grid......................................................................................................................... 49 4.4.7 The Managerial Grid .................................................................................................................................................. 50 4.5 Summary....................................................................................................................................................................... 51 4.6 References .................................................................................................................................................................... 51 Chapter 5: Types of Intellectual Property .......................................................................................................................................... 53 5.1 Introduction.................................................................................................................................................................. 54 5.2 Copyrights..................................................................................................................................................................... 54 5.3 Trademarks.................................................................................................................................................................... 55 5.4 Trade Secrets................................................................................................................................................................ 56 5.5 Patents.......................................................................................................................................................................... 57 5.6 Obtaining a Patent........................................................................................................................................................ 60 5.7 Design Patents.............................................................................................................................................................. 61 5.8 Plant Patents................................................................................................................................................................. 61 5.9 Utility Patents............................................................................................................................................................... 61 5.10 Major Elements of Patent Application........................................................................................................................ 62 Chapter 6: What is a Patent?.............................................................................................................................................................. 65 6.1 Where Patents are Held................................................................................................................................................ 66 6.2 What are the Laws and Rules in the U.S.?.................................................................................................................... 67 6.3 The Courts.................................................................................................................................................................... 69 6.4 Reading a Patent........................................................................................................................................................... 71 6.4.1 Sample Drawing......................................................................................................................................................... 74 6.4.2 Typical First Two Pages of the Specification.............................................................................................................. 75 6.4.3 Claims Section........................................................................................................................................................... 79 6.5 What Can You Do with a Patent?.................................................................................................................................. 80 Chapter 7: Leading Individuals and Engineering Project Teams ........................................................................................................ 83 7.1 Introduction ................................................................................................................................................................. 84 7.2 The Leader.................................................................................................................................................................... 84 7.3 Leading Individuals....................................................................................................................................................... 85 7.4 Leading Teams.............................................................................................................................................................. 87 7.4.1 Forming..................................................................................................................................................................... 87 7.4.2 Storming.................................................................................................................................................................... 87 7.4.3 Norming..................................................................................................................................................................... 88 7.4.4 Performing................................................................................................................................................................. 88 7.4.5 Adjourning................................................................................................................................................................. 88 7.5 Leading......................................................................................................................................................................... 88 7.6 A Leadership Development Model ............................................................................................................................. 89 7.6.1 Assessment................................................................................................................................................................ 89 7.6.2 Challenge .................................................................................................................................................................. 90 7.6.3 Support...................................................................................................................................................................... 90 7.7 Closing Thoughts.......................................................................................................................................................... 91 7.8 References.................................................................................................................................................................... 91

ix

Chapter 8: Managing the Multi-Generational Knowledge Based Workforce..................................................................................... 93 8.1 Introduction ................................................................................................................................................................. 94 8.1.1 Overview................................................................................................................................................................... 94 8.2 Generations.................................................................................................................................................................. 94 8.2.1 Baby Boomers............................................................................................................................................................ 95 8.2.2 Generation X (Gen X’ers)........................................................................................................................................... 95 8.2.3 Generation Y (Gen Y)................................................................................................................................................. 96 8.3 Management Impacts................................................................................................................................................... 96 8.3.1 Baby Boomers............................................................................................................................................................ 96 8.3.2 Generation X.............................................................................................................................................................. 96 8.3.3 Generation Y.............................................................................................................................................................. 97 8.4 Management Strategies for Leaders and Followers..................................................................................................... 97 8.5 Optional Content Commitment ................................................................................................................................... 97 8.5.1 Commitment and the Generations............................................................................................................................ 98 8.6 Recommendations for the Management Discipline..................................................................................................... 99 8.6.1 Understanding........................................................................................................................................................... 99 8.6.2 Bias............................................................................................................................................................................ 99 8.7 References.................................................................................................................................................................. 100 Chapter 9: Operations Research....................................................................................................................................................... 105 9.1 Introduction to Operations Research Modeling......................................................................................................... 106 9.1.1 Importance of Operations Research for Engineering Managers.............................................................................. 106 9.1.2 History of Operations Research............................................................................................................................... 106 9.1.3 Operations Research Methodology......................................................................................................................... 106 9.2 Deterministic Models ................................................................................................................................................ 108 9.2.1 Linear Programming................................................................................................................................................ 108 9.2.2 Basic Problem Formulation...................................................................................................................................... 108 9.2.3 Sensitivity Analysis.................................................................................................................................................. 110 9.2.4 Duality Theory......................................................................................................................................................... 111 9.2.5 Applications............................................................................................................................................................. 112 9.2.6 Integer Programming............................................................................................................................................... 112 9.2.7 Solution Techniques................................................................................................................................................. 113 9.2.8 Binary and Auxiliary Binary Variable........................................................................................................................ 113 9.2.9 Applications............................................................................................................................................................. 113 9.3 Non-Linear Programming........................................................................................................................................... 113 9.3.1 Solution Techniques................................................................................................................................................. 113 9.3.2 Separable Programming.......................................................................................................................................... 113 9.3.3 Applications ............................................................................................................................................................ 114 9.4 Dynamic Programming............................................................................................................................................... 114 9.4.1 Solution Techniques................................................................................................................................................. 114 9.4.2 Applications............................................................................................................................................................. 114 9.5 Stochastic Models ...................................................................................................................................................... 115 9.5.1 Markov Chains......................................................................................................................................................... 115 9.5.2 Discrete-time Markov Chains.................................................................................................................................. 115 9.5.3 Semi-Markov Processes........................................................................................................................................... 117 9.5.4 Queuing Theory....................................................................................................................................................... 118 9.5.5 Stability of Queues................................................................................................................................................119 9.5.6 Little’s Rule............................................................................................................................................................119 9.5.7 Single-Server Single-Channel Queues....................................................................................................................119 9.5.8 Multiple-Server Queues........................................................................................................................................121 9.6 Advanced and Other Topics......................................................................................................................................121 9.6.1 Meta-heuristics......................................................................................................................................................121

x

9.6.2 Advanced Stochastic Models................................................................................................................................122 9.6.3 Brownian Motion...................................................................................................................................................124 9.6.4 Discrete-Event Simulation.....................................................................................................................................125 9.7 Big Data and Operations Research...........................................................................................................................126 9.7.1 An Introduction to Big Data and the Internet of Things........................................................................................126 9.7.2 Big Data Value Chain..............................................................................................................................................127 9.7.3 Big Data Case Studies............................................................................................................................................128 9.8 References ............................................................................................................................................................... 128 Chapter 10: Simulation............................................................................................................................................................131 10.1 Introduction ............................................................................................................................................................. 132 10.1.1 Importance of Simulation..................................................................................................................................... 132 10.1.2 Key Terms of Simulation ...................................................................................................................................... 133 10.1.3 Simulation Theory for Engineers.......................................................................................................................... 134 10.1.4 Simulation Applications for Engineers.................................................................................................................. 134 10.1.5 Simulation Engineering......................................................................................................................................... 135 10.1.6 Modeling and Simulation as a Discipline............................................................................................................... 135 10.2 Simulation Theory.................................................................................................................................................... 136 10.2.1 Mathematical Foundations................................................................................................................................... 136 10.2.2 Computer Science Foundations............................................................................................................................ 138 10.2.3 Discrete Event Simulation..................................................................................................................................... 139 10.2.4 Data Analysis........................................................................................................................................................ 141 10.2.5 Monte-Carlo Simulation and Continuous Simulation........................................................................................... 141 10.3 Simulation Applications .......................................................................................................................................... 143 10.3.1 Simulation as an Engineering Method.................................................................................................................. 143 10.3.2 Simulation with ARENA........................................................................................................................................ 145 10.3.3 Agent-based Modeling......................................................................................................................................... 146 10.3.4 Simulation and Systems Engineering Models........................................................................................................ 148 10.4 References................................................................................................................................................................ 149 Chapter 11: Decision Analysis.................................................................................................................................................151 11.1 Introduction ............................................................................................................................................................. 152 11.1.1 What is Decision Analysis?................................................................................................................................... 152 11.1.2 Why Use Decision Analysis?................................................................................................................................. 152 11.1.3 When Do You Use Decision Analysis?................................................................................................................... 152 11.1.4 Who Uses Decision Analysis? .............................................................................................................................. 152 11.2 Decision Processes................................................................................................................................................... 153 11.2.1 Challenges ............................................................................................................................................................ 153 11.2.2 Analytical Process.................................................................................................................................................. 153 11.2.3 Decision Conference Process................................................................................................................................. 153 11.2.4 Dialog Decision Process......................................................................................................................................... 154 11.2.5 Advantages and Disadvantages of Decision Processes.......................................................................................... 155 11.3 Decision Elements.................................................................................................................................................... 155 11.3.1 Values and Outcomes ........................................................................................................................................... 155 11.3.2 Uncertainty ........................................................................................................................................................... 155 11.3.3 Decisions ............................................................................................................................................................... 156 11.4 Decision Modeling—Illustrative Product Development Example............................................................................. 156 11.4.1 Basic Influence Diagram ....................................................................................................................................... 156 11.4.2 Basic Decision Tree ............................................................................................................................................... 156 11.4.3 Basic Risk Profile.................................................................................................................................................... 157 11.4.4 Value of a Test ....................................................................................................................................................... 158 11.4.5 Value of Imperfect Information about Market Success ....................................................................................... 159

xi

11.4.6 Value of Perfect Information About Market Success ............................................................................................ 160 11.4.7 Value of Control..................................................................................................................................................... 161 11.4.8 Sensitivity Analysis................................................................................................................................................ 161 11.4.9 Comparison of Influence Diagrams and Decision Trees ........................................................................................ 161 11.5 Single Attribute Utility.............................................................................................................................................. 161 11.5.1 Utility .................................................................................................................................................................... 161 11.5.2 Risk Preference ..................................................................................................................................................... 162 11.5.3 Utility with Decision Trees .................................................................................................................................... 162 11.6 Multiple Objective Decision Analysis (MODA).......................................................................................................... 162 11.6.1 Additive Value Model............................................................................................................................................ 162 11.6.2 Value Functions .................................................................................................................................................... 163 11.6.3 Swing Weights ...................................................................................................................................................... 163 11.6.4 Swing Weight Matrix ............................................................................................................................................. 164 11.6.5 Multiple Objective Decision Analysis with Decision Trees .................................................................................... 165 11.7 Role of Engineering Manager................................................................................................................................... 165 11.8 Advanced and Other Topics...................................................................................................................................... 166 11.9 References................................................................................................................................................................ 166 Chapter 12: Multi-Criteria Analysis.........................................................................................................................................169 12.1 Introduction to Multi-Criteria Analysis .................................................................................................................... 170 12.1.1 Background............................................................................................................................................................ 170 12.1.2 Overview of Multi-Criteria Analysis....................................................................................................................... 170 12.1.3 Relevance of MCA to Engineering Management................................................................................................... 170 12.2 Analytic Hierarchy Process....................................................................................................................................... 171 12.2.1 Overview of Analytic Hierarchy Process................................................................................................................ 171 12.2.2 The AHP Process................................................................................................................................................... 171 12.2.3 Advantages and Limitations of AHP as a Multi-Criteria Tool................................................................................. 174 12.2.4 Conclusion............................................................................................................................................................. 175 12.3 Analytic Network Process......................................................................................................................................... 175 12.3.1 Overview of Analytic Network Process.................................................................................................................. 175 12.3.1.1 ANP Structure..................................................................................................................................................... 175 12.3.2 The ANP Process.................................................................................................................................................... 175 12.3.4 Benefits and Limitations of ANP as a Multi-Criteria Tool ...................................................................................... 178 12.3.3 Conclusion ............................................................................................................................................................ 178 12.4 Multi-Attribute Analysis (MAA)................................................................................................................................ 179 12.4.1 Overview............................................................................................................................................................... 179 12.4.2 The Multi-Attribute Analysis (MAA) Process......................................................................................................... 179 12.4.3 Utility Theory (Utility Analysis).............................................................................................................................. 180 12.4.4 Conclusion............................................................................................................................................................. 181 12.5 References................................................................................................................................................................ 182

xii

Chapter 13: Engineering Informatics – State of the Art and Future Trends............................................................................183 13.1 Introduction ............................................................................................................................................................. 184 13.2 Overview of Engineering Information Integration.................................................................................................... 187 13.2.1 IIIE-A New Discipline of Industrial Information Integration.................................................................................... 187 13.2.2 Engineering Integration......................................................................................................................................... 188 13.3 Enabling Technologies.............................................................................................................................................. 191 13.3.1 Business Process Management............................................................................................................................. 191 13.3.2 Information Integration and Interoperability........................................................................................................ 194 13.3.3 Enterprise Architecture and Enterprise Application Integration........................................................................... 195 13.3.4 Service-oriented Architecture (SOA)..................................................................................................................... 196 13.4 Summary and Challenges......................................................................................................................................... 196 13.5 References................................................................................................................................................................ 197

Chapter 14: Basic Accounting and Finance.............................................................................................................................199 14.1 Introduction ............................................................................................................................................................. 200 14.1.1 Importance of Accounting to Engineers................................................................................................................ 200 14.1.2 Accounting and Engineering Economics................................................................................................................ 200 14.1.3 What is Accounting? ............................................................................................................................................. 200 14.1.4 Users of Accounting Information .......................................................................................................................... 200 14.2 Basic Accounting....................................................................................................................................................... 200 14.2.1 Introduction........................................................................................................................................................... 200 14.2.2 Financial Accounting.............................................................................................................................................. 201 14.2.3 Transactions........................................................................................................................................................... 202 14.2.4 Financial Condition................................................................................................................................................ 202 14.2.5 Financial Statement Terminology.......................................................................................................................... 202 14.2.6 Financial Performance........................................................................................................................................... 204 14.2.7 Accounting Equation.............................................................................................................................................. 204 14.3 Income Statement.................................................................................................................................................... 204 14.3.1 Introduction........................................................................................................................................................... 204 14.3.2 The Income Statement......................................................................................................................................... 205 14.4 Balance Sheet........................................................................................................................................................... 206 14.4.1 Introduction............................................................................................................................................................ 206 14.4.2 The Balance Sheet................................................................................................................................................. 206 14.5 Stockholder’s (Owner’s) Equity................................................................................................................................ 209 14.5.1 Introduction........................................................................................................................................................... 209 14.5.2 Stockholder’s Equity.............................................................................................................................................. 210 14.5.3 Paid-In Capital ....................................................................................................................................................... 210 14.5.4 Retained Earnings.................................................................................................................................................. 211 14.5.5 Example of Retained Earnings .............................................................................................................................. 211 14.6 Cash Flow Statement................................................................................................................................................ 211 14.6.1 Introduction........................................................................................................................................................... 211 14.6.2 The Cash Flow Statement...................................................................................................................................... 212 14.6.3 Example of Cash Flow Statement.......................................................................................................................... 213 14.7 Depreciation............................................................................................................................................................. 214 14.7.1 Introduction........................................................................................................................................................... 214 14.7.2 Depreciation Terminology..................................................................................................................................... 214 14.7.3 Depreciation Methods........................................................................................................................................... 215 14.7.4 Gains and Losses from the Disposal of Assets....................................................................................................... 217 14.8 After Tax Analysis...................................................................................................................................................... 217 14.8.1 Introduction .......................................................................................................................................................... 217 14.8.2 After Tax Analysis and Cash Flow.......................................................................................................................... 218 14.8.3 After Tax Cash Flow from Depreciation Charges.................................................................................................... 218 14.8.4 After Tax Cash Flow from Investment Tax Credit .................................................................................................. 218 14.8.5 After Tax Cash Flows from Loans........................................................................................................................... 219 14.9 Accounting Process................................................................................................................................................... 219 14.9.1 Introduction........................................................................................................................................................... 219 14.9.2 The Accounting Process......................................................................................................................................... 219 14.9.3 Double Entry Accounting....................................................................................................................................... 219 14.10 Financial and Managerial Accounting..................................................................................................................... 220 14.10.1 Introduction......................................................................................................................................................... 220 14.10.2 Breakeven Analysis.............................................................................................................................................. 220 14.10.3 Contribution Margin............................................................................................................................................ 221 14.10.4 Contribution Margin Ratio................................................................................................................................... 221 14.10.5 Breakeven Sales in Dollars................................................................................................................................... 221 14.10.6 Target Net Profit.................................................................................................................................................. 222

xiii

14.10.7 Sales to Achieve Target Return on Sales.............................................................................................................. 222 14.10.8 Degree of Operating Leverage............................................................................................................................. 222 14.11 Advanced and Other Topics.................................................................................................................................... 222 14.12 Summary................................................................................................................................................................ 223 14.13 References.............................................................................................................................................................. 223 Chapter 15: Engineering Economics........................................................................................................................................225 15.1 Capital Expenditures................................................................................................................................................. 226 15.1.1 Importance of Capital............................................................................................................................................ 226 15.1.2 Capital Selection Process....................................................................................................................................... 226 15.1.3 Minimum Attractive Rate of Return ...................................................................................................................... 226 15.2 Mathematics of Finance........................................................................................................................................... 226 15.2.1 Rates of Return....................................................................................................................................................... 226 15.2.2 Simple and Compound Interest ............................................................................................................................ 226 15.2.3 Effective Interest Rates.......................................................................................................................................... 229 15.2.4 Compounding and Discounting............................................................................................................................. 230 15.2.5 Cash Flow Patterns................................................................................................................................................ 230 15.2.6 Loan Programs and Personal Finance..................................................................................................................... 232 15.3 Figures of Merit (FoM).............................................................................................................................................. 232 15.3.1 Present Worth ....................................................................................................................................................... 232 15.3.2 Annual Worth........................................................................................................................................................ 233 15.3.3 Future Worth......................................................................................................................................................... 233 15.3.4 Capital Recovery.................................................................................................................................................... 234 15.3.5 Capitalized Cost..................................................................................................................................................... 234 15.3.6 Internal Rate of Return (IRR) ................................................................................................................................ 235 15.3.7 Benefit Cost Analysis (BCA)................................................................................................................................... 235 15.4 Retirements and Replacements................................................................................................................................ 237 15.4.1 Types of Retirement and Replacement Problems................................................................................................. 237 15.4.2 Total Costs and Economic Life................................................................................................................................ 237 15.4.3 Capitalized Costs.................................................................................................................................................... 238 15.4.4 Operating and Maintenance Costs........................................................................................................................ 239 15.5 Inflation.................................................................................................................................................................... 240 15.5.1 Causes of Inflation.................................................................................................................................................. 240 15.5.2 Types of Inflation................................................................................................................................................... 240 15.5.3 Using Price Indices................................................................................................................................................. 241 15.5.4 Inflation and MARR............................................................................................................................................... 241 15.5.5 Cash Flows and Inflation........................................................................................................................................ 241 15.5.6 Common Problems Relating to Inflation................................................................................................................ 242 15.6 After Tax Analysis...................................................................................................................................................... 242 15.6.1 Net Cash Flow from Operating Activities .............................................................................................................. 243 15.6.2 Net Cash Flow from Capital Related Line............................................................................................................... 243 15.6.3 After Tax Cash Flow from Depreciation Charges.................................................................................................... 244 15.6.4 After Tax Cash Flow from Investment Tax Credit .................................................................................................. 244 15.6.5 After Tax Cash Flows from Loans........................................................................................................................... 244 15.6.6 After Tax Cash Flow for Salvage/Disposal of Assets .............................................................................................. 245 15.6.7 Total Cash Flow Discounted................................................................................................................................... 245 15.6.8 ATA Example.......................................................................................................................................................... 245 15.7 Decision Analysis..................................................................................................................................................... 248 15.7.1 Types of Problems................................................................................................................................................. 248 15.7.2 Choosing Among Alternatives Using ATA............................................................................................................... 248 15.7.3 Decision Model...................................................................................................................................................... 249 15.8 References................................................................................................................................................................ 250

xiv

Chapter 16: Project Management’s Role in Engineering Management..................................................................................251 16.1 Introduction.............................................................................................................................................................. 252 16.2 Project Management................................................................................................................................................ 252 16.2.1 Initiating Processes................................................................................................................................................ 254 16.2.2 Planning Processes................................................................................................................................................ 255 16.2.3 Executing Processes............................................................................................................................................... 259 16.2.4 Monitoring and Controlling Processes.................................................................................................................. 259 16.2.5 Closing Processes................................................................................................................................................... 263 16.3 Summary.................................................................................................................................................................. 263 16.4 References................................................................................................................................................................ 264 Chapter 17: Systems Engineering............................................................................................................................................265 17.1 Introduction.............................................................................................................................................................. 266 17.1.1 What is Systems Engineering?............................................................................................................................... 266 17.1.2 Why Did Systems Engineering Originate?............................................................................................................. 267 17.1.3 The Systems Engineering Lifecycle....................................................................................................................... 268 17.1.4 The Role of Systems Modeling and Systems Simulation........................................................................................ 270 17.2 Stakeholder Requirements Definition...................................................................................................................... 270 17.2.1 Use Cases and Scenarios....................................................................................................................................... 271 17.2.2 Performance Criteria............................................................................................................................................. 271 17.2.3 Inputs and Outputs................................................................................................................................................ 271 17.2.4 Conclusions............................................................................................................................................................ 272 17.3 Requirements Analysis............................................................................................................................................. 273 17.4 Architectural Design................................................................................................................................................. 274 17.5 Implementation........................................................................................................................................................ 274 17.5.1 Intermediate Specifications................................................................................................................................... 275 17.5.2 Peer Reviews......................................................................................................................................................... 275 17.5.3 Quality Inspections................................................................................................................................................ 276 17.6 Integration................................................................................................................................................................ 276 17.6.1 Big Bang Approach................................................................................................................................................ 276 17.6.2 Bottom-Up Approach............................................................................................................................................ 277 17.7 Verification............................................................................................................................................................... 277 17.7.1 Inspection.............................................................................................................................................................. 277 17.7.2 Demonstration....................................................................................................................................................... 278 17.7.3 Analysis.................................................................................................................................................................. 278 17.7.4 Test........................................................................................................................................................................ 278 17.8 Transition.................................................................................................................................................................. 278 17.9 Validation.................................................................................................................................................................. 278 17.10 Operation and Maintenance.................................................................................................................................. 278 17.11 Disposal.................................................................................................................................................................. 279 17.12 Conclusions............................................................................................................................................................. 279 17.13 References.............................................................................................................................................................. 280 Chapter 18: Systems Thinking.................................................................................................................................................281 18.1 Introduction ............................................................................................................................................................. 282 18.1.1 Changing Domain for Engineering Managers........................................................................................................ 283 18.1.2 Systems Thinking Challenges for Engineering Managers....................................................................................... 285 18.1.3 Implications of Systems Thinking (ST) for Engineering Management (EM)........................................................... 286 18.2 Overview of Systems Thinking ................................................................................................................................. 287 18.2.1 Nature of Systems Thinking................................................................................................................................... 287 18.2.2 Hard and Soft Systems Thinking............................................................................................................................ 288 18.2.3 Roles for Systems Thinking in Engineering Management....................................................................................... 290

xv

18.3 Philosophy and Central Concepts for Systems Thinking .......................................................................................... 291 18.3.1 Philosophical Basis for Systems Thinking............................................................................................................... 291 18.3.2 Foundation Principles for Systems Thinking.......................................................................................................... 292 18.4.3 Using the Foundations for Systems Thinking......................................................................................................... 299 18.4 Systems Thinking: Dealing with Complexity ............................................................................................................ 300 18.4.1 Nature and Challenge of Complexity for Engineering Managers........................................................................... 300 18.5 Systems Thinking Capacity for Engineering Managers............................................................................................. 301 18.6 Systems Thinking as a Responsive Strategy for Dealing with Complexity................................................................. 303 18.7 Methods and Tools for Systems Thinking ................................................................................................................ 304 18.7.1 Role of Tools for Systems Thinking........................................................................................................................ 305 18.7.2 Taxonomy of Tools for Systems Thinking............................................................................................................... 308 18.7.3 Selected Tools for Systems Thinking...................................................................................................................... 309 18.8 Systems Thinking Implications for Engineering Management ................................................................................ 309 18.8.1 Using Systems Thinking in Engineering Management........................................................................................... 309 18.8.2 Limitations for Systems Thinking........................................................................................................................... 310 18.9 Conclusion ............................................................................................................................................................... 311 18.10 References ............................................................................................................................................................. 313 Chapter 19: Risk Management................................................................................................................................................317 19.1 Introduction ............................................................................................................................................................. 318 19.1.1 What are Risks, Hazards and Accidents?............................................................................................................... 318 19.1.2 Why is Managing Risk Important for Engineering Managers?............................................................................... 319 19.1.3 What is Risk Management?................................................................................................................................... 319 19.1.4 What Are the Basic Activities in Risk Management?............................................................................................. 320 19.2 Scenario Identification ............................................................................................................................................. 321 19.2.1 Identify Desired Scenarios..................................................................................................................................... 322 19.2.2 Identify Risk Scenarios........................................................................................................................................... 322 19.2.3 Characterize Risk Scenarios Via Causalities and Correlation................................................................................. 322 19.3 Consequence and Likelihood Estimation.................................................................................................................. 326 19.3.1 Consequence Estimation....................................................................................................................................... 326 19.3.2 Likelihood Estimation............................................................................................................................................ 326 19.4 Risk Ranking.............................................................................................................................................................. 328 19.4.1 Risk Ranking Using Consequence and Likelihood.................................................................................................. 328 19.4.2 Risk Ranking Using Other Properties..................................................................................................................... 328 19.5 Generation and Tradeoff of Mitigation Alternatives................................................................................................. 329 19.5.1 Mitigation Strategies............................................................................................................................................. 329 19.5.2 As Low As Reasonably Practicable......................................................................................................................... 330 19.5.3 Comparative Analysis for Tradeoffs....................................................................................................................... 331 19.5.3 Analytic Hierarchy Process (AHP).......................................................................................................................... 332 19.6 Potential Problem Analysis....................................................................................................................................... 333 19.6.1 Stone-in-the-Pond Analogy................................................................................................................................... 333 19.7 Implementation, Documentation and Monitoring................................................................................................... 333 19.7.1 The Need for Documentation and Monitoring...................................................................................................... 333 19.7.2 Lessons Learned for the Engineering Managers.................................................................................................... 333 19.8 Advanced and Other Topics..................................................................................................................................... 334 19.9 References................................................................................................................................................................ 335 Chapter 20: What is Quality Management?............................................................................................................................337 20.1 A Definition of Quality Management........................................................................................................................ 338 20.2 A Brief History of Quality Management................................................................................................................... 338 20.3 Quality Planning....................................................................................................................................................... 345 20.4 Quality Assurance..................................................................................................................................................... 348

xvi

20.5 Quality Control......................................................................................................................................................... 349 20.6 Quality Improvement............................................................................................................................................... 349 20.7 The Seven Quality Tools............................................................................................................................................ 352 20.7.1 Process Maps......................................................................................................................................................... 352 20.7.2 Checklists............................................................................................................................................................... 353 20.7.3 Histograms............................................................................................................................................................. 353 20.7.4 Pareto Charts......................................................................................................................................................... 353 20.7.5 Ishikawa Diagrams................................................................................................................................................. 355 20.7.6 Control Charts........................................................................................................................................................ 356 20.7.7 Scatter Charts........................................................................................................................................................ 357 20.8 Quality Leadership and Teams.................................................................................................................................. 358 20.9 References................................................................................................................................................................ 359 Chapter 21: Strategic Management........................................................................................................................................361 21.1 Introduction ............................................................................................................................................................. 362 21.1.1 Importance of Strategic Management to Engineering Managers......................................................................... 362 21.1.2 What is Strategic Management? .......................................................................................................................... 362 21.2 Strategic Management Process................................................................................................................................ 363 21.2.1 Introduction............................................................................................................................................................ 363 21.2.2 Core of the Strategic Management Process.......................................................................................................... 363 21.2.3 Strategic Management Process Functions............................................................................................................. 365 21.2.4 Setting Strategic Intent Through Strategic Planning.............................................................................................. 366 21.2.5 Deploying the Strategic Intent............................................................................................................................... 368 21.2.6 Setting Strategy Through Implementation Planning.............................................................................................. 369 21.2.7 Deploying Resources............................................................................................................................................. 371 21.2.8 Executing the Strategy........................................................................................................................................... 372 21.2.9 Deploying Results.................................................................................................................................................. 374 21.2.10 Reviewing Performance Through Performance Evaluation................................................................................. 376 21.2.11 Deploying Learnings............................................................................................................................................ 376 21.3 Bibliography and References.................................................................................................................................... 377 Chapter 22: Innovation and Entrepreneurship .......................................................................................................................379 22.1 Introduction.............................................................................................................................................................. 380 22.2 Entrepreneurial Mindset and Qualities of an Entrepreneur..................................................................................... 380 22.2.1 So What Exactly is Entrepreneurial Mindset?........................................................................................................ 380 22.3 Functions of an Entrepreneur................................................................................................................................... 381 22.4 Intrapreneur............................................................................................................................................................. 382 22.4.1 Differences between Entrepreneur and Intrapreneur........................................................................................... 382 22.5 Innovation and Its Importance in Entrepreneurship................................................................................................ 383 22.6 Innovation: The Likely Cause of Successful Sustainable Businesses........................................................................ 385 22.6.1 Types of Innovation............................................................................................................................................... 385 22.6.2 Factors Affecting Innovation.................................................................................................................................. 386 22.7 Metrics to Measure Innovation................................................................................................................................ 386 22.8 The Design Thinking Process.................................................................................................................................... 387 22.9 The Entrepreneurial Process.................................................................................................................................... 388 22.10 How Can Small/Medium Enterprises (SMEs) Incorporate Innovation.................................................................... 389 22.11 Continuous Improvement....................................................................................................................................... 390 22.12 Competitive Advantages for the U.S....................................................................................................................... 391 22.13 References ............................................................................................................................................................. 391 Chapter 23: Supply Chain Management for Engineering Managers ......................................................................................393 23.1 Introduction.............................................................................................................................................................. 394

xvii

23.1.1 Definition of a Supply Chain.................................................................................................................................. 394 23.2 Supply Chain Management...................................................................................................................................... 395 23.3 Lean Supply Chain.................................................................................................................................................... 396 23.4 Importance of Supply Chain Management for Industries........................................................................................ 397 23.5 Supply Chain Management Risks.............................................................................................................................. 398 23.5.1 Definition............................................................................................................................................................... 398 23.5.2 Risks....................................................................................................................................................................... 398 23.6 Role of Engineers in the Supply Chain...................................................................................................................... 401 23.7 Application of Supply Chain in Industries................................................................................................................. 401 23.7.1 Examples................................................................................................................................................................ 401 23.7.2 Best Practices in Supply Chain Management......................................................................................................... 402 23.8 Innovation in Supply Chain Management................................................................................................................ 403 23.9 Conclusion................................................................................................................................................................ 403 23.10 References.............................................................................................................................................................. 404 Author Biographies .................................................................................................................................................................407

xviii

LIST OF FIGURES Chapter 1 Figure 1.1. Figure 1.2. Figure 1.3. Figure 1.4.

Management and Educational Trends That Have Affected the EM Field..........................................................4 Engineering Management as the Bridge Between Engineering and Management..........................................6 Challenges for the Technical Organization and Engineering Manager............................................................11 Five Knowledge Roles of the EM Discipline....................................................................................................12

Chapter 2 Figure 2.1. Organizational Stakeholders...........................................................................................................................20 Chapter 4 Figure 4.1. Figure 4.2. Figure 4.3. Figure 4.4. Figure 4.5. Figure 4.6.

The Five Elements of the Integrated Management Model.............................................................................36 Maslow’s Hierarchy of Human Needs............................................................................................................39 Traditional Organization Structure.................................................................................................................47 Team-Based Organization..............................................................................................................................48 Likert’s Basic Model.......................................................................................................................................49 The Managerial Grid......................................................................................................................................50

Chapter 6 Figure 6.1. Figure 6.2. Figure 6.3. Figure 6.4.

Page of U.S. Patent.........................................................................................................................................71 Sample Drawing for a U.S. Patent...................................................................................................................74 Typical Pages of the Specifications.................................................................................................................76 Claims Section of a Patent..............................................................................................................................79

Chapter 9 Figure 9.1. Figure 9.2. Figure 9.3. Figure 9.4. Figure 9.5. Figure 9.6. Figure 9.7.

Operations Research Methodology..............................................................................................................107 Basic Linear Programming Formulations......................................................................................................108 Feasible Region and Iso-Objective Lines.......................................................................................................109 Corner Point Optimal Solution......................................................................................................................110 Dual Linear Programs for Basic Linear Programs..........................................................................................112 Piecewise Approximation to a Non-linear Function.....................................................................................114 Big Data Value Chain.....................................................................................................................................127

Chapter 10 Figure 10.1. Modeling and Simulation as a Discipline....................................................................................................135 Figure 10.2. Visualization of the Principle of the Inverse Transform Method/Algorithm...............................................138 Figure 10.3. Time Advance in Discrete Event Simulation...............................................................................................140 Figure 10.4. Phases for Conducting a Study as Recommended in the NATO COBP........................................................144 Figure 10.5. Executing an ARENA Model.........................................................................................................................145 Figure 10.6. Architectural Frame Addressing Main Agent Characteristics......................................................................147 Figure 10.7. Agents, Environment, and Societies...........................................................................................................148

xix

Chapter 11 Figure 11.1. Decision Conference...................................................................................................................................154 Figure 11.2. Dialog Decision Process..............................................................................................................................154 Figure 11.3. Basic Influence Diagram.............................................................................................................................156 Figure 11.4. Basic Decision Tree.....................................................................................................................................157 Figure 11.5. Basic Cumulative Risk Profile......................................................................................................................157 Figure 11.6. Influence Diagram with Test.......................................................................................................................158 Figure 11.7. Decision Tree with Test...............................................................................................................................158 Figure 11.8. Influence Diagram with Test and Market Survey........................................................................................159 Figure 11.9. Probability Calculations..............................................................................................................................159 Figure 11.10. Decision Tree with Test and Market Survey..............................................................................................160 Figure 11.11. Three Utility Functions.............................................................................................................................162 Figure 11.12. Four Types of Value Functions..................................................................................................................163 Chapter 12 Figure 12.1. A Simple Three-Level AHP Model...............................................................................................................172 Figure 12.2. Simple ANP Network for a Decision-Making Process.................................................................................176 Figure 12.3. The Shape of a Hypothetical Utility Function.............................................................................................181 Chapter 13 Figure 13.1. The Scope of Engineering Informatics Proposed........................................................................................186 Figure 13.2. IIIE Discipline History..................................................................................................................................187 Figure 13.3. Discipline Structure of IIIE..........................................................................................................................188 Figure 13.4. The Relationship Between Engineering Integration, Manufacturing Integration, Customer Integration, and Enterprise Integration....................................................................................................................................................189 Chapter 14 Figure 14.1. Concept of Breakeven Analysis..................................................................................................................221 Chapter 15 Figure 15.1. Figure 15.2. Figure 15.3. Figure 15.4. Figure 15.5. Figure 15.6. Figure 10.6.

Shorthand Notation Used for Engineering Economics...............................................................................230 Cash Flow Notation ...................................................................................................................................231 Constant Payments - Interest and Principle Combined (amortization).......................................................232 Cash Flow Diagram for Purchasing an Automatic Welder .........................................................................233 Capitalized Cost.........................................................................................................................................234 Generalization of an R&R Problem.............................................................................................................238 General Decision Model to Determine Economic Feasibility......................................................................249

Chapter 16 Figure 16.1 Project Management Related to Engineering Management.......................................................................252 Figure 16.2. Project Management Process Groups........................................................................................................253 Figure 16.3. Work Breakdown Structure........................................................................................................................257 Figure 16.4. Linear Responsibility Chart.........................................................................................................................258 Figure 16.5. Project Performance Target - Scope, Time, and Cost..................................................................................260 Figure 16.6. Earned Value Graph....................................................................................................................................261 Chapter 17 Figure 17.1. Mind Map of SE Key Concepts....................................................................................................................266 Figure 17.2. U.S. Army Corporal Sounding Rocket..........................................................................................................267 Figure 17.3. Hall’s Book on Systems Engineering............................................................................................................268 Figure 17.4. Common Systems Engineering Lifecycles in Use Today..............................................................................268 Figure 17.5. ISO/IEC 15288 Systems Engineering Processes..........................................................................................269

xx

Figure 17.6. Sample Hybrid SUV Operational Use Case..................................................................................................271 Figure 17.7. Input/Output Matrix...................................................................................................................................272 Figure 17.8. DoDAF Architecting Guidance....................................................................................................................275 Figure 17.9. Big Bang Approach to Integration...............................................................................................................276 Figure 17.10. Bottom-Up Approach to Integration.........................................................................................................277 Chapter 18 Figure 18.1. Figure 18.2. Figure 18.3. Figure 18.4. Figure 18.5.

Using the Foundations of Systems..............................................................................................................290 Philosophic Level Spectrum........................................................................................................................292 Observation Through Intervention Activities Loop.....................................................................................305 Making Sense of “Methodologies,” “Methods” and “Tools”......................................................................306 Levels of Integrated Systems Thinking Application.....................................................................................312

Chapter 19 Figure 19.1. Schematic of the Activities in the Risk Management Framework..............................................................321 Figure 19.2. Simplified FTA Showing How Causalities of Risk Scenarios Can Be Represented by Events and Gate Symbols.............................................................................................................................................326 Figure 19.3. Estimating Probabilities of Damages..........................................................................................................327 Figure 19.4. Example of Risk Matrix to Rank Risk Scenarios Based on Consequence and Likelihood ...........................331 Chapter 20 Figure 20.1. Process Map Example.................................................................................................................................352 Figure 20.2. Histogram Examples...................................................................................................................................353 Figure 20.3. Pareto Chart................................................................................................................................................354 Figure 20.4. Ishikawa Diagram........................................................................................................................................355 Figure 20.5. Control Chart 1............................................................................................................................................356 Figure 20.6. Control Chart 2...........................................................................................................................................357 Figure 20.7. Scatter Chart Reflecting Positive Correlations.............................................................................................357 Figure 20.8. Scatter Chart Reflecting Negative Correlations...........................................................................................358 Chapter 21 Figure 21.1. Strategic Management Process..................................................................................................................363 Figure 21.2. Goals, Objectives, Strategies......................................................................................................................365 Chapter 22 Figure 22.1. Functions of an Entrepreneur.......................................................................................................................................381 Figure 22.2. Innovation Process - Six Stages....................................................................................................................................384 Figure 22.3. Design Thinking Process.................................................................................................................................................387 Figure 22.4. Stages of Entrepreneurial Process...............................................................................................................................389 Figure. 22.5. Continuous Improvement Process.............................................................................................................................390 Chapter 23 Figure 23.1. Supply Chain Organizational Pyramid.........................................................................................................................394 Figure 23.2. SIPOC Diagram................................................................................................................................................................396 Figure 23.3. Attributes of Lean Supply Chain...................................................................................................................................397 Figure 23.4. Definitions of Supply Chain Agility by Executives ...................................................................................................400 Figure 23.5. Agile Supply Chain...........................................................................................................................................................401

xxi

xxii

LIST OF TABLES Chapter 1 Table 1.1. Table 1.2. Table 1.3. Table 1.4. Table 1.5. Table 1.6.

ABET accredited and ASEE EM Related Programs..............................................................................................3 Common Characteristics of EM Definitions........................................................................................................5 Professional Societies Associated with the EM Discipline..................................................................................7 Journals Associated with the Engineering Management Discipline...................................................................8 Professional Conferences Associated with the EM Discipline............................................................................9 EM Discipline’s Stakeholder Needs..................................................................................................................13

Chapter 4 Table 4.1. Overlapping Concepts......................................................................................................................................44 Table 4.2. Theory of Motivating Knowledge Workers.......................................................................................................51 Chapter 8 Table 8.1. Influence of Commitment on Workplace Impact.............................................................................................98 Chapter 11 Table 11.1. Advantages and Disadvantages of Decision Processes................................................................................155 Table 11.2. Elements of the Swing Weight Matrix..........................................................................................................164 Table 11.3. Topics Referenced to Standard Texts in the Field........................................................................................167 Chapter 12 Table 12.1. Table 12.2. Table 12.3. Table 12.4. Table 12.5. Table 12.6. Table 12.7. Table 12.8. Table 12.9.

Scale for Pair-wise Comparison Using AHP1.................................................................................................172 Pair-Wise Comparison Between Attributes..................................................................................................172 Pair-Wise Comparison Between Attributes with Totals................................................................................173 Normalized Values of Pair-Wise Comparison Between Attributes...............................................................173 Pair-Wise Comparison Between Brands with respect to Maintenance Cost................................................176 ANP Supermatrix—Un-weighted.................................................................................................................177 ANP Supermatrix—Weighted.......................................................................................................................177 ANP Limit Supermatrix.................................................................................................................................177 Final Alternative Values based on ANP........................................................................................................178

Chapter 14 Table 14.1. Comparison of Types of Business Entities....................................................................................................201 Table 14.2. A Typical Income Statement.........................................................................................................................206 Table 14.3. Assets, Liabilities and Equity for ASEM LLC..................................................................................................208 Table 14.4. Balance Sheet for ASEM LLC.........................................................................................................................209 Table 14.5. Two Sources of Equity Capital for Shareholders Equity...............................................................................210 Table 14.6. Stockholder’s Equity Based Upon Two Sources of Equity Capital................................................................210 Table 14.7. ASEM Corporation Stockholder’s Equity Statement on December 31, 200X..............................................211 Table 14.8. Stockholder’s Equity Statement for ASEM Corp as of December 31...........................................................211 Table 14.9. Cash Flow Statement....................................................................................................................................212 Table 14.10. Revenue and Expenses for Merino Realty.................................................................................................213 Table 14.11. Cash Flow from Merino Reality Income Statement...................................................................................214

xxiii

Table 14.12. Table 14.13. Table 14.14. Table 14.15. Table 14.16.

SL Depreciation Example...........................................................................................................................216 MACRS Depreciation Example...................................................................................................................217 Loan Balance / Amortization Table.............................................................................................................219 T-account Method of Recording Debits and Credits..................................................................................220 Location of Typical Items That are Posted..................................................................................................220

Chapter 15 Table 15.1. Demonstration of Simple vs. Compound Interest.......................................................................................228 Table 15.2. Effects of Compounding Period...................................................................................................................230 Table 15.3. Salvage Value and O&M Costs by Year.........................................................................................................238 Table 15.4. Capitalized Costs by Year..............................................................................................................................239 Table 15.5. EVAC Calculation for O&M Costs..................................................................................................................239 Table 15.6. Total EVAC for an O&M Example..................................................................................................................239 Table 15.7. Net Cash Flow from Operating Activities.....................................................................................................243 Table 15.8. Net Cash Flows from Capital Related Activities............................................................................................243 Table 15.9. Loan Balance / Amortization Table...............................................................................................................245 Table 15.10. Total Cash Flow (Operating and Capital) Discounted.................................................................................245 Table 15.11. Economic Data for the ATA Problem..........................................................................................................246 Table 15.12. ATA Loan Amortization Table......................................................................................................................246 Table 15.13. Depreciation Expenses for the ATA Problem..............................................................................................247 Table 15.14. Net Cash Flows from Operating Income....................................................................................................247 Chapter 16 Table 16.1. Project Management Tools and Techniques.................................................................................................257 Chapter 17 Table 17.1. Input/Output Matrix....................................................................................................................................272 Table 17.2. Requirements Traceability Matrix................................................................................................................273 Chapter 18 Table 18.1. Table 18.2. Table 18.3. Table 18.4. Table 18.5. Table 18.6.

Multiple Perspectives of Systems Thinking..................................................................................................288 Distinctions between Hard and Soft Systems Thinking................................................................................289 Guiding Systems Principles for Systems Thinking.........................................................................................293 Systems Thinking Characteristics.................................................................................................................302 System-based Methodologies.....................................................................................................................307 Jackson’s “Creative Holism” Taxonomy........................................................................................................307

Chapter 19 Table 19.1. Table 19.2. Table 19.3. Table 19.4. Table 19.5. Table 19.6. Table 19.7. Table 19.8.

PHA Worksheet............................................................................................................................................323 JSA Worksheet..............................................................................................................................................324 FEMA Worksheet..........................................................................................................................................325 Basic FTA Symbols........................................................................................................................................325 Example of Risk Matrix to Rank Risk Scenarios Based on Consequence and Likelihood .............................328 Exploratory Questions That Can Be Used to Describe Risk Scenarios..........................................................329 Risk Analysis Tools .......................................................................................................................................332 References for Advanced Risk Analysis .......................................................................................................335

Chapter 20 Table 20.1. Pareto Results Example ...............................................................................................................................355 Chapter 23 Table 23.1. Varying Definitions of Supply Chain as Provided by Different Authors........................................................394 Table 23.2. Outsourcing Risks.........................................................................................................................................399

xxiv

Engineering Management: Past, Present, and Future

1 Engineering Management—Past, Present, and Future Timothy G. Kotnour University of Central Florida

John V. Farr United States Military Academy

1

Engineering Management Handbook

1.1 Introduction 1.1.1 Overview of Engineering Management With the globalization of the manufacturing base, outsourcing of many technical services, the efficiencies derived from advances in information technology (and the subsequent decrease in mid-management positions), and the shifting of our economy to be service-based, the roles of the technical organization and the engineering manager have dramatically changed. The 21st century technical organization must be concerned with: 1. Maintaining an agile, high quality, and profitable business base of products or services in a fluctuating economy, 2. Hiring, managing, and retaining a highly qualified and trained staff of engineers, scientists, and technicians in a rapidly changing technological environment, and 3. Demonstrating a high level of capability maturity. Engineers often enter the job market not as traditional engineers but as project managers, technical sales, and lead systems engineers (especially within the defense and information management arenas) involved with conceiving, defining, architecting, designing, integrating, marketing, and testing complex and multi-functional information technology centric systems (Abel, 2005). Within five years, for most engineers this has become their primary job function. Combined with the fact that the modern engineering enterprise is now characterized by geographically dispersed and multi-cultural organizations, engineering management (EM) is more relevant than ever. Because of the blurring of boundaries between technical and management roles, engineers must continue to redefine their roles to remain relevant in the modern economy. Like all technical professions, EM has evolved dramatically because of the information age and the interdisciplinary nature and complexity of modern systems.

1.1.2 The History of the Engineering Management Discipline According to Kocaoglu (1984), EM as a formal degree has existed since the mid 1940s. However, we know that courses in business and management aspects of engineering have been taught since the 1900s. For example, Stevens Institute of Technology founded a Department of Business Engineering in 1902 with the aim to teach students “to become efficient managers” (Clark, 2000). The Massachusetts Institute of Technology offered a degree in industrial management around 1913 (Kocaoglu, 1989). Several EM or EM-type programs grew out of the post World War II industrial expansion to include the University of Washington (1947) and Michigan Technological University (1949). The major growth occurred in the 1960s and 1970s. The first EM department was founded at the University of Missouri – Rolla (UMR now known as Missouri University of Science and Technology) in 1967. UMR also awarded the first PhD in EM in 1984 (Murray and Raper, 1997). The UMR contribution is further discussed by Babcock (2000). Today, there are probably in excess of 85 universities offering undergraduate and graduate degrees in programs named EM in the United Sates. Most EM programs can be categorized as being embedded within an industrial engineering department/program or combined with systems engineering departments/programs. Few undergraduate education EM programs exist because industrial engineering departments have been reluctant to embrace the profession at the undergraduate level. If you include the international programs, those embedded as concentrations within industrial engineering degrees, concentrations within MBAs, and hybrid programs such as engineering administration, systems EM, there are probably hundreds of universities that offer an EM-type degree. Given the recent downturn in MBAs degrees awarded in many programs (Triad Business Journal, 2004), EM degrees/programs/department should continue to grow. At the undergraduate level, there has also been growth in terms of related classes, minors, and certificates that are embedded within traditional degrees. However, the number of undergraduate EM programs has seen little growth. As shown in Table 1.1, the ABET website lists 11 accredited undergraduate programs in the US and five internationally with the word “management” in the program name and only 2

Engineering Management: Past, Present, and Future

one has been accredited in the US for the first time in the last five years. Only five use the term “engineering management” exclusively for the program name. A recent American Society for Engineering Education (ASEE) publication on domestic engineering programs lists 23 EM undergraduate programs, which also are summarized in Table 1.1. Table 1.1. ABET accredited and ASEE EM Related Programs (from Kaufman et al., 2015) ABET Accredited EM Programs* Domestic University of Arizona** (2003) Clarkson University*** (2009) University of Connecticut (1978) Missouri University of Science and Technology ** (1979) North Dakota State University (1971) Oklahoma State University (1936) University of the Pacific**(2003) Rensselaer Polytechnic Institute (1978) South Dakota School of Mines and Technology (1991) Stevens Institute of Technology** (1990) United States Military Academy** (1985) International Arab Academy for Science and Technology and Maritime Transport (2009) Istanbul Technical University (2009) Kuwait University (2006) Universidad Autonoma de San Luis Potosi (2012) University of Sharjah (2010)

ASEE Listed EM Undergrad Programs University of Arizona Arizona State University California State, Long Beach California State, Northridge University of California – Santa Cruz Christian Brothers University The College of New Jersey Colorado School of Mines Gonzaga University Illinois Institute of Technology Mercer University Miami University Missouri University of Science and Tech. University of North Carolina - Charlotte University of the Pacific NYU Polytechnic School of Engineering University of Portland Southern Methodist University St. Mary’s University Stevens Institute of Technology University of Tennessee - Chattanooga United States Military Academy University of Vermont

* Programs with “Management” in the name, ** “Engineering Management” programs, *** “Engineering and Management” programs. The number in parenthesis under ABET accredited programs is the year that the program was first accredited.

The EM profession mirrors both trends in business and education. Early business engineering focused on the civil and mechanical engineering disciplines. As shown in Figure 1.1, with the work Taylor (1911) contributed to the early focus on manufacturing that dominated the discipline through the 1990s. Rapid advances in information technology in the 1980s and organizational changes in all engineering practices led to decline in the specialist engineer and a rise in the generalist engineer. To reflect the shift from manufacturing to turn-key systems integrators in a global economic environment many EM programs are now aligned with systems engineering programs (Farr and Buede, 2003).

3

Engineering Management Handbook

Statistical Quality Control, (Shewart, 1924)

Principles of Scientific Management (Taylor, 1911)

Industrial Engineering (1909) (Jacko, 2005)

Department of Business Engineering, Stevens Int of Tech (1902) (Clark, 2000)

Figure 1.1. Management and Educational Trends That Have Affected the EM Field

Operations Research (1937) (ORMS, 2005)

1900-1925

Systems Thinking Emery and (Trist, 1960)

1925-1950

Process Reengineering (Hammer and Champy, 1993)

ISO 9000 (1991) (ISO, 2005)

Six Sigma (1986) (Motorola, 2005)

Total Quality Management (Deming, 1982)

Learning Organization, (Argyris and Schon, 1978)

1950-1975

1975-2000

1.1.3 Definition of Engineering Management In the literature you find few definitions of EM. Table 1.2 summarizes some the key characteristics common to all definitions of EM. We like the definition presented by Omurtag (1988) or Farr (2008). The EM field has its roots in the traditional engineering and management disciplines (Waters, 1994). This evolution has helped define the field. In the next section, we discuss the “knowledge” basis for the disciplines.

4

Engineering Management: Past, Present, and Future Table 1.2. Common Characteristics of EM Definitions Definition Engineering management is designing, operating, and continuously improving purposeful systems of people, machines, money, time, information, and energy by integrating engineering and management knowledge, techniques, and skills to achieve desired goals in technological enterprise through concern for the environment, quality, and ethics.

Reference Omurtag (1988)

The engineering manager is distinguished from other managers because he or she posses both the ability to apply engineering principles and a skill in organizing and directing people and projects. He or she is uniquely qualified for two types of jobs; the management of technical functions (such as design or production) in almost any enterprise, or the management of broader functions (such as marketing or top management) in a high technology enterprise.

Babcock and Morse (2002)

Engineering management is the discipline addressed to making and implementing decisions for strategic and operational leadership in current and emerging technologies and their impacts on interrelated systems.

IEEE (1990) and Kocaoglu (1991)

Engineering management is the art and science of planning, organizing, allocating resources, and directing and controlling activities which have a technological component.

American Society for Engineering Management

In today’s global business environment, engineer managers integrate hardware, software, people, processes and interfaces to produce economically viable and innovative products and services while ensuring that all pieces of the enterprise are working together.

Farr (2011)

1.2 Present State of the Engineering and Technology Management Field The present state of the EM field is described by understanding four elements: (a) the contributing disciplines, (b) professional societies, (c) relevant journals, and (d) professional conferences. Through the analysis of the present state conclusions for the future direction are offered: (1) the integration of the three core contributing disciplines of EM needs to continue and (2) the integration of the diverse set of professional societies, journals, and conferences that needs to take place.

1.2.1 The Connection of the Engineering Management Discipline to Other Disciplines To understand the EM discipline we need to understand how the discipline relates to other disciplines. In reviewing the history of EM, we assert that EM has evolved from the engineering and management disciplines. EM is the bridge between the engineering and management disciplines. Consistent with the definitions provided in the previous section, we view engineering manager as the “bridge” (Hicks, Utely, and Westbrook, 1999) between the traditional disciplines of science/engineering and management (see Figure 1.2).

5

Engineering Management Handbook Figure 1.2. Engineering Management as the Bridge Between Engineering and Management Engineering

Traditional Engineering Discipline

Management Within an Engineering Discipline

Engineering Management

Management Across Engineering Disciplines

MBA

Management of Technology

General Management

In reviewing the journals, professional societies, and conferences, five disciplines contribute to defining three different perspectives on the EM field. The five discipline groups are as follows: 1. Engineering disciplines. The core engineering disciplines in which the discipline focuses on the engineering and design process unique to a domain (e.g., civil, traditional industrial, mechanical, electrical). 2. Discipline specific engineering management. The EM discipline that focuses on the management process for a specific engineering discipline (e.g., management of the civil engineering process, management of the industrial engineering process). 3. Generalist engineering management. The EM discipline that focuses on the fundamental EM process across many engineering disciplines. 4. Management of technology. The business or management discipline that focuses on managing the creation, development, and use of technology (Badaway, 1998). 5. General management. The management discipline that focuses on the management of any organization. Given these descriptions, three perspectives to EM are: (1) discipline specific EM, (2) generalist EM, and (3) management of technology. Industrial engineering could be considered to be part of the overlap between engineering and EM in Figure 1.2. As will become evident in the rest of this section, the EM field continues to support this view. The EM discipline emerges from five unique sets of journals, professional societies, and conferences to provide three unique perspectives to the field.

1.2.2 Engineering Management Related Professional Societies Consistent with the three perspectives of the EM field we categorize the different professional societies related to EM. As has been completed before (Sarchet and Baker, 1995), Table 1.3 summarizes the different professional societies. In addition to the three perspectives to EM we have added three other categories for completeness: (1) disciplines associated with processes and tools used by the engineering manager, (2) general management, and (3) engineering education. Engineering disciplines and societies associated with the Accreditation Board for Engineering and Technology (ABET) were used as the source for the engineering programs. All of the engineering discipline professional societies are not included, just the societies with an associated EM group or division. We share these professional societies to help the reader understand the different avenues for actively participating and contributing to the profession. The EM discipline is supported with six groups of professional journals.

1.2.3 Engineering Management Related Journals Consistent with the three perspectives of the EM field, we review and categorize the different journals related to EM. Table 1.4 summarizes the journals related to EM. For completeness, in addition to the three perspectives to EM we have added three other categories: (1) disciplines associated with processes and tools used by the engineering manager, (2) general management and (3) engineering education. We share these related journals to help the reader understand where to go to for knowledge and to contribute to the knowledge of the profession. This list is not meant to be an exhaustive list. The EM discipline emerges from six unique sets of journals. 6

Engineering Management: Past, Present, and Future Table 1.3. Professional Societies Associated with the EM Discipline Group

Professional Societies

Engineering Management within an Engineering Discipline

• • • •

Disciplines Associated with Processes and Tools Used by the Engineering Manager

• Association for the Advancement of Cost Engineering (AACE) (aacei.org) • International Council of Systems Engineering (INCOSE) (www.incose.org) • Project Management Institute (PMI) (www.pmi.org)

Engineering Management Across Disciplines

• American Society for Engineering Management (ASEM) (www.asem.org) • Canadian Society for Engineering Management (CSEM) (www.csem-scgi.ca/index.html)

Management of Technology

• International Association for Management of Technology (IAMOT) (www.iamot.org) • Product Development Management Association (PDMA) (www.pdma.org)

General Management

• Academy of Management (AM) (www.aomonline.org) • Institute for Operations Research and the Management Sciences (INFORMS) (www.informs.org)

Engineering Education

• American Society of Engineering Education (ASEE) (www.asee.org)

American Society of Civil Engineers (ASCE) (www.asce.org) IEEE Engineering Management Society (IEEE EMS) (www.ieee.org/ems) Institute of Industrial Engineers (IIE) (www.iienet.org) Institute of Industrial Engineers (IIE)- Society for Engineering & Management Systems (SEMS) (www.iienet.org) • Society of Petroleum Engineers (SPE) (www.spe.org) • Society of Manufacturing Engineers (SME) (www.sme.org) • American Society for Mechanical Engineering (ASME) (asme.org)

7

Engineering Management Handbook Table 1.4. Journals Associated with the Engineering Management Discipline Group

8

Journals

Engineering Management within an Engineering Discipline

• Journal of Management in Engineering • Leadership and Management in Engineering • The Journal of Construction Engineering and Management

Disciplines Associated with Processes and Tools Used by the Engineering Manager

• • • • •

Cost Engineering International Journal of Project Management Journal of Systems Engineering Project Management Journal The Engineering Economist

Engineering Management Across Disciplines

• • • •

IEEE Transactions on Engineering Management Engineering Management Review Engineering Management Journal (ASEM) The Engineering Management Journal (IEE IN UK)

Management of Technology

• • • • • • • • • •

International Journal of Technology Management Journal of Engineering & Technology Management Journal of High Technology Management Journal of Product Innovation Management Technological Forecasting and Social Change Technovation R&D Management Research Policy Research Technology Management Technological Analysis and Strategic Management

General Management

• • • • • • • • • • • • • • • • • •

Academy of Management Review Academy of Management Journal Administrative Science Quarterly California Management Review Decision Analysis Harvard Business Review Information Technology & People Interfaces International Journal of Operations & Production Management International Journal of Quality & Reliability Management International Journal of Service Industry Management Management Decision Management Review Management Science Manufacturing & Service Operations Management National Productivity Review Organization Science Sloan Management Review

Engineering Education

• Journal of Engineering Education • IEEE Transactions on Engineering Education

Engineering Management: Past, Present, and Future

1.2.4 Engineering Management Related Conferences Consistent with the three perspectives of the EM field we reviewed and categorized the different professional conferences related to EM. Table 1.5 summarizes these conferences. We would like to share these related conferences to help the reader understand where to go to for knowledge and to contribute to the knowledge of the profession. This list of conference is not meant to be exhaustive, rather a starting place. The EM discipline emerges from six unique sets of conferences. Table 1.5. Professional Conferences Associated with the EM Discipline Group

Conferences

Engineering Management within an Engineering Discipline

• American Society of Civil Engineers (ASCE) (www.asce.org) • Institute of Industrial Engineers (IIE) (www.iienet.org)

Disciplines Associated with Processes and Tools Used by the Engineering Manager

• International Council of Systems Engineering (INCOSE) (www.incose.org) • Project Management Institute (PMI) (www.pmi.org)

Engineering Management Across Disciplines

• American Society for Engineering Management (ASEM) (www.asem.org) • IEEE Engineering Management Society (IEEE EMS) (www.ieee.org/ems) • PICMET (www.picmet.org)

Management of Technology

• International Association for Management of Technology (IAMOT) (www.iamot.org) • PICMET (www.picmet.org) • Product Development Management Association (PDMA) (www.pdma.org)

General Management

• Academy of Management (AM) (www.aomonline.org) • Institute for Operations Research and the Management Sciences (INFORMS) (www.informs.org)

Engineering Education

• American Society of Engineering Education (ASEE) (www.asee.org) • Masters of Engineering Management Programs Consortium (http://www.mempc.org) • Accreditation Board of Engineering and Technology (http://www.abet.org)

1.2.5 The Future of the Engineering Management Discipline The intent of this section is to develop a framework to continue the conversation about the future of EM. The intent is not to define the agenda, but rather provide the structure from which further conversations can be developed. In reviewing the past and present of EM and the emerging issues facing the world, the discipline of EM offers a unique ability to make lasting contributions (Sarchet and Baker, 1989). To define strategic issues we first understand three items: (1) a description of trends and challenges facing the EM organization, (2) a model of the EM discipline from a perspective of knowledge roles, and (3) a description of global outcomes for the stakeholders of the EM discipline. By taking these three perspectives we can better understand and define the emerging issues facing the discipline.

1.3 Emerging Engineering Management Related Trends, Drivers, and Challenges Barkema, Baum, and Mannix (2002) defined a set of trends defining management challenges. These challenges included: greater diversity; greater synchronization requirements; greater time pacing requirements;

9

Engineering Management Handbook

faster decision-making, learning, and innovation; faster newness and obsolescence of knowledge; more frequent environmental discontinuities; faster industry life-cycles; greater risk of competency traps; and faster newness and obsolescence of organizations. The challenges are being driven by the increased globalization of the knowledge economy and the increasing complexity of the systems. Technology managers are facing challenges managing in this domain. Engineering managers face challenges that include: (1) strategic planning for technology products, (2) new product project selection, (3) organizational learning about technology, and (4) technology core competencies (Scott, 1998). During the 2003 annual conference of the American Society for Engineering Management (ASEM), a session was held with both practicing and academic EM participants on defining the challenges associated with EM. During this session the participants identified challenges in three groups: (1) business environment trends and challenges, (2) organizational trends and challenges, and (3) engineering management/manager trends and challenges (Utley, Farrington, and Kotnour, 2003). The business environment trends and challenges included: • Globalization, • Short-term profit focused, • Increased regulatory/environmental stewardship/ethical focus, and • Changing demographics of the workforce. • • • • •

These trends create further trends and challenges for the technical organization: Forging partnerships, Operating networks of relationships, Implementing a process-based organization, Continuously managing change, and Gaining/maintain employee loyalty and commitment.

The engineering manager then faces of the challenges of operating in this environment. Specific challenges include: • Managing and leading teams, • Understanding and managing uncertainty, • Managing and leading the workforce, • Changing culture, • Using tools and metrics to manage, and • Developing the needed management and leadership skills and behaviors. Figure 1.3 summarizes these challenges. These trends and challenges offer the strategic context for the EM discipline. For example, the discipline needs to become more global and integrative across disciplines. The EM discipline must define a body of knowledge that provides the knowledge needed by the engineering manager to be successful in the challenging environment.

10

Engineering Management: Past, Present, and Future Figure 1.3. Challenges for the Technical Organization and Engineering Manager

We Defined EM Challenges Business Environment Trends & Challenges • • • •

Globalization Short-term profit focused Regulatory, environmental and ethical Demographics (age of the workforce, diversity, attitudes of the workforce)

• • • • •

Forging partnerships (with competitors) Operating network relationships Implementing a process-based organization Continuously managing change Gaining and maintaining employee loyalty and commitment

Organizational Trends & Challenges

Engineering Management/Manager Trends & Challenges • • • • • •

Managing and leading teams Understanding and managing uncertainty Managing and leading the workforce Changing culture Using tools and metrics to manage Developing management and leadership skills and behaviors

1.4 Engineering Management Discipline’s Knowledge Roles The EM discipline plays five knowledge roles (Boyer, 1990; Kotnour, 2001). The roles are based on the knowledge management function (i.e., generate, assimilate, or communicate) and application of the knowledge (i.e., generalist/across many organizations or organization specific). As can be seen in Figure 1.4, each of these roles supports the other roles. The challenge for the EM discipline is in integrating these five roles. The five roles are: 1. Research: The process of generating generalized knowledge. This knowledge can be applied to many different domains and does not necessarily solve an organization’s unique problem. This knowledge serves as the content and basis for the other roles. 2. Education: The process of teaching students knowledge that can be applied to many different domains or applications. The education roles pulls content from the other roles. 3. Training: The process of transferring knowledge to a unique domain, application, or organization. In training, the discipline’s knowledge is used to provide specific application insights. 4. Technical assistance: The process of working with an organization to solve a specific performance challenge. This technical assistance support creates knowledge unique to an organization. This unique knowledge can be used to generalize from for research or used as case studies in training or education classes. 5. Service: The set of activities to provide support to the university, profession, and society. The service role also provides an overarching or governance function for the discipline. The service or professional society role helps to assimilate the knowledge through conferences and journals. These five knowledge roles are needed to provide positive outcomes for the EM discipline’s stakeholders. The strategic issue facing the EM discipline is on how to integrate these five roles across the global and diverse set of contributing disciplines, professional societies, journals, and conference of EM. The intent of the rest of this chapter is to define specific challenges facing the EM discipline.

11

Engineering Management Handbook Figure 1.4. Five Knowledge Roles of the EM Discipline

Knowledge Function

Generate

Technical Assistance

Assimilate

Communicate

Research

Service

Training

Education

Organization Unique

Generalized

Application Level

1.5 Engineering Management Discipline Stakeholder Needs To raise a set of questions to help determine the agenda for the future of the EM discipline, we must first understand the discipline’s stakeholders and needs. The stakeholders are the set of individuals or groups who impact and are impacted by the profession. Table 1.6 summarizes the needs of the EM disciplines stakeholders. These outcomes can provide the overarching guidance or goals for the discipline.

12

Engineering Management: Past, Present, and Future Table 1.6. EM Discipline’s Stakeholder Needs Stakeholder

Desired Outcome

Engineering Management Discipline’s Contribution in Helping the Stakeholder Achieve their Desired Outcome

Society

• Strong, stable society • Useful products and services

• Provide graduates who are functional and make a difference

High-tech organizations

• Success in growing their business

• Provide educated graduates • Provide real-time knowledge to improve organizational performance

Profession

• Enhanced professionalism and profession

• Provide service to the professional societies and active students/graduates

Practicing engineering manager and engineering team

• Success in the workplace

• Provide real-time knowledge to improve individual, team, and organizational performance

Professional engineer

• Maintain professional certification

• Provide real-time knowledge to improve individual performance • Offer opportunities to complete professional registration requirements

University community

• Enhance the reputation of the university

• Provide an outlet (i.e., conferences and scholarly journals) for faculty to professionally grow and gain recognition for academic programs

Student

• Productive, working member of society

• Provide educational and work experiences to enable them to be a life-long learner • Provide a connection to employers and graduate schools

Faculty

• Enhanced reputation and freedom to intellectually explore

• Provide the infrastructure and outlets for conducting teaching, research, and service

Accreditation institutions

• Meet the desired outcomes of the accreditation process

• Define the bodies of knowledge and characteristics of the EM discipline • Systematically implement the accreditation process

1.6 Conclusions and Summary The intent of this chapter was to review the history and current state of the EM discipline as a foundation to help define the future of the discipline. We have presented a review of the history of the profession and also presented several definitions. To further describe the current state of the profession we have summarized relevant professional organizations, publications, and technical societies. However, the main contribution of this chapter is to present emerging trends, knowledge roles, and stakeholder needs for the profession along with strategic issues that will affect the future of EM and engineering education. We offer four conclusions from this work. First, the EM profession is at a critical juncture in its maturation. Unlike many traditional engineering professions, EM has been agile and responsive to changes in

13

Engineering Management Handbook

the global economic community. This can mainly be attributed to our main role as continuing education for engineers and scientists. In practice, we have had to be on the leading edge of managerial trends to produce competitive products and services. In order to remain relevant, we have had to adapt our skill sets. However, the role of EM is changing from both an educational and practical perspective. Most EM programs are run very similar to MBA programs with adjunct faculty. EM education is becoming more accepted within most universities. Unfortunately, few universities have standalone EM programs at the undergraduate and graduate levels staffed with mainly full-time faculty. The number of undergraduate programs has experienced steady growth. From a practicing EM perspective, the challenges in many ways are more daunting. Rapid changes in business practices require a continual self-evaluation and retraining to remain relevant. Second, the EM profession needs to build an integrated approach of teaching, research, technical assistance, training, and service. From this integration, the discipline will continue to grow and make significant contributions. Third, to draw this synergy, the EM profession must also recognize the complementary perspectives that different contributing fields can bring. These complementary perspectives will help develop and transfer the knowledge needed to address the challenges of the technical environment and technical organization. Fourth, the EM professional societies offer a key mechanism to foster collaboration across disciplines. The leadership for the profession needs to come from active participation from the discipline itself and the leadership of the professional societies.

1.7 References Abel, K., An Analysis of Stevens Engineering Management Graduates, 1990 – 2004, Hoboken, NJ: Stevens Institute of Technology, 2005. Argyris, C., and Schon, D. A., Organizational Learning: A Theory of Action Perspective, Reading, MA: Addison-Wesley, 1978. Babcock, Daniel, “Management Divisions of Engineering Societies,” Engineering Management Journal, vol. 1, no. 3, Sept. 1989, pp. 9-14. Babcock, Daniel, “Tribute to Bernie Sarchet,” Engineering Management Journal, vol. 12, no. 1, March 2000, p. 2. Babcock, Daniel, and Morse, Lucy C., Managing Engineering and Technology, 3rd Edition, Upper Saddle River, NJ: Prentice Hall, Inc., 2002. Badaway, M. K., “Technology Management Education: Alternative Models,” California Management Review, vol. 40, no. 4, 1998, pp. 94-116. Baldridge, D. C., Floyd, S. W., and Markoczy, L., “Are Managers from Mars and Academicians from Venus? Toward an Understanding of the Relationship Between Academic Quality and Practical Relevance,” Strategic Management Journal, vol. 25, no. 11, 2004, pp. 1063-1074. Barkema, H. G., Baum, J. A., and Mannix, E. A., “Management Challenges in a New Time,” Academy of Management Journal, vol. 45, no. 5, 2002, pp. 916-930. Boudreau, J. W., “Organizational Behavior, Strategy, Performance and Design in Management Science,” Management Science, vol. 50, no. 11, 2004, pp. 1463-1476. Boyer, E. L., Scholarship Revisited: Priorities of the Professoriate, The Carnegie Foundation for the Advancement of Teaching, 1990. Clark, Geoffrey W., History of Stevens Institute of Technology – A Record of Broad Based Curricula and Technogenesis, 1870-2000, Jersey City, NJ: Jensen/Daniels Publishers, 2000. Collins, T. R., Berivudes, M. G., Youngblood, A. D., and Pazos, P., Professional Development Training for Engineering Managers,” Engineering Management Journal, vol. 16, no. 3, Sept. 2004. Deming, W. E., Out of Crisis, Massachusetts Institute of Technology, Center for Advanced Engineering Study, Cambridge, MA, 1982. Dorf, R. C., The Technology Management Handbook, CRC Press, 1999. Emery, F. E., and Trist, E. L., “Socio-technical Systems,” In C.W. Churchman and M. Verhulst (editors), Management Sciences: Models and Techniques, Oxford: Pergamon, 1960. 14

Engineering Management: Past, Present, and Future

Farr, J., and Bowman, B., “ABET Accreditation of Engineering Management Programs: Contemporary and Future Issues,” Engineering Management Journal, vol. 11, no. 4, December 1999, pp. 7-13. Farr, John V., and Buede, Dennis, “Systems Engineering and Engineering Management: Keys to the Efficient Development of Products and Services,” Engineering Management Journal, vol. 15, no. 3, September 2003, pp. 3-11. Farr, John V., Systems Life Cycle Costing: Economic Analysis, Estimation, and Management, CRC Press, January 2011. Hammer, M., and Champy, J., Reengineering the Corporation, New York: Harper Business, 1993. Hicks, P., Utely, D., and Westbrook, J. “What are we teaching our engineering managers?” Engineering Management Journal, vol. 11, no. 1, March 1999, pp. 29-34. IEEE Editorial, “Research and Education Characteristics of the Engineering Management Discipline,” IEEE Transactions on Engineering Management, vol 37, no. 3, August 1990, pp. 172-176. International Organizations for Standardization, http://www.iso.ch/iso/en/iso9000-14000/index.html, accessed January 24, 2005. Jacko, Julie A., http://www.isye.gatech.edu/lhci/hci_role.pdf, accessed January 24, 2005. Johnson, J. H., Micro Projects Cause Constant Change, presented at Second International Conference on Extreme Programming and Agile Processes in Software Engineering, (20-23 May 2001), Cagliari, Italy, Kocaoglu, Dundar, “Engineering Management Education and Research,” Engineering Management Conference/International Congress on Technology and Technology Exchange, Pittsburgh, PA, October 8, 1984. Kocaoglu, Dundar, “Strategic Opportunities for Engineering Management,” Engineering Management Journal, vol. 1, no. 1, March 1989, pp. 8-10. Kocaoglu, Dundar, “Education for Leadership in Management of Engineering and Technology,” PICMET 91 – Portland International Conference on Management of Engineering and Technology, pp 78-83, Portland OR, 1991. Kotnour, T. G., “Building Knowledge for and about Large-Scale Organizational Transformations,” International of Operations and Production Management, 2001. Motorola Inc., http://www.motorola.com/content/0,,3074-5804,00.html, access January 24, 2005. Murray, Susan L., and Raper, Stephen A., “Engineering Management and Industrial Engineering: Six One Way, A Half Dozen the Other,” American Society of Engineering Education Annual Conference, Session 2542, 1997. OR/MS Today, http://www.lionhrtpub.com/orms/orms-2-01/nps.html, accessed January 24, 2005. Ramos-Rodrigues, A. R. and Ruiz-Navarro, J. “Changes in the Intellectual Structure of the Strategic Management Research: A Bibliometric Study of the Strategic Management Journal, 1980-2000,” Strategic Management Journal, vol 25, no. 10, 2004, pp. 981-1004. Sarchet, Bernie, and Baker, Merl, “Engineering Management—Key to the Future,” Engineering Management Journal, vol. 1, no. 1, March 1989, pp. 4-7. Sarchet, Bernie, and Baker, Merl, “Defining the Boundaries of Engineering Management,” Engineering Management Journal, vol. 7, no. 2, March 1995, pp. 7-10. Scott, G. M., “The new age of new product development: Are we there yet?” R & D Management, vol. 28, no. 4, 1998. Shewhart, Walter A., Bell Laboratory Memorandum, Issued May 16, 1924, http://www.itl.nist.gov/ div898/handbook/pmc/section1/pmc11.htm, accessed January 24, 2005. Taylor, Frederick Winslow, The Principles of Scientific Management, 1911. Triad Business Journal, http://triad.bizjournals.com/triad/stories/2004/03/29/focus2.html?t=printable, accessed March 1, 2005. Utley, D., Farrington, P., and Kotnour, T. G., “Understanding the Challenges of the Engineering Manager,” working paper, Author, University of Alabama Hunstsville, 2003. Waters, Bob, “Engineering Management Tradition and Education: Past, Present, and Future,” Engineering Management Journal, vol. 6, no. 3, Sept. 1994, pp. 5-8. 15

16

Professional Responsibility, Ethics, and Legal Issues

2 Professional Responsibility, Ethics, and Legal Issues William J. Daughton Missouri University of Science and Technology

17

Engineering Management Handbook

2.1 Introduction 2.1.1 Relevance and Importance As a profession, engineering must adhere to the highest standards of integrity and honesty. Engineering has a direct impact on society in terms of safety and quality of life so engineers must be vigilant in adhering to the highest principals of ethical conduct in conducting their professional work. Engineering is often directly involved with the creation of technology-based work product that has significant value to the employer or client. The value of this work product must be protected leading to patents and copyrights. Engineering managers must be aware of their responsibilities in this domain to ensure the proper protection of company work product assets.

2.1.2 What are Ethics? Ethics is concerned with the kind of values and morals an individual or a society finds desirable or appropriate (Northouse, 2016). Northouse also points out that ethical theory guides individuals or organizations in decision-making about what is right or wrong in given situations. It should be noted that ethics and legal requirements are different. Often, the real ethical dilemmas are choices among alternative that are all within the law but have different impacts on constituents.

2.1.3 What Constitutes Intellectual Property? For the engineering manager, intellectual property (IP) can be divided into two parts: industrial assets that are the result of invention or design and the creative work of individual authors. Both have the potential to need protection and the source of the protection in each case is different.

2.2 Engineering Code of Conduct 2.2.1 Introduction to the NSPE Ethical Canons The National Society of Professional Engineers (NSPE) list fundamental canons that form the basis for ethical conduct (NSPE, 2009): Engineers in the fulfillment of their professional duties shall: 1. Hold paramount the safety, health, and welfare of the public. 2. Perform services only in areas of their competence. 3. Issue public statements only in an objective or truthful manner. 4. Act for each employer or client as a faithful agent or trustee. 5. Avoid deceptive acts. 6. Conduct themselves honorably, responsibly, ethically, and lawfully so as to enhance the honor, reputation, and usefulness of the profession.

2.2.2 Safety, Health, and Welfare of the Public From a practical standpoint this canon requires engineering managers to be focused on ensuring that the work of their engineers is in compliance with all requirements set forth by their employer or client and with all established standards for workmanship and safety. If a violation report is received of either of these requirements, an engineering manager has an obligation to obtain all facts pertinent to the situation, and if the facts support the violation, report it to higher-level management of the employer or to the client. It is critical that the information be fact-based and not based on rumor or speculation. The credibility of the reporting engineers and the engineering manager is at stake in making such reports. Any information related to such disclosures cannot be revealed without the approval of the employer or client except as required by law or these canons. In holding the health, safety, and welfare of the public as paramount, this canon also requires that engineering managers report the unlawful practice of engineering by any person or company and to cooperate with proper authorities investigating such unlawful practice. 18

Professional Responsibility, Ethics, and Legal Issues

2.2.3 Professional Service Only in Qualified Areas Engineering managers must ensure that the engineers assigned to a specific job or projects indeed have appropriate credentials to qualify them to do the work. There may be a hierarchy of engineering experience and qualifications in a group of engineers assigned to a job or project, and this allows for at least one senior engineer with review and validation authority to oversee the technical work of less experienced or apprentice engineers. Ultimately, the work must be approved by qualified engineering and technical management individuals.

2.2.4 Objective and Truthful Public Statements As indicated in section 2.2, fact-based reporting of information is essential for the credibility of the engineers, engineering managers, and their employers. Objective and relevant facts must be included in technical reports and presentations, and the engineering manager must ensure that such facts are not selectively removed. Established and recognized technical professionals are often asked to provide public comment and opinion on technical matters. There are two critical factors to be considered if this situation arises: 1. Such comments and opinions are based on knowledge of the facts and competence in the subject matter. 2. If such comments or opinions are inspired or paid for by interested parties, that should be disclosed prior to any public statements.

2.2.5 Faithful Agents for Employers or Clients This canon relates directly to conflict of interest. A situation that is perceived to create a conflict should be fully disclosed to the employer or client. Such conflicts include: 1. Compensation by more than one party for services on the same project. 2. Financial or other considerations of more than trivial value from outside parties in connection with work being performed. Any consideration from an outside party that can influence the judgment, decisions, or work on an engineering manager represents a potential conflict. 3. Acceptance of a contract from a governmental body (local, state, or federal) on which a principal or officer of the organization serves as a member. 4. Participating in a decision as a member of a governmental body that involves the individual’s employer. The right course of action here is for individuals in this situation to disclose the conflict and recuse themselves until the matter is resolved.

2.2.6 Avoidance of Deceptive Acts Although this may seem obvious, there are several specific ethical and potentially legal situations worth mentioning. The first situation deals with misrepresenting or exaggerating individual or group capabilities or knowledge for the purpose of winning a contract or procuring professional work. Engineering managers must be particularly careful to ensure capabilities or knowledge are not misrepresented or even misconstrued by potential clients. Engineering managers would often be the name-of-record in validating work completed. There could be serious legal liabilities as well as ethical issues. The second situation involves behind-the-scenes offerings of payments or other considerations to a public authority to influence decisions favorably on the awarding of contracts or work. The bidding process for public work must be able to withstand public scrutiny.

2.2.7 Enhancing the Profession Through Ethical Behavior This may seem like a catch-all statement but it really points to the fact that each engineer and engineering manager is part of larger whole and any unethical or illegal activities of any sort by a single individual tarnishes the reputation and credibility of everyone in the profession. Seeing individuals involved in unethical or illegal behavior encourages slippery-slope thinking; e.g., “because others have done it, why not me?”. To protect the reputation of the profession, serious attention must be paid to ethics and legality.

19

Engineering Management Handbook

2.3 Ethical Decision-Making 2.3.1 Introduction There are four principles for making ethical decisions that engineering managers can use to analyze the impact of their decisions on the key stakeholders on an organization. As the label implies, stakeholders have a stake in the success of the organization, and all decisions that are made by the organization have potential consequences on one or more of the stakeholder groups. As a result, it is imperative to consider the impact of such decisions from an ethical standpoint. Traditionally, the organizational stakeholders are shown in Figure 2.1. Figure 2.1. Organizational Stakeholders

Suppliers

Employees

Customers

Organization

Communities

Stockholders

For simplicity, Employees include everyone, managers as well. Some sources separate employees and managers due to their different roles (Jones and George, 2006), but purposes of this treatment they are combined. Here, the term Communities refers to the local communities in which the organization has a presence as well as the broader content of national or international communities. The four ethical principles that can be applied to analyze the impact of managerial decisions on the above group of stakeholders are (Jones and George, 2006): 1. Utilitarian Rule 2. Moral Rights Rule 3. Justice Rule 4. Practical Rule These rules are in practice useful guidelines to guide decision-making by balancing the self-interest of the organization with those of the stakeholders.

2.3.2 Utilitarian Rule This rule is based on the concept that an ethical decision is one that produces the greatest good for the greatest number of stakeholders. So in applying this rule, engineering managers should consider how various alternatives would benefit or harm the stakeholder group. The principle is to choose an alternative that provides the most benefit or least harm for all stakeholders in equal measure. The practical problem with this rule is that such decisions may result in no one being satisfied with the outcome.

2.3.3 Moral Rights Rule

20

Under this rule, an ethical decision is one that protects the inalienable rights of the people affected by the decision. Issues of health and safety relative to Employees and Communities would be paramount here.

Professional Responsibility, Ethics, and Legal Issues

In addition rights of free speech and privacy arise as well. The practical difficulty with this rule is that decisions that protect the rights of some stakeholders may hurt the rights of others.

2.3.4 Justice Rule This rule is based on fair and equitable distribution of benefits or harm. It requires engineering managers to consider alternatives based on the degree to which they will impact all these stakeholders. The dilemma is what constitutes “fair and equitable.” Engineering managers must determine procedures for distributing outcomes fairly to all stakeholders.

2.3.5 Practical Rule The three previous rules may be difficult to readily apply in complex or fuzzy situations. However, this rule is one that is easy to apply and is one that no one should have any hesitation to use. Basically, the rule asks how a typical person would react to a decision from an ethical standpoint. The practical statement of the rule often appears as “Would you want this decision to be on the front page of your local newspaper?” Alternatively, “Would you want your mother to know what you did?” The point is whether the ethics of a decision or action could withstand public scrutiny. This rule is actually very powerful and easy to apply.

2.3.6 Implementing the Ethical Principles The principles cited in Sections 2.3.2 through 2.3.5 provide foundations for making ethical decisions. The challenge is evaluating an ethical dilemma so that one or more of the principles may be fairly applied. One approach (Morse and Babcock 2010) focuses on a series of steps for facilitating solutions to ethical dilemmas. • Step 1. Determine the facts—While an engineer might consider this obvious, this step takes on special significance in ethical dilemmas where emotion and passion often run high. Time constraints may limit this step, but still it is important to take the time to verify accusations and not be overrun by innuendo and rumor. • Step 2. Define the stakeholders—As pointed out in Section 2.3.1, it is important to understand which stakeholder groups may be affected and how by implementing one or more of the ethical decision rules. • Step 3. Assess the motivation of the stakeholders—Once the groups who have a stake in the decision have been identified, it is important to assess why they would care and how they might react to a decision. • Step 4. Formulate alternative solutions—Using the organizations core ethical values as a guide, develop alternatives to resolve the dilemma. • Step 5. Evaluate purposed alternatives—Using the information gathered about the affected stakeholders and applying the ethical decision principles, determine which alternatives provide the most positive or least negative effects on the stakeholders. • Step 6. Seek additional assistance, as appropriate—Review codes of ethics and previous similar cases, and seek guidance from knowledgeable individuals. • Step 7. Select the best course of action—Determine the optimal solution utilizing information from steps 4-6. • Step 8. Implement the selected solution—Ensure the solution conditions are properly and fully implemented. • Step 9. Monitor and assess the outcome—Monitor how the implemented solution is received by the affected stakeholder groups and archive this information for use in future dilemmas.

2.4 Global Considerations in Ethical Conduct 2.4.1 The Global Environment Engineering managers today work in a global environment. Whether it is R&D, manufacturing, technical marketing and sales, or financial engineering, work is occurring in a globally distributed

21

Engineering Management Handbook

environment. Managers need to understand how to ethically behave when conducting business outside the United States. Moral beliefs, social customs, and culture create different ethical standards in other parts of the world. Engineering managers must recognize these moral, social, and culture influences define what is considered appropriate and acceptable in doing business. Further, companies themselves must let their engineering managers know what is expected of them when working in foreign locations.

2.4.2 Laws and Codes for International Business The U.S. Foreign Corrupt Practices Act (FCPA) makes it illegal to knowingly corrupt a foreign official. However, even the FPCA allows for small payments to foreign government employees who are primarily working in administrative or clerical roles when such payments are considered part of the business culture in that country. The UN Global Compact describes principles for doing business globally in the areas of human rights, labor, the environment, and in anticorruption. This compact has been signed by a large number of CEOs around the world and does represent an effort to establish international ethical guidelines. The best advice is to seek the guidance and counsel of the organization prior to starting a business relationship in another country. Or, if that relationship already exists, determine what practices have been considered ethical in the past in these relationships. An organization’s legal office may have specific guidelines and the HR department may offer training for new managers. The one parting caveat is do not enter into an international relationship unprepared for ethical dilemmas.

2.4.3 Ethical Decision Making in the Global Environment As seen in the preceding sections, there are a number of challenges when faced with ethical dilemmas that have a global component. However, by addressing a few basic questions, engineering managers may minimize ethical problems: • Is it legal? • Does it in anyway violate basic human rights? • Is it consistent with the norms and culture of the society affected? • If there are seemingly ethical conflicts, can those be reconciled? Basic human rights include the right to good health, good standard of living, and the opportunity for economic advancement. In essence, they represent human dignity. Cultural-specific norms refer to the norms in the culture or cultures being affected by the decision. Clearly the term “reconciled” may raise some eyebrows as it appears to be a way to circumvent ethical obligations. However, the reference here is to find ways to work within the culture of a given country or part of the world while still respecting the legal and ethical requirements of the home country. This often arises with respect to gifts. As an example, in a country where exchanging gifts are the norm, a large company-to-company gift may be considered while small, nominal value personal gifts are exchanged. This would meet the requirements of both the U.S. and the other country.

2.5 Protecting Employees Who Raise Ethical Issues 2.5.1 Introduction Engineers working on new technology or products may become aware of potential adverse impact to the health and safety of the end user or to society at large. They also may see the behavior of other individuals as violating ethical principles.

2.5.2 Creating the Right Culture For engineering managers it is important to develop a healthy, open working relationship with the engineering staff. This will enable potential ethical issues to be discussed openly and misconceptions, rumors, 22

Professional Responsibility, Ethics, and Legal Issues

and lack of complete information to be addressed. There are occasionally stories circulating through a laboratory or engineering office about some type of unethical situation or behavior, often totally false or perhaps with a grain of truth. Unless these can be dealt with openly, they will take on a life of their own. What may have been a something almost trivial can grow to be a source of major discontentment and suspicion. There must be a working culture that encourages this kind of dialog and one that does not imply threats to an individual’s career for raising these issues.

2.5.3 Sarbanes-Oxley Act However, if employees perceive that their concerns are being ignored or covered up, or they feel that they cannot discuss the concerns with management, then what may occur is a whistle-blower case. Whistle-blowers now have some protection under the Sarbanes-Oxley Act. Retaliation against an employee for reporting ethical or legal issues may result in a 10-year prison sentence for the offending individual. However, in practice examples of retaliation can still be found. The current act is definitely a progressive step forward but it is not a guarantee that a whistle-blower will completely shielded from retaliation.

2.6 Responsibilities for Intellectual Property 2.6.1 Why Protect Intellectual Property? Work product generated by engineering groups or individuals may have significant strategic, competitive, or economic value to the organization. For industry assets such as inventions or designs, the value may be strategic such as being able to penetrate a new market, competitive, such as being able to offer a product with new or enhanced features, or economic, such as simplifying a design to reduce manufacturing costs. Individual and group creative work may contribute new ideas, methods, or approaches to enhance or improve any operational aspects of an organization, not just work product related. Credit for and use of these ideas, methods, and approaches can be protected to limit outside use and dissemination.

2.6.2 Invention Disclosure Processes Most commercial organizations have well-defined steps for disclosing inventions and those steps may differ somewhat from one another. However, to successfully plug into that process, engineering managers should follow a few simple guidelines: 1. Ensure all engineers and other technical contributors understand the importance of documenting their work. 2. Ensure all engineers and other technical contributors have a notebook with permanent binding so that pages cannot be added or removed in which their technical work can be recorded. 3. Emphasize the need to record and document technical activity with dates of work accomplished. Engineers engaged in any research or development as part of the normal work activity must be routinely assigned notebooks. Other engineers and technical contributors who from time to time work on projects with potential for protected work product should have notebooks available to record such work. 4. Store and maintain notebooks pending resolution of any disclosure claims. 5. Serve as an interface with organization IP liaisons and legal departments to simplify the disclosure of ideas. 6. Ensure that inventors are properly recognized for their contributions.

2.6.3 Patents If an organization determines that an invention disclosure is novel and valuable, it may choose to purse a patent to protect the invention. Patents in the U.S. are held for 20 years for an invention (14 years for a design) and provide the organization with exclusive rights of use and the opportunity to license the invention to others. Typically, patent attorneys employed by the organization manage the application process. Patents provide the owners with the opportunity to license the invention to others for their use in exchange for payment, which may create a significant revenue stream. 23

Engineering Management Handbook

2.6.4 Copyrights Copyrights can be awarded to owners of works such as books, pamphlets, lectures, professional papers, and training materials. The owner of the work may be the author, or in a professional setting, the employer if the work was created as part of the author’s assignment or job duties. The owner may see value in protecting such materials from free use and dissemination for strategic, competitive, or economic reasons and choose to identify the work as copyright protected. Effectively, a created work is considered protected as soon as it exists, and no public filing for copyright protection required. There are several limitations to the use of copyrighted material without authorization: • Single copies for private, non-commercial use. • Free use provided the source is properly referenced. This exception also allows for the use of protected works for teaching purposes or for news reporting. • Fair use depending on such things as the nature of the use, the amount of the total work used, and the impact the use has to the potential value of the whole work. • Non-voluntary licenses when compensation is paid to the author in respect of the use. Copyrights may be transferred by the owner. All economic rights may be transferred to a third party and such transfers may include selling those rights in return for payment. Transfers take two forms: 1. Assignments or the transfer of property rights. The party to whom the property rights are transferred becomes the new copyright owner. Often the publication of technical papers in proceedings or journals is contingent upon the assignment of copyright to the publisher. 2. Licenses are granted to third parties to use the copyrighted materials but the copyright remains with the original owner. License agreements may include payment to the original owner for use by the third party.

2.7 References C. M. Chang, Engineering Management, Challenges in the New Millennium, Pearson/Prentice Hall, 2005. Jones, G. and George, J., Contemporary Management, McGraw-Hill Irwin, 4th edition, 2006. National Association of Professional Engineers (NSPE), Code of Ethics, http://www.nspe.org/Ethics/ CodeofEthics, 2009. P. Northouse, Leadership Theory and Practice, Sage, 7th edition, 2016.

2.8 Other Sources of Information A Guide to the Engineering Management Body of Knowledge, Domain 11, ASEM publication, 3rd edition, 2012. World Intellectual Property Organization (WIPO), http://wipo.int/freepublications/en/intproperty/, 2009.

24

Management Theory and Concepts

3 Management Theory and Concepts

Jerry Westbrook University of Alabama, Huntsville

Chris Edmonds Roddies’ Code and Roddie Edmonds Foundation

25

Engineering Management Handbook

3.1 Introduction The objective of this chapter is to layout the evolution of the EM theory in order to better define the current thinking about EM. The end result may be a more standardized approach as to what is taught in EM academic programs and what is practiced by EM professionals in this country and around the world.

3.2 Historical Perspective

26

The logical beginning place for EM theory is with the Industrial Revolution. Prior to the Industrial Revolution, manufacturing was done by craftsmen, making one item at a time. Practically everything was made to order. There is some mention of enterprising craftsmen hiring workers to do the simpler parts of the total tasks and the craftsmen performed the skilled work in order to increase the number of items produced. Transportation was so primitive that it was difficult to secure raw materials consistently and it was equally difficult to identify a sufficient number of customers to make such efforts worthwhile. The invention of the steam engine did much to change this. The steam engine spawned transportation systems not dreamed of previously. In addition, the steam engine provided the potential for powering manufacturing industries. This fact was not lost on a myriad of inventors who used the steam engine to develop ways of manufacturing quality goods that previously were made only by craftsmen. James Watt invented the steam engine and formed a partnership with Matthew Boulton to manufacture and sell them. “In 1776, Watt’s first engine was sold to John Wilkinson for use in his iron works. Not knowing what price to charge, an agreement was made that the steam engine would be ‘rated’ at the equivalent of how many horses could do the same amount of work: hence the derivation of ‘horsepower’ for mechanical engines” (Wren, 1979). Another inventor, Richard Arkwright, is credited as being first to develop the concept of the factory. He organized all of the equipment required to make cotton cloth in one building. This model of efficiency was copied and improved on for many years. But it takes more than equipment and buildings to make products. It takes workers, preferably skilled workers. There was an initial effort made to recruit the skilled craftsmen to work in factories. There were not enough of them so farmers were also recruited. These two groups proved to be difficult to deal with. They were independent by nature and resented the factories which, in some cases, caused them to lose their former professions, and in every case attempted to regiment them and tell them what to do all of the time. As a result, there are many incidents recorded where factory equipment was destroyed by these discontented workers. These workers came to be known as “Luddites,” named after a youth in Ludlam had smashed his knitting frame when his father had been too harsh with him (Wren, 1979). Because of the shortage of skilled labor, the independence of craftsmen and farmers and the problems with them destroying machinery, factory owners turned to another labor source: women and children. It has been estimated that by the year 1800, 75% of the factory workforce consisted of women and children. Management talent was just as scarce as skilled labor. There was very little known about how to successfully run a factory. Abuses of women and children were widespread. Fourteen-hour workdays were common. The English Parliament investigated and attempted to establish a 10-hour workday for children. This effort went on for 20 years but was never passed. Yet, during this same time Robert Owen started and operated the New Lanark factory in Lanark, Scotland (George, 1968). Children’s work hours were limited to 10 and ¾ hours per day. Both school time and teachers were provided by the company, and workers were provided with homes at moderate cost. Company meetings and outings were held on a regular basis. And most importantly, the company was very profitable. After the invention of the steam engine, the second most significant development of the Industrial Revolution was the adjustment the early factory owners made to accommodate the large proportion of unskilled labor. They broke the complex tasks down into a myriad of simple tasks. They developed “division of labor.” Division of labor, considered the first EM theory, was widely written about during this time as a necessary principle for success in manufacturing. All of the decisions were made at the top of the organization by management. Workers only had to concentrate on the small task in front of them. Even so, some factories experienced problems with workers not paying attention to their work. Incentive plans were instituted so that workers were paid for only the good pieces they produced.

Management Theory and Concepts

Frenchman Henri Fayol (1949) is generally credited with being the first to develop general management principles. Fayol published his management principles in 1916 but they were not translated into English until 1949. He was an engineer who rose to the position of general manager in a mining firm. Fayol made two significant contributions to management theory. He was the first to propose management principles and he was the first to define elements of management. His 14 principles include: 1. Division of work 2. Authority and responsibility (relationship) 3. Discipline 4. Unity of command 5. Unity of direction 6. Subordination of individual to general interest 7. Remuneration (fairness of ) 8. Centralization (degree of appropriateness) 9. Scalar chain (of command) 10. Order 11. Equity (loyalty and fairness) 12. Stability of tenure (unnecessary turnover) 13. Initiative (motivation of subordinates) 14. Espirit de corps Fayol’s (1949) elements of management are: planning, organizing, commanding, coordinating, and controlling. These are considered to be fundamental concepts that are still being taught ninety years after they were first published.

3.3 Scientific Management Frederick W. Taylor was a contemporary of Fayol. While Fayol’s background was in mining, Taylor’s was in processing (steel) and construction. The next major development in management theory was Frederick W. Taylor’s Scientific Management. This is presented as the third major EM theory. Taylor’s (1911) methodology is contained in his four principles: 1. Develop a large collection of knowledge about the process under study. Use this knowledge to determine the one best way to perform the work. 2. Scientifically select workers who are most able to perform the work by the specified method. 3. Train the workers to do the work using the “one best way.” Provide incentives for using the correct method. 4. Let management and workers collaborate on decisions so that the unique knowledge that each has can be used toward the solution of organizational problems. It can be seen that division of labor is implied in these four principles. There is an overriding assumption that management divides the work and makes decisions affecting the way work is to be done. Taylor (1911) believed that if any task is studied sufficiently, management can determine the one best way for doing anything and can optimize productivity. He further believed that the variation introduced by the workers could be reduced to insignificance through training and incentives. Workers and machines were seen as only slightly different.

3.4 The Bureaucracy The next major theory to be discussed was developed by the German economist Max Weber. Weber (1947) became sensitive to the abuses of both bad and unscrupulous managers. He sought to develop a management system which would protect the worker while at the same time require managers to use accepted management practices. Weber was one of the first to make a clear distinction between managers and owners. He saw owners as those who routinely hired without regard to abilities and qualifications. They were also likely to promote workers to higher level positions similarly. The principles Weber chose to accomplish his goals were as follows (1947):

27

Engineering Management Handbook

1. 2. 3. 4. 5. 6.

A well-defined hierarchy of authority with centralized decision making (by top management) A clear division of work (labor) Rational program of personnel administration Rules and regulations as to how each job was to be done and the acceptable rate of production Written records A staff of experts to assist managers in solving complex problems

According to Weber’s (1947) concept, the manager represented authority. The manager was at the top of the organizational pyramid and made decisions based on “his sphere of competence.” “Rules and regulations” (Weber, 1947) prescribed as many decisions as possible, thus ensuring fair treatment of employees. The purpose of the “rational program of personnel administration” (Weber, 1947) was to “pre-make” decisions so that every employee is treated exactly like all others. Job descriptions and production quotas would ensure that only reasonable work would be expected of employees. Complex problems were to be solved by the manager and his staff of experts, not by the workers. Weber felt that not only would workers be protected by such a system but that the organization would be more productive also. Again, we see the familiar division of labor. Division of labor helped to train managers for each division of the process. This proliferation of managers adds levels to the organization. The many levels of the organization also contribute to the primary attribute of this system control. So, multi-levels of structure and a small span of control are characteristic of Weber’s design. Weber failed to see that his system would only function adequately in a stable environment where neither competitors nor technology were changing rapidly. If either of these were to begin to change, the organization bound by rules and a strict chain of command could not adapt to the changing environment. Weber’s system was designed to control, to prevent abuses. It was not designed to innovate, to develop new products or processes. It was called by a familiar name—it is the bureaucracy and it created a set of problems never envisioned by Weber.

3.4.1 A Critique The problem is that most industries in the U.S. use some version of the bureaucratic management principles just discussed. Chris Argyris (1957) wrote perhaps the most accurate critique of these management principles. First, he researched the common characteristics of personality development. They are as follows: 1. Man develops from a passive infant to an increasingly active adult 2. Goes from a state of dependence to independence 3. Changes from simple behavior to complex with maturity 4. From shallow interests, man develops deep commitments 5. Goes from short time frames to long time frames—more affected by the past than the future 6. Develops from family subordinate to peer or leader 7. Goes from a lack of awareness of self to the development of self control Argyris (1957) further identified four common classical organization concepts and compares the result of using them with the traits of normal personality development listed above. • Division of labor—The individual sells skills rather than total abilities • Chain of command—This tends to make individuals dependent, passive • Unity of direction—This is leader oriented, not a function of workers • Span of control (usually four to eight)—Adds levels to the organization, thus increases dependence Argyris (1957) hypothesized three results of using classical organization concepts: 1. There is a lack of congruency between normal personality development and classical organization concepts. 2. This lack of congruency generates frustration, short-term perspective and conflict. 28

Management Theory and Concepts

3. The result will be inter-subordinate hostility, rivalries, and a focus on parts of the organization rather than the whole. Argyris (1957) found that management frequently (at the time of the writing of his article) used concepts designed for use on uneducated women and children in manual labor factories.

3.5 Behavioral Approaches The first years of the twentieth century saw a multitude of management concepts developed. In addition to Fayol’s Principles’ Scientific Management and the bureaucracy, Frank and Lillian Gilbreath developed methods analysis and Henry Gantt developed the Gantt chart. The idea that management practice could improve productivity had many organizations actively searching for additional management concepts that would give their organizations a competitive advantage. Western Electric conducted a wide range of experiments with management practice at its Hawthorne works. They experimented with lighting, work breaks, incentive systems, organization communication and other concepts. The general conclusion reached was that the attitude of workers had much to do with organization productivity. They did not reach firm conclusions on how to develop those positive attitudes. Theories on workforce motivation required another thirty-five years to develop.

3.6 Quantitative Methods The quantitative methods of management were developing at the same time as the qualitative concepts. George Dantzig published a description of the simplex method of linear programming in 1947. Other optimizing techniques soon followed. The operations research (OR) movement formed and grew fast in the 1950s and 60s. There was a general feeling of the time that as computers become faster, management will be able to solve most of its problems mathematically using a variety of OR concepts. The development of decision trees, game theory, dynamic programming, and chaos theory are examples of concepts that would enhance the ability of managers to make optimal decisions. Engineering economy was first promoted within AT&T in the 1920s as a way to make better financial decisions. They developed the first textbook in the field and taught their managers and engineers in company sponsored classes. Engineering economy continued to evolve and became a course common to most industrial engineering curricula in the 1960s and is now a part of most EM programs. Engineering economy is a way of making economic decisions in terms of current currency valuations or taking time value of money for future resources into account. Engineering economy combined with cost and managerial accounting provide managers with powerful tools to aid decision-making. The tendency for bureaucratic organizations with decision-making concentrated at the top, the overuse of quantitative decision-making without first-hand knowledge of organizational processes led to a term labeled as “the rational model” by Peters and Waterman in their book In Search of Excellence (1982). The “rational model” was associated with low productivity organizations. Even powerful management tools and concepts can be used to the disadvantage of an organization.

3.7 Summary Hallmarks of classical management are division of labor, unity of command, and chain of command. Argyris (1957) found that applications of these concepts tend to cause problems for the organization. Taylor (1911) proposed the development of a body of knowledge on all-important tasks; there is one “best” way for doing a task, and management and workers should work together to solve production problems. This solved some production problems but contributed to others. Assortments of quantitative methods are available to assist managers, and there are concepts to motivate workers to do the jobs that need to be done. There are probably more examples of poor management than good. How, then, do we choose the best management practices under a specific set of circumstances? 29

Engineering Management Handbook

3.8 Attempts at Integration How does the behavioral information relate to the overall management knowledge base? Koontz (1961) made an attempt to put much of this information into perspective. He formulated six “schools of management thought”. Six schools are a bit unwieldy. They can easily be narrowed to three: 1. The Management Process School (including The Empirical School). The Management Process School describes management activities as planning, organizing, communication coordinating, and controlling. Focusing on these activities will improve the skills of the individual manager and that of the organization. Research by Mintzberg (1971) indicates that these activities do not adequately describe what a manager does in organizations he studied. The “management activities” do seem to be helpful in providing a conceptual framework to describe managerial activities. In other words, they form a good starting point in describing management. The Empirical School is promoted by the Harvard Business School. It uses case studies of actual situations to train and educate future managers and organizational leaders. Principles of management are formulated based on experiences either actual or resulting from studies of real situations. The case study approach allows students to learn from managers’ successes and failures. Studies of cases allow students to begin forming their own “principles” of management. 2. The Behavioral School (including both individual and group processes).This school infers that management is getting people wanting to get the work done versus just expecting them to get the work done. Individual theories include the motivation research of Herzberg (Motivators and Hygienes), Maslow (Hierarchy of Human Needs), McGregor (Theory X and Theory Y), McClelland (The Urge to Achieve) and others. Methods of achieving success through group or team processes has been developed by many, including Blake and Mouton (The Managerial Grid), Likert’s “Four Systems”, and Katzenbach and Smith (The Wisdom of Teams). 3. The Mathematical School (including all quantitative methods of solving management problems). One part of this school includes optimization concepts such as linear programming, decision probabilistic theory. Then the question becomes one of balance between the concepts and their appropriate relationship to each other. The EM field is dominated by knowledge workers, professionals, and talented technical personnel. Classical management concepts (as Argyris pointed out) were developed for unskilled workers in an environment controlled by upper management. Not included in the Schools of Management Thought is the impact of the organization structure. Burns and Stalker (1961) discovered that organizational success showed a relationship between the level of technology used and the type of structure employed. “Organic” structures were better adapted to organizations using moderate or high technology as a critical part of the enterprise. “Mechanistic” structures were used by organizations producing commodities with low technology processes. This 1961 study was done at a time when the use of low technology was a real option. Today, there are few organizations with that luxury. The implication drawn is that the Human Behavior School is more important to most organizations, especially to those employing the highest levels of technology.

3.9 What Is Working?

30

The most important concept to recognize is that in a technology-driven organization, the most valuable asset is the collective knowledge and abilities of employees. If facilities are degraded or destroyed, they can be rebuilt, at a significant cost, but it can be done. If key employees leave, they take significant amounts of knowledge with them. This knowledge may be more difficult to replace than facilities as well as more costly. It would be worthwhile to examine the practices of successful medium- to high-tech organizations. Article after article has some combination of the following characteristics: • Fayol and the Management Process School—Managers must be knowledgeable about the functions of management and how the processes work. • Scientific Management—There is a body of knowledge of the processes required to do the primary work of the organization. Management must have this information and understand how to continu-

Management Theory and Concepts

•

• •

ally improve them. There are too many instances where company executives attempt to manage with financial data without process knowledge. Behavioral Approaches—Capabilities of knowledge-workers must be harnessed to achieve success in the era of the global economy. Willing, capable employees solve problems and create solutions and opportunities. Talented workers must participate in decisions affecting their work. Decisions must be made close to the situation by managers most familiar with the situations. Complex work is done in teams that coordinate tasks as a normal team function. Training is expected of all employees. The organizations cannot improve unless its members improve. Quantitative Approaches—Mathematical models and probabilistic approaches have much to offer in the solution of complex problems but they are not a substitute for a positive, productive culture. Organization Structure—Flat organization structure, fewer levels, relatively high employee to manager ratio is the norm. Management layers add control when flexibility is more valued. Imposed controls are counterproductive. Team developed goals are part of an effective control system.

There are literally dozens of “systems” being used by industries that use some combination of these factors. Some of the systems in current use are: Total Quality Management, Statistical Process Control, Just in Time Inventories, Team Management, Management by Objectives, etc. Other companies, not wanting to be left out, have attempted to use these systems with widely varying success. These systems in themselves are no panaceas. Educators in EM should not be tempted to base programs in the classical theories that have limited use in the typical EM environment; nor should they be tempted to over-commit to the latest management “fads” such as TQM. If properly implemented, some of these “fads” may be productive. If they are used in an appropriate structure with the knowledge of behavioral theories, their probability of success goes up dramatically.

3.10 Conclusion There is a place for both classical management concepts, new techniques in EM curricula. Additional information on the nature of the external and internal environments has much to do with the way each is to be applied. Other necessary ingredients for a successful management strategy are: an appropriate structure and a knowledge of behavioral theories that underlie the new techniques.

3.11 References Argyris, Chris, “The Individual and Organization: Some Problems of Mutual Adjustment,” Administrative Science Quarterly, vol. 2, June 1957. Burns, Tom, and Stalker, G. M., The Management of Innovation, London, 1961. Fayol, Henri, General and Industrial Management, Pitman Books Limited, 1949. George, Claude S., Jr., The History of Management Thought, Englewood Cliffs NJ: Prentice Hall Inc., 1968. Herzberg, Frederick, “One More Time: How Do You Motivate Employees,” Harvard Business Review, January- February, 1968. Koontz, Harold, “The Management Theory Jungle,” Journal of the Academy of Management, December 1961. Maslow, Abraham H., “A Theory of Human Motivation,” Psychological Review, 1943, p. 50. McGregor, Douglas M., “The Human Side of the Enterprise,” Management Review, November 1957. Mintzberg, Henry, Managerial Work: Analysis From Observation, Management Science, vol. 18, no. 2, October 1971. Peters, Thomas J., and Waterman, Robert H. Jr., In Search of Excellence, New York: Warner Books, 1982. Taylor, Frederick W., Shop Management, New York: Harper and Brothers, 1911. Weber, Max, The Theory of Social and Economic Organization, McMillan Publishing Co., 1947. (A translation of the original document.) Wren, Daniel A., The Evolution of Management Thought, New York: John Wiley and Sons, 1979. 31



32

Managing Knowledge Workers

4 Managing Knowledge Workers Jerry D. Westbrook, PhD University of Alabama, Huntsville

Chris Edmonds Roddies’ Code and Roddie Edmonds Foundation

33

Engineering Management Handbook

4.1 Introduction It is generally agreed that management is the way an organization functions to get needed things done. That is about all that is agreed on. Management has been an issue for thousands of years. The Egyptians organized thousands of workers to build the pyramids and hundreds of impressive monuments. Romans built aqueducts to provide its cities with fresh water. They built roads that are still visible today. Much of the ancient accomplishments were built with slave labor or other kind of forced labor. This is the day of the knowledge worker (Drucker, 1959). People who work with their brains are critical to the success of most, if not all current organizations. Koontz (1961) attempted to summarize most management schemes that are in use today. He called them “Schools of Management Thought.” They are the: • Management Process School • Empirical School • Social Systems School • Human Behavior School • Mathematical School • Decision Theory School

4.1.1 Attempts at Integration How does the behavioral information relate to the overall management knowledge base? As discussed in the previous chapter, Koontz (1961) made an attempt to put much of this information into perspective. He formulated six “schools of management thought.” Six schools are a bit unwieldy. Three of the schools have much in common with another the six schools can be narrowed to three. The original six schools are described below. 1. The Management Process School—The Management Process School describes management as a process that can be taught and learned and consists of such activities as planning, organizing, staffing, leading, and communicating and controlling. Focusing on these activities will improve the skills of the individual manager and those of the organization. There was an old argument over whether management can be taught or is something one is born with. This school infers that management principles can be taught and learned, and further, such principles can guide management decisions, thereby making the organization more productive. Typical management principles center around the following: • Chain of Command—Communication is primarily vertical, guided by direct reporting relationships. • Division of Labor—Work is divided into relatively small tasks so that lower skilled workers can be trained to do these tasks repetitively. This concept was developed during the Industrial Revolution where unskilled women and children were employed to do work of that time. • Narrow Span of Control—Number of direct reports per manager. Span of control depends on many factors such as the skill level and number of tasks supervised. The question was “how many workers could one manager supervise”? Division of labor tends to reduce that ratio to as little as an average of four workers to one supervisor. • Unity of Command—One worker has one supervisor to minimize any confusion as whom to take orders from. Research by Mintzberg (1971) indicates that these activities do not adequately describe what a manager does in organizations he studied. Mintzberg found more communication activities such as “figurehead” and “spokesperson” and other activities such as “disturbance handler and resource allocator. The “management activities” do seem to be helpful in providing a conceptual framework to describe managerial activities. In other words, they form a good starting point in describing management. 34

Managing Knowledge Workers

2. The Empirical School—The Empirical School is promoted by the Harvard Business School. It uses case studies of actual situations to train and educate future managers and organizational leaders. Principles of management are formulated based on experiences either actual or resulting from studies of real situations. The case study approach allows students to learn from managers’ successes and failures. Studies of cases allow students to begin forming their own “principles” of management. 3. Social Systems School—This school of management thought examines how workers perform in groups or teams. The Hawthorne Experiments was the initial study for this way of looking at management. The Hawthorne Experiments showed that both external conditions such as lighting and work rules affect productivity, but so does how teams relate to organizational goals. Achieving success through group or team processes has been developed by many more researchers and theorists, including Blake and Mouton (The Managerial Grid), Likert’s “Four Systems,” and Katzenbach and Smith (The Wisdom of Teams). 4. The Human Behavioral School—This school infers that management is providing the environment where people want to get the work done versus just expecting it knowledge workers work with their minds. How they feel about their job influences their effectiveness. Individual theories of this school include the motivation research of Herzberg (Motivators and Hygienes), Maslow (Hierarchy of Human Needs), McGregor (Theory X and Theory Y), McClelland (The Urge to Achieve) and others. 5. The Mathematical School—This school attempts to model a significant portion of an organization’s systems. Mathematical techniques such as linear or non-linear programming are then used to optimize each system to get maximum productivity. Modeling a significant portion of the organization’s functions is a formidable task. Regardless, proponents of this school forge ahead. Several university EM programs focus on this school. 6. Decision Theory School—Decision Theory seeks to determine strategies for unique situations likely to be encountered by an organization. Decision rules that yield the best result for the broadest array of situations are used to select the best strategies. Probability of the occurrence of each situation is estimated and enters in the decision process. This school is prevalent during election years. Research programs are considered depending on which candidate may win a major office. For example, if a Democratic candidate should win, environmental research may be funded at a higher level. If the Republican candidate should win, new weapons systems may receive additional funding. Both universities and private contractors use decision theory to assist in resource allocation for potential future projects and project proposals.

4.1.2 Summary Each school of management thought makes significant contributions to the study and practice of management. Many of these schools are taught without referring to the others. The case study approach is a valuable tool for teaching management concepts. Linear and non-linear programming can solve problems other schools cannot approach. It is important to know when any one school is applicable to a situation existing in an organization.

4.2 How It All Works Together 4.2.1 The “Integrated” Part All of the schools of management thought have contributions to make. Those of us with engineering backgrounds can appreciate the availability of management principles. They provide a guide where there is little management experience or training in our backgrounds. It is the “people” part where there are the most unknowns. We have little preparation for dealing with talented professionals who know more about aspects of their jobs than their supervisors. In order to put all of the schools together where they make sense and provide a road map for our management careers, a model has been put together to assist in this effort. 35

Engineering Management Handbook

The Integrated Management Model is shown below in Figure 4.1. It has five elements. The overall interpretation of the model is that management is composed of management systems, the organization structure and a people (knowledge worker) orientation. These are inter-related building blocks. The external and internal business environments impact them. Figure 4.1. The Five Elements of the Integrated Management Model

• External Environment

• • • • • •

Customers Competitors Suppliers Community Regulators Funding agencies

• Internal Environment

Management Systems Org. Structure People Orientation

» Management » Employee demographics (age, etc.) » Skills » Facilities » Structures (levels)

Integrated Management Model 4.2.2 The External Environment The External Environment includes elements that impact the organization externally. They have influence on an organization from the outside. They include: • Customers • Competitors • Suppliers • Community • Regulators • Funding agencies Each of these must be known and dealt with strategically. For example, a strong customer may vertically integrate and become a competitor. A supplier may give competitors better prices. Competitors’ research and development programs may yield a better, cheaper product. The community may support a local organization or may attempt to put unfavorable regulations in place. Maintaining favorable local relations is important. These entities are a fundamental part of a strategic plan. Threats and opportunities are developed from an analysis of the external environment.

36

Managing Knowledge Workers

4.2.3 The Internal Environment The Internal Environment is made up of the internal resources that are available to allow the organization to operate successfully. It must be determined if the resources available are adequate for the challenges the organization faces. Composition of the internal environment include the following: • The size and skill base of the workforce • Key employee demographics • Adequacy, age, and condition of facilities • Management style and organization culture • The facilities available to meet customer demands A strategic analysis of the internal environment yields strengths and weaknesses. The resources available to an organization impact the management of that organization.

4.2.4 Management Systems Management Systems are the way an organization gets things done. There are many types of systems: everything from accounting, purchasing, and production to strategic efforts to coordinate work flow. It is the latter that is of most interest. Many powerful and useful systems have been implemented in the past twenty-five years. Further back, one of the first management systems was Management By Objectives (MBO). Then Zero Defects, Total Quality Management (TQM), and Business Process Re-engineering (BPR) became popular. Each had a half life of approximately two years and was gone. They were good systems that didn’t last. Why? Many were never implemented very well. Middle management gave lip service to some but never really supported them. Upper management felt like the systems were for workers, not them. The structures of some organizations were too complex to support the system. The reasons were systemic. The whole organization failed to adopt and use the new system. Lean Enterprise and Six Sigma are the management systems currently in use. Will they survive or be discarded like the rest? That will be determined, but they can succeed if the organizations using them take subtle but fundamental actions.

4.2.5 Organizational Structure Organizational Structure is the silent force behind an organization’s success or failure. Structure can be defined in several ways. • The number of levels in the organization that communication must travel between and among members • The number of employees reporting to one manager—this is called span of control • How members of the organization relate to each other—formally or informally, vertically or omni-directionally The higher the technology level in use, the faster communications must be. This is associated with relatively flat structures. Complex, tall structures do not tend to thrive in rapidly changing business environments.

4.2.6 People (Orientation) Much of a knowledge worker based organization’s success depends on how the organization relates to its employees. Most organizations proclaim that their employees are their most valued assets but make decisions taking them for granted. This thought is inscribed on plaques on the walls of many corporate headquarters that have moved most of their operation to another country. Acting in a manner that generates trust allows a knowledge worker organization to succeed. In other words, what hangs on the wall must happen in the hall. Failure to do so is the reason so many of the systems mentioned above did not work very long. 37

Engineering Management Handbook

4.2.7 The Model All elements of the model work together. Systems fail when they do not act in a coordinated manner. The internal environment must have the necessary resources. The external environment must present opportunities and manageable threats. Appropriate systems must be used and the structure must support those systems. The knowledge workers must be in an environment that challenges and appreciates their contributions. • Integration of these elements is presented in the Strategic Management chapter. • Assessment of appropriate structures is presented in the Organization Structure chapter. • Concepts for managing and motivating knowledge workers are presented in the People Orientation chapter. • Integration of organization systems are discussed in the Systems Management chapter. • Determining the conditions that allow teams to be effective is established in the Team and Group Management chapter.

4.3 The “People” Orientation The focus here is on the “People” block of the Integrated Management Model previously discussed. It is the block that forms the base of the model. It interacts with both the internal and external environments. Observations have been made that indicate that employee representatives tend to treat customers much the same way as the organization treats them. The Organization Structure and Systems blocks can only function with motivated and productive knowledge workers. Today’s economy is being driven by knowledge workers. Inventors, innovators, process developers, and process improvers are all knowledge workers. This is a term coined by Peter Drucker (1966). He recognized that the world had changed from a manufacturing economy to an information economy where manufacturing is an integrated element in a larger system. Knowledge workers changed the rules of management. No longer can a manager observe a (knowledge) worker and assess the effectiveness of the work being done, or even if the worker is working. When work is mental, the value of the work may not be known for a significant period of time. If the knowledge worker does not feel as if he or she is being treated fairly, rewarded adequately or supported appropriately, the worker may not be able to concentrate on the complex issues at hand. This slowing of work may go without detection. It is more important than ever to understand behavioral concepts of management. The knowledge worker wants to be a part of the organization, not just occupy a “slot” on the organization chart. The model previously introduced will continue to guide the thought process in these chapters. In this chapter, we will study the “People” Orientation, the least known and practiced of the concepts in the model.

4.3.1 Background on Behavioral Approaches The first years of the twentieth century saw a multitude of management concepts developed. A French engineer, Henri Fayol, developed the first recorded principles of management. Frederick W. Taylor developed Scientific Management to increase productivity. Frank and Lillian Gilbreath developed methods analysis and Henry Gantt developed the Gantt chart for scheduling large-scale projects. The idea that management practice could improve productivity had many organizations actively searching for additional management concepts that would give their organizations a competitive advantage. Western Electric conducted a wide range of experiments with management practice at its Hawthorne works. They experimented with lighting, work breaks, incentive systems, organization communication and other concepts. The general conclusion reached was that the attitude of workers had much to do with organization productivity. They did not reach firm conclusions on how to develop those positive attitudes. Theories on workforce motivation required another thirty-five years to develop. Harvard professor Chris Argyris performed a seminal study of many previous studies of management. He compared management principles with normal personality development and concluded that modern management tended to treat workers as children while expecting adult behaviors. The results of the study 38

Managing Knowledge Workers

raised concerns about what is considered normal management practice. If there is any semblance of truth in Argyris’ work, those guiding EM programs must be very careful about what they teach as being acceptable practice. After World War II, many behavioral theories were developed. The ill-fated “human relations” movement spawned much research which has proved to be beneficial but not the total answer. We will explore the major behavioral concepts that apply to knowledge workers by engineering managers. Concepts developed by Douglas McGregor, Abraham Maslow, Frederick Herzberg and David McClelland will be discussed in depth.

4.3.2 McGregor’s Theory X and Theory Y Douglas McGregor was a Harvard professor and business consultant. He had many clients over an extended period of time. McGregor (1957) observed managers making assumptions about workers in their decisions. He labeled these assumptions about workers as: • Theory X—Assumes workers must be coerced to work, they are lazy and want security above all. • Theory Y—Assumes that workers will exercise self-control to achieve organizational objectives to which they are committed, seek responsibility and are innovative in solving organizational problems. Neither assumption should be assumed to be good or bad. They were assumptions that formed the bases of decisions. McGregor observed that management made these assumptions about its workers and made decisions based on the assumptions. If the assumptions were in error, workers developed resentment that management never understood. Problems arose when assumptions became reality. If a manager assumed that workers could not be trusted and made decisions based on that assumption, workers responded to how they were treated. They became like the assumption. Likewise, if there are assumptions that the workforce strongly supports the organization, workers tend to value the relationship and respond positively. McGregor observed that management made work rules governing access to facilities and adherence to job descriptions as if the workers needed close supervision or could not be trusted. In most cases the Theory X assumption became a self-fulfilling prophecy. Knowledgeable, capable workers acted as if they needed to be told what to do when in fact they knew very well what needed to be done, many times better than management. McGregor exposed hidden potential that organizations are missing due to their unwarranted assumptions.

4.3.3 Maslow’s Hierarchy of Human Needs Abraham Maslow was an American psychologist who theorized the five levels of the “hierarchy of human needs” (1943), which are shown in Figure 4.2. He studied exemplary “healthy” people such as Albert Einstein, Jane Addams, Frederick Douglass, and Eleanor Roosevelt and developed a theory of psychological health predicated on fulfilling innate human needs in priority, culminating in self-actualization. Figure 4.2. Maslow’s Hierarchy of Human Needs

selfactualization morality, creativity, spontaneity, acceptance, experience purpose, meaning and inner potential self-esteem confidence, achievement, respect of others, the need to be a unique individual love and belonging friendship, family, intimacy, sense of connection safety and security health, employment, property, family and social stability physiological needs breathing, food, water, shelter, clothing, sleep

39

Engineering Management Handbook

The first is the physiological level. Maslow observed that unless the physiological needs are met, no other needs motivate an individual. Once these needs are met, then the individual seeks to meet safety and security needs. When any need is substantially (approximately 85%) met, then individuals seek the next higher level. The higher level needs are more long-term than physiological needs. A good job and adequate housing normally meet the safety and security needs. After those needs are met, Membership (love and belonging) needs will be sought. Because employment meets the Safety and Security needs, the new employee will likely seek to join a group. A union will be attractive to many individuals. Others may be satisfied by being a part of a close-knit project team or by participating in company-sponsored sports leagues. After Membership needs are substantially met, we are motivated to seek Esteem needs; first self esteem, then public esteem. At this point, it is not enough to be a member of an organization; we want to be recognized as a leader or as a superior member in some way. Finally, after Esteem needs are met, we desire to achieve Self Actualization. At this level we are contributing not only to the organization but also to the good of the industry, mankind or some other higher level good. We work to contribute and are recognized as an expert in some field. According to Maslow, workers are motivated to achieve the next level in the hierarchy. The organization must recognize this and initiate efforts to assist this process. The organization benefits when its members are advancing up Maslow’s Hierarchy. In 2011, researchers from the University of Illinois put the hierarchy to the test. They found that fulfillment of needs strongly correlated with happiness. Between 2005 and 2011, via the Gallup World Poll, they surveyed people from 155 countries representing every major region of the world. They concluded that while Maslow’s “hierarchy of needs” suggests that basic needs must be fulfilled before higher needs are pursued, their study indicated that people benefit from satisfying any of these needs in any order. They further reported that self-actualization and social needs were important even when basic needs were unfulfilled. There are many cautions for management by understanding the Hierarchy. Management is most likely at one of the higher levels. Most workers are at a lower level and seeking different needs than management. Workers may be seeking to fit in at the membership level whereas management is dedicated to the organization at the self-actualization level. Management must recognize where workers are on the hierarchy and assist them in reaching the next level. This should promote better understanding throughout the organization. Positive actions by management should come from this increased understanding.

4.3.4 Herzberg’s Motivators and Hygienes Frederick Herzberg (1968), a noted industrial psychologist and pioneer of job enrichment, did research on job satisfaction and published his findings initially in 1959. He used accountants and engineers in his original study. He asked participants if they would think of a time they were very satisfied with their jobs and identify what they considered were the causes and contributing factors to that satisfaction. He repeated that question two more times. Then he asked for the factors associated with a time of significant dissatisfaction. He also repeated that twice. He and his researchers recorded the responses and grouped them according to similarities or “thought units.” Herzberg found one set of factors that primarily dissatisfied workers and another set that acted as satisfiers. The satisfiers that Herzberg later called motivators were found to be: • Recognition • Achievement • Possibility of growth • Advancement • Responsibility • The job itself The dissatisfiers later called hygienes were found to be: • Working conditions • Company policies 40

Managing Knowledge Workers

• Relations with the supervisor • Relations with peers • Pay Dr. Herzberg determined that because only motivators motivated employees, the best that could be expected from hygienes was neutrality. Hygienes are primarily dissatisfiers but if are at a reasonable level, they may be able to lead to neutrality in employee motivation. If management observes a problem in employee motivation, consideration might be given to a higher focus on motivators such as more equitable recognition or the opportunity to advance—that is to learn how to accomplish more challenging tasks or to increase responsibility. Additional hygienes are short-term and will likely make matters worse. Herzberg observed that management frequently attempts to use hygienes to motivate the workforce but an increase in hygienes only raises the anticipation of further increases. Costs rise but motivation and productivity do not. The motivators are more difficult for management to apply but are not as expensive as hygienes. He further observed that hygienes must be maintained at an appropriate level to prevent dissatisfaction but they cannot motivate. The Herzberg study has been replicated many times with similar results for subjects in different professions, countries, and cultures. A study of blue-collar workers showed similar overall results but hygienes were of more importance and motivators were of slightly less importance than in other Herzberg studies.

4.3.5 McClelland’s Need to Achieve Harvard professor David McClelland (1966) studied workers involved in a plant shutdown in Erie, Pennsylvania. A few of those laid off immediately set about to find jobs in nearby towns. They used their connections to explore jobs in the closest cities. Most found equivalent jobs within six weeks. The majority checked with their union several times; inquired if another company would buy the company and reopen the factory and read the local want ads. They gathered in small groups to discuss the situation to see if anyone knew of available jobs. The majority of this group was still unemployed after six months. McClelland studied and interviewed both groups. The first group was described as follows: • They set challenges for themselves. • They didn’t take chances, preferring to take action to solve problems rather than leaving them to chance. • They preferred to receive concrete feedback on how they were doing. These workers were labeled nAchievers (nAch) because they exhibited a need to achieve. The second group was described as follows: • They stayed home for a while. • The checked the newspaper for jobs similar to the ones they lost. • They checked with state and federal and state employment agencies. • They met in small groups to see if anyone had heard that the plant was reopening or if another company had bought it. McClelland labeled this group as nAffiliators (nAff) for their need to affiliate. There was a group discovered in another study—the nPower (nPow) group. They were motivated by the need for power independent of achievement. McClelland discovered that organizations tend to take on the characteristics of its workers. Some are achievement oriented while some are more affiliating organizations. Organizations need to examine their recent recruits to determine what kind of employees they are attracting. Organization productivity and success are products of its workforce but may not be achievable if the organization has the wrong kind of employees—nAffiliators, and fewer nAchievers. 41

Engineering Management Handbook

McClelland has had success in training company workforces to become nAchievers. Apparently, people can be trained to become nAchievers. McClelland further suggests that these concepts can be used to describe whole countries. There is little doubt that the US was, at one time, populated by nAchievers. One can but wonder if that preponderance of achievers is still in evidence.

4.3.6 Pink’s Motivational Concepts After all of the preceding works of McGregor, Maslow, Herzberg, McClelland, and Skinner were completed; Daniel Pink (1995) continued research into motivation. His studies concluded that motivation is stimulated by either external or internal forces. External forces include incentive plans where workers are paid by the piece for their production. Management pressure to produce is also an external factor. He found that as long as jobs are relatively simple and required a minimum of mental activity, external motivation works as intended. But when a modicum of mental activity is required by the worker, external motivators are ineffective. He described work as being algorithmic or heuristic. Algorithmic work follows detailed instructions that require discipline but little originality. Heuristic work requires mental activity to determine how to proceed to accomplish a complicated task. This type of activity requires intrinsic motivation. Pink (1995) found intrinsic motivation factors include autonomy, mastery and purpose. Skilled knowledge workers require strong participation in determining the work to be done. Workers are more committed to projects they helped define and plan. Mastery includes the opportunity to improve skills required to do complex work to accomplish significant goals. Purpose relates to the meaning of the work. The knowledge worker must believe in the value of the project and that value must mesh with the worker’s own values. Pink cited many examples of successful organizations using his concepts.

4.3.7 B. F. Skinner’s Operant Conditioning Theory B. F. Skinner (1953) won the Noble prize for science in 1963 for his work on behavioral research. He termed the results of his research The Operant Conditioning Theory. His theory is as follows: • Behavior that is rewarded tends to be repeated. • Behavior that is ignored tends to be extinguished. • Behavior that is punished generates a negative, fragmented response. According to Skinner’s theory, if a manager sees a behavior that is favorable to the organization, the behavior should be rewarded. The reward need not be financial, but that is not excluded. A favorable comment, a write-up in the company newsletter, a mention before employees in the same department, etc. can be used to good effect by the alert manager. The second statement of the theory is significant. If a manager ignores positive accomplishment, he or she runs the risk of “extinguishing” it. Negative behaviors may be ignored and thereby discouraged as long as positive behaviors are rewarded. This does not mean ignoring violation of rules or failure to follow prescribed methods or procedures. They are dealt with according to policy. It is assumed that many negative comments and attitudes are an attempt to get attention. If the attention is not given, the behavior is likely to be “extinguished.” Behavior that is punished generates a negative, fragmented response. This is interpreted that the person being punished perceives that the punishment is not deserved, above what others doing similar things get. This type of punishment causes a backlash, perhaps not immediately but sooner or later. That is not the intent of the action (punishment) but such is likely to happen. The manager has to always make sure that the offense merits the punishment. The offender should expect the action taken. 42

Managing Knowledge Workers

4.3.8 Multiple Generations in the Workplace For the first time in history there are five generations working side by side in the workplace. Generation Z, the newest and youngest generation, has begun working as summer and part-time employees. They join the oldest generation, the Traditionalists, who continue to enjoy healthy lives and productive careers. This age phenomenon adds to the complexity of managing people and teams. Since each generation is influenced by different life experiences, managers need knowledge of the different generations (or generational intelligence) to effectively lead and motivate people. Here is a brief summary of the five generations. 1. Traditionalists—Also known as the Silent Generation was born between 1900 and 1945. They comprise 2% of the workforce (Pew Research Center). • They have experienced the Great Depression and World War II • They prefer face to face meetings in structured workplaces, have a respect for rules, and seek hard work with bright people • They are loyal to their employer, tend to live frugal lives in the “burbs” and tolerate technology though it is sometimes hard to grasp 2. Baby Boomers—Comprise 29% of the workforce and were born 1946 to 1964. • They grew up during the Vietnam War, Woodstock, assassinations, landing on the moon, Civil rights and the 1960s social changes • These workaholics prefer adrenaline charged assignments, teamwork, innovation, and few rules • They are optimistic in nature and work to live instead of live to work like their co-workers in Generation Y and X • Calculators and calendars are second nature including note cards and note pads • It is estimated that 70 million baby boomers will retire over the next decade (Pew Research Center) 3. Generation X—Currently they are the second largest generational workforce at 34% of workers and were born between 1965 and 1980. • They experienced government corruption, environmental disaster, the fall of the Berlin Wall, the Gulf War and the global AIDS epidemic • They are self reliant latch key kids who are results oriented and fun • They are realists with a strong sense of personal values many having experienced the good and bad in relationships from divorced parents and mothers entering the workforce in droves • They enjoy independence and operating as a free agent and often gravitate to being an entrepreneur rather than an employee 4. Generation Y—Also known as Millennials they are now the majority population in the workplace with more than 36% and were born between 1981 and 1995. • This group is tech savvy, socially conscious and were influenced by doting parents • They are competitive, confident, diverse, and concerned with community service • They are motivated by worklife balance, specific instructions, and work best when they get R.E.S.P.E.C.T. 5. Generation Z—Also known as Linksters, 911’s or the Facebook Crowd and were born after 1995 and comprise 1% of the workforce. • This young generation is very optimistic, has high expectations, is incredibly tech savvy, and are social media wizards • They are motivated by new technology, social activism, green causes and specific instructions • They have the ability to text fast and efficient and program new technology before any of the other generations Each generation brings its own view of the world, which creates both threats and opportunities for organizations. Effective managers will learn to work with these diverse groups by understanding work styles, considering generational values, sharing perceptions, finding commonality, and learning from each other. 43

Engineering Management Handbook

4.3.9 Summary Maslow’s Hierarchy of Human Needs, McGregor’s Theory X and Theory Y, Herzberg’s Motivators and Hygienes, and McClelland’s Achievers and Affiliators may at first glance seem to be independent concepts. They are different approaches that are very complementary. For example, Maslow’s Self Actualization and Esteem Needs are closely related to Herzberg’s Motivators. Theory Y is a natural way of thinking about such employees who tend to be nAchievers. Those having membership and security needs tend to focus on hygienes, to be nAffiliators and perhaps assumed to be described by Theory X. These overlaps are shown in Table 4.1. Table 4.1. Overlapping Concepts Maslow Self Actualization/Esteem Membership/Security

McGregor Theory Y Theory X



Herzberg Motivators Hygienes



McClelland nAchievers nAffiliators

In order to motivate knowledge workers, Theory Y must be assumed; self actualization must be encouraged with opportunities; there must be a focus on motivators and achievement. In order to limit dissatisfaction, membership activities should be assisted; Theory X assumptions must be avoided and hygienes should be provided at as high a level as can be reasonably afforded.

4.4 People Orientation—Team Management The “people” part of management is too big to attempt in one chapter. The focus of this module is managing knowledge workers in groups or teams. The background for this module is a meta-study by Chris Argyris (1957). He reviewed literally hundreds of management studies that linked management practices with human behavior. Most organizations use some version of the standard management practices described in the Management Process School of Management Thought described in the first module. This School espouses the chain of command, unity of command, division of labor, vertical communication channels, authority in accordance with responsibility, etc. Argyris (1957) wrote perhaps the most accurate critique of these management principles. First, he researched the common characteristics of personality development: 1. Man develops from a passive infant to an increasingly active adult 2. Goes from a state of dependence to independence 3. Changes from simple behavior to complex with maturity 4. From shallow interests, man develops deep commitments 5. Goes from short time frames to long time frames—more affected by the past than the future 6. Develops from family subordinate to peer or leader 7. Goes from a lack of awareness of self to the development of self control. Argyris further identified four common classical organization concepts and compares the result of using them with the traits of normal personality development listed above. • Division of labor—The individual sells skills rather than total abilities • Chain of command—This tends to make individuals dependent, passive • Unity of direction—This is leader oriented, not a function of workers • Span of control (usually 4 to 8)—Adds levels to the organization, thus increases dependence Argyris hypothesized three results of using classical organization concepts: 1. There is a lack of congruency between normal personality development and classical organization concepts. 2. This lack of congruency generates frustration, short-term perspective and conflict. 44

Managing Knowledge Workers

3. The result will be inter-subordinate hostility, rivalries, and a focus on parts of the organization rather than the whole. If there is any truth to these hypotheses, and many believe that there is, there must be management concepts that use the whole ability of individuals that will not generate such problems. The concepts that use total abilities are discussed in the next sections.

4.4.1 Likert—An Integrating Principle Rensis Likert (1961) was an international management consultant, theorist, and author. Three of his noteworthy concepts will be discussed in this section. • Principle of Supportive Management • Team Management • Four Systems He documented the concepts used by the most successful organizations and continuously compared them with those of less successful organizations. He concluded that high producing organizations and their managers managed differently than managers of low producing organizations. Managers using classical management theories were less successful than those managers who managed in a way that will be discussed.

4.4.2 Characteristics of High Producing Organizations Likert studied the management of many organizations and characterized how they operated and their success over a period of time. His findings are noted in a concise form below. • There are favorable attitudes of members of an organization toward superiors, toward the work, toward the organization. There is mutual confidence and trust throughout the organization. • There is a high sense of involvement in the achievement of high goals and there is a sense of dissatisfaction if goals are not met. • The organization effectively harnesses all of the major motivational concepts including: • Ego motives • Security motives • Creativity and curiosity • Economic motives • • • •

Other characteristics of high producing organizations are: The organization consists of a tightly knit, effectively functioning social system. Employees want to work together, solve problems and make the organization successful. The system is made up of interlocking groups with a high degree of group loyalty and with favorable attitudes and trust between subordinates and superiors. Measurements of organizational performance are primarily used for self guidance rather than for super-imposed control. (This is more of a Theory Y approach within an nAchieving workforce.)

4.4.3 Characteristics of Low Producing Organizations • Motivation is achieved by the exercise of control through authority. (Traditional management) • Jobs are organized, methods are prescribed, standards are set, performance goals and budgets are set by management. (Remember Argyris’ findings on worker dependence and the result?) • Compliance is sought through the use of hierarchical and economic pressure. This is a Theory X assumption that this is the only way to get workers to produce. In short, those managers who demand success don’t get it. Those managers who allow employees to be successful are in successful organizations. 45

Engineering Management Handbook

High Producing Managers Likert also looked at individual managers and assessed their effectiveness. Characteristics of the high producing managers are shown first. An employee in an productive organization holds the following perceptions toward his or her superior: • He or she is supportive, friendly, helpful, rather than hostile. • He or she shows confidence in my integrity and ability. • He or she has high expectations of subordinates. • He or she coaches and assists employees whose performance is below expectations. This type of manager is with his or her employees rather than just being above them. This manager is heavily involved with both the work and the employees. The manager’s work (in high producing organizations): • Plans and schedules work, trains employees, supplies subordinates with required material, tools and information required for them to be successful • Provides technical competence when needed • Develops subordinates into a working team, uses participation and group leadership practices.

The Integrating Principle This concept is somewhat vague but that is similar to the way Likert wrote it. It is still possible to grasp the central concept. Subordinates react favorably to experiences that they feel are supportive and contribute to their sense of importance and personal worth. Employees react unfavorably to situations that are threatening and decrease their dignity and feeling of personal worth. The employee perception must be positive in such work experiences. The leadership and other processes of the organization must ensure, that as much as possible, that in all interactions and relationships, each member, in light of his background, values and expectations, views the experience as supportive and one which builds and maintains his or her sense of personal worth and importance. The worker must feel that he or she is making a worthwhile contribution to the organization and that contribution is recognized by the organization.

4.4.4 Team Management High producing organizations as described by Likert communicated differently than low producing ones. Likert observed and wrote of this and characterized the pattern of communication as Team Management. Each manager leads a team but is a member at the next higher level. This manager is thus a linking pin between levels. That is only true if the group functions and communicates as a team. Most work groups do not do that. Both normal work groups, which operated with managers working one on one with subordinates, and team managers are depicted in Figure 4.3.

46

Managing Knowledge Workers Figure 4.3. Traditional Organization Structure

A A

C C B

C

D

The Traditional Organization Structure • The boss has a one-on-one relationship with each subordinate. • Each subordinate attempts to use whatever means available to extract more resources from the boss than other subordinates receive. • Communication from subordinate to boss is highly filtered. The boss hears only what the subordinate wants the boss to know in order to receive favor for his or her unit. • There is mistrust between subordinates who each consider that other subordinates are better treated. • The good of the organization as a whole receives little consideration. Decisions are made in a vacuum with each unit competing for resources. • Each unit staffs for the maximum contingency. There is little sharing of resources between units. This is the way unenlightened managers deal with their employees. This is frequently true organization wide. They know no other way of supervising. A simplified representation of a team-based organization is shown in Figure 4.4. Characteristics of a team-based organization include: • The good of the organization as a whole is easy to relate to. • Communications are with the whole group, filtering is not possible. • Vigorous debate focused on the issues generates better decisions. • Sharing of resources is looked upon with favor as it allows unnecessary unit cost to be reduced by loaning employees to units in greater temporary need. • Decisions are better supported. Even those whose recommendations are not followed had input and know the other positions and can generally support the decision.

47

Engineering Management Handbook Figure 4.4. Team-Based Organization

The results of an organizational communications study was released several years ago. Typically there is communication loss from the top level down. When top management communicates: • Middle management receives 67% • General manager receives 50% • Supervisor receives 33% • Worker receives 20% This indicates that those at the base of the organization receive very little accurate information that emanates from the top. Likert’s Team Management concept seeks to improve on this kind of organization communication. Another significant question is: “What are the similar percentages going from the bottom of the level to the top?” That is easy. It doesn’t happen, so it is zero. Team Management can improve that situation also.

4.4.5 Likert’s System IV Rensis Likert (1961) developed management systems through observation and determined that only System IV achieves normal productivity goals consistently. He developed the model shown next.

Likert’s Four Systems • • • •

System I: Exploitive - Authoritative System II: Benevolent - Authoritative (we have your best interest at heart) System III: Consultative - Democratic System IV: Participative - Democratic

Likert’s systems are evaluated in six areas: (see Figure 4.5) 1. Leadership processes—The extent to which superiors have confidence and trust in subordinates. 2. Character of motivational forces—This varies from physical security and economic needs to use of ego and achievement needs which come from group achievement. 3. Character of Communication Process—The amount and direction of communication aimed at achieving organizational objectives. 48

Managing Knowledge Workers

4. Character of Interaction Influence Process—How do different levels of an organization work together to solve problems and achieve objectives. 5. Character of Decision-Making Process—At what level in the organization are decisions made? Productive organizations make them throughout the organization. 6. Character of Goal Setting—Where are goals established and from what level do orders normally come? Figure 4.5. Likert’s Basic Model Causal Variables Management systems

Intervening Variables Attitudes

End Result Variables Productivity, profits

No one has actually surveyed the population of current organizations, but it is widely believed that most organizations are somewhere in System II. A few organizations can be described as using System III. Organizations using System IV are rarely found. Likert believed that in assessing the value of an organization that the value of the human asset must be assessed. A layoff is a liquidation of a valuable asset just as surely as selling equipment, land, facilities or inventories; except that the latter can be easily replaced and the human asset cannot. Valuation of the human asset involves: • Recruiting costs • Training costs • Familiarization costs • Capability costs • Development costs All these are lost when human resources are liquidated. Many of these costs will have to be paid again, at a higher rate, without the guaranty that the work will be done as well. Likert’s philosophy: Attitudes and skill generate productivity. He observed that the management systems that an organization uses drive attitudes. Attitudes generate productivity and profits. In other words, attitudes impact productivity and profits. System IV obviously has much in common with Theory Y, Self-Actualization, Motivators, etc.

4.4.6 Blake and Mouton’s Managerial Grid In the late 1960s, Dr. Robert Blake of the University of Texas studied the management of a major corporation at their corporate headquarters. The commonly accepted theory of the day was that management style cold be categorized on a continuum from autocratic to participative. Blake and Mouton (1964) sought to either validate this theory of develop a new one. They observed that some managers were very successful and managed productive units. They were observed to have similar management styles. Other managers that were in charge of less successful units also had similar management styles. Neither management style fit into the autocratic-participative continuum. Further observations were made of some successful managers who were transferred to units with productivity problems. These units improved over a period of time to be similar to the unit the successful managers had left. Blake and Mouton concluded that management style was a major influence on productivity. Problems with a particular group of task difficulty were overcome with the new management style. Next, Blake and Mouton’s challenge was to describe the style of successful managers in a way that could be adaptable by others. The Managerial Grid is the theory that they developed to describe the results that they had observed, which include. • High performance groups were headed by similar managers • Some low performance managers focused on performance • Some low performance managers focused on making employees “happy.” 49

Engineering Management Handbook

• • • •

These managers: Used teams extensively Confronted issues to resolve conflict Got extensive input from employees Developed employee capabilities.

4.4.7 The Managerial Grid There was not sufficient difference between positions to define all positions on a 9x9 matrix. As shown in Figure 4.6 there are five basic positions where adequate to demonstrate the major management styles in common use. 1-1 Avoids decisions and uses “policies” instead of positions. Do such managers really exist? Unfortunately, they do. They hide behind policies and avoid confrontation. They certainly aren’t oriented to toward production either. They stay in their offices a lot. Conflict resolution is to ignore the conflict. 1-9 The country club manager tries to keep people happy. These managers hope that production will take care of itself if employees are “happy”. They are attempting to meet goals by manipulating employees. Conflict resolution is to sweep problems “under the rug.” 9-1 This is the autocrat position. He solves problems by edict. It is production first. People are hired to get the work done. Conflict resolution is “my way or the highway.” 5-5 This type of manager views people and production equally but is not strong on either. This is a bureaucratic, status-quo position. There are more managers in this position than all others according to standardized tests. They tend to solve problems by compromise. 9-9 This is the team manager position. She uses teams to gather and share information and to identify and solve productivity problems. There is a natural balance between people and production. She works to solve problems by confronting them. Figure 4.6. The Managerial Grid

concern for people

1,9

50

9,9 5,5

1,1

concern for production

9,1

Managing Knowledge Workers

Results Only managers with strong concerns for both people and work were in high productivity units. The managers: • Used teams extensively • Confronted issues to resolve conflict • Received extensive input from employees • Developed employee capabilities

4.5 Summary Notice that all of the studies that have been discussed reach similar conclusions. Argyris shows that traditional management doesn’t work very well. As shown in Table 4.2, efforts to get high productivity under classical principles are not successful. Likert’s system IV had similar constructs and results as Blake and Mouton’s Managerial Grid 9-9 position. Table 4.2. Theory of Motivating Knowledge Workers How is all this related? Systems III, IV, 9-9 on Managerial Grid Theory Y, Self Actualization, Motivators, nAchievers Self Esteem Systems I, II, 1-9, 9-1, 1-1 on Managerial Grid Hygienes nAffiliators Safety, Security

It should also be noted that concepts in this chapter correlate with those in Chapter 2. Managers making a Theory Y assumption and a focus on motivators will likely use Likert’s System IV and have a 9-9 management style on the Managerial Grid.

4.6 References Argyris, Chris, The Individual and Organization: Some Problems of Mutual Adjustment, Administrative Quarterly, vol. 2, June 1957. Blake, Robert, and Mouton, Jane, The Managerial Grid, Houston: Gulf Publishing Co., 1964. Drucker, Peter, The Effective Executive, The Definitive Guide to Getting the Right Things Done, Harper Business Essentials, 1966. Fry, Richard, Millennials Surpass Gen Xers as the Largest Generation in U.S. Labor Force, Pew Research Center, 2015. Herzberg, Frederick, “One More Time: How Do You Motivate Employees,” Harvard Business Review, January- February, 1968. Koontz, Harold, “The Management Theory Jungle,” Journal of the Academy of Management, December 1961. Likert, Rensis, New Patterns of Management, McGraw Hill, Inc., 1961. Maslow, Abraham H., “A Theory of Human Motivation,” Psychological Review, 1943, p. 50. McClelland, David, “That Urge To Achieve,” Think Magazine, 1966 by the IBM Corporation. McGregor, Douglas M., “The Human Side of the Enterprise,” Management Review, November 1957. Mintzberg, Henry, Managerial Work: Analysis From Observation, Management Science, vol. 18, no. 2, October 1971. Pink, Daniel H., Drive, The Surprising Truth About What Motivates Us, The Penguin Group, 1995. Skinner, B. F., Science and Human Behavior, New York: McMillan, 1953. Tay, L., and Diener, E., Needs and subjective well-being around the world, Journal of Personality and Social Psychology, 2011.

51



52

Types of Intellectual Property

5 Types of Intellectual Property Donald W. Merino Transpacific Advisors

53

Engineering Management Handbook

5.1 Introduction There are at least four types of intellectual property that are commonly used today. They are copyright, trademarks, trade secret, and patents. In order to understand the value of patents and how they fit into the landscape, a review of the various types of intellectual property is required.

5.2 Copyrights The first type of intellectual property is copyrights. A copyright is a legal right created by the laws of a particular country. This grants the creator of the work exclusive rights to use and distribute this work. Typically, this is a limited time right. The right is applicable to any express representation of a creative work. It can be shared among multiple authors, where each one has the right to use the licensed work. These are called “rights holders.” Other rights frequently included in copyrights include control over derivative work, distribution, public performance and what is called “moral rights”. Moral rights are unique and typically seen in Europe but are now becoming more common around the world. Basically, a moral right is a right that preserves the integrity of the work and bars alteration, distortion or mutilation of the work. Much of the fight over the colorization of black-and-white films was a battle over “moral rights.”Copyrights came about with the invention of the printing press. The legal concept originated in England in the 17th century with the “Licensing of the Press Act” in 1662, which was an act by Parliament. Since then the law has been further modified and codified. One of the best examples of where copyright protection is found is in the U.S. Constitution. In 1787, the U.S. Constitution included a clause: “to promote the progress of science and useful arts” found in Article 1, Section 8, Clause 8 of the U.S. Constitution. This clause gives Congress the right: “to promote the progress of science and useful arts, by securing for limited time to authors and inventors the exclusive right to their respective writings and discoveries.” This clause in the U.S. Constitution is the basis of copyright and patent law. What is the justification for creating copyrights? It is to enable the creators of intellectual property to create wealth to financially support themselves and, hopefully, to continue to create new works. Examples of copyrighted materials include songs, paintings, books, movies and photographs. One distinction of copyrights compared to patents for example, is that copyrights do not cover the idea or the information itself but rather how those are expressed. For example, in a typical story where boy meets girl, boy loses girl, boy gets girl back, the “Arc of the Story” is not copyrightable but the actual story and characters are. Since copyrights are protected by law, there are exclusive rights usually attached to them. These include the right to: • Produce copies and to sell those copies • Import or export the work • Create derivative works • Perform or display the work publicly • Sell the rights to others • Transmit or display the work by radio or video

54

Because of these rights, copyright holders can become extremely wealthy. For example, it is often the songwriter who is the wealthiest member of any band. The Beatles member, Paul McCartney, is far wealthier than Ringo Starr because he wrote many of the famous songs that the Beatles and others have performed. Every time you hear songs like “Hey Jude” or “Let it Be” on the radio you can be assured that Paul McCartney is obtaining payment for that broadcast. Interestingly, the richest songwriter today is Andrew Lloyd Webber who rarely performs at all but only writes the songs that others perform. The duration of copyrights varies by country. In the United States copyrights have increased in duration initially starting at 28 years and currently running the life of the author and +70 years after the author’s death.

Types of Intellectual Property

In the case of joint works, a copyright lasts 70 years after the last of the joint owners dies. And in the case of what is called “anonymous works” or “works made for hire” the term is 120 years from the year of creation. For many years software inventions were considered un-patentable and instead were protected by copyright. As the wealth of software companies grew software became patentable. It wasn’t until 1996 that the U.S. Patent and Trademark Office (USPTO) issued guidelines for examining computer-related patents. Over the last 30 years the number of patents granted per year in USPTO class codes that relate to computer-implemented inventions has grown from less than 1,000 to over 17,000.

5.3 Trademarks The next form of intellectual property is the trademark. A trademark is a recognizable sign or design or an expression that identifies a product or service from a particular source. Most sources agree that trademarks were first used by blacksmiths when they made marks to identify which blacksmith made which sword. During the reign of Henry III in 1266, the English Parliament required all bakers to use a distinctive mark for the bread they sold. This could probably be considered the first trademark legislation. Modern trademark laws began to emerge in the late 19th century particularly in France when they passed the “Manufacture and Goods Mark Act.” England in 1862 passed a similar law called the “Merchandise Marks Act.” In the United States Congress first attempted to establish a trademark system in 1870 relying on Article 1, Section 8, Clause 8 of the United States Constitution. The U.S. Supreme Court struck down this first attempt because it did not agree that a trademark would promote the progress of science and useful arts. In 1881 Congress instead passed the new trademark act that was pursuant to the commerce clause powers. This basic act has been modified through the years. The law considers trademark to be a form of property. These rights are established through actual use in the marketplace or through registration with a trademark office. Depending on the country, a trademark can be established by either or both means. Some jurisdictions do not recognize trademarks through use of the trademark. In other countries the holder of the right will not be able to enforce his or her right through trademark infringement proceedings unless the trademark is registered. In the United States the trademark rights registration process requires the examination of the trademark for compliance. Once a trademark is obtained there are a series of rights that are conferred upon the owner. These rights typically include the right to exclusively use of the mark for products and services created. In most countries the law also allows the owner of a trademark to stop others from using his or her mark in relation to products or services that are identical or very similar to his or her product. The test is usually if there is any sort of confusion on the customer side as to the source of the product. One interesting aspect of trademarks is that trademarks must be maintained by actual use. In most countries if the trademark is not used for five years it is considered to be abandoned. Additionally, in many jurisdictions it is required that the trademark owners defend their trademark from other infringers. When the author was working at Intel Corporation he had a front-row seat to one of the best examples of trademark and trademark enforcement. One of the issues for Intel was that, as a component manufacturer, its ability to distinguish its product was limited. For example, how many of us open the computer we use and look inside at the chips? Not many, unless we are engineers. In 1991 Intel created the “Intel Inside” brand program. Many case studies can be found that describe the details of this program and its success. Intel then set about to convince the computer manufacturers that if they featured an Intel microprocessor the customer would see value in that. Part of the strategy was to “co-market” with their customers to highlight the “Intel Inside” brand. Separately, Intel also created very aggressive trademark enforcement. Intel was able to establish through its protection process that anything that was “blank - Inside” would create customer confusion and therefore would need to change its trademark. For example, there was a group in California called the “Yoga Inside Foundation,” which provided yoga instructions to at-risk juveniles. Intel was concerned that “yoga inside” would likely cause confusion and/or dilute Intel trademark rights. Ultimately, “Yoga Inside” changed its name to “Yoga on the Inside” and reached an agreement with Intel to stop legal action. Why would Intel risk the wrath of the public by being so aggressive? The answer is the value of the Intel trademarks. The Intel name and the Intel Inside program were considered one of the four pillars that

55

Engineering Management Handbook

kept Intel’s margins so high. The Intel trademark and brand recognition was probably the strongest pillar of Intel’s business. While there are many ways to measure the value of the brand, Intel has for many years been one of the top 10 most valuable brands. In 2013, Intel was listed as the ninth most valuable brand and the value of the brand was in excess of $37 billion. This would make up between 20% and 30% of Intel’s total market capitalization, a very significant number indeed.

5.4 Trade Secrets Trade secrets are the next form of intellectual property discussed. What is a trade secret? A trade secret is an invented formula, practice, process, design or method or information that is not generally known of which the business obtains an economic advantage from. There are many examples of trade secrets in our daily life; one of the most famous is the recipe for Coca-Cola, Kentucky Fried Chicken, and even WD-40. But it is important to note that not all trade secrets are recipes. Other trade secrets can be the methods and practices that a company uses to make its product. In some countries trade secrets are referred to as confidential information. These trade secrets are generally referred to as classified information in the U.S. because that designation is typically reserved for governmental secrets. How trade secrets are defined often depends on the country or jurisdiction in question. For example, in the U.S., a trade secret is defined under 18 U.S.C. Chapter 1839: “the term ‘trade secret’ means all forms and types of financial, business, scientific, technical, economic, or engineering information, including patterns, plans, compilations, program devices, formulas, designs, prototypes, methods, techniques, processes, procedures, programs, or codes, whether tangible or intangible, and whether or how stored, compiled, or memorialized physically, electronically, graphically, photographically, or in writing if— • (A) the owner thereof has taken reasonable measures to keep such information secret; and • (B) the information derives independent economic value, actual or potential, from not being generally known to, and not being readily ascertainable through proper means by, the public;” Similarly, the 1996 Trade-Related Aspects of Intellectual Property Rights (TRIPS) agreement on trade-related aspects of intellectual property rights that define trade secrets as follows: “Is secret in the sense that it is not, as a body or in the precise configuration and assembly of its components, generally known among or readily accessible to persons within the circles that normally deal with the kind of information in question; Has commercial value because it is secret; and Has been subject to reasonable steps under the circumstances by the person lawfully in control of the information to keep it secret.” There are number of similarities between the two definitions. In 1970s the US Supreme Court allowed states to freely develop their own trade secret laws. In response to this many states adopted the Uniform Trade Secrets Act (UTSA). This law has been further amended and currently 47 states in the US have adopted it as the basis of their trade secret law. Thus, in the U.S. trade secret law is a law both on a state-by-state basis and at the federal level. At the federal level the US developed the Economic Espionage Act of 1996 (18 U. S. C. Chapters 1831-1839). This is a law that has used the example definition of a trade secret above. Most countries in the world are signatures to the international agreement on TRIPS. TRIPS is an international agreement administered by the World Trade Organization (WTO) and attempts to set a series of standards for intellectual property regulation for all members of the treaty. While the TRIPS agreement calls for standardization for copyright, patent and trademark protections, it also calls for the protection of undisclosed information in Section 7 Article 39. While the TRIPS agreement is rather general, it does form the basis for protection of trade secrets throughout the world. 56

Types of Intellectual Property

So what does one have to do to create a trade secret? Distinguished from other types of intellectual property there is no registration requirement for a trade secret. Rather it is up to the company or individual to keep the information secret and to protect it. When one says that it must be kept confidential, it does not mean that the information cannot be shared, but that a company must take care in disclosing that information. For example, the use of non-disclosure agreements and non-compete agreements helps companies disclose trade secrets to others, for example vendors and employees, while maintaining their confidentiality. Often companies need to decide what information should be patented and what information should be kept as a trade secret. If a trade secret is successfully kept as a secret there is no time limit imposed and the trade secret can last for an indefinite period of time. But a trade secret can be legally reverse engineered and/or if that trade secret becomes common knowledge due to other people’s research, then it will no longer be a trade secret. For example, the recipe for Coca-Cola has been kept secret for over 100 years and to date no one has been able to reverse engineer the Coca-Cola recipe. Interestingly, in 2006 a Coca-Cola employee tried to sell the recipe to Pepsi. To Pepsi’s credit they notified Coca-Cola officials and the employee was arrested. So, while recipes are a clear example of a trade secret, what are some of the other examples? Certain methods and processes can be considered trade secrets. For example, how the New York Times determines which books will be on its bestseller’s list is a method that the New York Times considers a trade secret. Additionally, many kinds of business information can be considered trade secrets; for example, how a company determines its pricing for certain products, a company’s marketing strategy or customer lists. Also, designs and methods can be considered trade secrets. When a trade secret is stolen or misappropriated there are number of penalties that can be imposed. Many states as well as the U.S. federal law include both imprisonment and fines for stealing trade secrets. Additionally, the federal law and many state laws also call for forfeiture of any profits obtained from the stolen trade secret. This law applies not only to the person who steals the trade secret, but also to anyone who tries to buy the trade secret knowing that it was stolen. In many trade secret cases today the person who is the violator is the person who wants to sell the stolen trade secrets. One interesting problem that most companies face is the question “should I file a patent on an invention or should I keep it a trade secret?” There are pros and cons to both courses of action. Much depends on the product and some depends on the ease of reverse engineering. In the chemical and pharmaceutical industries most companies elect to patent inventions because reverse engineering is relatively simple. In the electronics industry most companies chose a mixed strategy. In other words, companies will patent certain inventions and at the same time keep some inventions as trade secrets. The more complex the product the more inventions are usually needed to make the product work.

5.5 Patents What is a patent? According to the World intellectual Property Organization (WIPO), a patent is an exclusive right granted for an invention. The invention is a product or process that provides, in general, a new way of doing something or offers a new technical solution to a problem. To receive a patent, technical information about the invention must be disclosed to the public. The principle of a patent is that the patent owner obtains from a government agency the exclusive right to prevent or stop others from using the patented invention. In other words, patent protection means an invention cannot be commercially made, used, distributed, imported or exported by someone without the patent owner’s consent. Patents are territorial rights that are the exclusive rights applicable in the country or region where a patent has been filed and granted. Generally, a patent is granted for a limited period of time, mostly 20 years from the date the application is filed. Why do countries grant patents? In general it is because patents represent a bargain between the government and the inventor. That bargain is: the inventor is willing to donate his or her invention to the public; the government, in return, will grant a monopoly right for the invention. Why does this make sense? It makes sense because throughout history there have been technologies that have been lost to mankind. These include Stradivarius violins, Damascus steel, Roman cement and a number of herbal drugs. Let’s look at a few of these cases.

57

Engineering Management Handbook

58

Stradivari violins are considered some of the best musical instruments ever built. Approximately 1000 of these were produced by the Stradivari family between 1650 and 1750. Today many of these violins are worth millions of dollars and there are only around 600 of these instruments left. The name Stradivari is one of the most prized brand names even today. The problem is that the technique for building a Stradivari violin has been lost. Another example is Roman cement. Modern concrete was developed in the 1700s and is a simple mixture of cement, sand, water and rocks. But this recipe, invented in the 1700s, was not the first time it was invented. In fact, concrete was widely used by the Romans over 2000 years ago. The Roman mastery of concrete allowed them to build many of the famous structures that we associate with the Romans like the Coliseum, the aqueducts and the Roman baths. Like many technologies of the Greeks and Romans this technology was lost during the dark ages. Why it was lost we will never know; some believe it was a trade secret of stonemasons. What is the most interesting is that Roman cement has certain qualities that are different and better when compared to traditional cement that we use today. Structures like the Coliseum in Rome have managed to last for thousands of years while many modern structures are known to wear down much faster. Many believe that this is the result of different chemicals the Romans added to their cement, including things like milk and blood. It is believed that these different chemicals allow the material to better expand and contract with the heat and cold. Unfortunately, we will never know because the written recipe of the method is lost. The history of patents can be traced back to the concept of “Letters Patent.” A “Letters Patent” was a royal decree that granted something to somebody. Throughout history, Letters Patent were used to grant property to individuals. Additionally, in the 1300s Edward III decided to grant “letters of protection” to various artisans to encourage them to come to England and develop their skills within England. The first evidence of a “letters of protection” appeared in 1331 when King Edward III granted one to John Kempe, a Flemish weaver, as part of the effort to encourage foreign artisans to settle in England. Apparently, other craftsmen were encouraged to immigrate to England and were offered similar privileges that included clockmakers from Delft and craftsmen in the metalworking field who could help manufacture cannons and gunpowder. Another example was the grant by Henry VI in 1449 to John of Utynam, a Flemish man, who was skilled in the art of making stained-glass. Other examples outside of England include a patent awarded by the Republic of Florence to an architect for a barge and hoisting gear that was used to carry marble along the rivers. That patent was a three-year patent. By the 1450s the growing city state of Venice began issuing patents for new and inventive devices; the period of protection was for 10 years and was mostly in the field of glassmaking. As these Venetian craftsmen began emigrating to new countries they sought similar patent protection in their adopted countries. The king of France introduced the concept of publishing a description of the patent a century later. The English patent system evolved from those humble beginnings into the first modern recognizable patent system. In the 16th century it was common for the British crown to grant letters of patents or monopolies to their favorites or to people who were prepared to pay for them. The power to grant monopolies was used to raise money for the crown and was widely abused as the king granted patents to all sorts of common goods like salt and starch. After a certain amount of public outrage, James I of England was forced to revoke all existing monopolies and declared that the grant of “Letters Patent” could only be used for “projects of new invention.” This was incorporated into the “statute of monopolies,” which was passed in 1624. While the statute has been considered a key foundation of patent law, it also marks the transition of England’s economy from a land-based agrarian feudal system to that of industrial capitalism. Through the process of judicial interpretation of the law important developments emerged in the th 18 century. During the reign of Queen Anne patent applications were required to supply a complete specification of how the invention worked. This complete specification would be available to the public. Legal battles in 1796 occurred over the steam engine. James Watt was a Scottish inventor and mechanical engineer who made significant improvements to the Newcomen steam engine. He had realized that the current engine designs wasted a great deal of energy by repeatedly cooling and heating the cylinder. Watt introduced a design enhancement that used a separate condenser and this radically improved the power efficiency and cost-effectiveness of steam engines. Eventually, he got his engine to produce rotary motion which greatly increased the use of the steam engine beyond just pumping water. Edward Bull, the origi-

Types of Intellectual Property

nal manufacturer that Watt used, started making his own design of the steam engine and then challenged Watt’s patent on enforceability when Watt tried to stop him from producing his design. Bull believed that Watt had not invented the steam engine but had just improved the existing steam engine and the question whether or not an improvement could be patented became the issue of the case. The judges in 1799 issued a decisive opinion in favor of Watt and the concept of patenting an improvement was accepted. As discussed in the section on copyrights the “patent and copyright clause” of the U.S. Constitution was proposed in 1787 by James Madison in the Federalist papers (Federalist No. 43). Madison wrote: “the utility of the clause will scarcely be questioned. A copyright of the author’s has been solemnly adjudged, in Great Britain, to be a right of common law. The right to useful inventions seems with equal reason to belong to the inventors. The public good fully coincides in both cases with the claims of the individual”. The first patent act of the U.S. Congress was passed April 10, 1790 and is titled “An Act to Promote the Progress of Useful Arts.” The first patent was granted July 1790 to Samuel Hopkins for producing potash. The act called for the Secretary of State, the Secretary of War and the Attorney General or any two of them to decide if the invention is useful and important and the cause of the letters of patent to be issued. Additionally, in this act it set up the need to have a specification in writing and a model to be delivered and filed in the office of the Secretary of State. The requirement of the model was dropped later in the century. The act called for the monopoly to last for 14 years once issued. From these humble beginnings the patent system was borne and a separate patent office to examine patents was created in 1802. Patent rights became more international upon the signing of the “Paris Convention for the Protection of Industrial Property” in 1883. This convention is still in force. As of September 2014 it had 176 member countries. The convention called out three main ideas: national treatment of intellectual property, priority right and common rules. The first of these is “national treatment,” which means that when an application is filed for a patent in a foreign country, the applicant and application will receive the same treatment as if they lived and were born in the country that they are seeking the patent in. For example, if an American citizen files a patent in Germany, the German government cannot treat that American citizen and his patent application any differently than it would treat a German citizen. Also, if the patent is granted, the owner will benefit from the same protections and legal remedies as if the owner were a national in the country where the infringement occurred. The next element of the convention is the “priority right.” Here the convention calls for the priority right (or priority date) as the first filing date that would have the same priority date in another country so long as the application is filed within 12 months of the patent. From this first convention a number of other treaties, international treaties and agreements have been established. The “Patent Cooperation Treaty” (PCT) signed in 1970 establishes an international patent filing system that makes it possible to seek patent protection for invention simultaneously in a number of countries. (Currently, there are 148 countries that are signatories to the treaty.) The “Patent Law Treaty” (PLT), signed in 2000, establishes common and maximum requirements regarding many procedures and formalities related to national and regional patent applications and patents. (Currently, there are 59 countries and the European Patent Organization who are signatories to the treaty.) At this time there is a proposed international patent treaty called “The Substantive Patent Law Treaty” (SPLT). This treaty, in contrast to the “patent law Treaty,” goes beyond formalities and seeks to harmonize requirements such as novelty, inventive step, non-obviousness, and sufficiency of disclosure, utility of an invention and claim drafting and interpretation. As of this writing this treaty is on hold. The “Budapest Treaty” signed 1977 (currently there are 79 countries and the European Patent Organization that are signatories to the treaty) concerns the international disclosure of biotechnological inventions. It stipulates that for the purpose of patent the applicant must deposit the microorganism with an international “depository authority.” This will ensure that there are examples of the organisms for future generations. These treaties and others are administered by the World Intellectual Property Office (WIPO). As was stated earlier, a patent is an exclusive right for invention. This exclusive right to the product or process is only available if it is a new way of doing it or if it offers a new technical solution to a problem. To receive a patent, the invention must be disclosed to the public in the patent application. Once a patent

59

Engineering Management Handbook

has been obtained the patent owner has the permission to license other parties to use the invention. The owner may also sell the rights to someone else. If this is done then the new owner of the patent has the same rights as the original inventor. Once a patent expires the protection ends and the invention enters the public domain. In other words, after the patent expires anyone can commercially exploit the invention without fear of infringement. During the period of protection it is the patent owner who has the right to decide who may or may not use the patented invention. Patents can be granted in many fields of technology—everything from computer chips, nanotechnology, chemical compounds as well as plants and everyday kitchen devices. Patent protection is limited to a period of 20 years from the filing date of the application. Patents are territorial rights; in general, the exclusive right is only applicable to the country or region where the application has been filed and granted and only in accordance with the laws of that country or region. To enforce his or her rights the patent owner has to take someone to court. In most systems a court of law has the authority to stop patent infringement, but it is up to the patent owner to monitor, identify and take action against infringers. In order to obtain patent protection the inventor needs to meet certain key conditions. While it is not possible to compile an exhaustive list there are five key conditions that typically must be met. First, the invention must show novelty. That is, it needs to show some new characteristic that was not known before and it must be different from the “prior art.” The invention must be “non-obvious”; it needs to be something that someone who has ordinary skill in the art would not have known. The invention must be useful. It has to go beyond a mere theoretical phenomenon and the subject matter of the invention must be disclosed in a manner that is clear and complete and would enable someone with ordinary skills in the art to replicate the invention. And lastly, a patent must be patentable under the law of the country. In some countries scientific theories, mathematical methods, plant or animal varieties or computer programs are not patentable, but in other countries they are. There is no universal international system for granting patents. Rather, patents are granted by national offices or by regional offices that carry out that task for a number of countries. An example of a regional office is the European Patent Office (EPO). Under these regional systems an applicant will request protection from one or more member states of the original organization. The regional office accepts a patent application, examines the patent and if all criteria are met, it ultimately grants the patent.

5.6 Obtaining a Patent How do you obtain a patent? The first step in getting a patent is to determine if you have sometime new and novel. So, typically, you would do a search to see if the invention has already been patented or disclosed. This is called a Prior Art Search. Once you determine that you have something that can be patented, then you need to fill out a patent application. Many patent offices actually provide specific forms to fill out and in some patent offices you can even fill out a patent application online. In the patent application you generally must describe the title of the invention as well as what technical field the invention is part of. You will need to include a background and a description of the invention in clear language and enough detail that a person with an average understanding of the field could use or reproduce the invention. Many of these descriptions are accompanied by visual material such as drawings or diagrams. Most importantly you must clearly and concisely define the matter that you are hoping to patent. This is called “the claims” of the patent application. More than anything else it is “the claims” that will determine the scope of your invention. Many great inventions have been “mis-claimed” and the resulting claims are far narrower than the invention deserves. This frequently leads to patents that are far less valuable than they should be. Once a patent application has been completed, the next step is for the patent office in the country or region where the application is filed to examine the patent. One of the first steps in the examination process is to determine if the invention has already been patented or invented. The patent office will conduct its own prior art search to augment the prior art search of the applicant. The patent office will determine if the inventor/applicant had missed anything in the disclosure to the patent office. Next, the application is examined by a patent examiner. The patent examiner acts as an advocate for the public. His job is to en60

Types of Intellectual Property

sure that the application develops into a clear and complete record. He tries to act cooperatively with the inventor to investigate the patentability of the idea according to patent laws. Ultimately, the examiner will serve as a judge as to the patentability of the inventions claimed in the patent application. Based on that the examination process is there to make sure that the examiner can understand the invention set forth in the specification and determine if the application and specification is adequate to define the “metes and bounds” of the claimed invention. Additionally, the examiner will determine if the patent application is novel, useful, non-obvious and patentable. The examiner will communicate with the applicant by writing Office Actions (OA), which identify and analyze the patentability of claimed inventions. The applicant has the ability to respond to these Office Actions and argue his case with the patent examiner. This back and forth is called “patent prosecution.” In the end, the examiner is trying to answer certain questions about the application: what subject area is most related to the applicant’s invention; what are the existing inventions the applicant identifies; what problems did the applicant identify with the existing inventions; how does the applicant propose to solve the problem; how does the applicant implement the solution and do the claims incorporate the applicant’s solution. In the U.S. Patent Office, the applicant needs to decide what type of application to file.

5.7 Design Patents A “design patent” is an ornamental characteristic of a product. One of the best examples of a design patent is the form of the Coca-Cola bottle D48, 160. Other countries have a similar concept called a registered design or industrial design. With a design patent an object with a design that is substantially similar to the design claimed cannot be made, used, copied or imported into the United States. The copy doesn’t have to be exact; it only has to be substantially similar. A case in point here is the Apple vs. Samsung case that Apple won because Apple was able to show that the face of the Samsung GalaxyS 4G was substantially similar to Apple’s iPhone design patent D593,087.

5.8 Plant Patents The second type of patent is a “plant patent,” which describes a new variety of asexually reproduced plants. In the United States, the Plant Patent Act of 1930 was enacted as part of the Smoot-Hawley Tariff. This was inspired by the work of Luther Burbank, the famed botanist. The legislation made it possible to patent new varieties of plants. Plants can also be protected internationally under the International Union for the Protection of New Varieties of Plants. The object of the convention was to protect new varieties of plants with an intellectual property right. These rights are known as plant breeders’ rights and give the breeder exclusive control over the propagating materials including seeds, cuttings, tissue cultures, etc. and the harvest material of a new variety for number of years. There are more than 72 party members to this convention.

5.9 Utility Patents The last type of patent to discuss is the “utility patents.” In general, the utility patent protects the way an invention is used and/or works. It can be granted to anybody who invents a new and useful method, process, machine device, manufactured item, system, or chemical compound. Utility is actually a patentability requirement. The US law 35 U.S. Code § 101 (Inventions Patentable) states: “Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent.” As recently as the late 1990s few patents were ever challenged on patentability and this law was used to limit inventions that could never possibly work like perpetual motion machines. More recently, however, this law has become more controversial and has seen rulings of unpatentability of electronic signals, medical diagnostic or treatment and certain “business methods.” The concept of utility is mostly an American one. European patent law does not consider utility as one of its criteria for patentability. Instead, it requires that an invention must have industrial applicability. In other words, an invention is eligible for a patent if it can be made or used in some kind of industry. This is

61

Engineering Management Handbook

a similar requirement to utility but there is a difference. Generally, to fulfill the requirement of utility for patents, there are three main issues to review: the operability of the invention, the use of the invention that must be beneficial, and the practical use of the invention. Operability is satisfied by enabling the invention in the detailed description of the patent. The beneficial use must be that the patent is capable of providing some benefit. In the past it was believed that beneficial utility established in a patent should not be granted if frivolous or immoral or against public policy. But recently the patent office has taken the following position: “A rejection under 35 U.S.C. 101 for lack of utility should not be based on grounds that the invention is frivolous, fraudulent or against public policy” (Manual of Patent Examining Procedure 706.03(A) III). The last issue is practical use; in other words, the invention must have some real-world use. With utility patents there are two different types of applications that can be filed in the United States. The first is the traditional non-provisional application. The second is the provisional application. Since 1995 the USPTO has offered inventors the ability to file provisional applications for a patent. A provisional application pendency lasts 12 months from the date it is filed. It cannot be extended and at the end of the 12 months the provisional application expires. While a provisional application is inexpensive one must understand that in order for it to be in force a non-provisional application will need to be filed also. There are a few benefits to the filing of provisional but also some risks. The benefits include a simplified filing at a lower investment that gives the inventor 12 months to assess the invention and determine if additional investment should be made. Another benefit is that while the provisional establishes the application filing date for the invention it also permits the usage of “patent pending” notice for 12 months after the description of the invention is filed. A provisional application begins the Paris Convention priority year and it also allows an inventor to begin to promote the invention while having some degree of protection. The risks are that the benefit of the provisional application cannot be claimed after or if the 12month deadline is passed and the provisional application cannot result in a U.S. patent unless the corresponding non-provisional application is filed within 12 months. Provisional application cannot be filed for design patents and provisional applications are not examined. Oftentimes provisional applications have not completely disclosed the invention and therefore the parts of the non-provisional application cannot claim benefit to what is not disclosed in the provisional application. Inventors must understand that the provisional application will not mature into granted patent without further submissions and cost by the inventor. Some unscrupulous invention promotion firms have misused the provisional application process leading inventors to believe that they can obtain a patent inexpensively. These firms instead take the money from the inventor and leave the inventor without a patent. Different from a provisional application is a standard application. The standard application requires all the necessary parts of the patent including a written description of the invention as well as the claims. It should include all the information required to grant the patent. The application may or may not result in a grant of the patent; it depends on the outcome of the examination of the patent office. Here we are talking about application filing, preparation and prosecution, sometimes called “prep and process.”

5.10 Major Elements of Patent Application Let us review some of the major elements of the patent application. As previously mentioned, the patent specification is the detailed description of the invention and will set out the scope of the patent. Generally, this section includes a background or an overview of the invention, a description of the invention and any drawings. It also includes the claims that are very important. Other things typically included in application are any known “prior art” that the inventor knows about, the filing date, the priority claim, and an abstract of the invention. For example, rule 5 of the PCT states: G. The description shall first state the title of the invention as appearing in the request and shall: i. Specify the technical field to which the invention relates; ii. Indicate the background art which, as far as known to the applicant, can be regarded as useful for the understanding, searching and examination of the invention, and, preferably, cite the documents reflecting such art; 62

Types of Intellectual Property

iii. Disclose the invention, as claimed, in such terms that the technical problem (even if not expressly stated as such) and its solution can be understood, and state the advantageous effects, if any, of the invention with reference to the background art; iv. Briefly describe the figures in the drawings, if any; v. Set forth at least the best mode contemplated by the applicant for carrying out the invention claimed; this shall be done in terms of examples, where appropriate, and with reference to the drawings, if any; where the national law of the designated State does not require the description of the best mode but is satisfied with the description of any mode (whether it is the best contemplated or not), failure to describe the best mode contemplated shall have no effect in that State; vi. Indicate explicitly, when it is not obvious from the description or nature of the invention, the way in which the invention is capable of exploitation in industry and the way in which it can be made and used, or, if it can only be used, the way in which it can be used; the term “industry” is to be understood in its broadest sense as in the Paris Convention for the Protection of Industrial Property. In some patent offices, patent applications can also be filed as continuations to previous applications. These are convenient methods to include material from previous applications in a new application. In some cases the new application will take the priority year from the earlier patent. These are called “continuations.” Additionally, in the U.S., there is a method called a continuation-in-part application that allows the applicant to add new material to the application. In this case an applicant would add subject matter not disclosing the original patent but will repeat significant portions of the patent specification and will share at least one inventor with the original patent. This is the way to claim enhancements that were developed on the original invention. One of the odd situations is that in a “continuation-in-part” the applicant is likely to end up with two priority dates—one for the original patent and another priority date for the added material. The courts will use these two dates to determine the priority date of the patent and therefore what prior part is applicable. In some patent offices it is also possible to file what is called a “divisional patent application.” This application also claims the priority of the filing date from the parent application but is a divisional and claims distinct independent claims that are different from the parent application. Oftentimes divisionals are filed when the patent attorney determines that the specification covers more than one invention. Additionally, the patent examiner may issue what is called a “restriction requirement” because a patent can only claim a single invention. This would cause the applicant to file a divisional on all but one of the inventions disclosed. It should be noted that continuations and continuations-in-part are a USPTO phenomenon and generally not available in other jurisdictions. Divisional patent applications on the other hand are often found in other countries.

63

64

What is a Patent?

6 What is a Patent? Donald W. Merino Transpacific Advisors

65

Engineering Management Handbook

6.1 Where Patents are Held Patents are granted on a country-by-country basis. Additionally, when patents are discussed it would be remiss to not introduce the concept of a patent family. A patent family consists of the patents that relate to the original invention. When deciding to file a patent one must also decide in which countries to file that patent. As recently as the mid-1990s the main focus of most major companies were U.S. patents. Very little thought in the electronics industry was given to non-U.S. patents. At that time, China was a nascent market and was just beginning its economic rise. In 1997 most companies did not believe that Chinese patents would ever be enforceable and there they were considered not valuable. At that time most companies preferred to focus on the United States and somewhat also on Europe. The reason for this was due to market considerations and the legal landscape in those regions. In 1997 the United States Court system was considered a reasonably predictable system that enabled the patent holders to receive fair compensation for their patents as well as the possibility for injunctive relief. Europe, at that time, was somewhat of an afterthought for most electronics companies because, while the European Union (EU) as an entity was a large market, it was subdivided into many countries (Germany, France, England, Italy, Spain, and the Low Countries). Japan was also sometimes considered but usually as an afterthought. While Japan had a small sales market compared to the U.S. (approximately 1/5 the size of the U.S. market) there were many Japanese manufacturing companies and a substantial manufacturing base in Japan. Korea was an even smaller market than Japan and had fewer manufacturing companies. Additionally, in both the case of Korea and Japan, it was felt that the “home-field advantage” was significant for the manufacturing companies. In other words, most people believed that a non-Japanese or non-Korean company would be treated fairly in the court system if they were seeking infringement claims against a company based in that country. Since a patent can be enforced for the manufacturer’s sale and use of the invention, it was felt that most jurisdictions were either just too small a market or had “home-field advantage” for the local manufacturing companies. During this timeframe many patent families had relatively few foreign counterparts. Most people focused on the United States since that was four to five times the size of any other local jurisdiction. Also, depending on the technology market, many companies wanted to file where their foreign competitors were based. This was Germany and/or Japan and, as a result a patent might be filed in those countries. Rarely was it the case that more than 3 to 5 jurisdictions would be covered. Many of the European companies as early as the late 1990s had a pretty aggressive strategy for filing patents in the EU and would mainly file in what was considered the big three European countries—England, Germany and France. By 2010, most major companies were filing significant numbers of patent applications in China and other major countries in Asia due to the economic growth of that region. But with all of that said, even in 2015 finding broadly filed portfolios with good worldwide coverage was difficult at best and many portfolios are only U.S. patents. Even in Asia you often find companies that file patents in their home jurisdiction plus the U.S. and China. However, there are signs of this changing with the new applications companies are filing. A combination of factors has led to this change. In the United States we have seen a systematic weakening of rights for the patent holder. The United States has also gone through quite a significant change in how damages are calculated, lowering the potential outcome of a successful litigation. Lastly, the United States has made injunctive relief extremely hard to obtain. In the 2015 market some of the most valuable patents were European patents and, in particular, German patents. Conventional wisdom is that the German courts are friendlier to patent holder rights and that injunctive relief is available in German courts. Compounding this issue is the cost of litigation. In the United States, litigation is both time-consuming and costly due mostly to discovery that does not occur in many jurisdictions. Also it is reasonable to expect that a patent litigation will cost between $5 and $10 million and take up to five years. Conversely, in Germany the cost of litigation is much less expensive, approximately $1 million for a case and relatively quick, between one and two years. Additionally, China appears to become a very important market for patents. First, China is a very large market and is beginning to rival the U.S. in the sale of high-end electronics, computers, smartphones and networking equipment. Second, much of the world’s manufacturing base has moved to China. While Chinese patent law and patent courts are relatively new it does appear that the courts are moving in a direction to protect the rights of patent holders. Additionally, the Chinese government has 66

What is a Patent?

made creating patents a priority and urges companies and individual inventors to invent and file patents. While not a perfect market, many people believe that China will continue to strengthen its patent laws in court and ultimately become a location where getting patents will become imperative. The problem in architecting any portfolio is that there is no one simple solution. One needs to consider a case-by-case basis filing strategy. A company needs to look at what its goals are, what market it is in and who its competitors and or IP threats are.

6.2 What are the Laws and Rules in the U.S.? U.S. patent law is a combination of three different elements. First, the U.S. Constitution allows Congress to pass patent laws. Congress writes those laws and then the courts interpret those laws. In order to understand patent law in the United States it is important to understand where it comes from, what it is and how it is interpreted. The USPTO (United States Patent and Trademark Office) publishes the Manual of Patent Examining Procedure (MPEP). This manual describes the procedures that the patent office must follow when examining a patent application. The first MPEP was published in 1920; it is still used by patent attorneys and patent agents to make sure that they are following the appropriate regulations. In order to become a qualified patent attorney or a patent agent one must take an examination given by the USPTO that tests the knowledge of the applicant of the MPEP and the underlying laws and regulations. A copy of the MPEP can be found on the USPTO website and downloaded, but beware, it is very dense. For a non-patent lawyer it is important to know that the MPEP exists but it is primarily a document for the lawyers who are trying to obtain patents for their clients. U.S. patent law was created in the United States Constitution in Article I, Section 8, Clause 8. This is known as the Patent and Copyright Clause. Section 8 of the Constitution lays out what is called the Enumerated Powers of Congress. This section starts with the phrase “The Congress shall have power” and then lists the powers that the Congress shall have including the collection of taxes, duties, imports and excises, to pay the debt and provide for the common defense. The eighth clause is the clause that we are most interested in as it gives Congress the power: “To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.” What is interesting and often misinterpreted in this clause is that it grants the power to Congress to do these things it does not require that Congress actually do is these things. In order for there to be patent laws the U.S. Congress needed to take affirmative action and pass laws. The substantial law in the United States is found under Title 35 of the United States Code. Title 35 has five parts or sections, four of the parts are related to patents and one section is related to industrial designs. The first part establishes the United States patent and trademark office. As part of this section the Congress allows the United States patent and trademark office to set the procedures for granting a patent. The second part is titled Patentability of Inventions and Grant of Patents. Key in this part are sections 101, 102 and 103. In section 101 the law describes what is patentable and states: “Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.” Section 101 has recently been under scrutiny as to what can and cannot be patented. For example, can software be patentable under section 101? Or can an electromagnetic wave shape be patentable under this section? Section 102 covers the condition for patentability; novelty. In particular, this section sets up the idea of novelty as compared to the prior art. It states: “(a) Novelty; Prior Art. — A person shall be entitled to a patent unless— (1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention;” 67

Engineering Management Handbook

This section has become one of the bases for determining if a patent is valid. In particular, it requires of the person be the first to have invented the patent. Section 103 covers conditions for patentability; non-obvious subject matter. This section states that a patent must be not obvious to somebody having ordinary skill in the art. Specifically, this section says: “A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.”

68

Some years ago a section 103 objection was considered a rather easy bar for a patent holder to overcome. More recently the courts have interpreted section 103 much more aggressively and now 103 objections are not only common but many times prevail in patent cases either in District Court, the Appeals Court and/or in the Patent Office. The next chapter under part 2 describes the law for the application of patents. Of particular note is section 112, the specification section. This section states that a patent must have a written description of the invention and must set forth what is called the “best mode” contemplated by the inventor. It also states that the description has to be in such a way to “enable” a person skilled in the art to make and use the invention. Often times patents will be challenged based on this section. The challenge is that the patent is not patentable under section 112 because either the inventor did not describe the “best mode” or the patentee did not “enable” the invention in the specification. There has been a recent argument around what are called “paper patents.” These are patents only of ideas for which there are no working models or prototypes. The argument goes that many “paper patents” should not be granted because they do not enable someone to make use of the invention. This has become a common challenge now in District Court, the Appeals Court and in the Patent Office. Other chapters in Part 2 include who can review patent and trademark decisions, how patents are issued, the rules governing plant patents, designs and patent rights inventions made with government assistance, etc. Part 3 of U.S. code 35 is titled Patents and Protection of Patent Rights. This part covers a number of things; most important for us is chapter 28, which discusses infringement of patents. Section 271 describes what it means to infringe a patent, how to file an infringement case and what some of the rules are about infringement. Chapter 29 of part 3 is titled Remedies for Infringement of Patent, and Other Actions. This chapter sets out the remedies for infringement of the patent, the presumption of validity and available defenses. Section 282 states that the defenses for infringement are either non-infringement, invalidity based on section 101, 102 or 103 or that a patent fails to comply with section 112 which, as we discussed earlier, is the best mode and enablement section. Section 283 relates to injunctions and states that courts may grant injunctions on such terms that the court deems “reasonable.” Section 284 covers damages and states that the floor of damages is a “reasonable royalty” for the use of the invention by the infringer. Three important chapters in part 3 are chapters 25, 31 and 32, which establish what is called a reissue procedure, “Inter Parties Review” (IPR) and the post-grant review processes. Chapter 25 allows the patent owner to reissue effective patents; in other words, the patent owner has the ability to go back to the patent office and have the patent looked at again by the patent office. This chapter lays out the proceedings for making simple corrections as well as corrections in the title and/or the inventorship. It also allows a request for supplemental reexamination of a patent. If the reexamination is done within two years of the issue of the patent the patent owner can apply for a broader set of rights by broadening their claims but, if the reexamination request is done after two years, then the patent owner can only have the patent reexamined and only the same or narrower rights can be issued. Reexamination is usually done for one of two reasons. First, it is an attempt by the patent holder to broaden the claims of the patent. In this case the patent owner should review the patent on a regular basis and determine if the specification of the patent can yield broader claims. Since this can be done up to two years after the patent application issue, the

What is a Patent?

patent owner now has a chance to understand better how the technology is developing and craft a better set of claims. The goal is to develop claims that would be easier to prove infringement of. The second case is if the patent owner realizes that there is perhaps additional prior art that was not cited by the patent office. Thus, he now wants to put the prior art in front of the patent office and preserve claims that would still be useful for an infringement case. The second case is typically called a Narrowing Reexam, while the first case is called a Broadening Reexam. The IPR is a procedure that allows defendant companies to request the patent office to review the patentability of the patent that they are accused of infringing. This is a relatively new procedure and is a process that lets the patent office determine the validity of a patent after the patent has already been issued. In the first years of this new proceeding almost 80% of all patents brought to the patent office were found to have defects of validity and were declared invalid. It will be interesting to see how this process changes over the years. Many believe that this new process is just a burden for patent holders while others believe that determining the validity of a patent at an early stage will not only decrease litigation but will also ensure that litigation only goes forward on valid substantial patents. Once infringers initiate an IPR they are enjoined from further challenging the validity.

6.3 The Courts As stated previously, the courts in the United States have the important function of interpreting U.S. patent law. The courts also decide patent cases and it is because of the requirements of these patent infringement cases that the courts interpret the laws. In the U.S. there are basically three courts to contend with: District Court, Federal Circuit Court, and the U.S. Supreme Court. The first stop for infringement cases is U.S. District Court. The District Court will usually decide if the patent is valid and if the patent is infringed. (Note: previously discussed was that validity can be determined in the USPTO via an IPR procedure.) This can be done by both, the judge or the jury. When judges are hearing patent infringement cases the lawyers will often use arguments that require the judge to make a determination as to what the law means in the particular case. These rulings on the law are rulings that are frequently appealed to the next highest court. In a civil case either side may appeal the verdict. In a criminal case, the defendant may appeal a guilty verdict, but the government may not appeal if a defendant is found not guilty. A litigant who files an appeal, known as an “appellant,” must show that the trial court or administrative agency made a legal error that affected the decision in the case. Prior to 1982 the Court of Appeals from the District Court was the local Circuit Court. Circuit courts were defined by geographic boundaries. In other words, the circuit courts would decide cases and law based on rulings of District Courts in their jurisdiction. In 1982 Congress passed the Federal Courts Improvement Act, which merged the United States Court of Customs and Patent Appeals and the Appellate Division of the U.S. Court of Claims. They created the U.S. Court of Appeals for the Federal Circuit. This is known either as the Federal Circuit or by the initials C.A.F.C. This court hears the appeals from all the District Courts on patent matters as well as appeals from certain administrative agencies and appeals that arise under certain statutes. The Federal Circuit has jurisdiction over appeals from the U.S. Court of Federal Claims, U.S. Court of Appeals for Veterans Claims, the Board of Contract Appeals, U.S. International Trade Commission and the United States Patent Trial and Appeals Board as well as the U.S. Trademark Trial and Appeals Board. One of the unique aspects of this is because the Federal Circuit hears all appeals of patent cases their rulings having a binding precedent throughout all the District Courts. This is different than a typical Federal Circuit whose decisions are only binding precedent for the District Courts within the Circuit. Finally, after the Federal Circuit the next highest court is the Supreme Court. One of the important aspects of appeals courts is that they hear appeals on lower court opinions on law but not opinions on facts. In most cases the judge determines the law while the jury determines the facts. The case law around patents will be discussed but not in detail. This subject is, in itself, a whole book. Also for most practitioners understanding the case law is something that is probably best left to the lawyers. With that said it is important for a person in the patent business to be aware of what some of the new case law is and how that affects the overall business. For example, probably the biggest change to patent law in the last few years was the Supreme Court’s eBay Inc. vs. MercExchange, L.L.C. decision. This 69

Engineering Management Handbook

was an important case because the Supreme Court unanimously decided that an injunction should not automatically be issued on finding a patent infringement. It basically gave the District Court judge much more latitude to decide not to issue an injunction than was previously the purview of the District Judge. The court described a four-factor test to determine if an injunction should be issued. The factors are that the patent holder must demonstrate: 1. That the plaintiff suffered irreparable injury;That the remedies available in the law such as monetary damages are inadequate to compensate for the injury; 2. That, considering the balance of hardships between the plaintiff and defendant, a remedy in equity is warranted; and 3. The public interest would not be disserved by a permanent injunction. … These familiar principles apply with equal force to disputes arising under the Patent Act. This was a pretty radical departure from what most people thought the law was and it also radically affected patent valuations in the marketplace. In the past most companies and individuals felt that if they had a valid and infringed patent they could enjoin the other side from selling its infringing products. Because of the possibility of injunction most companies considered this a severe penalty that they wanted to avoid. This led to numerous settlements of patent cases just before trial as few companies wanted to risk the possibility of an injunction being issued. With this threat removed most companies, unless they have a directly competing product, only have to worry about monetary damages. Therefore they feel that time is on their side and that even if they lost a District Court case they can appeal it to the Federal Circuit and only delay the payment. It is obvious how this change in law had a direct impact on patent values as the payment time was lengthened. It also shows the importance of knowing the current case law and what cases are being decided that can drastically change the business environment for those involved with patents.

70

What is a Patent?

6.4 Reading a Patent Figure 6.1 illustrates the front page of a U.S. patent. Figure 6.1. Page of a U.S. Patent

For those who are unfamiliar with patents, the parts of a patent will be discussed. Note that how patents look changes depending on the source you are viewing it from. This example shows how the patent looks when sent to the inventor or printed from the USPTO image database. If one uses an online database, the same information appears but frequently the information shows up in different areas. For example, on the USPTO website, the claims are not written at the end of the patent but rather right after the bibliographic information. 71

Engineering Management Handbook

Below are the numbers of the patent in Figure 6.1. These will be used to identify the sections. (12) Type of Document: For an issued patent, this will just read “United States Patent” (or, if appropriate, “United States Design Patent”, “United States Plant Patent” or “Patent Application Publication”). ”Reissued Patent”: This patent is based on another patent, which was previously issued. Earlier in the chapter I discussed the process for reexamination and reissue was discussed. The Reissue Patent has been examined a second time because there are usually modifications to the claims. (10) Number: The patent number. This is a serial number of the patent – patents are issued and given the consecutive number in the series. Patent 6,000,000 was issued December 1999; Patent 7,000,000 was issued February 2006 and patent 8,000,000 was issued April 2011. Note that there is also a prefix. The prefix tells a little more about the patent. Sometimes there are prefixes to the number prefixes before the number, which indicates the kind of document: •B - Reexamination Certificate, issued after a patent was issued and has been examined again by request of the patentee. The numeric portion of the number is the number of the patent to which the reexamination certificate refers. (For design, plant, reissue patents, the Reexamination Certificate will include the normal designation letter - BD for design, BP for plant, etc.) •D or Des. - Design Patent - 14-year term, covers ornamental appearance of a useful object•PP or Plt. - Plant Patent - covers certain plants•RE - Reissue patent (RD for reissued design patent, RP for reissued plant patent) Documents also have a letter or letter/number after the number. These appear on the face of the patent only after January 2, 2001; these suffixes have meaning and tell us a little more about the patent. Here are a few of the common suffixes: •A1 - Published Patent Application - (if application is published more than once, A2 for second, etc.) •B1 - Utility Patent, not previously published •B2 - Utility Patent, previously published as an application •C1 - Reexamination Certificate (if more than one certificate, then C2, C3...) (B1 ... is used for pre-2001 Reexamination Certificates) •P1- Plant Patent (before 2001) •P1 - Published Plant Patent application (after 1/2/2001 - additional publications are P4, P5...) •P2 - Issued Plant Patent without pre-grant publication (after 1/2/2001) •P3 - Issued Plant Patent that was previously published (after 1/2/2001) •S - Design Patent (45) Date of Patent: The date the patent was issued by the USPTO. This is important because only after this date is the patent enforceable; sometimes this is the date that the term of the patent is measured from. These dates always fall on a Tuesday. (54) Title: This is the full title of the patent. Sometimes the title is very clear and one can quickly tell what the invention is. But sometimes the title is much more obscure and understanding the technology is very difficult for a generalist. (75) Inventors: All of the inventors will be listed on the patent. A practicing patent attorney said that if there were more than three inventors he felt it was important in a litigation to interview each inventor and ask what he contributed to the invention. It was his feeling that often companies will add names to a patent based on the group of people who were working on the project and therefore misrepresent the true inventor. This could have dire consequences in litigation.

72

(73) Assignee: If the patent is owned by an entity it is listed here. There will be no assignee field if the assignee is the inventor. It is important to note that the assignee name is filled in when the patent issue fee

What is a Patent?

was paid and, therefore, if there is a change of ownership after that time, the cover of the patent will not change. There is a separate database at the PTO that shows who the current assignee of the patent is but it is also important to know that it is not required assignees register their new assignment in the patent office. For example, when one company buys another company they it transfer all the patents legally in the contract for the sale but may not record the assignment in the patent office. This occurs many times. (*) Term extension notice: the term extension notice is relatively new and it will indicate if there are additional days of life to the patent. Basically, the patent office does not penalize the applicant for time the patent application spends in the queue at the patent office. This has been as much as three additional years; knowing the life of the patent is critical in determining the value. (21) Application Number: The identifying number of the application on which this patent was based. The serial number is always six digits, assigned sequentially as applications are received by the USPTO, prefixed by a two-digit series number. When the number reaches 999,999, they start a new series. There are two unique series that should be mentioned: series that begin with 90 represent patents that have been reexamined under Ex Parte Reexamination procedure and series that begin with 95 indicate patents reexamined under Inter Parties Reexamination procedure. (22) Filing date: the filing date is the date that the patent application was filed. Note the time the application was originally filed if the patent is a continuation and this may not be the “first filing date.”(65) Prior Publication Data: If this patent was published while it was a pending application, the publication number and date will be listed here. (60) Related U.S. Application Data and (30) Foreign Priority Data: If this application is related to any other applications or patents, they will be listed here. This field will often tell you what the “first filing date” was. (51) International Patent Classification: Patents are classified based on the IPC for ease of searching. Sometimes patents have multiple IPC numbers. All patents are classified by subject matter for ease of searching. The classifications in which a patent is indexed are listed in these sections. (53) U.S. CL: The USPTO uses its own U.S. Patent Classification System (USPC) in which all inventions are first put in a class having a three-digit number, then in a numbered subclass under the class. The subclasses are arranged in hierarchical form but not necessarily in numerical order. (58) Field of Search: These are the U.S. classes/subclasses that the examiner searched when he or she reviewed the patent. (56) References Cited: This is the list of prior art that was cited in this case. Some of the prior art may have been found by the examiner when he or she did the patent search or the prior art may be listed by the patentee when the patentee filed the application. Both U.S. and foreign patents may be listed as well as non-patent literature. In the U.S. and other countries such as Japan and Canada there is a “duty to disclose.” It is found in 37 C.F.R. 1.56. Duty to disclose information is material to patentability. “A patent, by its very nature, is affected with a public interest. The public interest is best served, and the most effective patent examination occurs when, at the time an application is being examined, the Office is aware of and evaluates the teachings of all information material to patentability. Each individual associated with the filing and prosecution of a patent application has a duty of candor and good faith in dealing with the Office, which includes a duty to disclose to the Office all information known to that individual to be material to patentability as defined in this section.” The number of references cited will sometimes indicate the importance of the patent. In other words, if the references cited are very long, the inventor and the inventor’s patent counsel thought that this

73

Engineering Management Handbook

pattern is particularly valuable. With that said it is not fair to say that because there are few references the patent is not valuable. There are many valuable patents that have very few references. Conversely, there are many patents that are not valuable that have a great number of references. Primary Examiner, Assistant Examiner: These are the USPTO examiners who examined the patent. (74) Attorney, Agent or Firm: When the Issue Fee is paid for the patent, one or more patent attorneys, patent agents or law firms may be listed on the cover sheet. (57) Abstract: A brief summary of the invention, with the emphasis on “brief ” (less than 150 words). Number of claims and drawing sheets: Lists the number of claims and drawing sheets in the patent. This is useful to determine if the copy of the patent is complete. Representative drawing: The examiner will pick one of the drawing figures and put it on the first page. This is usually figure number one but other figures can be picked.

6.4.1 Sample Drawing Shown in Figure 6.2 is a sample drawing for a U.S. patent. Figure 6.2. Sample Drawing for a U.S. Patent

74

What is a Patent?

Most U.S. patents have drawings. But some chemical patents do not have drawings. The drawings are numbered from “fig(ure) 1” up and sometimes a figure will be divided into several subfigures. The subfigures are identifiable because there is a letter suffix to the figure number. (fig. 1A, fig. 2B, etc.). It is sometimes necessary to show the “prior art” as a figure. These figures will be labeled as “prior art” so that one knows by looking at the patent that it is not part of the patented invention. You will notice on each of the figures that there are often reference numbers on the label. These are used so that the reference numbers can easily be called out for the specification of the patent. Frequently, there are several kinds of drawings for mechanical patents that will be a 3-D view, cutaway view, or an exploded view of the device. Cutaway views are used for semiconductor process patents. A cutaway view can show the layers of the semiconductor or the structure of a transistor after the process. In many electrical patents circuit schematics as well as block diagrams will be shown as figures. Block diagrams are used to show the system in general terms and how they work in relationship to each other. Sketches as well as flowcharts are often shown. Method patents often have flowcharts that show the steps of the method. After the drawings the “specification” of the patent appears. The specification is the written part of the patent. It has a number of parts:

6.4.2 Typical First Two Pages of the Specification Figure 6.3 shows the typical first two pages of the specification.

75

Engineering Management Handbook Figure 6.3. Typical Pages of the Specification

76

What is a Patent?

77

Engineering Management Handbook

Reading down the first column of the first page there are a number of sections: (A) First is the Title of the Invention: (See 37 CFR 1.72(a).) (B) Next is Cross-Reference to Related Applications: (See 37 CFR 1.78.) For the sample patent shown this part is not presented but if the panel is related to other patents and or earlier filed applications it will be stated right after the title. This will include if a provisional application has been filed or if the patent is a continuation or continuation of part of other applications or if the patent is a divisional of another application. Often foreign priority applications are listed here also but this is not required. Sometimes, if the invention was developed using federal research and development money, there will be a Statement Regarding Federally Sponsored Research or Development: (C) Next is the Background of the Invention: Most specifications set forth the background of the invention into different parts. (D) Following the Background of the Invention is the Field of the Invention: The field of the invention is usually a very broad description of the area of technology of the invention. Sometimes reading the section will give a clue to what the patent is actually useful for. (E) Then there is the Description of the Related Art: The description of related Art section talks about some of the prior art that was known before the invention occurred. This description may have references to specific patents and/or other documents. For example, if the patent was developed as part of a standard-setting organization, the standard number would be referenced here. Often this Background of the Invention is rather short. This was often done because patent attorneys feared that the Background of the Invention could be used to invalidate the patent or limit the scope of the claims and litigation. (F) Next is the Brief Summary of the Invention: This is a summary of the invention (see 37 CFR 1.73). The summary should give you the idea of the invention and hopefully is useful in directing people to understand what technology the patented invention covers. Sometimes this is the case but many times it is not and the summary of the invention is as dense and incomprehensible as the abstract of the invention. The summary should be broad; in fact, it should be broader than the broadest claim. (G) Then there is a Brief Description of the Drawing: This is typically a one-sentence description of the figures. (H) The last major section before the claims is the Detailed Description of the Invention this typically is the largest section of the patent: this is the description of the preferred embodiment of the invention (see CFR 1.71). What is most important here is that the description needs to be able to accurately describe the invention and ensure that the invention is actually enabled. Frequently the detailed description of the invention will end up defining some of the terms that are used in the claims. In this section of the specification each of the figures will be explained. Sometimes a detailed description will include the method for making the invention or the method for using the invention. Often there will be more than one example. This section of the specification is used to describe the various embodiments of the invention in great detail.

78

What is a Patent?

6.4.3 Claims Section Figure 6.4 shows the Claims Section. Figure 6.4. Claims Section of a Patent

In a printed patent the claims usually come last. With online databases, they are usually shown first immediately after the bibliographic information in the abstract. As mentioned before, the claims are the most important part of the patent because they define the metes and bounds of the patent grant. One way of thinking about patent claims is to use a property analogy. The patent claims represent the description of what is the piece of property that you own. Understanding these claims is key in determining if a 79

Engineering Management Handbook

product or process infringes the patent. This is oftentimes not an easy thing to do and we will discuss this in more detail later. A few points about claims: • If one claim is infringed the patent is infringed. • All elements of the claim must be used by the product or process in order to be infringed. • Many of the elements of a claim are part of the prior art. Only one element of the claim needs to be a novel one.

6.5 What Can You Do with a Patent?

80

Now that you have a patent, what do you do with it? There are many inventors with the misconception that a patent grants them the right to build something. What most people do not understand is that, basically, a patent is a “negative right”—in other words, it is a right to exclude people from practicing your invention. As we discussed earlier, recent rulings, in the U.S. in particular the eBay case, have made the “right to exclude” harder to obtain. So perhaps the “right to exclude” is a bit broad but at least we can say that a person needs to be compensated for his invention if someone else uses it. The idea that a patent does not grant the inventor the right to build something may be difficult to understand. Let’s look at this hypothetical. Assume that you have an invention for a new innovative chair. The chair has four legs, a seat, a back and armrests. But your invention is the tilt of that chair. Your innovation is not so much in the chair but in the mechanism of how the chair tilts. So in order to build a chair that tilts one would also need to have the rights to build a chair. Those patents to build a chair may be owned by someone else. The fact that you have a patent on the tilting mechanism does not give you all the rights to build chairs. While this is a rather simple example for most people, it is important to see this in the context of the complexity of modern products. The vast majority of patents are improvements to others’ ideas. It has been said that inventors stand on the shoulders of giants. In other words, to create the next brilliant invention, one often needs many previous inventions in order to make the product. The first microprocessor needed the basic invention of the transistor and it also needed the first patent of an integrated circuit. In fact, there was significant litigation between Intel and Texas Instruments as to who actually invented the first integrated circuit. So, if you can’t build something with a patent, then what can you do with the patent? The short answer is you can stop someone and/or have someone pay you for the rights to use that invention. Frequently, startups wonder if they should get patents or not. Sometimes they think that their new products are so unique that there is no competition and other times it is that they just don’t want to worry about the cost. Most startups should begin a patenting program sooner rather than later, obtaining a patent should be part of everyone’s business plan not because it grants you a right but because of the negative right that is granted to you. One of the things that often happens with patents is that patents are already licensed. In a startup situation there are frequently other competitors who might not be direct but more tangential competitors. They don’t offer the exact same product but similar products that use similar methods. Having a number of patents that relate to your technology area is critical when it comes to licensing. If you are able to get patents that somebody else uses and, in particular, that someone else is a person who wants you to pay a license fee, you have the ability to trade licenses. This is called “cross-licensing” in the industry. What it means is that I will license you my patents if you will license me your patents. Most licensing professionals believe these are the most numerous types of licenses that occur in industry today. While the number of cross licenses agreements may be fewer than the number of total licenses agreements, the number of patents covered by a cross license are significantly greater than the number of patents covered by noncross licenses. Actually, many cross licenses in the electronics industry cover the entire patent portfolio of the companies. Most major companies require a license back if they are going to grant the license. The reason is that at the point of the negotiation license they have the maximum leverage. What no one wants to do is to go back to their boss and say “by the way I just licensed our patents to company X” but now company X is suing us for patent infringement. This would be a career-limiting situation. When talking about licensing there are typically two types of licenses that most companies grant. They are either nonexclusive licenses or exclusive licenses. A nonexclusive license can be thought of as

What is a Patent?

granting someone the right to use your invention but you are also telling the person that you will grant others the same rights. To put this in a property context imagine that you have a plot of land. You may grant people the right to travel over your land to get to the beach on the other side. You may even decide to charge them for that right. But you will grant that right to many different people. You may also grant the right to many different people at the same time. Imagine a town that owns the beach and charges people a certain amount of money to park near the beach. This is a nonexclusive right to use the beach parking facilities. The other option is an exclusive license. With an exclusive license one only grants the rights to use the invention to one person or company. That entity has the “exclusive” rights to use the invention. Back to the example: If you wanted to be the only person to have rights to park at the beach and no one else could park at the beach, you would then have “exclusive rights.” While this may sound excessive in our example, there are examples in the property industry were “exclusive rights” are commonly found. For example, if your property has certain natural resources on it, minerals from mining or timber for harvesting then you may grant an entity “exclusive rights” to mine the minerals or harvest the trees. In a technology setting exclusive rights frequently depend on the specific technology. Exclusive licenses are somewhat rare in the high-tech industries of computer science and electronics. Many of these products require tens of thousands or hundreds of thousands of patents to work and, therefore, granting an exclusive license is not necessarily beneficial. In other industries like pharmaceuticals, where a specific formulation of the drug might be covered by one or two patents, exclusive licensing is much more common. Other than licensing, what else can one do with a patent? The other two options are: selling the patent or abandoning the patent. To abandon the patent, not paying the maintenance fees that are due on the patent accomplishes this. Many large companies in the electronics industry routinely abandon a large number of their patents. These companies will periodically look at their portfolio to see if the patents they are keeping in force are still useful. Patents that are neither in the current marketplace nor relevant to the potential intellectual property competitors should be considered for sale or abandonment. Frequently, because of the market dynamics of the company and who the company’s customers are, it is not feasible for the company to assert its patents against one of its customers. Since the company has developed a valuable portfolio and one of its customer’s businesses is not relevant to its business, perhaps there is somebody else who may want to buy the patent. It would make sense at this time to test the market and see who needs those patents. Often companies will buy patents because they are entering into new markets where they feel they have inadequate patent coverage. Oftentimes companies need to buy patents when they have few patents of their own but have very successful products. Recent history abounds with companies that fall into that category: Cisco, Broadcom, Google, HTC and Xiaomi have all faced this issue. In each case, the very successful product company had not developed a strong enough patent portfolio quickly enough to fend off those who thought they were due license fees. This led to litigation and a need for these companies to buy patents as well as to file more patents quickly. Since filing a patent takes a long period of time it was very difficult for these companies to gain protection from an internally generated patent portfolio if the patents took between three and five years to issue. Additionally, when generating a new patent portfolio the inventions that are generated often do not become commercially acceptable until many years after the patent issues.

81

82

Leading Individuals and Engineering Project Teams

7 Leading Individuals and Engineering Project Teams Donna Brazil United States Military Academy

Darcy L. Schnack United States Military Academy

83

Engineering Management Handbook

7.1 Introduction As an engineer chances are very good that you were hired for your technical skills and expertise. You studied hard in school, you are technically savvy, and you certainly know your area of expertise. You worked your way up in the organization and now they want you to lead an engineering project team. Initially, you were excited about this promotion and new challenge, but now after just a few meetings you are beginning to think twice about that excitement. As you think back to your preparation, you suddenly realize that while you were promoted because you were a good engineer, no one really trained you to lead. A great definition of leadership is that “Leadership is providing purpose, direction and motivation while operating to accomplish the mission and improve the organization” (U.S. Army, 2012). It gives the leader a starting point – providing purpose, direction and motivation as well as dual end states – to accomplish the mission and improve the organization. Let’s look at the starting points first and return to the end states later. This chapter will begin by introducing the basics of a systematic way to approach leadership. We will start with understanding yourself as a leader and then look at the characteristics of followers, motivational issues that might arise when leading individuals, and some issues that are common to all groups. We will then turn to theories of leadership that take the situation and the task into consideration.

7.2 The Leader

84

Before you can expect to lead anyone else, it is important to assess where you yourself are as a leader. Commonly referred to as self-awareness, this can be seen as broad individual self assessment. Quite simply, becoming self-aware involves developing a clear picture of yourself through self-assessment, peer and superior feedback, and formal and informal 360-degree assessment tools. Personality inventories, critical thinking tests, and emotional intelligence quizzes are all examples of self-awareness tools that are available online and can be used as part of this assessment process. These tools can make you aware of your preferences, leadership and learning styles, biases, character strengths, and much more. Through the self-assessment process, you can begin to determine what you see as your strengths and weaknesses and what areas you would like to develop more fully. One way to think about self-awareness is to imagine yourself as a combination of a number of possible “selves” (Markus and Wurf, 1987). The makeup of these selves comes from a number of different sources. Who do you think you are? Who does your boss think you are? Who do your subordinates think you are? Who does your family think you are? A final self you should consider is what Markus and Wurf calls your ideal self – the person you would like to be – what characteristics would you like to own and display as a leader? Through this self-assessment process, you can piece together this information to determine where you are in terms of your current self and where you would like to be in terms of your ideal or desired self. In this way you can identify strengths you would like to build on and gaps that you would like to address. You can then develop a personal action plan for working to close those gaps. Often this plan will include a systematic attempt to use new behaviors and an assessment to determine if they fit better with who you would like to be as a leader. Perhaps through your self-assessment process you discover that although you believe that you are approachable and open to new ideas, your subordinates and friends tell you otherwise and list occasions where they have felt that their suggestions have been ignored. You might work hard to listen to those around you and incorporate their ideas into your next project. Such attempts, if successful can help to boost your self-confidence and desire for further growth and development. Just as taking the time to get a complete picture of yourself is important before you set out to lead your team; this is also a good time to take stock of what biases and preferences you might bring to the table as a leader. Each of us has a history and a comfort zone of behaviors that we will fall back on as a default. Identifying these preferences so that we are aware of them is critical to approaching situations with an open mind. One such preference that is very common is for people to be more comfortable with others who are similar to them. This similarity can be across a number of dimensions including age, race, gender, and technical specialty. It is important that you think about this simple preference as a potential area of bias and what this might mean to your work situations.

Leading Individuals and Engineering Project Teams

A fascinating tool for discovering your preferences is the Implicit Associations Test (IAT) developed by researchers at Yale, Harvard and other institutions. Available at https://implicit.harvard.edu/implicit/ This tool offers you the opportunity to measure the strengths of your preferences on a number of dimensions from gender and race to religious and political affiliation as well as current social issues such as self esteem, anxiety and mental health (Project Implicit, 2011). The results do not predict behavior but they might provide you with personal insight into your preferences. Keep in mind that having a preference or bias is not the same as acting on that bias, but having thought about them beforehand is critical to minimizing the potential impact. Acknowledging these preferences upfront will make you more self aware and better prepared to manage and leverage diversity on your team and to try new behaviors in this new situation. Some key self-awareness issues to think about from the perspective of a leader: What are my predispositions toward: • People of other races? • People of the opposite gender? • People of a different sexual orientation? • People from a different academic or technical discipline? • • • • •

What is my preferred method of interacting with others? Do I tend to keep my ideas to myself? Do I prefer to work alone? Do I tend to delegate everything? Do I tend to micromanage? Do I seek consensus on decisions?

• • • •

What are my beliefs about people in general? Do I believe that people are inherently good? Do I believe that people are inherently bad? Do I believe that I must motivate people through rewards and punishments? Do I believe that the best way to have something done right is to give very explicit directions?

As you can see from this short list, there are many facets to being a leader and it is helpful to take stock of yourself and your preferences and determine if they are the best fit for your particular situation. If they are not, you might consider how you can develop your capabilities in these particular areas. Taking time upfront to consider these issues before you are under a deadline or in the middle of a crisis will certainly be time well spent.

7.3 Leading Individuals Just as each leader brings to the table different skills, attributes and needs, so does each individual member of the group. As a team leader it is your job to bring together this collection of individual skills and attributes to work toward a common task. A great place to start is in understanding your team members, their strengths, their weaknesses and what each of them can and seek to contribute toward the goal. No two people are exactly the same and the skills and desires of two people are the same either. Many of the predispositions and preferences that you evaluated within yourself will be present in your team members. Asking them what leader and management style they prefer would be a great first step in making sure you understand and meet their needs and desires. As identified in the definition of leadership at the start of this chapter, a significant part of leadership is providing purpose, direction and motivation to accomplish a task. For engineering management team members, purpose and direction are easy – skilled technicians are good at that. Providing motivation can be a bit trickier. There are many theories of motivation that can assist you in leading your team and each of them requires that you as the leader look at each individual in order to determine if they are motivated or not and if they lack motivation, then to determine what exactly is it that they need. 85

Engineering Management Handbook

86

The common way to look at motivation is to focus separately on intrinsic and extrinsic motivation; intrinsic motivation comes from within the individual or task itself while extrinsic comes from an external force. We will come back to intrinsic motivation later, but let’s start with extrinsic. Extrinsic motivation is an external force, normally a reward or threat of a punishment that compels an individual to accomplish a task. The more they want the reward (or want to avoid the punishment), the more motivated they are work to accomplish the task. Sounds simple, right? So why doesn’t it always work? Researchers have discovered a number of explanations that are worth your consideration as you begin to diagnose the motivation issues of your team members. Stacy Adams (1965) explained a lack of motivation through the lens of fairness or equity. He suggested that individuals seek to get their “fair share” out of most situations. In a simple explanation, equity theory suggests that group members look at their inputs (work, effort) and their outcomes (compensation, praise, self-worth) and determine if their ratio is equal to or better than the other members of the team. If they feel that their input/outcome ratio is lower than their peers’ then they will determine that the situation is inequitable and will work to resolve that inequity (Adams, 1965). When team members think that they are being treated unfairly in a situation, they might lower their effort or in extreme situations, they might quit altogether. As the team leader, you can work with the members and try to see the situation as they see it. Sometimes they will identify a valid inequity in the system that you were not aware of. Often, you will find that you and the team members have different perceptions about the quality and quantity of both inputs and outcomes and all that is required is talking about the differences so that you all have a clearer picture of the situation. The bottom line in these situations is to understand that perceptions matter, and addressing the perceptions of your team members will make them feel valued and will in turn, increase their motivation. Another theory that explains possible sources of motivation issues is expectancy theory (Mitchell, 1974). Based on expectancy theory, individuals are motivated to perform a specific behavior to a specified standard in order to receive a specific reward. This framework is helpful in understanding motivation as it enables a leader to break a situation down into three components: the specific individual behavior that you want the member to perform, the standard that you want the performance to reach and the reward or outcome that the team member will receive for meeting this standard. Motivation therefore comes into play in any one of three ways: the team member might lack the confidence that he or she can perform the task to the established standard; the team member might not trust that he or she will actually get the promised reward or finally that the team member might not value the promised outcome or he or she believes that the energy expenditure required to meet the standard is not worth the outcome. Using this framework as you approach the situation of a seemingly unmotivated subordinate, you might first ask yourself a few questions. What is the task I want completed? What standard have I established for this task? What will the team member get for his or her effort? Since a motivation issue can involve any one or more of these components, you should then talk to the team member to determine which linkage is weak. Your leader action will follow directly from the weak link. If the team member lacks the confidence to meet the standard, you might break the task down into smaller subtasks or you might ensure that the team member gets additional training, or if necessary, you might revaluate the standard. If the team member doesn’t believe that you or the organization will provide the promised outcome, you must work on trust issues and ensure the team member that you will in fact make good on your promises. Most trust issues are based on past experience so you will need to convince the team member that you are different than their past leaders and that you will see to it that the reward is realized. Finally, if the team member does not value the reward or thinks that it is not worth the effort, you can discuss what they would like as the reward and try to come to a compromise that will suite you both. Equity and expectancy are two ways to look at extrinsic motivation. While most of us desire and indeed need compensation for our work, many professionals stay in their particular field for intrinsic reasons. Certainly, we could earn more money, have a bigger house or expense accounts; but we like what we do, it gives us meaning. When these team members lack motivation, another way to diagnose the situation is to through the components of intrinsic motivation (Deci and Ryan, 1985). In terms of Deci and Ryan’s theory, individuals who are motivated intrinsically are thought to find completing the task

Leading Individuals and Engineering Project Teams

enjoyable in and of itself. Intrinsically motivated individuals enjoy tackling a new task and finding a new way to solve the challenges. They are motivated by being given the opportunity to challenge themselves and the freedom to determine their own course of action. Understanding these conditions will help you as the team leader of intrinsically motivated members. These members enjoy taking on and accomplishing new challenges and being allowed the freedom to figure it out. These individuals find the challenge rewarding and might feel manipulated when offered a reward as an incentive for completion. Any rewards for these individuals should be given after completion of the task and will serve as information – namely that their performance was exceptional and appreciated. You are probably thinking that intrinsic motivation and extrinsic motivation are at odds with each other. In some ways, you are correct and this highlights the importance of your initial diagnosis of the situation and your team members. Promising a reward to one person will assist in raising motivation, while for another, it could make him or her feel manipulated and therefore lower motivation. According to expectancy theory, challenging a team member who feels uncertain about his or her performance might overwhelm and cause a drop in motivation while an intrinsically motivated team member will excel when challenged. Obviously, the key to selecting the best actions as a leader start with truly understanding your team members. It is important to take the time to know what their strengths and weaknesses are as well as what they value and expect from you as their leader. Remember, just as you are different than every past leader, your team members are different from each other. Each comes with a history, with varying levels of experience and with different personal needs. The better you understand them, the better you will be able to bring them together as a team.

7.4 Leading Teams

Just as humans develop in fairly predictable patterns and go through developmental stages, so do groups and teams. Understanding Tuckman’s (1965) model of group development and reflecting on the actions that a leader can take to facilitate a team’s progression through the stages of development can be invaluable to you as a leader. The steps of group development include, forming, storming, norming, performing and adjourning. Let’s take a closer look at each step.

7.4.1 Forming In this first stage of group development, members are more concerned about personal characteristics and issues than they are about actually performing the task. This is a time when group members attempt to figure out how they fit in, who has status, what skills everyone has and so forth. Many leaders come to the first team meeting with a full agenda planning to jump right into a meaty discussion of the project at hand. A leader who understands group development and who has thought through the “first encounter” will facilitate the needs of his or her team by allowing some time up front for the informal interaction that the team needs. You can also assist the group in working through this stage of development by clarifying roles and facilitating their desire to share information among each other.

7.4.2 Storming As groups start to get to know each other and work together, it is normal for them to enter a period of conflict. In this stage, members are assigned to roles and differences in status begin to emerge. Some members might believe that they are better qualified to perform a task than the member who was assigned to it; others might be concerned about where they fit in. Informal leaders might begin to build alliances with other members in order to solidify their position in the group. While it can appear chaotic and messy, this is an important time in the development of a group. You should not think that you are failing because the group begins this period of conflict; on the contrary, most teams that never go through this stage have probably never really been tested and when they do hit rough times, the issues that were never resolved will come to the surface. You can facilitate the group’s movement through this stage by working to build consensus, investing the time to assist with role clarity and negotiation, and openly discussing 87

Engineering Management Handbook

conflict issues. If you can work to eliminate or minimize these sources of conflict, you can facilitate the group’s movement to the norming stage.

7.4.3 Norming Groups really start to click during the norming stage. They have achieved consensus on how tasks should be accomplished, they have established norms that are self-enforced and individual status has been identified and agreed to. There may be some cycling back through storming and norming as new members or tasks are added but as the norms solidify, the cycles become shorter and the smoother. In this stage you as the leader can start to focus more on the social and emotional needs of group members and you might serve the group best by keeping them focused on the task by ensuring their other individual needs are met. Leaders of groups at this stage need only give guidance and allow the systems that the group has developed to take over. That said, there might be a tendency in this stage to allow the “systems” to substitute for thinking and decision making. You must be attentive to rigid adherence to a system that has been agreed on but is no longer functional. You can assist the team’s development by playing or appointing a devil’s advocate, encouraging deviance from time to time and sharing the leadership decisions of the group. Groups who successfully negotiate this stage move on to the performing stage.

7.4.4 Performing In this stage, group members are very comfortable with themselves and each other. They are interdependent and creative. They are not wedded to any particular structure or system and instead are expected to suggest improvements and point out potential problems. As the leader you are able to fade into the background and are completely free to work strategic issues that might facilitate the group’s current task or future task. By now you are probably thinking to yourself “well that’s a great theory, but my REAL teams never get that far.” Well, take comfort in knowing that you are not alone in your experience. Despite the name of the “performing” stage, groups will perform – and if forced to, will produce something– no matter what stage they are in. In fact, some groups never get out of the storming stage as they continue to go about their task.

7.4.5 Adjourning Few groups exist in perpetuity. Recognizing this and reviewing subsequent literature, Tuckman and Jensen (1977) updated an earlier model of group development to address the feelings of separation a group experiences as it disbands. Members may feel a sense of loss for group relationships or distress over accomplishment of group goals. As a leader, it is important to focus on communication within the group and to prepare for this stage if the group’s dissolution is anticipated. Take the time to celebrate the group’s accomplishments and the teamwork that you have all shared. As a leader, understanding the normal stages of development and the issues faced by group members at each stage can assist you in facilitating the needs of the group. Quite simply, a group performs better at a higher stage of development. If you are able to recognize the issues and move your group forward, all the better. Having discussed the importance of understanding yourself as a leader and the individual and group characteristics of the followers, it is now time to turn our attention to leading.

7.5 Leading

88

The issues and discussions above are all absolutely critical functions of a leader and are almost prerequisites to being able to enact the traditional theories of leadership. For these theories to be effective, a foundational understanding of yourself and your team members is essential. Perhaps the most useful leadership theories for an engineering management team leader are contingency theories that take into consideration both the characteristics of the team member and the characteristics of the situation. These theories, developed in the early 1970s focus on understanding the level of structure of the task and the team members’ skill with regard to the task as well as their motivation

Leading Individuals and Engineering Project Teams

toward completing the task. The most well known of these theories is Hersey and Blanchard’s Situational Leadership Theory (Blanchard, Zigarmi and Zigarmi, 1985). This theory requires that the leader first understand both the task requirements and the development level of his or her team members and then, taking these into consideration apply the appropriate amount of task and relationship behaviors. Leader behaviors that are high on task direction and low on building relationships are appropriate for team members who are low in competence but high in their commitment to task completion. As team members become more competent and perhaps begin to waiver in their commitment to the task, the leader must begin to ease off on providing direction and begin to focus more on building supportive relationships. At the far extreme, team members who are extremely competent and highly motivated need little direction or external motivation. While the tenants of Situational Leadership Theory might seem simplistic and intuitive, many of us can recall a leader who failed to recognize that we were at either extreme and insisted on providing inappropriate leadership at the critical moment. Either the leader continued to be very direct despite your skill development or the leader took a delegating approach when you really did not have all of the required skills. Either mismatch can have serious negative repercussions for both the team member and the team. According to Situational Leadership Theory, as the team members’ progress from low skill with high motivation through varying skill with varying motivation to high skill with high motivation, the leader should at first be directive, then take on a coaching role, move into a fading supportive role and finally realize that it is time to delegate. Until this point in the chapter we have been concerned primarily with the first part of the definition of leadership – namely providing purpose, direction and motivation in order to accomplish the mission. While certainly required of a good leader, your job is not yet complete. We must now turn our attention to the last end state of that definition: improving the organization. While you personally might have felt unprepared for your first leadership position, it does not (and should not) have to be the same for your team members. Part of your job as a leader should be to develop your replacement.

7.6 A Leadership Development Model Many Engineering Leader Development Programs (ELDP) focus on a specific skill set or competency that will enable a young engineer to better negotiate a particular new position or a specific new challenge. While these programs might be effective in the short term, very few models incorporate development across a broad spectrum of situations and developmental levels. What follows is a discussion of one such model that can and should span an entire career. This discussion is based on the leader development framework developed by the Center for Creative Leadership (CCL) (McCauley and VanVelsor, 2004). The foundation of this model was derived from the result of hundreds of developmental sessions with executives, educators, business managers, and military officers. Their work provides a skeleton for understanding the impact of various developmental programs and for beginning to integrate these initiatives into a coherent whole. The CCL defines leader development as the “expansion of a person’s capacity to be effective in leadership roles and processes” (McCauley and VanVelsor, 2004, p. 2). This definition is especially applicable to team members in technical fields because the emphasis is on the individual. It seeks to increase capacity, not meet a preestablished set point and it acknowledges that there are many leadership roles along the route from follower to chief executive officer. The expansion of a person’s capacity can relate to any type of development, from technical expertise to leadership skills. The framework is quite simple and is based primarily on three components: assessment, challenge and support.

7.6.1 Assessment The assessment component calls for leaders to become self-aware. As your team members begin to take on more responsibility and prepare to become leaders themselves, they should seek out feedback from others and should begin to assess themselves in terms of their strengths, weaknesses, preferences and predispositions. The model calls for the same introspection and self-assessment that we looked at in the beginning of this chapter. As their leader, you will be in a unique position to provide feedback help to identify

89

Engineering Management Handbook

developmental gaps that they should address. If possible, gathering feedback from peers and subordinates will round out the assessment picture and give the individual a 360-degree assessment.

7.6.2 Challenge A second key component of leader development is challenging experiences. Your engineers must be directed or encouraged to undertake experiences that will challenge them and push them out of their comfort zone, experiences from which they will grow and develop. Runners will never get faster if they only run at a comfortable pace. They may stay in shape and maintain good cardio fitness, but they will never get faster if they do not push and challenge their ability by going faster and faster on training runs. The same principle holds true for leadership. Engineers will never improve their interpersonal, communication, managerial, etc., skills if they stick to other aspects of the project that they feel comfortable with. For example, engineers stick to the design and the production, and never accepts the challenge to speak with the other group members or the customer, they will never grow in that area. Too often, engineers remain in their technical comfort zone and do not cultivate other elements of the “whole” leader. Individuals develop by taking on stretch assignments, situations, and experiences that offer them a challenge outside of their comfort zone. These are not assignments that are completely outside their area of expertise – a runner does not attempt to be a better runner by learning to scuba dive – nor would we necessarily want an engineer student to attempt the stretch assignment of public relations. The challenging experiences should be based on what you or the individual determined to be a gap in their development during the assessment phase. Truly challenging experiences make individuals uncomfortable and create a disequilibrium that they must work to resolve. They are forced to develop and try new skills when their tried-and-true favorites do not work. This is true for all areas of development and especially true for leadership development. Leaders in technical organizations and management teams must be encouraged and rewarded for seeking out challenging leadership experiences.

7.6.3 Support Support comes in many forms. Universities, corporations, etc., must recognize the need for this development and allow their engineers the time and resources required. In academics, this may require restructuring graded requirements in preexisting courses or developing entirely new course goals and objectives. In corporations, this might involve formal mentoring, rotational training, professional coaching, and professional development activities. Another form of support comes from those surrounding the leader. Leaders must have a person or group of individuals that they can turn to in order to help them make sense of the experiences they have had and the feedback they have received. Far too often young engineers live through challenging experiences and simply throw it them into their files, never to be seen or evaluated again. These valuable experiences can be powerful but without reflection, no growth takes place and, as a result they are not significant in their development. The real promise for growth and development is in the processing of that experience, either alone or with the help of a trusted friend, peer or mentor. In these after- action reviews that look at the engineer’s actions, inactions, decisions and interactions are rich learning opportunities that will expand their capacity to do the same or better the next time they are presented with a similar issue. In our fast-paced, just-in-time culture, it is often difficult to take time out to reflect on our experiences and seize the developmental opportunity. As a team leader you must recognize this tendency and purposefully set aside time and resources to enable and assist your team members in making the best of each growth opportunity. Along with support must come the freedom to fail. Teachers, peers, mentors, coaches, and superiors must understand that not all challenging experiences will be met with complete success. What truly matters from a developmental perspective are the lessons that the individual takes away from the experience and his or her ability to own that experience and the lessons learned. As a team leader, you must set up learning and developmental opportunities for your team members and provide them the support to make all of their outcomes opportunities for growth. 90

Leading Individuals and Engineering Project Teams

7.7 Closing Thoughts When striving to improve the organization and the members of the organization, another leadership theory comes to mind as very applicable. Much of the discussion above focuses on the leader applying the appropriate leader action that the follower needs for a specific situation. Transformational Leadership Theory (TLT), on the other hand, focuses on the relationship between the leader and the team members as well as the long-term development of the follower. TLT was first identified by the political historian James M. Burns (1978) who sought to explain the power behind great leaders who seemed to transform their followers in support of a cause. Social scientists built on this original idea and have attempted to isolate the behaviors of these leaders as well as the outcomes that are engendered in the followers. Bass and Avolio (1994) suggested that leaders who have a vision of where they want their team to be, communicate high expectations and confidence in their team members, demonstrate individualized concern and use innovative and unconventional strategies when interacting with their followers develop team members who in turn are inspired, empowered and motivated to do more than they thought they could. Transformational leaders still manage the details and still seek to know themselves and their subordinates. They will occasionally deal with feelings of inequity and problems with motivation but over time the as they live and demonstrate the behaviors mentioned above, their team members identify with and internalize the assignment, the cause, or the task at hand and the leader can fade into the background.

7.8 References Adams, J. S., “Inequity in Social Exchange,” In L. Berkowitz, Advances in Experimental and Social Psychology (pp. 276-299), New York: Academic Press, 1965. Blanchard, K. Z., Zigarmi, P., and Zigarmi, D., Leadership and the One Minute Manager: Increasing Effectiveness Through Situational Leadership, New York: William Morrow, 1985. Bass, B. M. and Avolio, B. J., (eds.), Improving Organizational Effectiveness Through Transformational Leadership, Thousand Oaks, CA: Sage, 1994. Burns, J. M., Leadership, New York: Harper & Row, 1978. Deci, E., and Ryan, R. M., Intrinsic Motivation and Self-Determination in Human Behavior, New York: Plenum Press, 1985. Goleman, D., Boyatzis, R., and McKee, A., Primal Leadership: Realizing the Power of Emotional Intelligence, Boston: Harvard Business School Press, 2002. Project Implicit, Implicit Associations Test, 2011, Retrieved September 8, 2015, from Project Implicit: https://implicit.harvard.edu/implicit/ Markus, H. and Wurf, E., “The Dynamic Self Concept: A Social Psychological Perspective,” Annual Review of Psychology, vol. 38, 1987, pp. 299-337. McCauley, C. D., and VanVelsor, E. (eds.), Handbook of Leadership Development, San Francisco: Josey Bass, 2004. Mitchell, T. ,“Expectancy Models of Job Satisfaction, Occupational Preference and Effort: A Theoretical, Methodical, and Empirical Appraisal,” Psychological Bulletin, vol. 81, 1974, pp. 1096-1112. Tuckman, B., “Developmental Sequence in Small Groups.” Psychological Bulletin, vol. 63, 1965, pp. 384399. Tuckman, B. and Jensen, M., “Stages of Small-Group Development Revisited,” Group & Organization Management, vol. 2, 1977, pp. 419-427. US Army, Army Doctrinal Publication (ADP) 6-22, Washington, DC: US Army, 2012.

91

92

Managing the Multi-Generational Knowledge Based Workforce

8 Managing the Multi-Generational Knowledge Based Workforce Gene Dixon East Carolina University

93

Engineering Management Handbook

8.1 Introduction In order to create a productive work environment, generational gaps have to be approached in a way that benefits employers and employees. To have a positive affect with the differences in behaviors between the three majority generations currently in the workplace, a general understanding of generational difference is needed.

8.1.1 Overview Employers and employees of all ages must work proactively across all generations to create an effective work dynamic. Engineering managers and engineers can benefit from knowing how generational norms impact the work dynamic. Management’s understanding of the diversity of values and beliefs of generations will facilitate effective management and create a productive workforce. Silver (2011) described diversity as a value of different perspectives. For some countries, the multi-generational workforce reflects a range of employee age that has not previously been experienced, a demographic remix. In America, the generational remix means that (Silver, 2011): • Currently, there may not have enough workers to take care of older Americans • By 2023, minorities will comprise half of all children; 62% by 2050 • By 2030, 1 in 5 Americans will be over 65 • By 2042, there will no longer a majority race • By 2050, the Hispanic population is expected to triple • By 2050 the 18 – 64 age workforce will decline from 63% to 57% • • • •

From a broader perspective, Silver (2011) reported that: Educational levels are trending down More children are being born to unwed mothers Marriage rates among young adults (25-34) is in decline Multi-generational households are increasing

All of these trends are expected to have some impact on emerging generational norms that will also impact the workplace.

8.2 Generations

94

Research studies do not draw arbitrary and abrupt lines between generations (Jorgensen, 2003). For convenience, generational age brackets are identified to support the research agenda. Jorgensen (2003) marked generations by particular historical events while Shaw and Fairhurst (2008) defined a generation as starting with an increase in the birth rate and ending with a birth rate decline. Birth rates often trend with societal or historical shifts. Events are not momentary; history unfolds over periods of time and therefore the definitions have considerable overlap. In a sense then, a generation is a demographic cross-section that possesses commonality related to defining social or historical events. Common life experiences are theorized to create commonalities of perspectives, attitudes, and assumptions within a generation (Blythe et al., 2008). Generational groups develop distinct values and workforce patterns according to Blythe et al. (2008). Common generational values are attributed to generations: Baby Boomers, those born between 1946 and 1964; Generation X, born between 1965 and 1979; and, Generation Y also known as millennials were born after 1980 (Keepnews, Brewer, Kovner and Shin et al., 2010). As Baby Boomers age and move out of the workplace, Generation X progress through the work hierarchy, and the Generation Y/millienalists enter into the workforce. Increasingly, engineering managers find themselves addressing the values and patterns of a multigenerational work environment. This environment requires an understanding of generational differences in order for the workplace to remain attractive to employees. The work environment preferences of the various generations and the impacts on motivation, productivity, and other basic workplace cultural and structural pediments must be understood and leveraged. Engineering managers are often responsible for

Managing the Multi-Generational Knowledge Based Workforce

creating a productive environment, productive processes, and supporting systems that stimulate employees of all generations to high performance. The organization’s processes and systems provide a framework for building loyalty and commitment (Dixon and Knowles, 2013). Specific generations dominate the workplace. They are the Baby Boomer Generation, Generation X, and Generation Y. Each of these generations display distinct characteristics and make significant positive contributions toward a global economic landscape.

8.2.1 Baby Boomers The “Baby Boomers” are a generation defined by events such as the Vietnam War, the assignations of President John Kennedy and civil rights advocate Martin Luther King Jr., and the sexual revolution (Adams, 2000). The toll of these events on the Baby Boomer population contributes to their lack of respect for authority (Dixon and Mercado, 2011). Baber Boomers tend to be optimistic, as if the planet was theirs, and have a great sense of teamwork (Zemke, Raines, and Filipczak, 1999). They are dedicated to their work and are sometimes called workaholics due to their dedication to their jobs (Keepnews et al., 2010). Baby Boomers are entering the twilights of their careers and are retiring, moving into roles of corporate leadership or pursuing philanthropic endeavors in their communities (Dixon and Mercado, 2011). Baby Boomers will continue to influence the workplace through 2020. A summary of some general characteristics of Baby Boomers’ behaviors follows (Dixon and Mercado, 2011): • Willing to invest themselves in, and serve, the organization, • Need to distinguish themselves from peers, • Will not confront issues directly, • Resistant to change, and • Team players who believe in lifetime employment.

8.2.2 Generation X (Gen X’ers) Generation X experienced the oil crisis of the 1970s, the stock market crash of the 1980s, and the effects of those historical events on their family life during their formative years. The economic impacts of these events saw them face the experiences of their parents losing jobs, relocation, or lower incomes. This has created a norm where Gen X’ers are characterized as working to live versus living to work, a baby boomer/ parent norm (Dixon and Mercado, 2011). Being motivated drives their lifestyles (Loomis, 2000) and they look for career opportunities offering a collaborative work environment. Gen X’ers want challenging opportunities with flexibility and recognition (Loomis, 2000). Members of Generation X may skeptical, independent, and energetic, with less loyalty than Baby Boomers (Ansoorian, Good, and Samuelson, 2003). Jobs are viewed as openings for competency building, i.e., they value opportunities for learning and training over loyalty and pensions (Ansoorian, Good, and Samuelson, 2003; Bova and Croft, 2001). Hoerr (2007) stated that Gen X’ers respond well to change, are not intimidated by authority, and are less bound by structure and hierarchy than previous generations. A summary of some general characteristics of Gen X’ers (Dixon and Mercado, 2011): • Use teams to support individual efforts and relationships • Relationships take precedence over careers • Likely to challenge but expect friendly work relationships • Education is a necessary tool • Comfortable with diversity and change • Also known as Buster or Me Generation or latchkey kids • First of the technologically adept generations • Individualistic with a casual disdain for authority • Dislike being micromanaged • Value work/home balance • Emergence of creative class 95

Engineering Management Handbook

8.2.3 Generation Y (Gen Y) The Generation Y, or the “Millennial Generation,” has experienced historical events including the Persian Gulf wars and the 9/11 attacks. This generation has parents that are still involved in Millennials’ lives and who are referred to as “helicopter parents” as they are want to drop in and rescue their millennial at any time (Coley, 2009). Millennials have grown up with an electronic and wireless network of computers and smart phones, which they use for texting and personal networking via social media. Their linking behavior is expected to be permitted in the workplace. Millennials see education as a commodity that includes limitless options (Merritt and Neville, 2003). This generation values personal connections, want to know all about their contacts, and don’t mind revealing information about themselves (Coley, 2009). Millennials desire intensive support, and expect value-added experiences, clear investment outcomes, and diversity within their environments (Coley, 2009). Brown (2011) stated that a work-life balance is the Millennial’s primary concern. A summary of some general characteristics of Millennials include (Dixon and Mercado, 2011): • Accustomed to working in teams; will assume responsibility for team • Want clear direction; must be challenged • Value honesty • Embrace transformation • Will leave if not challenged or supported by their work environment • Will sacrifice personal security for attention • Needs constant feedback and attention • Not to be forced, they live on choice • Move from job to job • Achievement-oriented, team-oriented • More radically and culturally tolerant than previous generations • Prefers urban lifestyle; place matters, not just job • Environmentally conscience

8.3 Management Impacts Generational impacts on management systems and styles are trending toward the frontlines of management literature. In this section, the characteristics of the three generations are briefly examined.

8.3.1 Baby Boomers Dixon and Knowles (2013) studied Baby Boomer workplace tendencies in regards to loyalty and followership and found that they tended to be loyal to their employers while demonstrated the behaviors characterized by followers (Chaleff, 2009). Boomers are recognized for their positive attitudes toward work and their abilities in building consensus, mentoring, and effecting change (Smola and Sutton, 2002). McGuire, By and Hutchings (2007) reported on research showing that Baby Boomers have relative high productivity relative to their experience, organizational commitment and stability. Gibson et al. (2010) found Baby Boomers to be comfortable with change, loyal, security oriented, workaholic, and idealistic even to the point of allowing work life to come before family life (Keepnews et al., 2010). Sixty-six percent plan to remain active in the work place following retirement (Ansoorian, Good, and Samuelson, 2003).

8.3.2 Generation X Gen X’ers have high value for professionalism (Blythe et al., 2008), yet tend to be cynical and untrusting (Ansoorian et al., 2003). Gen X’ers entered the workforce during the popularity of workforce reengineering and organizational restructuring. As a result Generation X does not expect organizational stability and demonstrate a high tolerance for career risk (Blythe et al., 2008). Generation X may lack a sense of traditions and demonstrate a sense of individualism (Jurkiewicz and Brown, 1998) tempered with support from their network of colleagues (Kuperschmidt, 2000; Karp, Sirias, and Arnold, 1999). Gen X’ers have a 96

Managing the Multi-Generational Knowledge Based Workforce

need to be mentored (Jurkiewicz and Brown, 1998) and want immediate feedback. They bring practical approaches to problem solving and expect employers to listen, provide a facilitating culture, and pay fairly. Organizations that provide opportunities for improving knowledge, skills and work attitudes enable Gen X’ers mobility (Ansoorian et al., 2003; Chen and Choi, 2008). McGuire et al. (2007), reported that Generation X values working for themselves and capitalizing on employment opportunities.

8.3.3 Generation Y As the most technologically literate of the workforce (Blythe et al., 2008), and want their work to be meaningful and have opportunity to contribute to a higher purpose. Hira (2007) identified the Y-Gen as high maintenance needing supervision and feedback. Gen Y are capable multitaskers (Shaw and Fairhurst, 2008) seeking employment where they can experience: a fun environment, growth opportunities, a variety of work projects, chances to learn new skills, and flexible schedules that support of a balanced work-life (Kuperschmidt, 2000; Carver and Candela, 2008). The Gen Y is accustomed to teamwork and desires supervision and structure. They have an affinity for sustainability. If not challenged and supported, they will job hop (Carver and Candela, 2008). Retirement benefits are important in their job choices (McGuire et al., 2007). Retention is a function of commitment (Dixon, Mecado, and Knowles, 2013). In the next section, the correlation of commitment and generational influences are discussed. When employees are committed, turnover is reduced.

8.4 Management Strategies for Leaders and Followers Inter-generational conflicts are recorded throughout history (Tomkiewicz and Bass, 2008). When differences in generational norms affect working relationships, the resulting conflict can affect workplace performance. Generational interdependence correlates positively where generationally diverse employees share a common end state (McGuire et al., 2007). Intergenerational conflict may be serious during organizational reengineering when work groups are targeted by seniority. The severity of these conflicts can be a function of cross-generational distrust and animosity as the struggle for jobs becomes acute. When times are good, generations are tolerant at least, cooperative at best (Dixon et al., 2013). Generationally diverse talent pools are an important step in for sustaining organizational cultures (McGuire et al., 2007). Programs for employee development should relate generational norms and work tasks to strategic initiatives. Stretch assignments requiring diversity in skills, knowledge and abilities related to intergenerational capabilities, knowledge, and skills are powerful methods for increasing commitment across generations. Anderson (2010) suggested that intergenerational employee development start with younger employees. This may be impractical as the body of work knowledge is usually held by older generations. Members of older generations must recognize that the organization’s future belongs to younger engineers. Members of younger generations flourish on teamwork that lends itself to employee development, particularly when attention is given to having generational or age-diverse team members. The different set of network skills that younger generations possess must also be guided toward understanding administrative processes, building relationships across generations, and following established procedures before openly advocating for a more hierarchal-free workplace; i.e., know it before you change it.

8.5 Optional Content Commitment Organizational commitment is a term used to describe an employee’s psychological connection to the organization (Dixon et al., 2013). Each generation holds different beliefs and values, which vary across the generations and affect the generations’ norms related to organizational commitment. McGuire, By and Hutchings (2007) indicated that the X and Y generations exhibit less organizational commitment than boomers. Blau (1985) reported that both age and tenure are positively related to organizational commitment. Joo and Park (2009) found commitment was related to behavioral investments in the organization and likelihood to stay (loyalty) with the organization. Carver and Candela (2008) expanded the commitment construct by relating organizational commitment to an employee’s dedication to the values of an

97

Engineering Management Handbook

organization. Commitment has been positively correlated to higher organizational learning and developmental feedback from supervisors (Joo and Park, 2009). Allen and Meyer (1990) identified three constructs that describe commitment: affective, continuance, and normative. Affective commitment refers to an employee’s emotional attachment to, involvement in, and identification with, an organization. Watsi (2005) stated that stronger affective commitment results from positive work-related experiences. Continuance commitment is related to the costs an employee associates with leaving an employer. Employees with strong continuance commitment remain because they feel they have to do so (Meyer and Herscovitch, 2001). Tenure and benefits (accrued vacation, etc.) represent a “sunk cost” of employment (Sinclair, Leo and Wright, 2005) that induces loyalty. Normative commitment describes an employee’s feeling of obligation to remain with an organization as a general sense of obligation to fellow employees. Normative commitment develops from experiences that emphasizing loyalty to an employer (Wiener, 1982), what Kondratuk et al. (2004) refered to as “corporate loyalty.” Work by Meyer et al. (2010) demonstrated that how an employee behaves on the job is influenced jointly by commitment to the organization and to the occupation. Table 8.1 summarizes the relationship of the three commitment categories and generally recognized factors of workplace impacts; e.g., higher levels of commitment result in lower turnover rates. Table 8.1. Influence of Commitment on Workplace Impact Commitment Categories

Affective

Continuance

Normative

(Kondratuk et al., 2004)

Job performance

Positive

Negative

None

(Watsi, 2005)

Job outcomes

Positive

Positive

Positive

(Meyer and Allen, 1991)

Turnover

Negative

Negative

Negative

(Meyer et al., 2002)

Organizational Citizenship

Positive

Negative

Moderately

(Meyer et al., 2002)

8.5.1 Commitment and the Generations Boomers are characterized as having a sense of ownership in the organization. This sense of ownership correlates positively with all three categories of commitment, affective, normative and continuance as reflected in the underlying construct definitions related to want to, ought to and have to, respectively. The organization is seen as a means to an end; the end being a need to demonstrate success in their personal pursuits. As such, Baby Boomers will have a tendency to migrate to opportunities for accomplishment and therefore will reflect modest measures of affective commitment. Having seen parents recover from the impacts of economic depression, Baby Boomers will demonstrate strong normative and continuance commitment as they seek to maintain employment. In serving the organization, Baby Boomers demonstrate conviction for a shared purpose that supports high commitment across all three categories. This is also manifested in a willingness to challenge any deviations from the integrity of the pursuit of that purpose. Support for the shared or common purpose would be reflected in high measures of commitment. When the shared purpose no longer supports values, boomers are hypothesized to have a willingness to change themselves and/ or the organization for the good of the organization an indication of high measures of loyalty. Generation X, while reflecting a self-centered approach to work and commitment is has a stronger need to invest in relationships than careers and therefore measures of commitment will tend to demonstrate modest levels. Similarly, having been on their own (latchkey generation), Gen X’ers are expected to form strong bonds with colleagues on a personal basis rather than an organizational basis. As Gen X’ers consider their employer as a tool to be used for skill development they would have modest measures of commitment. The Generation X is focused on personal growth and personal relationship stability and will work to maintain that even if proactivity in seeking a new employer is part of their path toward fulfillment. The youngest generation, the Generation Y, represents a group newest to employment. Gen Y’ers are used to teams, teamwork, and social networking. Raised in daycare with their peers they quickly assume 98

Managing the Multi-Generational Knowledge Based Workforce

responsibilities associated with their work group(s) resulting in high measures of affective commitment. They expect to be managed well and challenged in their work assignments as a rite-of-passage. Gen Y’ers seemingly demonstrate high performance when properly challenged. Lacking work challenges, the Gen Y’ers are expected to insist on work conditions that meet their requirements and expectations and not vice versa.

8.6 Recommendations for the Management Discipline 8.6.1 Understanding Understanding how generations perceive their careers and life challenges can lead to high performance in the workplace. Dixon, Mercado and Knowles (2013) offered recommendations for the engineering manager and the engineering management discipline: 1. Take time to learn about candidates for hire or promotion. Asking questions pertaining to workplace behaviors and commitment constructs could provide clues as to know whether or not the candidate for hire or promotion is ideal. Determining things that may have had a major impact on someone’s life could also make known a person’s perceptions and expectations. Picking up (non) appropriate attributes early could facilitate proper job placement, increase retention, and reduce turnover. 2. Develop younger employees. Give them responsibility and encourage initiative early in their careers. Many Gen X’ers and Y’ers prefer stretch assignments requiring development of new knowledge and skills. The generational norm is to leave if not challenged. Providing regular guidance and feedback along the way increases retention. 3. Recognize commitment levels and capitalize on them. Do not let them go unrecognized. 4. Support employees in their need for knowledge and skills acquisition. Generation X in particular sees an employer as a means for skill building and tends to use a job as an extended education. If allowed to continually learn within the organization, they tend to be loyal. 5. Believe in change and embrace transformation. Be able to recognize that Baby Boomers can resist change and deal with them accordingly. 6. Retain corporate knowledge. Older engineers have obtained substantial business knowledge and skills that will be lost with attrition or retirement (Mraz, 2009). To prevent losing this body of knowledge, engineering managers should codify methods and processes. A referenceable body of knowledge can be used by younger engineers as needed.

8.6.2 Bias According to Karp (2012) managing generational bias is an issue all managers face. Generational biases certainly exist and are manageable only when the will and the means are available. Hiding bias is not a viable solution. Leveraging generational-difference bias as a source of and for energy, drive, and determination that is useful in harnessing the differences that reflect personal bias. The struggle is harnessing the differences for the good of the organization and the competitive posture of the business’ strategic focus. 1. The first consideration would be the engineering manager’s ability to manage personal bias with respect to generational diversity. Engineering managers can struggle with their bias throughout their career and life-stages. According to Karp (2012) as the manager matures focus changes from personal achievement, to career development, and culminates with contributing to society on some level. This latter career stage is the time when senior engineer managers are best able to lead the integration of multigenerational collections into a cohesive teams. These mature engineering managers can spur inter-generational cohesiveness when they bring the focus on the performance of the team using their wisdom, courage, justice and temperance that has moved past their own advancement (Peterson and Seligman, 2004). 2. Xu and Thomas (2011) conjectured leadership overlaps the construct, engagement. Engagement is defined as the degree to which employees make full use of their cognitive, emotional, and physical resources to perform role-related work (May, Gibson, and Harter, 2004). Engaged employees is the goal of every engineering manager who is responsible for performance measures tend to be highly subjective and

99

Engineering Management Handbook

prone to diverse interpretations; e.g., engineers and engineering. Engineering managers work to create an environment where engineers feel psychologically safe in the face of generational diversity. This is an acute need for inexperienced engineers. Two objectives compound the influences of a multigenerational workforce. (1) Developing new engineers into productive employees and (2) maintaining the morale and performance of experienced engineers, typically Baby Boomers and Gen X employees. Satisfying these objectives requires the engineering manager to address the norms and values of the generations in the context of the workplace environment. This environment requires younger engineers to recognize that work-place wisdom is held by the older engineers and must be mined or it will have to be recreated. Networking across the generations will develop respect within the younger engineers for the experienced engineers and will enable their recognition for engaging as team players. An understanding of engagement allows the engineering manager from any generation to place emphasis on the classic team-development methods such as development of the individual and rewarding work group successes. Enhancing engagement also will require goals and metrics associated with monitoring task-oriented behaviors. Engineering mangers must also provide appropriate resources and facilities, challenging tasks, effective task management, displaying integrity and open, honest communications along with mentoring all engineers (Xu and Thomas, 2011). Engineering managers must reflect work habits and related attitudes that they expect from their engineers. This is sets the example that each work activity is part of the organization’s strategic mission, a classic example of leading by example. 3. All engineering managers recognize that beyond satisfying regulatory requirements, there is limited return for mandated training. “Engaged” training–training that enhances satisfies the employees’ need for education and skill development consistent with organizational objectives–will recognize the differences in generations (Hotho and Dowling, 2010). Engineers interpret training based on their personal orientations, norms, values and situational context including the influences represented in the generations. Training is interpreted based on individual and group bias. A classic training failure is when a one-size-fits-all training intervention is required of employees without recognition of the individual trainees’ motivation, ability, personality, and work context (Hotho and Dowling, 2010). Training for development should be developed through discussions with the engineering manager, the candidate and the training designers (Haskins and Shaffer, 2010) and should focus on the individuals attributes, capabilities, needs, potential, and return for the organization. The design for developmental training should focus on desired strategic behaviors, self-awareness, change and change barriers, within organizational and professional contexts. 4. Any cross-generational integration initiatives should be augmented with dispersion tactics to leverage any training initiative across the organization. The diffusion of learning is best accomplished by applied team activities or learning projects (Atwood, Mora and Kaplan, 2010). This requires that integration initiatives be permeated with communicating the knowledge, skills, and abilities (KSA) gained and facilitating an environment where the generations can adapt the new KSAs into both their social and work groups. This is often called organizational learning where training includes a process of KSA transference or supporting learning in others. As part of the organization learns and shares the need for additional training interventions is reduced. As the organizational learning spreads, individuals in each generation will begin developing their own supportive behaviors.

8.7 References

100

Adams, S. J., “Generation X: How understanding this population leads to better safety programs,” Professional Safety, 45, 2000, pp. 26-29. Allen, N. J., and Meyer, J. P., “The measurement and antecedents of affective, continuance and normative commitment,” Journal of Occupational Psychology, 63, 1990, pp. 1-18. Allen, N. J., and John P. M., “The Measurement and Antecedents of Affective, Continuance and Normative Commitment to the Organization,” Journal Of Occupational Psychology, vol. 63, no. 1, 1990, pp. 1-18. Anderson, Jamie, “Customized executive learning: a business model for the twenty-first century”, Journal of Management Development, vol. 29, no. 6, 2010, pp. 545-555.

Managing the Multi-Generational Knowledge Based Workforce

Ansoorian, Andrew, Good, Pamela, and Samuelson, Dave,“Managing Generational Differences,” Leadership, vol. 32, no. 5, 2003, p. 34. Atwood, Meredith A., Mora, Jordan W. and Kaplan, Abram W., “Learning to Lead: Evaluating Leadership and Organizational Learning,” Leadership & Organization Development Journal, vol. 31, no. 7, 2010, pp. 576-595. Beutell, Nicholas J., and Wittig-Berman, Ursula, “Work-Family Conflict and Work-Family Synergy for Generation X, Baby Boomers, and Matures: Generational Differences, Predictors, and Satisfaction Outcomes,” Journal of Managerial Psychology, vol. 23, no.5, 2008, pp. 507-523. Blau, Gary J., The measurement and prediction of career commitment, Journal of Occupational Psychology, vol. 58, 1985, pp. 277-288. Blythe, Jennifer, Baumann, Andrea, Zeytinoglu, Isik U., Denton, Margaret, Akhtar-Danesh, Noori, Davies, Sharon, and Kolotylo, Camille. “Nursing Generations in the Contemporary Workplace,” Public Personnel Management, vol. 37, no. 2, 2008, pp. 137-159. Bova, Breda and Kroth, Michael. “Workplace Learning and Generation X,” Journal of Workplace Learning, vol. 13, no. 2, 2001, pp. 57-65. Brown, Patrick, “Reaching out to Generation Y,” Nuclear Engineering International, vol. 56, no. 680, 2011, p. 42. Carver, Lara and Candela, Lori. Attaining organizational commitment across different generations of nurses, Journal of Nursing Management, vol. 16, 2008,pp. 984-991. Chaleff, Ira, The Courageous Follower, Third Edition, Berrett-Koehler, 2009. Chen, Po-Ju and Choi, Youngsoo. “Generational Differences in Work Values: A Study of Hospitality Management,” International Journal of Contemporary Hospitality, vol. 20, no. 6, 2008, pp. 595-615. Coley, D. C., “Leading Generation Y,” Education Digest, vol. 74, no. 9, 2009, pp. 20-23. Dalakoura, Afroditi, “Differentiating Leader and Leadership Development,” Journal of Management Development, vol. 29, no. 5, 2010, pp. 432-441. Dixon, Eugene N. An Exploration of the Relationship of Organizational Level and Measures of Follower Behaviors, A Dissertation. Huntsville AL: The University of Alabama in Huntsville, 2003. Dixon, Gene and Westbrook, Jerry. “Followers Revealed,” Engineering Management Journal, vol. 15, no. 1, 2003, pp. 19-25. Dixon, Gene. Emotions and Follower Behaviors in a Time of Crisis, Society of Automotive Engineers, Proceedings of the 2009 World Congress, Detroit MI, 2009a. Dixon, Gene. “Followers: The Rest of the Leadership Process,” SAE International Journal of Materials and Manufacturing, vol. 1, no. 1, April, 2009b, pp. 255-263. Dixon, Gene, and Mercado, Ashley. “Followers and Commitment in the Workplace,” Society of Automotive Engineering Conference Proceedings, 2011. Dixon, Gene, and Westbrook, Jerry D., “Followers Revealed,” Engineering Management Journal, vol. 15, no. 1, 2003, pp. 19-25. Dixon, Gene and Knowles, Brady. “A Study of the Correlation of Follower Behaviors and Levels of Commitment Across Generations,” 2013 Proceedings of the ASEM International Annual Conference, October 2013. Downing, K., “Next Generation: What Leaders Need to Know About the Millennials,” Leadership in Action, vol. 26, no. 3, 2006, pp. 3-6. Ferguson, Karen L., and Reio Jr., Thomas G. “Human Resource Management Systems and Firm Performance,” Journal of Management Development, vol. 29, no. 5, 2010, pp. 471-494. Gibson, Jane Whitney, Jones, J. Preston, Celia, Jennifer, Clark, Cory, Epstein, Alexandra, and Haselberger, Jennifer,“Ageism and the Baby Boomers: Issues, Challenges and the TEAM Approach,” Contemporary Issues In Education Research, January vol. 3, no. 1, 2010, pp. 53-59. Hira, N. ,“You Raised Them, Now Manage Them,” Fortune, vol. 155, no. 9, 2007, pp. 38-48. Hoerr, T. R., “Supervising Generation X,” Educational Leadership, vol. 65, no, 2 , 2007, pp. 85-86. Hotho, Sabine and Dowling, Martin, “Revisiting Leadership Development: the Participant Perspective,” Leadership & Organization Development Journal, vol. 31, no 7, 2010, pp. 609-629. 101

Engineering Management Handbook

Johnson, James P., Korsgaard, M. Audrey, and Sapienza, Harry J. “Perceived Fairness, Decision Control, and Commitment in International Joint Venture Management Teams,” Strategic Management Journal, vol. 23, no. 12, 2002, pp. 1141-1160. Joo, Baek-Kyoo and Park, Sunyoung, “Career Satisfaction, Organizational Commitment, and Turnover Intention,” Leadership and Development Journal, vol. 31, no. 6, 2009, pp. 482-495. Jorgensen, Bradley, “Baby Boomers, Generation X, and Generation Y?: Policy Implications for Defense Forces in the Modern Era,” Foresight, vol. 5, no. 4, pp. 41-49, 2003. Jurkiewicz, C. E., and Brown, R. G. “GenXers vs. Boomers vs Matures: Generational Comparisons of Public Employee Motivation,” Review of Public Personnel Administration, vol. 18, 1998, pp. 18-37. Karp, H., Sirias, D., Arnold, K. “Teams: Why Generation X Marks the Spot. The Journal for Quality and Participation, vol. 22, 1999, pp. 30-33. Karp, Tom, “Develop Oneself as a Leader,” Journal of Management Development, vol. 32, no. 1, 2012. Keepnews, David M., Brewer, Carol S., Kovner, Christine T., and Shin, Juh Hyun, “Generational Differences Among Newly Licensed Registered Nurses,” Nursing Outlook, vol. 58, 2010, pp. 155-163. Kelley, R. E., “Power of Followership: How to Create Leaders People Want to Follow and Followers Who Lead Themselves, New York: Double Currency, 1991. Kondratuk, Tammy B., Hausdorf, Peter A., Korabik, Karen, and Rosin, Hazel M., “Linking career Mobility with Corporate Loyalty: How Does Job Change Relate to Organizational Commitment?” Journal of Vocational Behavior¸ vol. 65, 2004, pp. 332-349. Kuperschmidt, B. R., “Multigeneration Employees: Strategies for Effective Management,” The Health Care Manager, vol. 19, 2000, pp. 65-76. Loomis, J., “Generation X,” Rough Notes, vol. 143, no. 9, 2000, pp. 52-54. Martin, Harry J., “Improving Training Impact Through Effective Follow-Up: Techniques and Their Application,” Journal of Management Development, vol. 29, no. 6, 2010, pp. 520-534. May, D. R., Gilson, R. L. and Harter, L. M., “The Psychological Conditions of Meaningfulness, Safety and Availability and the Engagement of the Human Spirit at Work,” Journal of Occupational and Organizational Psychology, vol. 77, no. 1, 2004, pp. 11-37. McGuire, David, By, Rune Todnem, and Hutchings, Kate, “Towards a Model of Human Solutions for Achieving Intergenerational Interaction in Organisations,” Journal of European Industrial Training, vol. 31, no. 8, 2007, pp. 592-608. Merritt, Stephen R., and Neville, Shelley, “Generation Y,” The Serials Librarian, vol. 42, no. 1-2, 2003, pp. 41-50. Meyer, J. P. and Allen, N. J., “A Three Component Conceptualization of Organization Commitment,” Human Resource Management Review, vol. 1, no. 1, 1991, pp. 61-89. Meyer, J. P., and Herscovitch, L., “Commitment in the Workplace: Toward a General Model.” Human Resource Management Review, vol. 11, 2001, pp. 299-326. Meyer, J. P., Stanley, D. J., Herscovitch, L., and Topolnytsky, L., “Affective, Continuance, and Normative Commitment to the Organization: A Meta-analysis of Antecedents, Correlates, and Consequences,” Journal of Vocational Behavior, vol. 61, 2002, pp. 20-52. Silver, Mitchell, Changing Demographics of Eastern NC in 2050, November 3, 2011, ECU-TV. Mraz, Stephen, J., “Changes in the Engineering Profession Over 80 Years,” Machine Design.com, 2009. Peterson, Christopher and Seligman, Martin, Character, Strengths and Virtues. A Handbook and Classification, Oxford University Press, 2004. Ray, Linda Keith. Follow the Leader: An Investigation of the Relationship Between Hierarchical Levels and Measures of Follower Behaviors of Selected NC Community College Employees, A Dissertation, Greenville, NC, East Carolina University, 2006. Rich, Theresa A., Becoming the boss whisperer: An Examination of the Relationship Between Employee Follower Behaviors and Supervisor Satisfaction With Employee Performance, A Dissertation. Minneapolis MN: Capella University, 2008.

102

Managing the Multi-Generational Knowledge Based Workforce

Ricketson, Rushton S., An exploration of the relationship of leadership styles and dimensions of courageous followership, A Dissertation, Virginia Beach, VA, Regents University, 2008. Riggio, Ronald E., Chaleff, Ira, and Lipman-Blumen, Jean, The Art of Followership, Jossey-Bass, 2008. Shaw, Sue and Fairhurst, David, “Engaging a New Generation of Graduates,” Education + Training, vol. 50, no. 5, 2008, pp. 366-378. Sinclair, Robert R., Leo, Michael C., and Wright, Chris, “Benefit System Effects on Employees’ Benefit Knowledge, Use, and Organizational Commitment,” Journal of Business and Psychology, vol. 20, no. 1, 2005, pp. 3-29. Smola, Karen Wey and Sutton, Charlotte D., “Generational Differences: Revisiting Generational Work Values for the New Millennium,” Journal of Organizational Behavior, vol. 23, no, 4, Special Issue: Brave New Workplace: Organizational Behavior in the Electronic Age, 2002, pp. 363-382. Tomkiewicz, Jospeh and Bass, Kenneth, “Attitudes of Business Students Toward Management Generation Cohorts,” North American Journal of Psychology, vol. 10, no. 2, 2008, pp. 435-444. Watsi, S. Arzu, “Commitment Profiles: Combinations of Organizational Commitment Forms and Job Outcomes,” Journal of Vocational Behavior, vol. 67, 2005, pp. 290-308. Wiener, Y. “Commitment in organizations: A normative view.” Academy of Management Review, vol. 7, 1982, pp. 418-428. Xu, Jessica and Thomas, Helena Cooper, “How Can Leaders Achieve High Employee Engagement?,” Leadership & Organization Development Journal, vol. 32, no. 4, 2011, pp. 399-416. Zemke, Ron, Raines, Claire, and Filipczak, Bob, Generations at Work, AMACOM Books, 1999.

103

104

Operations Research

9 Operations Research Scott E. Grasman Rochester Institute of Technology

Abhijit Gosavi Missouri University of Science and Technology

Katie McConky Rochester Institute of Technology

105

Engineering Management Handbook

9.1 Introduction to Operations Research Modeling Operations Research (OR) applies a rigorous methodology to managerial problems, typically with quantitative factors. With application to manufacturing, service, and military, OR is widely applied by engineering managers and consultants, and is seen as a valuable approach to aiding managerial decision-making. The following sections provide a brief introduction to the importance and history of OR, as well as an overview of the standard OR methodology.

9.1.1 Importance of Operations Research for Engineering Managers Operations Research focuses on the application of resources to the production of goods and services. Operations researchers optimize the use of capital, materials, technology, human skills, knowledge and information. Solutions provided by operations researchers have helped engineering managers design effective new systems, identify opportunities for improving systems, and make tradeoffs to coordinate policies from different functional areas. In order to successfully apply OR, engineering managers must have an understanding of the basic language and concepts of operations, the mathematical skills, including modeling, probability and statistics, as well as effective intuitive and critical thinking skills to synthesize all areas of operations. Since OR typically aids in the decision making process, in contrast to smart systems, managers must combine the analysis and recommendations, often generated through decision support systems, with other qualitative factors. In addition, engineering managers must use quantitative relations founded on theories from simple systems to build theories and intuition for more complex systems. Operations Research is generally used synonymously with Management Science, while less quantitative researchers often use the term Operations Management as well. Loosely, the spectrum may be defined from theoretical to applied, with OR on the theoretic end, Operations Management on the applied end, and Management Science in the middle. Operations Management is described in Chapter 15.

9.1.2 History of Operations Research Operations Research is a rich discipline with historical milestones involving world events, people, and technology. Many date the discipline back to the industrial revolution or to the turn of the 20th century, while other operations research historians look to military applications during the World War I and World War II. Shortly after World War II, operations research was propelled by George B. Dantzig, who derived the Simplex Method for solution of linear programs. Certainly, modern operations research has been greatly enhanced by the use of modern computational technology that has allowed for the solution to more and more complex problems. Annotated timeline of operations research is available (Gass and Assad, 2004). In the United States, the flagship professional society is the Institute for Operations Research and the Management Sciences (INFORMS), though many other professional societies have operations research divisions, councils or interest groups. Internationally, there are more than 50 societies organized under the umbrella of the International Federation of Operational Research Societies (IFORS). These professional organizations are purveyors of professional journals, international conferences, and professional stewardship.

9.1.3 Operations Research Methodology Operations Research provides a modeling process from which a collection of (mathematical) models that are appropriate for many real-world problems have evolved. These mathematical models are either descriptive in that they predict or explain an outcome in order to determine if a given system would meet specifications, or prescriptive and control or optimize an outcome in order to provide optimal design or control of a system. Importantly, prescribing first requires the ability to describe! The basic operations research process consists of eight phases, as shown in Figure 9.1, including: 1. Problem Definition which consists of stating the decision that needs to be made, including system objectives and constraints. 106

Operations Research

2. Data Collection specifies the data collection, processing, analysis, and management requirements. 3. Model Formulation translates the problem definition into mathematical form by stating decision variables, objective and constraint functions. 4. Model Verification checks to make sure the model is working as intended. This is especially important with the implementation of technology based solutions. 5. Solution is generating an answer to your defined problem using the verified model. The solution method is dependent on the model formulation. 6. Validation ensures that the model developed accurately represents the real system. 7. Presentation of Results combines the quantitative analysis and results with other factors such as company objectives. 8. Implementation applies the solution and recommendation to achieve the desired outcomes. Operations Research has lead to a variety of deterministic and stochastic models that are appropriate for use by engineering managers. Applications include traditional engineering disciplines such as manufacturing and technology management, but have also lead to process improvement in sectors such as health care, financial services, and entertainment. Specific models and applications are discussed in the following sections. It is noted here, that these models often involve the designation, “program.” Although computer programming has been instrumental in the implementation of solution algorithms for these models, the term originate from the concept of “order of operations.” Mathematical programs were researched and applied long before the invention of computers. Figure 9.1. Operations Research Methodology Define Problem

Collect Data

Formulate Model

No

Verify? Yes Solve Model

Present Results

Valid?

No

Yes Implement Solution

107

Engineering Management Handbook

9.2 Deterministic Models This section covers topics in OR that deal primarily with known or deterministic parameters. Topics include linear and integer programming, which provide a way to model and explore a system to find an optimal solution to an objective. Linear programs lend themselves nicely to a process called sensitivity analysis, which allows the researcher to understand the impacts of changes to model parameters. A discussion of non-linear programming follows and provides the ability to model more complex relationships than strictly linear programs. Finally, the section is concluded with a discussion of dynamic programming, a modelling framework often used for systems with recursive-based solutions.

9.2.1 Linear Programming Linear programming is the optimization of a linear function subject to linear constraints, expressed as equalities or inequalities. Generally, these problems either aim to maximize an objective, e.g., profit, subject to limited resources, or aim to minimize an objective, e.g., cost, subject to minimum requirements. Other models have fixed-requirement constraints, e.g., matching supply with demand or combinations of the above constraints. Linear programs (LP) are extremely useful to engineering managers because they are efficient to solve, with solution procedures that guarantee optimal solutions (if one exists). In addition, linear programs allow for the generation of useful sensitivity analysis information. By definition, linear programs have the characteristics of proportionality and additivity. Further basic linear programming has the additional characteristics of divisibility and certainty. All the characteristics can be relaxed, leading to other mathematical models. • Proportionality—If an activity level is multiplied by a constant, then the contribution of the activity to the objective function and constraints is multiplied by the same constant. • Additivity—The total contribution to the objective function and constraints is equal to the sum of the contributions of the individual activities. • Divisibility—Both integer and non-integer levels of activities are allowed. • Certainty—No random outcomes. Although many practical problems are essentially linear, linear programming can also be used to address more complex problems. These techniques will be described in later sections.

9.2.2 Basic Problem Formulation As described, linear programs consists of either a maximization problem or minimization problem subject to a set of constraints. Two basic forms are shown in Figure 9.2, though fixed-requirement problems and mixed-constraint problems are also common. Figure 9.2. Basic Linear Programming Formulations

maximize Z =

n

∑c j =1

jxj

subject to : n

n

∑c j =1

j

xj

subject to : n

∑ aij x j ≤ bi for i = 1,2,..., m

∑a

x j ≥ 0 for j = 1,2,..., n

x j ≥ 0 for j = 1,2,..., n

j =1

108

minimize Z =

j =1

ij

x j ≥ bi for i = 1,2,..., m

Operations Research

The following notation is used to formulate the models in Figure 9.2. Z n m xj aij

= objective value function = number of decision variables = number of constraints = decision variable j = resource/requirement for activity j and constraint i

bi cj

= resource/requirement i = objective value coefficient for activity j

The linear matrix nature of the formulation is evident from Figure 9.2. These characteristics are leveraged to obtain optimal solutions.

Solution Techniques The main technique for solving linear programs is the Simplex Method (Dantzig), or a variation on the Simplex Method. Other methods, e.g., Interior Point Methods (Karmarkar’s Algorithm), have also been successfully applied to linear programs or special cases of linear programs. While the details of these methods will not be provided here (see Hillier and Lieberman, 2010), small linear programs can be solved graphically, along with linear algebra, to provide insight into the solution techniques. Consider the following linear program. maximize Z = 2 x1 + 3 x 2 subject to 2 x1 + x 2 x1 + 2 x 2

≤ 4 ≤ 5

x1 , x 2 ≥ 0

The problem may be solved by graphing all constraints in order to establish the feasible region, i.e., the points that satisfy all constraints, and by plotting iso-objective lines that are used to find an optimal solution. The shaded region in Figure 9.3 is the feasible region, while the dotted lines represent the iso-objective lines. Figure 9.3. Feasible Region and Iso-Objective Lines x2

maximize Z = 2x1 + 3x2

x1

x1 , x2 ≥ 0 2x1 + x2 ≤ 4000

x1 + 2x2 ≤ 5000

109

Engineering Management Handbook

As the iso-lines in Figure 9.4 are moved “up and to the right”, the objective value increases, thus the optimal solution is obtained by moving an iso-line to the extreme of the feasible region.



Figure 9.4. Corner Point Optimal Solution x2 optimal solution (maximum Intersection of: 2x1 + x2 = 4000 x1 + 2x2 = 5000 x1 = 1000, x2 = 2000 Z = $8000

x1 x1 , x2 ≥ 0 2x1 + x2 ≤ 4000

x1 + 2x2 ≤ 5000

Once the corner point is identified, as in Figure 9.4, linear algebra can be used to solve for the optimal solution. The main observation is that an optimal solution will be at one of the corners or extreme points of the feasible region. More specifically, an optimal solution will exist at, at least one corner point solution; if the optimal iso-line is parallel to a constraint, then there are multiple optimal solutions. In other cases, it is possible that no feasible solutions exist, i.e., the combination of constraints is too restrictive, or that the objective function is unbounded, i.e., the combination of constraints is too relaxed. In these situations the model should be reformulated and validated. The main observation, that optimal solutions occur at extreme points of the feasible region, is the basis for most LP solution algorithms, including the Simplex Method. For more information on solution theory, the reader is referred to Hillier and Lieberman (2010). Of course, linear programs may be solved using algorithms implemented in computer software applications. These range from simple spreadsheet applications to advance commercial software packages.

9.2.3 Sensitivity Analysis Sensitivity Analysis asks questions related to the effect of changing model parameters. The most common analysis targets changes to objective value coefficients, e.g., profit/revenue or costs, and constraints, e.g., resources or requirements. For objective coefficients, two important questions relate to range of optimality and reduced costs. For constraints, two important questions relate to shadow prices and allowable ranges. It is necessary to note that, for changes to objective coefficients, though the optimal solution may not change, the value of the objective function certainly will change. For changes to binding constraints, both the solution and the objective value are likely to change. See Hillier and Hillier (2008) for further discussion. Range of Optimality—How much coefficients can change without changing optimal solution. Graphically, the range of optimality is found by changing the slope of the objective function line within the limits of the slopes of the binding constraint lines. At this point, the iso-lines will last intersect at two corners, causing multiple solutions; beyond this, the original optimal solution will no longer be optimal. 110

Operations Research

For a two variable example with objective function c1x1 + c2x2, and binding constraints, a11x1 + a21x2 = b1 and a12x1 + a22x2 = b2, the range of optimality is given by: –

a11 a21

≤ –

c1 c2

≤ –

a12 a22

.

Changes to a single objective value coefficient are fairly straightforward to handle, while multiple changes require further analysis. When changes are made to multiple objective coefficients, the percentage change relative to the range of optimality of each changing coefficient is calculated. If the sum of the percentage changes does not exceed 100%, then the original optimal solution will remain optimal. If the sum exceeds 100%, then the optimal solution may or may not change. The key is the relative change of the coefficients and their effect on the slope of the objective function. For example, multiplying all coefficients by a constant will not change the slope of the objective function, regardless of the magnitude. In most cases, computer software applications are used to solve and provide this information, though the same logic applies to higher dimension problems. Reduced Costs—How much a coefficient must change in order to enter the optimal solution. If an activity is relatively too costly, or does not generate enough revenue, then this activity will not be included in the optimal solution. The reduced cost indicates the change (reduction) in cost required to make the activity desirable. Conversely, if an activity is not relatively profitable, then the reduced cost indicates the required change (increase) in profit/revenue. The engineering manager can use this information for pricing decisions, as well as identification of cost reduction strategies. Reduced costs may be found in the same manner as the range of optimality. By leaving the other objective coefficients unchanged, the range of optimality of the coefficient of interest indicates the reduced cost. Shadow Prices—How much the optimal objective value changes with changes in a constraint (within limits). Shadow prices are perhaps the most useful sensitivity information because they indicate the linear change in objective value due to increases or decreases in available resources or requirements. The engineering manager can use this information to determine if additional resources should be obtained or if requirements should be adjusted. Shadow prices are determined by obtaining the slope/gradient of binding constraints. Again, this may be done mathematically, but is normally done using commercial software applications. Allowable Ranges—Valid Range of Shadow Prices. Allowable ranges indicate how much the constraint, resource, or requirement, can change without invalidating the shadow prices. Beyond the allowable range, the objective value may continue to improve; however, the rate of change will be different. The allowable ranges are determined by obtaining the point at which a binding constraint no longer becomes binding, or a nonbinding constraint becomes binding. The 100% Rule also applies in that shadow prices will remain valid as long as the sum of the percentage changes does not exceed 100%. Additional analysis is required if the change exceeds 100%.

9.2.4 Duality Theory Duality Theory allows for both the solution of linear programs and sensitivity analysis. The original, or primal, linear program may be converted to the dual linear program through a standard conversion process involving linear matrix transformation. 111

Engineering Management Handbook

For the standard form linear programs shown in Figure 9.2, the dual linear programs are shown in Figure 9.5. Figure 9.5. Dual Linear Programs for Basic Linear Programs





Once the dual has been formulated, complementary slackness may be used to obtain the solution and sensitivity analysis for both models. Generically, the solution to the dual problem provides the shadow prices for the primal constraints and can be used in decision-making as previously described. The dual solution can also be used to determine the range of optimality, but this process is a bit less transparent. For more information on this duality, see Hillier and Lieberman (2010).

9.2.5 Applications For application purposes, linear programs are often classified into one of four categories based on the objective function and the set of constraints. These classifications include resource-allocation, cost-tradeoff, fixed-requirement, and mixed problems (see Hillier and Hillier, 2008). Resource-Allocation—Linear programming problems involving the allocation of resources to activities so as to satisfy objective, e.g., maximizing profit, subject to limit resources. Thus, constraints are resource constraints, in that the amount allocated has to be less than or equal to the amount available. Typical examples include product-mix problems, activity-mix problems, and capital budgeting problems. Cost-Tradeoff—Linear programming problems where the various activities are chosen to satisfy requirements, typically at minimum cost. Constraints are requirement constraints, in that the amount allocated must be greater than or equal to the requirement. Typical examples include activity-mix problems, personnel scheduling problems, and others. Fixed-Requirement—Linear programming problems involving allocation of activities such that the amount provided is equal to the amount required. Functional constraints are fixed requirement constraints. Many common applications relate to network optimization models, such as transportation/assignment problems, shortest/longest path problems, maximum flow problems, and spanning trees. The engineering manager can address these applications through the implementation of standard procedures presented earlier, of course, with the assistance of software applications. A common example is the use of fixed-requirement models for critical paths in project networks.

9.2.6 Integer Programming

112

Integer programming models restrict the decision variable to integers, which violates the divisibility assumption of linear programs. A variety of integer models exist including pure integer programs, mixed integer programs, and binary integer programs. The formulation of basic integer programs is similar to that of linear programs, but the solution creates many more challenges. Additionally, the integer nature, duality theory does not hold and sensitivity analysis is much more challenging.

Operations Research

9.2.7 Solution Techniques Solution techniques are more complex than linear programming and include cutting plane methods and branch and bound algorithms (see Hillier and Lieberman, 2010). Unlike linear programming, not all integer programs can be solved in reasonable computational time, thus many heuristics have been devised to achieve approximate solutions. One such technique is linear programming relaxation, which ignores the integer constraints. The benefits of relaxation include more efficient solution procedures, as well as simplified sensitivity analysis. Of course, the applicability of relaxation depends on solution quality. Commercial software packages are available for practical application.

9.2.8 Binary and Auxiliary Binary Variable Use of binary and auxiliary binary variables is a powerful technique that aids in the formulation and application of mathematical models. Binary decisions, e.g., yes or no decisions, are quite common in practice, and auxiliary binary variables can be used to efficiently incorporate problem characteristics such as fixed (or setup) costs, mutually exclusivity or contingency, and either-or constraints. The ability to utilize binary variables leads to models that are easy to interpret and analyze. For example, the use of auxiliary binary variables may allow a non-linear model to be formulated as a linear integer program; thus aiding in the solution and sensitivity analysis.

9.2.9 Applications Though integer programming relates to any mathematical model with integer variables, there is much overlap with network optimization problems (transportation/assignment problems, shortest/longest path problems, maximum flow problems, and spanning trees). Facility location, inventory modeling, scheduling, and investment are additional popular application areas.

9.3 Non-Linear Programming Non-linear programming is similar to linear programming (and integer programming) except for the inclusion of non-proportional relationships. Depending on the nature of the problem, formulation of non-linear programs is generally more difficult, and solution is much more difficult, if possible. Non-proportional relationships may exist in the objective value function, in the constraints, or in both. For example, the objective function may include marginally increasing/decreasing revenues or costs, or discontinuities.

9.3.1 Solution Techniques Due to the complexity of non-linear models, numerous approaches have been used to attempt to reduce the computational complexity and quality of solutions. Generally, the solution challenges relate to the non-convexity of the feasible region or local optima. Solution approaches include gradient methods, branch and bound, dynamic programming and meta-heuristic approaches, including genetic and other evolutionary algorithms, tabu search, and simulated annealing (Hillier and Lieberman, 2010). Software packages implement a variety of algorithms.

9.3.2 Separable Programming Separable programming is a technique that may be used to convert a non-linear program to a linear program in order to take advantage of the efficiency of solution and sensitivity analysis. The approximation technique enhances the decision-making ability of the engineering manager, but quality of solution must be considered. A non-linear relationship, either an objective or constraint, may be approximated using piecewise linear functions as shown in Figure 9.6. 113

Engineering Management Handbook Figure 9.6. Piecewise Approximation to a Non-linear Function

The quality of approximation depends on the number of piecewise functions used in the approximation. Hewitt et al. (2015) provide a methodology to overcome this challenge by adapting a reformulation technique from non-convex optimization to model non-linear functions with a discrete domain using sets of binary and continuous variables and linear constraints.

9.3.3 Applications Some high profile applications of interest to the engineering manager include marketing and inventory management. In these areas, non-proportional relationships commonly exist among various profit/revenue or cost parameters. In some cases, these relationships may be made linear by using auxiliary binary variables (as described earlier). Other common non-linear relationships are prevalent in financial management and risk management due to compound interest, variance and co-variance.

9.4 Dynamic Programming Dynamic Programming involves the recursive solution to a set of sub-problems. These models apply to problems requiring (or allowing) a sequence of interrelated decisions. Solutions usually rely on network representations, where the nodes represent decision points and arcs represent the cost/reward of decisions, thus many of previously mentioned models can also be formulated using dynamic programming.

9.4.1 Solution Techniques A variety of solution methods are used to solve for optimal solutions to dynamic programming models. Most of these rely on the Principle of Optimality, which states that the optimal solution from this point (decision epoch) forward is independent of the past. This principle allows for significant computation savings over enumeration. As noted, a network representation is often used to represent the costs associated with decisions. Nodes, representing State Space, provide enough information for decision-making (cost computation), and the arcs, representing Action Space provide the set of feasible decisions. The Objective Value Function provides a recursive optimal cost path from any decision node to final node. A Boundary Condition is required as a starting condition for the recursion. Mathematically, optimal solutions may obtained using recursion (forward or backward), Dijkstra’s Algorithm, reaching, or other techniques (Denardo, 2003). Computationally, these algorithms are quite conducive to computer coding. Dynamic programs can also be transformed to linear programs and solved using standard linear programming methods. In addition, a number of commercial software packages are available.

9.4.2 Applications

114

Due to the network nature of dynamic programming models, the network models discussed earlier are prime candidates for dynamic programming formulations. Additionally, inventory modeling and invest-

Operations Research

ment decisions are popular applications. Dynamic programming may also be used to solve integer and non-linear models; however, the curse of dimensionality often hinders practical application due to long solution times.

9.5 Stochastic Models The previous section discussed system models with known or deterministic parameters. All too frequently systems, encountered by the operations researcher include random or stochastic variables. A system model that uses a collection of random or stochastic variables is termed a stochastic process. Stochastic processes occur frequently, and examples include the service time for a customer at a fast food counter, the loading and unloading times of parts on a machine, or the random decisions of amusement park patrons as they decide which attraction to experience next. Stochastic processes are frequently modelled with Markov Chains discussed next, which support queuing theory discussed later in this chapter.

9.5.1 Markov Chains Markov Chains are powerful tools useful in modeling systems that generally exhibit randomness in their behavior, i.e., random or stochastic systems. A so-called stochastic process is an abstract (mathematical) entity associated with the stochastic system. The mathematics underlying the events that occur in a stochastic system is often captured via the stochastic process. Here, we will focus on discrete-event stochastic systems, wherein the time between successive events that occur in the system is usually a discrete random variable. The “state” of the system is typically an n-tuple (with n = 1 for the simplest case), which defines the condition of the system at any given time. Consider, for example, a single-server queue that forms in a supermarket. Other examples of queues that are observed in real-world systems are: queues of jobs in front of machines in a production shop, virtual queues of customers on telephones that form in call-centers, and queues of passengers that form in the security checkpoint in an airport. The state of this “queuing system” could be defined by the number of customers in the system. An important fact about the state of the system is that it changes with time, and usually, it changes randomly. Predicting what the next state will be if the current state is known, usually, requires an analysis of the random variables that govern the system’s behavior. Also, analysis of this nature can help the engineering manager estimate parameters such as the average time spent by the system in a given state in the long run and the long-run cost of running the system. In other words, this kind of analysis can help formulate a suitable objective function for a stochastic system, which then can be used for optimizing the system. What is important to remember is that the analysis requires an understanding of the mathematical properties associated to the stochastic system and the stochastic process. The Markov Chain is one of the simplest examples of a stochastic process. We will study two types of Markov Chains (processes): (a) discrete-time Markov Chains and (b) semi-Markov processes.

9.5.2 Discrete-time Markov Chains We begin by introducing some notation. Let X(t) denote the state of the system at time t when viewed by the analyst. Also, P[event] will denote the probability of the occurrence of the “event” (within the square brackets). We will assume for the discrete-time Markov Chain model that the system is observed periodically after unit time. Hence time, t, will be increased in steps of 1 as follows: 1, 2, 3, …Note that the duration of the “unit time,” which is technically the time interval between successive views of the system by the analyst, is determined by the analyst, and it could be any finite quantity, e.g., 1 second, 4.5 hours, or 17 days. The fundamental property that defines the Markov Chain is:

𝑃𝑃[𝑋𝑋(𝑡𝑡 + 1) = 𝑗𝑗|𝑋𝑋(𝑡𝑡) = 𝑖𝑖] = 𝑓𝑓(𝑖𝑖, 𝑗𝑗),

where f(i,j) is a function of the states i and j. What the above implies is that for a stochastic process to be called a Markov Chain, if the current state of the system is known (i.e., the state of the system at time t), 115

Engineering Management Handbook

the probability that the system will transition to a given value of the next state (i.e., the state of the system at time t+1) should depend only on what the current state is and what the next state is. This means that the probability that the system will transition to a new state depends entirely on what the current state and the new state are; further, it is independent of where the system has been in the past. This is called the Markovian property, or the memoryless property. A number of system parameters can be easily measured for a system that can be modeled as a Markov Chain. This is because the Markov Chain lends itself to some straightforward mathematical analysis. We will discuss one simple aspect of the Markov Chain that is almost invariably of special interest to managers: calculation of the limiting probabilities, also called the invariant or steady-state probabilities, of the Markov Chain.

Limiting Probabilities The proportion of time spent by the Markov Chain in any given state in the long run is called the limiting probability of the state. This concept is best illustrated with an example. Consider a two-state Markov Chain. The transitions of the Markov Chain are defined by the so-called one-step transition probabilities. Here, P(i,j) will denote the one-step transition probability or simply the transition probability of going from state i to state j in one step. The transition probabilities can be conveniently stored in a square matrix, such that P(i,j) is the element in the ith row and the jth column in the matrix. This matrix is called the transition probability matrix (TPM). In a so-called regular Markov Chain, for every value of i and j, P n (i,j) > 0 for some finite value of n, where Pn(i,j) denotes the element in the ith row and jth column of the TPM of the Markov Chain raised to the nth power. For a regular Markov Chain, limn→∞P n (i,j) can be shown to exist and is independent of i, i.e., it only depends on j; we will use the notation to denote limn→∞P n (i,j). This limit is called the limiting (or steady-state) probability of the state j. The limiting probability of a state represents the proportion of transitions the chain makes to that state in the long run (in other words, if we observe the system for a long time). For the Markov Chain, it is usual to assume that the system spends the same amount of time in every state and that the transitions are instantaneous; hence the limiting probability of a state also represents the proportion of time the chain spends in that state. It is thus clear that the limiting probability provides us with useful information about the system. The limiting probability can be easily computed from the TPM via the solution of the following system of linear equations:

for j = 1, 2, ... , k (where k denotes the number of states) and

We illustrate this idea with a simple example below. Consider a 2-state Markov Chain, whose TPM is as follows:

116

Assume that the Markov Chain models a machine that can either be “up” (working) or “down” (under repair because of failure); state 1 stands for the up state and state 2 for the down state. Solving the linear system of equations provided above, we obtain that π (1) = 0.7273 and π (2) = 0.2727. This solu-

Operations Research

tion provides the manager with a critical piece of information: the machine is down 27.27% of the time. Analysis of this nature can lead to downtime reduction and process improvement. Much of the work in building a Markov-chain model is in computing elements of the TPM. Once the TPM is computed, it can be used elegantly to derive a number of system measures of interest to the manager. Also, if the manager has access to the costs associated with spending time in a system, these measures can also be useful for measuring the costs of operating a system under a given strategy. The computations needed for these measures can be easily performed with the help of commonly available software such as MS EXCEL or more sophisticated software programs such as MATLAB. The above was a very simplified example (with a small 2-state Markov Chain) presented only for illustrative purposes. However, we note that the principles of computing limiting probabilities can be easily extended to scenarios with a larger number of states. In practice, it is common to see large Markov-chain models with hundreds of states used by managers in production systems for inventory control, quality control, and preventive maintenance. Markov Chains are also very useful in analyzing queues, which is a topic that we will discuss in more detail next. Applications of Markov-chain-based models can be found in a very large number of other domains, including wireless communication, analysis of banks, airports, and amusement parks, robotics, speech recognition, forecasting, and biological modeling to name a few.

Absorbing Markov Chains An absorbing state, i, in a Markov Chain is one for which P(i,i)=1, i.e., once an absorbing state is entered, the chain does not leave the state. A Markov Chain in which at least one state is an absorbing state is called an absorbing Markov Chain. The states of an absorbing Markov Chain that are not absorbing are called transient states. A simple example of a Markov chain that is absorbing is:

In the above, the third state is the absorbing state. Examples of systems that can be modeled with absorbing Markov chains are a production process that moves through gradually deteriorating acceptable states but eventually enters an absorbing state from where it cannot return to an acceptable state and the process must be halted. Interesting questions that are relevant in absorbing Markov Chains are: (1) If there are multiple absorbing states, what is the probability that the system is absorbed by a given absorbing state? (2) What is the expected number of transitions that occur before a system is absorbed in a given absorbing state? Answers to these questions depend on the initial state of the system and the TPM. Some straightforward mathematical analysis can be used to answer the questions posed above. Absorbing Markov Chains are useful in analyzing biological processes and quality control processes. We refer the interested reader to Ross (2006).

9.5.3 Semi-Markov Processes The Semi-Markov Process (SMP) is a more general model than the Markov Chain. In an SMP, we assume that the time spent in any given state is a random variable with a given distribution. In other words, underlying the SMP is a Markov Chain, which is usually referred to as the “embedded” Markov Chain. The time spent in a state is usually called the sojourn time. If the sojourn time for every state is exponentially distributed, the SMP is called a Continuous-Time Markov Chain (CTMC). We will first discuss the CTMC, and general SMPs.

Continuous-Time Markov Chains (CTMCs) In a CTMC, the time spent in every state has the exponentially distribution. The limiting probabilities of the CTMC are usually found from the rate of transition from one state to another. While it is certainly possible to determine the transition probabilities from the distributions involved, it is easier to use the rates of transition from one state to another to determine the limiting probabilities, via the so-called bal117

Engineering Management Handbook

ance equations. It is easy to determine the rates from the underlying distributions. Note that use of these balance equations works only for CTMCs and should not be used for other types of Markov Chains. Let qij denote the rate of transition from state i to state j, and let vi denote the total rate of transition from state i to all the other states. Thus, vi = Σqim, where the summation is over all values of m. Then, the limiting probabilities can be determined by solving the following system of linear equations (balance equations): For i=1,2, … , k,

and

Countless examples of CTMCs can be found in EM and social sciences, including the analysis of certain types of queues, inventory control, and studying biological processes (cancer cells, HIV, and flu virus propagation). An attractive feature of the CTMC is that it can be analyzed very easily for its steady-state properties via the balance equations provided above. However, a major criticism of the CMTC model is that it works only if all the random variables for the sojourn times are exponentially distributed; this is not usually the case in real-world applications.

General SMPs The generalized model of the SMP in which the sojourn times do not have the exponential distribution exploits properties of the embedded Markov Chain. In particular, to determine the proportion of time spent by the SMP in any given state, one must first compute the limiting probabilities of the underlying Markov Chain. Let L(i) denote the proportion of time spent in state i by the system. Then, if s(i) denotes the mean sojourn time in state i, we have that:

for i=1,2, … , k and denotes the limiting probability of the ith state in the embedded Markov Chain. More interestingly from the standpoint of the engineering manager, if there are costs involved in operating the system, one can determine the average cost per unit time (ρ) of operating the SMP as follows:

where R(i) denotes the total cost of being in state i. The above is a very useful expression for quantifying the performance of a semi-Markov process (and hence any CTMC and any Markov Chain) in terms of dollars. Note that the expression developed above for the average cost of operating the system also works for the Markov Chain by setting s(i) =1 for all values of i. The above formulation has been widely used in the literature to develop objective functions of production systems (Tomasevicz and Asgarpoor, 2009; Solo, Kharoufeh, and Ulukus, 2010; Nodem, Kenne, and Gharbi, 2011).

9.5.4 Queuing Theory

118

The theory of queuing has been developed in order to study queues or waiting lines that form in service systems such as grocery stores, banks, airports, and amusement parks, production systems, e.g., machine shops or assembly systems, and electrical systems, e.g., Internet and ATM networks. Queues are usually

Operations Research

formed when there are one or more servers and a number of entities that need service from the server. Examples of queues that are observed in real-world systems are: queues of jobs in front of machines in a production shop, virtual queues of customers on telephones that form in call-centers, and queues of passengers that form in the security checkpoint in an airport. Long waiting lines generally imply long waits that typically result in frustration for those waiting. On the other hand, ensuring short waits oftentimes requires a large number of servers or highly efficient servers, who may charge higher salaries. Queuing systems are often studied for the mean (and higher moments) of the waiting time and queue length, average utilization of the server, and the probability that a customer reneges. Designing a queuing system appropriately is usually an important problem from the standpoint of managing the system efficiently. Under-staffed servers can lead to longer than expected waits driving customers away, while overstaffed servers can be expensive and can increase operating costs. We will discuss some basic notions related to queues in what follows.

9.5.5 Stability of Queues The arrival rate at any server station (which may have multiple servers) is the mean number of customers that arrive to the server station in unit time. The arrival rate will be denoted by λ in our discussion. The service rate is the mean number of customers served by the server station in unit time; it will be referred to as μ. If γ < μ, we say that the system is stable, i.e., the system does not get overloaded. If this condition is not true and if there is no restriction on how many customers can wait in line, gradually the system becomes unstable and the number of customers waiting in the line starts approaching infinity. This happens unless there is a restriction on how many customers can wait in the line. Usually, there is little point in analyzing unstable queues. However, unstable queues do form in the real world. When an Internet website crashes, it is usually because the server cannot handle the number of people trying to get on to that website.

9.5.6 Little’s Rule Any queuing system can be subjected to a well-known principle of queuing called the Little’s rule (see Gross and Harris, 1998 or any standard text on queuing for additional details). This principle says that:

𝐿𝐿 = 𝜆𝜆 𝑊𝑊

where L denotes the average length, W denotes the average wait, and λ denotes the rate of arrival of customers into the system. In other words, if one knows the arrival rate, then knowledge of the mean wait in the system can be computed from the mean length and vice-versa. This is a very useful principle widely used in analysis of queuing systems. A simple application is as follows. Let the mean length in a queue, usually denoted by Lq, be 25 and the mean rate of arrival be 5 persons per minute, the mean waiting time in the queue will be: Wq = 25/5 = 5 minutes. This kind of analysis can be applied to any queue or even a queue, which is a part of a queuing network (which has several interconnected queues). If the server is a machine in a production shop, the mean length in the system usually denotes the work-in-progress (WIP) inventory associated with the machine. It turns out that Little’s rule has been extensively used in analyzing lean manufacturing systems (e.g., Askin and Goldberg, 2002).

9.5.7 Single-Server Single-Channel Queues Single server single-channel queues are those in which all customers stand in one line and there is only one server. These are usually the simplest types of queues to model. Generally, the ratio of the arrival rate to the service rate in such queues is called ρ, i.e.,

where μ denotes the mean rate of service by the server. It can be shown that ρ also equals the proportion of time the server is busy, i.e., the utilization of the server. 119

Engineering Management Handbook

We will discuss four types of single-server queuing models. We will assume that the server is available at all times and that the queue is stable. To describe these models, we will first introduce some commonly used queuing notation, the Kendall-Lee notation. A queue is often described with the symbolic: X/Y/n/s, where X denotes the distribution of the inter-arrival time, Y denotes the same for the service time, n denotes the number of servers, and s denotes the capacity of the waiting line in the queue. When s is omitted, it implies that there is infinite waiting capacity in the queue. M is commonly used to represent the exponential distribution and G is used to stand for general, i.e., any given arbitrary distribution. Thus, an M/G/1 queue is a single-server queue whose inter-arrival time is exponentially distributed and service time is generally distributed (has any arbitrary distribution).

M/M/1 Queues The continuous-time Markov Chain model discussed above can be used to generate the following expression for the mean length of an M/M/1 queue:

where ρ denotes the utilization of the server (as defined previously). Via Little’s rule, one can compute the mean wait in the queue. The mean wait in the system, W, for this queue is given by:

which follows from the fact that the mean wait in the system is a sum of the mean wait in the queue, Wq, and the mean time in service at the server which is 1/μ. From W, via Little’s rule, it is easy to compute the value of L, the mean number in the system.

M/G/1 Queues This queue has been studied extensively over the years, but the most useful result related to this queue was developed by Pollaczek and Khintchine in the 1930s. Their result, usually called the Pollaczek-Khintchine formula, produced the following exact expression for the queue length:

where σS2 denotes the variance of the service time. Little’s rule (Equation 1) and Equation 2 can then be used to determine the mean wait in the queue and the system. This formula is very useful in production systems and in telecommunication systems.

G/G/1 Queues This is the most general model that has the greatest applicability. Unfortunately, no closed-form formula exists for its basic performance measures (queue length or wait). Marchal (1976) developed a very useful approximation that works extremely well in practice. We present it next, along with some additional quantities needed for this model. Let: • C2: coefficient of variation of a random variable, i.e., C2 = variance /(mean)2 • : variance of the inter-arrival time 2 • Cs : coefficient of variation of the service time • Ca2: coefficient of variation of the inter-arrival time

120

Operations Research

Clearly, then,

Marchal’s formula is:

The previous equation is used in measuring waits in telecommunication systems, analysis of waiting times in amusement parks, and in measurement of inventory in Kanban-controlled systems. It is to be noted that Little’s rule and Equation 2 can be used in the context of this queue as well. Under heavy traffic, i.e., roughly speaking when:

a much superior approximation, called the heavy traffic result, exists for the G/G/1 queue (see Medhi, 2002); this result makes use of a stochastic process called Brownian motion (discussed later in the chapter).

9.5.8 Multiple-Server Queues Not all queues have single servers. When a queue has multiple servers, it can be a multiple server single-channel queue, in which all customers wait in one line, or a multiple-server, multiple channel queue in which customers wait in different lines but are served by a pool of servers. Such queues are more difficult to analyze. Oftentimes, simulation is the only way to evaluate the performance of these queues. See Gross and Harris (1998) or Medhi (2002) for an in-depth discussion of topics related to multiple-server queues.

9.6 Advanced and Other Topics Traditional OR topics will only take the operations researcher so far before problems become so complex that they can no longer be solved in a reasonable amount of time. With the advent of Big Data systems, the problem is only compounded. Meta-heuristics are often employed to efficiently search a large solution space for integer and mixed integer programs. Markov processes can be further exploited in Markov Decision Processes and Brownian Motion. Finally, due to problem complexity, mathematical models are sometimes avoided altogether in favor of creating a system model using discrete event simulation. This section provides an introduction to and application areas for these advanced topics, and concludes with an introduction to Big Data and its impact on OR.

9.6.1 Meta-heuristics Meta-heuristics are relatively new heuristic techniques that attempt to solve those problems arising in discrete and combinatorial optimization which cannot be solved via exact methods such as branch and bound or dynamic programming. The field of discrete optimization is well-known for problems that do not have any nice structure. Methods that produce global optimal solutions usually break down on these problems due to their size or the structure. Another characteristic of these problems is that there are usually multiple optima and derivative-based techniques can get trapped in local optima, which are not necessarily the global optima. On such problems, a class of methods called meta-heuristics has emerged in recent times. Because a very large number of real-world problems share these characteristics, this remains an important challenge in the field of OR. The meta-heuristic solutions in general tend to start at any arbitrary solution and “move” to better solutions in their neighborhood. Hence, meta-heuristics are also sometimes called neighborhood search methods or local search methods. Some meta-heuristics, however, tend to search over domains of the objective function that are not in the neighborhood of the current solution. The three best-known meta-heuristics are: genetic algorithms (Holland, 1975), simulated annealing (Kirkpatrick, Gelatt, and 121

Engineering Management Handbook

Vecchi, 1983), and Tabu Search (Glover, 1986; Hansen, 1986). This list is not exhaustive, however, and other options include ant colony optimization, greedy randomized adaptive search procedure (GRASP), variable neighborhood search, and iterated local search to name only a few (Blum and Roli, 2003). Genetic algorithms tend to rely on a stochastic search in which the move from the current solution to the next is guided by a random process that uses ideas from evolution. Simulated annealing is also a stochastic search algorithm in which the move from the current solution to the next is inspired by a metallurgical process. Tabu search is a so-called “recency-based” search technique in which the move from the current solution to the next is based on the moves that have actually occurred in the recent history of the algorithm. All of these algorithms have their advantages and drawbacks. See Pham and Karaboga (2000) for a discussion of their details and applications in engineering. The genetic algorithm uses a strategy of cross-over that occurs in chromosomes during reproduction. The strategy for selecting an improved solution mimics that of the evolutionary processes such as mutation. Typically, in a genetic algorithm, one moves from a population of solutions to another. This is one of the first biologically inspired meta-heuristics. Simulated annealing as stated above relies on a technique used in metallurgy for strengthening metals to obtain high quality solutions. In the annealing technique in metallurgy, the temperature of the liquid metal is lowered at a sufficiently high rate so that it acquires some desirable properties, e.g., mechanical strength, as it solidifies. In simulated annealing, the role of temperature is played by the probability of moving to a solution that is not better than the current solution. As the algorithm progresses, this probability is reduced; in other words, the temperature is reduced. It is hoped that during the initial stages of the search process, the algorithm avoids all the local optima and escapes out of them, but after searching the solution space for a sufficiently long time period, eventually gets trapped in a local optima that is also a global optimum. Tabu search is different from the two techniques described above in that it is not usually a stochastic search. However, in tabu search moving from one solution to a better solution is quite flexible. The strategy to move from one solution to another can exploit the structure of the problem if it is known. What is important in tabu search is that a list is maintained of recent moves. If a move has been made recently and is in the list, also called tabu list, it is not permitted. This ensures that the algorithm does not repeatedly traverse the same set of solutions. We would like to note that meta-heuristics are very widely applied in industry for solving large-scale and complex combinatorial optimization problems. Some of the noteworthy applications of meta-heuristics have occurred in the area of machine scheduling, transportation routing of large supply chains, logistics management, layout designing, VLSI circuit design, and airline scheduling.

9.6.2 Advanced Stochastic Models Markov Decision Processes Markov Decision Processes (MDPs) are decision-making problems associated with systems that can be modeled via Markov Chains. Essentially, MDPs can be viewed as applications of control theory to optimizing stochastic systems. The area of MDPs have seen a great deal of research activity in the last 50 years, and there are now numerous applications of MDPs in a variety of fields ranging from marketing to medical disease treatment. Our goal here is to present the main ideas, and name a number of application areas for the engineering manager.

Mathematical Model We will explain the mathematical model underlying an MDP with a simple example. Consider a 2-state Markov Chain in which two actions are permitted in each state. For instance, in the example considered in the Limiting Probabilities section in this chapter, the two actions could be “Repair the Machine” and “Do Nothing.” Usually, a unique TPM is associated with every action. Also, in the MDP model, one has a so-called Transition Reward Matrix (TRM), which contains data for the immediate reward gained in going from one state to another. Consider the following MDP: 122

Operations Research

where TPMa and TRMa denote the TPM and TRM, respectively, associated with action a. The element in the ith row and jth column of the matrix TRMa denotes the reward gained from selecting action a in state i and transitioning to state j as a result. Like the TPM, a unique TRM is associated with each action. A typical goal of the MDP is to consider the evolution of the system over the long-run, i.e., an infinite time horizon. A policy is a map from the state space to the action space. A deterministic policy specifies a unique action for each state in the system. Using optimization techniques, it is possible to determine the “best” deterministic policy that optimizes some performance measure, provided an optimal deterministic policy exists. Using the TPMs and the TRMs, it is possible to develop expressions for the long-run average reward per unit time or the total discounted reward earned over the long run. We will present an expression for the first measure, average reward. Let p(i,a,j) denote the term in the ith row and jth column of TPMa and r(i,a,j) denote the corresponding term in TRMa. Then, the average reward, ρ, of a policy in which action a is chosen in every state is given by:

where denotes the limiting probability of state i when action a is chosen in every state. Clearly, ρ depends on the policy, and the optimal policy is one for which ρ is maximized. One way to determine the optimal policy is to evaluate the objective function of every policy, which can make the problem very difficult to solve because typically there is a very large number of policies; fortunately methods based on linear programming and dynamic programming have been devised, which are more efficient for solving these problems (Bertsekas, 1995). Dynamic programming methods originated from the work of Bellman (1957) and Howard (1960). Recent developments in the field of reinforcement learning (Bertsekas and Tsitsiklis (1996); Sutton and Barto (1998); Szepesvari (2010); Gosavi (2014) have allowed managers to solve an MDP within a simulator of the system without access to the TPMs, which may be notoriously hard to find for complex and large-scale systems.

Applications MDPs have been used widely in a variety of domains. A critical difficulty in applying the MDP model in the real world has always been the task of generating the transition probabilities. If this can be overcome or circumvented (possibly via simulation-based methods such as reinforcement learning), the MDP provides high-quality solutions that can outperform industrial heuristics. We will now discuss some key applications of MDPs in the industry. Inventory Control: Many of the fundamental inventory control models such as (S,s) strategies have been analyzed via MDPs. In the area of supply chains MDPs continue to provide useful insights that often lead to reduction of costs and increase of profits. The interested reader is referred to Fox, Barbueanu, and Teigen (2000) for a tutorial on how the MDP framework forms the underlying basis for agent-based architectures in modern supply chain management software. Communication Networks: Management of cell-phone networks and asynchronous transfer mode (ATM) networks requires the solution of routing problems and so-called admissions control problems. Many of these problems are usually set up as MDPs and solved via the techniques enumerated above (Altman, 2002). 123

Engineering Management Handbook

Revenue Management: Revenue management is a broad field that encompasses study of pricing problems in the context of resource allocation. The problem of allocating seats to “fare classes” is a central one for the airline industries and hotels that use revenue management extensively in their operations. The revenue management can be set up as a finite horizon MDP (Subramaniam, Stidham, and Lautenbacher, 1999). A related infinite horizon MDP has also been solved by reinforcement learning (Gosavi, 2004; Gosavi, Bandla, and Das, 2002). Total Productive Maintenance: Total productive maintenance is a management philosophy popularized by Toyota motors as a part of “lean enterprise management.” The problem of preventive maintenance of production-inventory systems has been studied via an MDP model in Schouten and Vanneste (1995) and a semi-Markov Chain model in Das and Sarkar (1999). MDPs provide the most accurate policies for preventive maintenance in production systems.

9.6.3 Brownian Motion Brownian motion was first discovered as a mechanism to describe the motion of particles in liquids in 1827 by Robert Brown. A number of other mathematical models were subsequently developed for the same, but the most successful models became available only as late as 1905 with the dissertation of Einstein (see Einstein, 1956) and the work of Smoluchowski (1906). Norbert Wiener is credited with having developed a rigorous definition of Brownian motion (also called a Wiener process in his honor). An important fact about Wiener’s definition is that it is abstract and allowed the model to be used as an operational model outside of physics. Although we will not delve into the mathematical details of Brownian motion here, it is important to point out that it can be explained as a type of limit of the random walk, which is a specific class of Markov Chains. We will limit our discussion to some of the key applications of Brownian motion in operations research. Financial Engineering: Black and Scholes (1973) developed the famous Black-Scholes formula that uses a Brownian motion model to describe the fluctuation of stock prices. (This work, in combination with parallel work completed by Merton (1973) has been awarded a Nobel Prize in economics.) These ideas have been subsequently extended to numerous other topics in the study of stock markets, options, and other monetary instruments offered by banks and financial institutions. It is the case that almost all the mathematical models in financial engineering rely on some sort of stochastic process. This is an area of ongoing research. Queuing Networks: A queuing network is a system in which customers from one queue join another queue(s) in the system. Usually the queues are inter-connected. Queuing networks are notoriously hard to analyze. Queuing networks are very common in manufacturing systems with serial or forked transfer lines. Queuing networks are common in particular in semi-conductor manufacturing. Starting with the work of Kingman (1962), Brownian motion has been used to study the single-server single channel queue under the so-called heavy traffic conditions. This body of work has now been significantly expanded and provides some very powerful results in the analysis of queuing systems including queuing networks (Chen and Yao, 2001). This is also an area of active research that holds tremendous promise of generating closed-form models for performance analysis of complex queuing systems and scheduling of workloads in queuing networks. Inventory Control: Brownian motion has been used to model inventory (Sethi and Thompson, 2000) in production systems. Brownian motion has also been used to model the behavior of a deteriorating quality control process. Although it is unclear how many of these models have been actually used by practicing engineering managers, because many of these works are still in the early stages of research and development, it is well-known that many operations research models in the stochastic area that started out as theoretical concepts are now widely applied and used in the industry, examples being the news vendor in 124

Operations Research

supply chain management (Wong, Qi, and Leung, 2009) and Littlewood’s equation (Littlewood, 1972) in revenue management.

9.6.4 Discrete-Event Simulation Discrete-event simulation is a numerical or computational approach for evaluating the performance of a discrete-event system; a discrete-event system is one in which the time between two successive events is usually a finite quantity. Some examples of events in a queuing system are: the arrival of a new customer, service completion of a customer in the system, breakdown of the server, etc. Typically, the time between such events is a random variable whose distribution is known, or it is a function of random variables whose distributions are known. This time could also be a deterministic variable, and the simulation is called deterministic simulation. Unless specified otherwise, discrete-event simulation is usually understood to be for a stochastic system. While simulation is the topic of another chapter, we will highlight its relationship with OR. In a discrete-event simulation, one samples values of the random variables (underlying the system) from their distributions. These random variables are generally called the input variables for the simulation. The samples generated are used to determine the time interval between successive events. Essentially, the mechanism underlying discrete-event simulation is to generate the time intervals between successive events and collect statistics about the system from every event generated in this process of “recreating” events within the simulator. Because this is a time-consuming and tedious process, it is best left to the computer. The time (in the simulation environment) of every event that can occur in the system is also determined and stored in the computer. The sequence of events or “event list” along with the “simulation clock,” which is simply a list of the actual times of each event stored in a chronological order, play a key role in ensuring that the events are regenerated properly within the computer. A simulation program is generally written with the intention of collecting statistics about one or more parameters of interest (performance measures). Oftentimes, these parameters are random variables themselves and are commonly called the output variables or response variables from the simulation. Usually, the value of the parameter is obtained by averaging over several values. Hence, in order to obtain the desired average value of the parameter(s), one must update the value of the parameter after every relevant event. This process of updating values of the output variables must occur simultaneously as the events occur. For more details on the inner workings of a simulation program, the reader is referred to Law and Kelton (2000). A large number of software programs are now available in the industry. These software programs are very user-friendly and are very popular in the industry. Simulation is a very powerful tool for evaluating the performance of a system. For analyzing stochastic systems, it is one of the most widely used operations research tools in the industry (Carson et al., 2004). Usually, if the distributions of random variables are available, it is possible to simulate the system. However, simulation is not used if a closed-form mathematical model is available. This is because the mathematical formula can be used easily within spreadsheet or some other software for analysis of the system. It is usually the case that writing the computer program in the simulation package requires a significant amount of effort and expertise, and it also takes a significant amount of time to run the program and obtain results. Hence, simulation is not the preferred model if a mathematical model can be developed. However, with complex systems, it is oftentimes very difficult to develop exact or even near-exact mathematical models. It is in cases like these that simulation becomes extremely useful. Needless to add, most real-world systems produce complex scenarios for which closed-form models are not available, and one must take resort in simulation.

Simulation and Optimization Simulation is very effective at answering what-if kind of questions about a given system, e.g., what will be the throughput if an additional machine is purchased. Even today, a majority of applications of simulation in the industry are geared toward this end. However, optimizing a system via simulation is a relatively difficult idea because simulation is like a black box and provides the numerical value of the objective function at any given point. Also, simulation takes a significant amount of time to generate the objective

125

Engineering Management Handbook

function value. Hence, combining simulation with optimization techniques can be challenging. However, recent breakthroughs in continuous optimization, e.g., simultaneous perturbation (Spall, 1992) and in control theory, e.g., reinforcement learning (Watkins, 1984) has made it possible to combine simulation with optimization (see Gosavi, 2003 for an overview of this topic).

Applications Simulation is very widely used in the military for planning their operations, by health-care service providers such as hospitals and ambulance services, for analyzing electrical communication networks, for analyzing the operations of an amusement park, and of course in production systems. In production systems, the uses of simulation are many: measuring the throughput of a system, measuring the inventory in a system and the lead time of a new product, analyzing the throughput and inventory of a system with automated guided vehicles (AGVs), and detecting the bottleneck in a system. Oftentimes, the analysis in the systems described above revolves around measuring the time taken to perform an activity, waiting times in queues involved in these systems, and the utilization of servers in the queues. Very importantly, simulation can provide critical answers to what-if questions that can lead to improvements in the design of the system.

9.7 Big Data and Operations Research Since the first published academic article in 2008, the topic of Big Data has seen an exponential climb in published literature (Wamba et al., 2015). This rise in popularity is due not only to the large number of challenges associated with Big Data, but also is significantly due to the potential for Big Data to provide new insights impacting all facets of business decisions. Big Data, and it’s closely related cousin the Internet of Things, fall directly into the existing OR Methodology discussed in Section 9.1.3, with the data collection phase impacted the most. This section covers a description of Big Data and how it relates to OR, the Big Data value chain, and two case studies of Big Data applications in industry.

9.7.1 An Introduction to Big Data and the Internet of Things

126

Big Data is often differentiated from large or massive amounts of data by discussing the Vs. Depending on the framework chosen there can be between three to five Vs, here we have chosen White’s (2012) 5 Vs representation: volume, variety, velocity, veracity, and value. Volume, as the name suggests, refers to the sheer quantity of data. The data volumes experienced in true Big Data situations should exceed the standard capacities and capabilities of traditional relational database data management systems (Chen et al. 2014). Minimally, there should be terabytes of data to process, though often the data quantity is orders of magnitude larger. Variety refers to the types and structure of data collected. Big Data datasets should have a mix of several data types including structured data such as database entries, unstructured data such as free text, time series data such as weather data or click streams, video, geospatial, and image data. The velocity characteristic of Big Data refers to the frequency of receiving new data points. Big Data datasets are constantly growing at rapid rates, making the development of real-time data analytics a prevailing challenge. The veracity of the data refers to the trustworthiness of the data. Big Data datasets may often be noisy, contain duplicate data, or due to the data sources themselves have underlying trust issues. The final attribute is value. Value describes the idea that there is economic value to be found in analyzing the data as a Big Data dataset, but the individual data points on their own provide little value. Coupled closely with Big Data is the idea of The Internet of Things, often abbreviated IoT. The Internet of Things is a Big Data instantiation where a significant proportion of the data comes from sensors embedded in devices and machines. By collecting and processing in real time, these sensor data and machines can work together to optimize process and system efficiencies. The machines and devices producing data could be cell phones, vehicles, a factory CNC machine, street lights, solar panels, wind mills, subway trains, busses, household appliances, rfid scanners, or stock market tickers to name only a few options. An example instantiation of the Internet of Things is the Smart Grid. In the Smart Grid, households have Smart Meters that receive price signals from the utility. These price signals can be relayed to appliances and HVAC systems

Operations Research

within the premise in order to change their settings to accommodate the new price of electricity. Meanwhile, the Smart Meters are reporting back to the utility usage data at frequencies of up to every 5 minutes. Data is collected and analyzed by the utility on these usage patterns. Optimal pricing strategies are developed in order to reduce peak demand on the electric grid. Smart Meters also relay outage information back to the utility. The utility can then use that information to optimize restoration efforts. Notice, even in this Smart Grid example, that despite the large amount of data, the data is still being used for traditional OR tasks: optimizing a restoration plan, optimizing price signals to produce a desired demand pattern, and optimizing appliance function to minimize electricity cost. Despite the large volume of data generated by the Internet of Things and enterprise Big Data systems, the same OR questions still apply to the system as a whole. The ORer still wants to maximize profits, minimize waste, and generally use resources most efficiently as possible. The connection to Big Data is that there is now more data to work with and with that comes more areas for optimization. Big Data frameworks have significant impact on Step 2: Data Collection of the OR Methodology presented in Section 1.3. Here the Operations Researcher can aid in identifying data collection requirements, and can benefit significantly from the data collected. The additional data available to Operations Researchers in a Big Data enterprise promises to improve the model validity and the value add for OR based analyses.

9.7.2 Big Data Value Chain Executing Big Data projects can be divided into four main steps, each requiring increasing effort, while simultaneously providing increased value. The steps involved in the Big Data value chain are data generation, data acquisition, data storage, and data analysis as outlined by Chen et al. (2014) and seen in Figure 9.7. Data generation requires the least effort and likewise provides the least value. Data is generated from diverse sources including sensors, video feeds, Internet click streams, social media, transaction information, logistics data, energy data, and vehicle data, to name only a few sources. Creating or generating the data is followed by the task of actually recording the data, so that it can be analyzed at a later point in time. Recording the data is part of the data acquisition stage. Data acquisition consists of data collection, data transmission, and data pre-processing, all of which occur prior to data storage. Of the three parts of data acquisition, data pre-processing plays the most significant role in affecting the quality and speed of the data analysis phase. Data can be pre-processed to remove noisy or anomalous data points, to remove data from faulty sensors, to fuse data with other sources, and to reduce data redundancy thereby minimizing storage requirements. Figure 9.7. Big Data Value Chain

127

Engineering Management Handbook

Once data is generated, acquired, and transmitted the next phase is data storage. Storing Big Data requires the use of new information technology infrastructure that may or may not exist in your current enterprise. Due to the size of Big Data, storage systems require data be distributed across multiple interconnected servers, and consideration needs to be given to consistency, availability and partition tolerance (Brewer, 2000). Distributed file systems are fairly mature and include open source options such as HDFS and Kosmosfs. Depending on the type of data being stored, you may want to have a key-value database such as Dynamo, a Column-Oriented Database such as HBase or Google’s BigTable, a document database such as MongoDB or CouchDB, or a combination of two or more of the above. Finally, after the data have been generated, acquired and stored comes the analysis phase. It is in the data analysis phase that OR resides. While common data analysis tools such as R or CPLEX can still be used on subsets of big data datasets, significant progress has been made in exploiting Big Data distributed frameworks, such as Map Reduce, to solve complex integer programs (Chandu, 2014).

9.7.3 Big Data Case Studies Big Data in practice requires the integration of disparate data sources often found across different departments of an organization. Bringing these sources together not only requires technical skills, but more importantly strong leadership and the ability to bring diverse groups together to see the good of a common vision. Here two case studies are presented that highlight the value that can be achieved by creating a Big Data vision for your company. Electricity Theft Detection: TROVE, a predictive data science company, in partnership with Teradata (2015), a commercial big data information management and analytics platform, provide details on the use of big data to identify incidences of electricity theft for a large west coast US utility. Like many progressive utilities, when this utility transitioned to Smart Meters, they lost their eyes in the field as meter readers no longer needed to visit meters. Prior to using a Big Data strategy, the utility’s primary theft detection system was focused on leads generated by customers calling in to report suspicious activity. In a single year, 15,000 such leads were investigated, generating a hit rate of only 30% true incidence of theft. By instead focusing on their many internal data sources, the utility was able to improve their performance. TROVE was able to increase the lead hit rate from 30% to 86% by fusing smart meter data with customer demographic information, premise data, and other available sources, and then applying segmentation and optimization algorithms. Emergency Management: Wamba et al. (2015) provided details of a longitudinal case study on the use of Big Data by New South Wales State Emergency Service (NSW SES), which integrated both structured and unstructured data to form a comprehensive picture of the state of their 17 regions. The data integrated information sources including real-time volunteer staff availability, GIS, equipment availability and location, weather services, police service data, fire and rescue readiness, rural fire readiness, public information services, public welfare services, health services, energy and utility services, and environmental services. By integrating all these systems, the NSW SES was able to enhance their situational awareness enabling them to evaluate and assess mission readiness. The system allowed for the improved coordination of emergency response by coordinating 17 regional headquarters, including current staffing and GIS systems, to track and monitor mission completion. Furthermore, the system allowed visibility into “who and where,” which allowed for the reallocation of assets across the region in order to maximize emergency readiness. Finally, with visibility into the entire system, NSW SES was able to optimize future investments by minimizing current vulnerabilities. Cited as crucial for the system to work was active engagement from both the implementation team and top management.

9.8 References

128

Altman, E., “Applications of Markov Decision Processes in Communication Networks,” In Handbook of Markov Decision Processes, edited by E. Feinberg and A. Schwartz, Chapter 15, 2002, pp. 489-536. Askin, R., and Goldberg, J., Design and analysis of lean production systems. New York: Wiley, 2002.

Operations Research

Banks, J, Carson, J, Nelson, D, and Nicol, D., Discrete-Event System Simulation, 4th edition, Prentice Hall, 2006. Bellman, R. E., Dynamic Programming. Princeton, NJ: Princeton University Press, 1957. Bertsekas, D. P., Dynamic programming and optimal control. Belmont, MA: Athena Scientific, 1995. Black, F. and Scholes, M., “The Pricing of Options and Corporate Liabilities.” The Journal of Political Economy, vol. 81, no. 3, 1973, pp. 637-654. Blum, C., and Roli, A., Metaheuristics in Combinatorial Optimization: Overview and Conceptual Comparison. ACM Computing Surveys, vol. 35, no. 3, 2003, pp. 268-308. Brewer, E., Towards Robust Distributed Systems. In PODC, pg 7, 2000. Chandu, D., “A Parallel Genetic Algorithm for Three Dimensional Bin Packing with Heterogeneous Bins,” International Journal of Computer Trends and Technology, vol. 17, no. 1, 2014, pp. 33-38. Chen, H. and Yao, D., Fundamentals of Queueing Networks: Performance, Asymptotics, and Optimization, New York: Springer, 2001. Chen, M., Mao, S., Zhang, Y., and Leung, V. Big Data: Related Technologies, Challenges and Future Prospects. New York: Springer, 2014. Das, T.K. and Sarkar, S., “Optimal Maintenance in a Single Machine Production Inventory Systems,” IIE Transactions on Quality and Reliability Engineering, vol. 31, no. 6, 1999, pp. 537-551. Denardo, E. V., Dynamic Programming: Models and Applications, New York: Dover Publishers, 2003. Einstein, A., Investigations on the Theory of Brownian Movement. New York: Dover Publishers, 1956. Fox, M., Barbueanu, M., and Teigen, R., Agent-Oriented Supply-Chain Management, International Journal of Flexible Manufacturing Systems, vol. 12, 2000, pp. 165–188. Gass, S. I. and Arjang, A., An Annotated Timeline of Operations Research: An Informal History, Robert H. Smith School of Business, University of Maryland, 2004. Glover, F., “Future paths for integer programming and links to artificial intelligence.” Computers and Operations Research, vol. 13, 1986, pp. 533-549. Gosavi, A., “A reinforcement learning algorithm based on policy iteration for average reward: empirical results with yield management and convergence analysis.” Mach Learning, vol. 55, no. 1, 2004, pp. 5-29. Gosavi, A., Simulation-Based Optimization: Parametric Optimization and Reinforcement Learning. Kluwer Academic Publishers (now Springer), 2003. Gosavi, A., Bandla, N., & Das, T., “Airline seat allocation among multiple fare classes with overbooking,” IIE Transactions, vol. 34, no. 9, 2002, pp. 729-742. Gross, D. and Harris, C. M., Fundamentals of Queueing Theory, Wiley Inter-science, 3rd edition, 1998. Hansen, P. “The steepest ascent mildest descent heuristic for combinatorial programming,” In Conference on Numerical Methods in Combinatorial Optimisation, Capri, Italy, 1986. Hewiit, M.R., Chacosky, A., Grasman, S. E., and Thomas, B. W., “Integer Programming Techniques for Solving Non-Linear Workforce Planning Models with Learning.” European Journal of Operational Research, vol. 242, no. 3, 2015, pp. 942-950. Hillier, F. S. and Hillier, M. S., Introduction to Management Science, 3rd edition, McGraw-Hill, 2008. Hillier, F. H. and Lieberman, G. J., Introduction to Operations Research, 9th edition, McGraw-Hill, 2010. Hillier, F. S. and Hillier, M. S., Introduction to Management Science, 4th edition, McGraw-Hill, 2010. Holland, J. H., Adaptation in Natural and Artificial Systems. Ann Arbor, MI: University of Michigan Press, 1975. Howard, R. A. Dynamic Programming and Markov Processes, The M.I.T. Press, 1960. Kharoufeh, J., Solo, C., and Ulukus, M. Semi-Markov models for Degradation-Based Reliability. IIE Transactions, vol. 42, 2010, pp. 599-612. Kingman, J. F. C., “On Queues in Heavy Traffic.” Journal of the Royal Statistical Society. Series B (Methodological), vol. 24, no. 2, 1962, pp. 383-392. Kirkpatrick, S., Gelatt, C. D. and Vecchi, M. P., “Optimization by simulated annealing.” Science, vol. 220, no. 4598, 1983, pp. 671-680. Law, A. M. and Kelton, W. D., Simulation Modeling and Analysis, Third Edition, McGraw Hill, 2000. Littlewood, K., “Forecasting and control of passenger bookings.” In Proceedings of the 12th AGIFORS (Airline Group of the International Federation of Operational Research Societies) Symposium, pp. 95-117, 1972.

129

Engineering Management Handbook

Marchal, W. G., “An approximate formula for waiting time in single server queues,” AIIE Trans, vol. 8, 1976, p. 473. Medhi, J., Stochastic Models in Queueing Theory, 2nd edition, Academic Press, 2002. Merton, R., “Theory of Rational Option Pricing,” Bell Journal of Economics and Management Science, vol. 4, 1973, pp. 141-183. Nodem, F., Kenne, J., and Gharbi, A. “Simultaneous Control of Production, Repair/Replacement and Preventive Maintenance of Deteriorating Manufacturing Systems.” Int. J. Production Economics, vol. 134, 2011, pp. 271-282. Pham, D. T. and Karaboga, D., Intelligent optimisation techniques. Springer, 2000. Ross, S. M., Introduction to Probability Models, Ninth Edition. Academic Press, 2006. Sethi, S. P., and Thompson, G. L., Optimal Control Theory. Kluwer Academic Publishers (now Springer), 2000. Smoluchowski, M. Zur kinetischen Theorie der Brownschen Molekularbewegung und der Suspensionen, Annalen der Physik, 21, 1906, pp. 756-780. Spall J. Multivariate stochastic approximation using a simultaneous perturbation gradient approximation. IEEE Trans. Automat. Contr., vol. 37, no. 3, 1992, pp. 332-34. Subramaniam, J. Stidham, S., and Lautenbacher, C., “Airline yield management with overbooking, cancellations and no-shows,” Transportation Science, vol. 33, no. 2, 1999, pp. 147–167. Sutton, R. and Barto, A., Reinforcement Learning: An Introduction, Cambridge MA: MIT Press, 1998. Szepesvari, C., Synthesis Lectures on Artificial Intelligence and Machine Learning: Algorithms for Reinforcement Learning. Morgan & Claypool Publishers, 2010. Teradata. TROVE Predictive Data Science - Revenue Protection Application. http://www.teradata.com/ brochures/TROVE-Predictive-Data-Science-Revenue-Protection-Application. Accessed August 19, 2015. Tomasevicz, C. and Asgarpoor, S., “Optimum Maintenance Policy Using Semi-Markov Decision Processes,” Electric Power Systems Research, vol. 79, 2009, pp. 1286-1291. The Institute for Operations Research and the Management Sciences (INFORMS) Website: http://www. informs.org/ The International Federation of Operational Research Societies (IFORS) Website: http://www.ifors.org/ Van der Duyn Schouten, F. A., and Vanneste, S. G., “Maintenance optimization of a production system with buffer capacity.” European Journal of Operational Research, vol. 82, 1995, pp. 323-338. Watkins, C. J. C. H. Learning from Delayed Rewards. PhD Thesis, Cambridge University, Cambridge, England, 1989. Wamba, S., Akter, S., Edwards, A., and Chopin, G. “How ‘Big Data’ Can Make Big Impact: Findings From a Systematic Review and a Longitudinal Case Study.” International Journal Production Economics, vol. 165, 2015, pp. 234-246. White, M., Digital Workplaces: Vision and Reality. Bus. Inf. Rev. vol. 29, no. 4, pp. 205-214, 2012. Wong, W. K., Qi, J. and Leung, S. Y. S., “Coordinating supply chains with sales rebate contracts and vendor-managed inventory,” International Journal of Production Economics, 2009.

130

Simulation

10 Simulation Andreas Tolk

The MITRE Corporation

131

Engineering Management Handbook

10.1 Introduction 10.1.1 Importance of Simulation Engineering Management bridges the gap between technical engineering processes and necessary administration and management processes. As such, engineering managers should be aware of developments that are happening in the technical domain as well as of possible support on the management side. Modeling and architecture methods have been in the scope of the body of knowledge for EM from its beginning, as models and architectures belong to the main tools of communication administrative and management needs in order to support technical solutions. With simulation, another logical element is added to the toolbox of engineering managers: models executed over time! Modeling, architecture, and simulation are connected in this chapter to support EM twofold. First, they are interpreted as a new category of quantitative tools and methods supporting EM. Second, they are increasingly the object of EM knowledge when applied as methods and tools in projects. Both aspects are important for EM, as the engineering manager needs to understand the formalisms, methodologies, and technology applied in order to utilize simulation in projects for which he or she is responsible. He or she needs to understand the technical engineering process of simulation as well as the administrative and management processes required for simulation. The general importance of simulation for engineering was featured among other publications in the 2006 NSF Report on “Simulation-based Engineering Science.” This report showed the potential of using simulation technology and methods to revolutionize the engineering science. Among the reasons for the steadily increasing interest in simulation applications are the following: • Using simulations is—as a rule—cheaper and safer than conducting experiments with a prototype of the real thing. One of the biggest computers worldwide is currently designed in order to simulate the detonation of nuclear devices and their effects in order to support better preparedness in the event of a nuclear explosion. Similar efforts are conducted to simulate hurricanes and other natural catastrophes. • Simulations can often be even more realistic than traditional experiments, as they allow the free configuration of environment parameters found in the operational application field of the final product. Examples are supporting deep water operation of the NAVY or the simulating the surface of neighbored planets in preparation of NASA missions. • Simulations can often be conducted faster than real time. This allows using them for efficient if-thenelse analyses of different alternatives, in particular when the necessary data to initialize the simulation can easily be obtained from operational data. This use of simulation adds decision support simulation systems to the tool box of traditional decision support systems. • Simulations allow setting up a coherent synthetic environment that allows for integration of simulated systems in the early analysis phase via mixed virtual systems with first prototypical components to a virtual test environment for the final system. If managed correctly, the environment can be migrated from the development and test domain to the training and education domain in follow-on life-cycle phases for the systems (including the option to train and optimize a virtual twin of the real system under realistic constraints even before first components are being built).

132

Especially for the engineering manager it is of particular interest that simulation starts to replace traditional experiments on a significant scale. The trend is most obvious in the military domain, but is also deeply rooted in industry traditionally using computer aided design, such as the automobile industry. Model-based design using virtual prototypes in realistic synthetic environments starts to emerge as a new engineering method. Another domain of interest is executable architectures. System architectures, as blueprint standards for complex systems, are an excepted tool. However, most of the current system architecture frameworks do not fully support the evaluation of the dynamic behavior of a system, as the artifacts used in these frameworks deliver more or less snapshots of the systems. However, software tools increasingly offer the option to “execute” a blueprint. This execution of a system’s architecture is a special case of applying the principles of simulation, as the execution of an architecture de facto simulates is functionality. This application

Simulation

requires EM knowledge in the disciplines of system architecture and modeling as well as in simulation. The tight connection between systems engineering processes, support system architecture artifacts, and simulation methods has been dealt with in detail by Tolk and Hughes (2014). As a result, simulation systems will be increasingly applied as tools in more and more engineering domains and require to be managed. Although the general domain of information systems is supported by good practices including those, documented in textbooks, such as Fuller, Valacich, and George (2008), simulation systems are more than just huge information systems requiring special knowledge by those who have to manage them over their complete life cycle. As models are purposeful abstractions of reality, and simulation systems are executed models via information systems, an additional layer of complexity is added to the information system challenge. In other words, Modeling & Simulation (M&S) opens a new challenging field for EM. In summary, M&S is a new way of understanding the interaction among parts of a system and the systems as a whole. Simulations allow engineers to dynamically change design decisions and immediately see the consequences. They can evaluate alternatives and options without creating risks or expensive prototypes. The level of understanding of complex systems supported by M&S surpasses most other disciplines. The U.S. Congress recognized the contribution of M&S technology to the security and prosperity of the United States and recognized M&S as a National Critical Technology in its House Resolution 487 in July 2007. All these points motivate the inclusion of simulation as a chapter into this book.

10.1.2 Key Terms of Simulation Every scientific discipline needs to establish a set of community-wide agreed key terms in order to formulate its body of knowledge in a coherent and concise manner. This is also the case for M&S. The application-oriented nature of this emerging discipline, however, sometimes results in the challenge, that the supported application domains themselves already have vocabularies in place that are not necessarily aligned between disjunctive application domains. This problem is comparable to the challenges of system architectures, where engineers face the same challenge. The terms introduced here are based on curricula definitions of these terms as captured in newer textbooks for M&S education, in particular Sokolowski and Banks (2008, 2010) and Yilmaz and Oren (2008). Modeling and Simulation can both be understood at least as sub-disciplines by themselves. Modeling is the purposeful abstraction of reality. These abstractions take place on several steps, starting with the high-level conceptualization of the real-world referents (the system or part of the real world that should be modeled) and resulting in coding decisions for the final implementation. The resulting artifacts must be captured in formal specifications of the conceptualization. Robinson (2008) defined the conceptual model as “a non-software specific description of the simulation model that is to be developed, describing the objectives, inputs, outputs, content, assumptions, and simplifications of the model.” Simulation is the execution of such a model. The simulation can be executed by hand, using physical models, and so on. In the context of this chapter, the focus lies on computer-based simulation. They can be written in general-purpose languages or in special simulation languages; or they can utilize configuration interfaces of high-level simulators. If the execution is done in parallel on several computers that may be spatially distributed, this becomes a distributed simulation. If a distributed simulation is based on several different simulation systems (using different underlying models), each simulation is referred to as a federate within a simulation federation. There are several simulation standards for distributed simulation. Simulation and simulation federations can be executed stand-alone or integrated into operational systems. For integrated systems, the application of system-of-systems engineering is good practice and topic of ongoing research. The body of knowledge for M&S identifies three simulation paradigms: Monte-Carlo simulation, discrete event simulation, and continuous simulation. All three will be dealt with in more detail in this chapter. It will be shown that these paradigms are not mutually exclusive, but they are often used in mixed forms to solve engineering problems in the real world. Simulation already is a critical component of many application domains. Although traditionally the military domain was the driving force behind standardization and applications for training, education, and even operational support, other domains are becoming equally important: medical applications, transportation, 133

Engineering Management Handbook

business, manufacturing, social sciences, and more. Whenever the term complex system can be applied, it is likely that simulation applications are used to understand the behavior of this complex system better.

10.1.3 Simulation Theory for Engineers Engineers in general and engineering managers in particular need to understand the theory of simulation in order to understand the valid range of applicability of simulation-based tools as well as to understand the challenges of managing simulation applications. There is a significant overlap with other disciplines, in particular operations research. On the one hand, simulation is often used as an additional powerful means of operations research. Mathematical descriptions of systems that cannot be solved mathematically can still be used to define the model for a simulation system enabling the engineer to gain insight into the system dynamics and generally the system behavior. Heuristics can be applied and sensitivity analysis can be conducted. On the other hand, simulations are also often used to produce quasi-heuristic data that needs to be evaluated by traditional means of operations research. In any case, the tight connection between both disciplines is obvious. The theory of simulation described in this chapter is divided into five categories. These five categories have been identified as the core aspects engineers and engineering managers must know when they want to apply simulation within their projects and maybe subject to change and additions over time, but they have been successfully used in engineering focused curricula development. The first category builds the mathematical foundations with focus on probability and random numbers that are essential for stochastic modeling approaches. The second category builds computer science foundations. There are several simulation languages and frameworks allowing students without a strong computer science background to utilize simulation methods as well, but some basic understanding in algorithms and data structures generally used in programming are essential to fully understand and utilize simulation. The third category deals with the theory of discrete event simulation, which is the predominant simulation paradigm in the engineering domain. Category four deals with the principles of data analyses needed for evaluation of results. Finally, the last category deals with Monte-Carlo simulation and continuous simulation as the two additional simulation paradigms of special interest to engineers. The objective of these contributions is to enable EM students to understand the theoretic foundations of simulation in order to enable them to manage their development and application within engineering projects and to be aware of potential and constraints of using simulation applications within his area of expertise. They should be able to write small simulation applications in a higher programming language, and should be able to set up a statistically meaningful simulation experiments and evaluate the results.

10.1.4 Simulation Applications for Engineers Although understanding the theoretic foundations is necessary, the predominate interest of engineering is to contribute to solutions by applying scientifically derived knowledge to solve real-world problems. The simulation applications are therefore an important part that comprises three categories. The first category describes simulation as an engineering method. This category relies on best practices on how to successfully use simulation in engineering projects. The second category focuses on simulation with ARENA. The main point is not to introduce a tool, but to focus on the engineering application using the tool as an example. The statistical experimentation design is an important aspect of this section. Finally, agent-based modeling is introduced in the third category. Individual agents and multi-agent systems are discussed and their application is shown. The objective of these contributions is to give examples for simulation applications for engineers. Student will not be experts in simulation applications, but will understand the principles in all three categories and will know how to apply the theory they learned in the first section in support of their engineering tasks. 134

Simulation

10.1.5 Simulation Engineering Finally it must be pointed out that simulation in the context of this chapter is clearly perceived as an engineering discipline. This does not exclude experience, heuristics, creativity, and the art of simulation, but the emphasis of this chapter is on methods and engineering principles supporting the development and application of simulations to solve engineering problems. As such, we focus on simulation as a tool in support of the engineer. In addition, simulation systems are perceived as “virtual” systems that require the same rigor in design, development, test, and application traditional systems need. Therefore, EM for virtual systems and systems of systems is in the scope as well, although this topic is emergent and topic of current research. In summary, simulation should be of high interest to every engineer due to its increasing practical relevance in projects and growing management demands within this professional support. The use of simulation in a project can positively influence schedule, performance, costs, and risks of engineering projects, if applied and managed correctly. This chapter outlines the topics each engineering manager should be aware of when utilizing simulation for engineering.

10.1.6 Modeling and Simulation as a Discipline Several universities are increasingly offering not only courses, but also degrees in Modeling and Simulation. Pioneers in these efforts were the Old Dominion University in Norfolk, VA, the Naval Postgraduate School in Monterey, VA, and the University of Central Florida in Orlando, FL. Simulation societies are supporting these efforts, in particular the Society for Modeling and Simulation (SCS) and the Special Interest Group on Simulation of the Association for Computing Machinery (ACM SIGSIM). The idea behind the curricula is to identify common principles and insights that allow to derive applicable methods and solutions that lead to solutions in the various application domains. This should foster reuse and dissemination of research results to the benefit of simulation users in all application domains. Figure 10.1 shows the important components as discussed by Padilla, Diallo, and Tolk (2011). Figure 10.1. Modeling and Simulation as a Discipline

EM is the art and science of planning, organizing, allocating resources, and directing and controlling activities that have a technological component. As engineering managers are responsible for recommending the best available engineering solutions, they have to understand this new technology and the underlying epistemological foundations of simulation. 135

Engineering Management Handbook

To be able to do this, a sufficient knowledge in the domain of modeling and simulation is needed. Kossiakoff and Sweet (2002) defined the role of a systems engineer in major system development projects as bringing specialist together that are characterized by different fields and disciplines with his or her own languages, experiences, and knowledge bases. The systems engineer needs to ensure that these diverse track converge in support of developing and producing a new system. Similarly, the role of the engineering manager dealing with simulation is to bridge the virtual worlds of simulation and the engineering world to allow successful application of modeling approaches and simulation systems to minimize negative environmental effects, either by avoiding unnecessary experiments and prototypes that could be supported by virtual systems, or by decision supports for the real system under development by clearly analyzing and communicating environmental challenges.

10.2 Simulation Theory 10.2.1 Mathematical Foundations One of the main applications for simulation in industry is to formulate a stochastic model to describe a real phenomenon. In order to use mathematics to analyze a situation using simulation, probability models are needed. One of the major elements in this probability models is the use of random number exhibiting required characteristics. In order to use the model, such random numbers must be generated for applications within the simulation. This section will introduce the main ideas of probability, random numbers, and how they can be generated. Among other valuable resources, these topics are well covered in Ross (2006), chapters 2 to 5. They are also explained in Kelton, Sadowski, and Sturrock (2007), chapter 12. A handbook chapter can never replace required in-depth studies of the referenced domain, and in the case of probability, this is definitely the case. Each applicant of simulation must be aware of the principles of probability in order to use simulation systems effectively. Even if the simulation system itself is deterministic, the simulation results may reflect elements of probability if statistical experimentation planning is used. This section can therefore be no more than a list of “things to know” and is a starting point or a checklist for the engineering manager. Several key terms need to be defined in order to deal with probability. Probability is always needed when the outcome of an experiment is not known in advance or uncertain. The set of all possible outcomes is the sample space. Each subset of this sample space is an event. For any event we define the complement of this event as all events that are in the sample space but not in the event. Two events are mutually exclusive if there are no outcomes that are a subset of both events. With these definitions, the three axioms of probability can be formulated. For each event A in a sample space S we denote the probability of A as P(A) in accordance with the following three axioms of probability:

Axiom 1:

0 ≤ P( A) ≤ 1

Axiom 2:

P( S ) = 1

Axiom 3:

For any sequence of mutually exclusive events A1, A2, …

 n  n P  Ai  = ∑ P( Ai ) ,  i =1  i =1

n = 1, 2, …, ∞

These axioms can be used to proof a variety of results about probabilities and are the basis for stochastic modeling. Of particular interest for simulation applications are furthermore conditional probability and independence. Conditional probability is defined as the probability that event A is observed under the condition that event B occurs. As we know that B occurred, B becomes the new sample space for the conditional probabil136

Simulation

ity. The events comprised in the conditional probability value must furthermore satisfy A and B, hence we need the probability of AB relative to the probability of B, so that the conditional probability is denoted by

P( A | B) =

P( AB) . P( B)

Following similar ideas, we define that two events A and B are independent if P ( AB ) = P ( A) P ( B ) . It can be seen that formulas for independent events are much easier to compute and preferred for simulation, but not for the sake of correctness and accuracy. The engineer must therefore know which events are independent and which events were assumed to be independent when the model for the simulation was developed. Within simulation, the probability is often only a means used to specify the value of some numerical quantity determined by the result of an experiment. Such numerical quantities are defined as random variables. Random variables can be discrete or continuous. If they are discrete, the overall probability can be determined by adding the single probability values. If they are continuous, intervals are needed that define the probability density function. Three of the most useful concepts in probability often used for simulation are the expectation of a random variable, the variance of a random variable, and the covariance of two random variables. The expected value is the weighted average of the possible values a random variable can take on weighted by the probability that the random variable assumes this value (or the integral over the random variable interval weighted by the probability density function). The variance measures the variation around the expected value by considering the average value of the square of the difference between the expected value and the observed value. The covariance measures the degree of similarity between two random variables by considering the expected value of the difference between the observed value and the expected value of the first random variable multiplied by the difference between the observed value and the expected value of the first random variable. The correlation is a very similar measure to the covariance, but the correlation has no dimension, as the covariance is divided by the square root of the product of the variances of both random variables. Ross (2006) introduced some selected discrete random variables that are useful for many simulation applications, such as • Binomial random variables: n independent events with two possible outcomes (yes/no, success/no success, etc.) with equal probability. If we are interested in one successful outcome (n=1), we are talking about Bernoulli random variables. This category of random variables is important for statistical significance testing. • Poisson random variables: They are used to approximate the distribution of successes in a large number of trials. They are also used to count the number of events that occur in a certain time interval. • Geometric random variables: If a series of n independent trials with the same probability p is conn −1 ducted, these variables can be used to compute the overall success probability by p (1 − p ) . Many simulation applications make use of these variables.

Most software development environments provide the user with a function called random or similar that returns a pseudorandom number. One of the most common approaches to generate pseudorandom numbers is to start with an initial value, called the seed, and then computing successive values by letting xnext = axcurrent mod m where mod is the modulo function and a and m are positive integers. This procedure generates values within the interval [0, 1, …, m-1] and is an approximation to the value of a uniform (0, 1) random variable. It is obvious that this is not a real random number, as the next upcoming number is completely determined by the formula and the series of numbers will repeat itself as soon as one number is repeated (which must happen at least after m executions).

137

Engineering Management Handbook

Once the user is able to generate uniform random variables, he or she can apply well-established methods to transform them into other random variables with known distributions (defined by the probability mass function for discrete random variables and the probability density function for continuous random variables). For discrete random variables, the inverse transform method is often used. For continuous random variables, the inverse transform algorithm is used. Both methods follow the same basic idea that can be explained best using the following figure. The graph shows the probability mass/density function of the targeted distribution. Both methods use a uniform (0, 1) random variable to generate a random number (1), determine where this event “hits” the desired accumulated probability mass/density function (2), and than determine which value corresponds this value on the desired random value scale (3) as shown in Figure 10.2. Figure 10.2. Visualization of the Principle of the Inverse Transform Method/Algorithm

Uniform (0,1) random variable

1 Desired Probability Mass/density function 1

0

2

min

3

max

Random variable with desired distribution

Ross (2006) describes both, the inverse transform method and algorithm, in detail and gives more established methods for special distributions. The interested engineering manager is referred to the references for more detailed introductions. For engineering managers, the current work on establishing a common general understanding of simulation theory based on common principles formulated in mathematical model theory can be of additional interest (Diallo, Padilla, Gore, Herencia-Zapana and Tolk, 2014).

10.2.2 Computer Science Foundations In order to understand simulations, it is highly recommended to be able to understand and write algorithms in a higher programming language, such as Java or C++. In the application section, we will deal with special simulation languages that are an alternative to general purpose programming languages; however, as it is necessary to understand the foundations of probability before being able to apply a statistics tool efficiently, it is necessary to understand the foundations of programming to understand simulation programs. Computers execute instructions. A computer program is a list of instructions including all necessary input data. An algorithm is an instruction-by-instruction procedure to solve a problem, such as ordering a set of numbers in ascending order, find the fastest way connecting two points, etc. 138

Simulation

The list of instructions is normally executed in sequential order, if this order is not modified using control flow statements that exist in all programming languages in one form or another. There are five categories of control flow statements: halt, choice, loop, jump, and subroutines that are defined as follows: • Halt statements stop the execution flow immediately and prevent any additional actions. • Choice statements execute a set of statements only if certain criteria are met. The most common choice statement is the conditional statement of the form IF condition THEN instruction. • Loop statements allow to repeat a set of statements for a given number or as long as a condition is true. Some of the most common loop statements are

REPEAT instruction UNTIL condition; WHILE condition DO instruction; FOR enumerator DO instruction.

• Jump or go-to statements go directly to a specified instruction in the program. This instruction is often labeled to make the reference easier. One of the most common jump statements is GOTO label. • Subroutines execute a set of instructions after which they potentially return to the calling instruction. Subroutines are normally called with a set of input parameters, such as SubroutineName (InputParameterList). Beside these control structures, a certain commonality of data structures can be found in all programming languages as well. The basic or primitive data item types are integer numbers, real or float numbers, and characters or strings. Often, the numbers are distinguished in long and short primitives, where long primitives require more storage space within the computer but also have a higher accuracy or bigger range. When two or more primitive data item types are composed into a new data item, they build a composite type. Primitive and composite types can be stored in ordered form, e.g., in lists and trees. Lists and trees are ordered collections of data items. Lists can be indexed allowing to access a certain element (these lists are often called array or vector), or they may only allow access to the first and/or last element of the list. When data structures and procedures working on these data are combined into a common construct, this is called an object (and the procedures are called methods). Many books have been published in the domain of data structures and algorithms that are normally used in undergraduate programs of computer sciences education. It is good practice to search for collaboration with a local college offering education in this domain. The EM body of knowledge enumerates the introduction to basic of Java or C++ as an important component. This topic alone fills books and each attempt to introduce these languages in a handful of sentences must be futile. Instead, the interested reader is referred to standard computer science literature as well as web sites offering introduction tutorials and overviews. In particular Java is supported by an active and competent online user community, such as http://java.sun.com and other websites.

10.2.3 Discrete Event Simulation While the mathematical and computer science foundation sections build the broad basis, discrete event simulation basics are the first topic that aims at the need of engineering managers looking specifically for simulation education. Discrete event simulation is one of the simulation paradigms. It models a system as it evolves over time as a series of system states. At discrete times, events occur that change the state of the system—or the state of a system component—instantly. In other words, discrete event simulation simulates events and state changes of the system connected with these events in chronological order. In order to be able to do so, the events must be stored and delivered in the right order. For this purpose, the simulation system uses at least one event list—or event queue— which stores events until they occur and distributes them when they are needed. Furthermore, a representation of time—a simulation clock—is needed that determines together with a time advance algorithm what timestamp is used to label the events and when and how to progress in

139

Engineering Management Handbook

time. In particular for distributed simulation system, where events can be produced in remote procedures, the synchronization of several event lists is an issue requiring time management. There are two principle approaches for time advance: fixed time steps and variable time steps. Figure 10.3 exemplifies both approaches. It shows discrete time steps tn, tn+1, and tn+2 that occur at predefined fixed points on the time scale, and the time stamps of three events te1, te2, and te3. Figure 10.3. Time Advance in Discrete Event Simulation Event 1

te1 tn

Event 2

Event 3

te2

te3 tn+1

time tn+2

If a next-event time advance method is used, the time advance starts with the first event in the event list (event 1) and uses the time stamp of this event to set the simulated time (te1). The system state connected with event 1 occurs, and the next event is request from the event list. If a fixed time advanced method is used, the time advance goes from time tn straight to tn+1. It the requests all events with a time stamp smaller or equal to tn+1; in the example, this are the events 1 and 2. When all events that fulfill this criterion are computed, the next time step is made to tn+2. In what order the events are computed is due to the implementing details, as the events principally are aggregated into one event composite that comprises all events that are happening in the interval within the current fixed time step. In both cases, an event may lead to the creation of a new event in the future. In this case, a new event is created, the parameters of the event are calculated, the timestamp for this event is calculated, and the event is handed over to the event list manager to be included at the right place. Challenges the simulation developer faces are those cases where the timestamp of the new event potentially lies in the past of the current simulated time. If in the example above a fixed time step approach is chosen, and event 1 creates a new event with a timestamp smaller then that of event 2, this may cause problems if the resulting state change caused by the new event is important for the computation of the correct effect of event 2. If the simulation systems are distributed, the necessity for time management arises, as the different simulation clocks must be synchronized. One possibility is to use real-time, but this creates two challenges: (a) when a simulation is computation intensive, it may be too slow to be able to run in real time; (b) when a simulation is very fast, the computation resources are idle most of the time. The same problem occurs when two simulation systems have to be synchronized, as the faster simulation has to wait for the slower simulation. Therefore, several different time management approaches have been designed, such as conservative synchronization—only continue simulation when all simulation systems reached the same time point—and optimistic synchronization—faster simulation may continue their work, but must be prepared to roll back in case of need. A good introduction to discrete event simulation in comparison to the alternative simulation paradigm of continuous simulation is given in Sokolowski and Banks (2008), chapter 3. For advanced simulation engineers it is recommended to evaluate the Discrete Event System Specification (DEVS) by Zeigler, Praehofer, and Kim (2000). DEVS one of the most rigorous simulation formalism and well rooted in engineering principles. It has been applied in many engineering contexts and is internationally applied on a broad basis. 140

Simulation

10.2.4 Data Analysis It is very interesting that in practical application a lot of effort goes into the development of simulation systems, but how to evaluate the simulation results is often neglected. The development of a hierarchy of measures of merit should be included in every simulation project. The development of operational, user driven definition of the measures, the development of instruments or definition of access points in the simulation, the application of these instruments or use of these access points to collect the appropriate data, and the establishment of relationships between the different forms of measure need to be agreed by the user of the simulation system, the managers of the project, and the simulation engineers responsible for the simulation. As these efforts are likely to vary in difficulty, costs, requirements for precision and accuracy, and other characteristics, engineering management is challenged to established best practices for their simulation applications. It is common to distinguish measures of performance (MOP) and measures of effectiveness (MOE). MOP focuses more on internal contributing components while MOE evaluates the system as a whole. Both measure categories rely on the availability of the appropriate data. As it has been recognized that a single definition for MOP and MOE does not exist, the term measures of merit (MOM) is often used to generalize the ideas. An alternative term for MOM is figures of merit. Engineering managers are used to this form of analysis requirements, as it can be directly mapped to the objectives hierarchy used in systems architecture and modeling. Once the access to the required data and their use for the agreed MOM is established, the resulting data can be analyzed following the principles for output data analysis, such as described in Ross (2006), chapters 7 and 8. The techniques that are essential for engineering applications of simulation recognized as best practice are statistical analysis, variance reduction techniques, and sensitivity analyses. The statistical analysis should at least result in means and variances and related interval estimates. This is well known from operational research methods where these analysis methods are used typically to evaluate the stability of optimization solutions. The difference to traditional operations research is simply that the data is produced by a simulation system. For the engineer, the view of a simulation as a computer-based statistical sampling experiment is beneficial: Whatever is applicable and necessary for other statistical sampling experiments is applicable and necessary for simulation result evaluation as well. Furthermore, variance reduction techniques fall into this category. However, while statistical analysis targets to increase the knowledge gained from a simulation experiment, variance reduction techniques are normally applied for improving the speed and efficiency of a simulation by intelligently choosing simulation runs that contribute to better estimates. The underlying view of a simulation system is that we utilize the performance measures by which the system is to be judged (the MOM), and parameters that may be adjusted to improve the system performance. The objective of variance reduction techniques is helping to find best combinations used for these parameters to maximize the knowledge gained by simulation runs. Finally, sensitivity analysis evaluates the stability of a recommended solution. Generally, sensitivity analysis helps to track how variations in the input parameters of a simulation system trigger variations in the output parameters. For many practical problems, a solution that is stable within the region of uncertainty or accuracy of the input parameters is preferred to a better solution that is unstable. Sensitivity analysis is therefore closely related to uncertainty analysis that answers the question: how certain is a given solution, or in which confidence intervals is the proposed solution can be applied without becoming invalid. In particular when simulation systems are used in support of decision-making, these questions are essential and need to be addressed by the responsible engineering manager. Simulation is also used to generate quasi-empirical data for data farming. In many domains in which conducting real-world experiments is too expensive, too dangerous, ethically not justifiable, or simply not possible, simulation experiments can be a surrogate that provides the data needed to support engineering managers with the information needed to make the best possible informed recommendation.

10.2.5 Monte-Carlo Simulation and Continuous Simulation Discrete event simulation as the predominant simulation paradigm has already been introduced earlier. This section describes the other two simulation paradigms of interest to the engineering manager: Monte-Carlo simulation and continuous simulation.

141

Engineering Management Handbook

• Monte-Carlo simulation uses a deterministic simulation model that maps input parameters to output parameters. In the standard case, time is not modeled. The simulation model is then used to iteratively evaluate the model by feeding random variables and evaluate the resulting outputs. The statistical analysis means shall be applied here. • Continuous simulation models system changes over time, but in contrast to discrete simulation the system is described for continuous simulation by state variables that change continuously with respect to time: state changes are not instant events. Before going into further explanations and examples, some additional simulation foundations need to be explained. The first is the difference between deterministic and stochastic models. In both cases, a simulation model is interpreted to map input parameters to output parameters. • In deterministic models, the same input will always produce the same output. Only if input parameters are varied, the output parameter will vary. • In stochastic models, the results when computing the output parameters are determined by using one or more random variable to represent uncertainty about a process in which a given input will produce an output according to some statistical distribution. Therefore, the same input may result in different outcomes. Another fundamental concept resulting from stochastic processes is the Markov chain, which is known to engineering managers from other applications. The characteristic of a Markov chain is that it operates without memory: the conditional probability of any future event given any past events and the present state is independent from any past events and only depends on the current state. This makes them of interest to simulation developers, as it reduces the implementation efforts tremendously as past states do not need to be tracked in order to compute the possible follow-on events. Following similar ideas, the Poisson distribution is of particular interest for developers of stochastic models as well. The basic idea and mathematical foundation was introduced earlier in this chapter: the Poisson distribution expresses the probability of a number of events occurring in a fixed period of time if these events occur with a known average rate and independently of the time since the last event. As before with Markov chains, this insight can facilitate the implementation: if a stochastic process can be approximated by this distribution, the likelihood for an event to happen in the next time step is independent from its history, so no past states need to be tracked. The use of Monte-Carlo simulation is quite obvious. It should be pointed out that a Monte-Carlo model can be part of a discrete event simulation system: whenever a procedure operates on random variables without advancing the simulation time—such as producing new events, calculating the system state change, etc.—all techniques supporting Monte-Carlo simulation can be applied. In particular when simulation systems need to be evaluated, or even validated and verified, this can be beneficial. The engineering manager responsible for such a project needs to be aware of these relations. Continuous simulation is often used to simulate physical systems or systems that involve mechanical, electrical, thermal, or hydraulic components. Such systems are most precisely described by differentials equations. When computed using digital computers, these differential equations need to be transformed into difference equations, or numerical approximation methods like the Euler method or Runge-Kutta method need to be applied. This automatically introduces a numerical error, contributing to artificial variances in the results of the system. It is of critical importance for the engineering manager to be aware of these approximations and how they are dealt with in the data analysis. In summary, this section on simulation theory shows that the engineering manager responsible for a simulation project—be it development or application of a simulation in the engineering context—needs a solid foundation to align the work of professional in the domains of statistic, computer science, numeric, and M&S. In practice, the engineering manager is often the first of these experts that has access to all these contributing elements, so he is the experts with the “big picture” of what is going on. The ability to understand the nature of challenges that can occur in all these contributing domains is therefore essential to successfully bridge the gap between these contributing experts as well as between technical experts, management, and the user community. 142

Simulation

10.3 Simulation Applications 10.3.1 Simulation as an Engineering Method Simulations are becoming ubiquitous. Related research continuously introduces new and improved formalisms, methodologies, and simulation tools that are intended to support the new, broader interpretation of engineering. The 2006 NSF Report on “Simulation-based Engineering Science” states that engineering and science communities have become increasingly aware that computer simulation is an indispensable tool for resolving a multitude of scientific and technological problems facing our country. The NSF view on simulation is summarized in the report as follows: “Computer simulation represents an extension of theoretical science in that it is based on mathematical models. Such models attempt to characterize the physical predictions or consequences of scientific theories. Simulation can be much more, however. For example, it can be used to explore new theories and to design new experiments to test these theories. Simulation also provides a powerful alternative to the techniques of experimental science and observation when phenomena are not observable or when measurements are impractical or too expensive. “Simulation-Based Engineering Science (SBES) is defined as the discipline that provides the scientific and mathematical basis for the simulation of engineered systems. Such systems range from microelectronic devices to automobiles, aircraft, and even the infrastructures of oil fields and cities. In a word, SBES fuses the knowledge and techniques of the traditional engineering fields—electrical, mechanical, civil, chemical, aerospace, nuclear, biomedical, and materials science—with the knowledge and techniques of fields like computer science, mathematics, and the physical and social sciences.” The American Society for Engineering Management definition of EM shows that this is a task in their realm: “Engineering management is the art and science of planning, organizing, allocating resources, and directing and controlling activities that have a technological component.” This can be mapped directly to the tasks identified for SBES in the NSF report. To fill this generic task description with applicable recommendations, best practices are needed. Simulation systems have a long and successful history in the military domain. They are used in analysis, procurement, acquisition, training, education, and in some instances even for support of operations. The North Atlantic Treaty Organizations (NATO) published the NATO Code of Best Practice (COBP) for Command and Control Assessment in 2002. This code represents over a decade of work by many of the best analysts from the NATO countries, conducted under the guidance of NATO’s Research and Technology Organization (RTO). While the original application domain of the COBP is military command and control, the code was written to give broad guidance on the assessment of complex systems in complex environments for a wide variety of decision-makers. As such, it comprises best practices and guidance on how to conduct a scientific study rooted in operations research ideas on complex systems in complex environments. The COBP is mandated for operations research studies conducted for the US Department of Defense (DoD) Assistant Secretary Defense (ASD) Networks and Information Integration (NII). Recent research work has shown that many traditional project management artifacts can be mapped to the recommended processes and products documented in the COBP. It has also been proven to be a valuable guide when conducting simulation-based studies, in particular for the US Joint Forces Command, in particular when supporting new research requiring increasingly the incorporation of geopolitics, culture, religion, and political economy to better understand how Diplomatic, Intelligence, Military, and Economic (DIME) factors affect military decision-making. A growing area of research for the Department of Defense is centered on related Political, Military, Economic, Security, Information, and Infrastructure (PMESII) aspects (for example, see Sokolowski and Banks (2008), chapter 1). To ensure that all relevant information is captured, the NATO COBP recommends an iterative approach captured in Figure 10.4.

143

Engineering Management Handbook Figure 10.4. Phases for Conducting a Study as Recommended in the NATO COBP

Problem Formulation Solution Strategy Sponsor’s Problem

Human & Organizational Issues Measures of Merit

Scenarios

Methods & Tools Data

Products

Assess Risk

This COBP is organized into four phases. The first phase deals with study dynamics, problem formulation, and the development of a solution strategy. This phase is the initialization phase, in which the objectives and requirements are analyzed and it is ensured that the sponsor’s problem is understood and a simulation-based solution is possible. The second phase identifies and discusses in depth the essential elements of assessment: measures of merit, scenarios, and human and organizational issues. In this preparation phase, applicable solutions are identified and the selection of best tools is prepared by in-depth analysis of systems and their environment. This leads to the definition of necessary data and selection of the best tools. The third phase addresses issues related to risk and uncertainty while the final phase describes the range of assessment products. As it is the case in all engineering task, the success starts with understanding the real needs of the sponsor and the environment in which the support has to be delivered. There may be political, economic, or even cultural constraints. Once this is understood, the problem formulation activities lead to specifications on what needs to be solved, and the solution strategy captures how these problems will be solved. The results are captured in a study plan as well as in a study management plan. These documents can be augmented by additional documents and artifacts supporting better project management, cost management, and risk management. Capturing the relevant system information regarding human and organizational issues becomes increasingly relevant. A system is commonly defined to be a collection of hardware and software, people, facilities, resources, and procedures organized to accomplish some common objectives. These aspects must be managed for real systems and modeled in virtual systems. An exhaustive variation of all parameters defining a solution space is normally not feasible for large and complex challenges. This can normally only be done for small problems that can be solved by closed mathematical solutions and not for the type of problems that are addressed by simulation-based solutions. It is therefore essential to capture operationally relevant scenarios to ensure that relevant parameter combinations are evaluated. A scenario can be defined as a description of the geospatial constraints, the environment, available resources, given objectives, past and current relevant events, and all other factors being related to the system to be evaluated during a specified time frame suited for satisfactory study objectives and the problem analysis directives. Quite often, the use of smaller vignettes is good practice, in particular for smaller scenarios and as excursions from the main scenario. Measures of merit have already been dealt 144

Simulation

with in the theory section. They should be directly linked to the elements of the scenarios and reflect the solution strategy. It should be pointed out that up to this point all decisions and evaluations have been made tool independent. Selection the methods and tools to be applied in the project, including selecting the best simulation tool, should always be based on these tool independent analysis. It is bad practice to build a study or a project around a preselected simulation tool, as this automatically limits the focus to “what can the tool do” instead of answering “what is the sponsor’s problem.” As a rule, the engineering manager should be able to select an orchestrated set of tools that are able to generate the required MOM and for which the necessary data can be obtained. It is good practice to include the management of tools selection—including simulation systems—and data engineering for obtaining the necessary input data and produce the required MOM in the management artifacts mandatory for simulation-based projects. Finally, risk assessment is recommended. Sensitivity analysis, as described earlier in this chapter, is recommended but not sufficient. In summary, simulation-based studies and projects require rigorous EM to be successful. Best practices exist that can be of help. However, it is still true that success comes from wisdom, wisdom comes from experience, and experience comes from making mistakes. Simulation as an engineering method can give guidelines and avoid obvious traps; but as so often in engineering, there is no golden rule that ensures success.

10.3.2 Simulation with ARENA As mentioned before, computer simulations can be written in general-purpose languages or simulation languages. ARENA is a widely distributed example for using a simulation framework. In general, simulation languages have become very popular and are often part of operations research curricula. Frameworks support the user by providing not only a simulation language, but also by providing composable simulation constructs that are often supported by intuitive graphical user interfaces. ARENA supports the user with a great variety of predefined constructs plus the ability to program user-created constructs. In summary, as stated in various overview papers, the philosophy of this approach is to treat simulation modeling as an in-vitro laboratory that facilitates the understanding of complex systems and experimentation with what-if scenarios in order to estimate their performance metrics. Figure 10.5 shows a screenshot of executing an ARENA model. Figure 10.5. Executing an ARENA Model

As shown in the screenshot, building a simulation in ARENA is pretty straightforward: users select constructs representing the necessary processes or process steps (and users can define such constructs, 145

Engineering Management Handbook

including their graphical representation) and connects the input of following activities with the output of producing activities. In addition, users can define data collection and visualization capabilities that can be used to produce statistical data and their display. In addition, ARENA supports several probability distributions. ARENA supports all three simulation paradigms, but the emphasis clearly lies on discrete event simulation. Trying to give a more detailed introduction in form of a short section would be futile, so the interested reader is referred the Kelton et al. (2007) for further reading. Another aspect in the focus of simulation applications for engineers is the statistical experimentation design. Some simulation frameworks are supporting their users with integrated tools for the definition of MOM or with predefined procedures supporting data analysis as defined in the theory part of this chapter. Nonetheless, the overview compiled by Kleijnen (2004) should be known by every engineer that applies simulations in his or her project. This journal article comprises best practices and theoretic foundations critical for simulation experiment design and completes the recommendations given in Kelton et al. (2007), chapter 12.

10.3.3 Agent-based Modeling Agent-based modeling is on the brink to becoming the fourth broadly applied simulation paradigm. As a metaphor, the use of agents is not new. The availability of necessary computing power and memory for execution and storage helped this powerful concept to become an emerging branch in simulation. Yilmaz and Oren (2008) compiled several chapters into a book focusing on agent-directed simulation for systems engineering. Yilmaz and Oren (2008, chapter 3) gives an overview comprising the important points of interest to engineers wanting to understand agent-based modeling for engineering applications. The working definition for an agent as proposed in Yilmaz and Oren (2008, chapter 3) defines agents by their characteristics as follows: • The agent is situated, it perceives its environment, and it acts in its environment. The environment includes typically other agents, other partly dynamic objects, and passive ones, that are, e.g., subject of manipulation by the agent. The communication with other agents is of particular interest systems comprising multiple agents, as agents can collaborate and compete for tasks. This later characteristic has also been referred to as social ability. • The agent should be autonomous, in the sense that it can operate without the direct intervention of humans or others and autonomy requires control about its own state and behavior. They must be guided by some kind of value system. • To be flexible for an agent means to mediate between reactive behavior, being able to react to changes in its environment, and deliberativeness to pursue its goals. A suitable mediation is one of the critical aspects for an agent to achieve its tasks in a dynamic environment. An agent can act upon its knowledge, its rules, beliefs, operators, goals, and experiences, etc. and adapt to new constraints and requirements— or even new environments—as required. New situations might ask for new goals, and new experiences might lead to new behavior rules. Furthermore, being mobile adds to the flexibility of an agent. The architectural frame shown in Figure 6.5 was proposed in support of discussing how the agent characteristics enumerated above can be realized. It is kept simple on purpose, as it is not intended to prescribe a solution but to make developers aware of domains that need to be taken into account.

146

Simulation Figure 10.6. Architectural Frame Addressing Main Agent Characteristics Information Exchange with other Agents

Perception

Decision Making Memory

Action directed at objects in the Situated Environment

Sense Making

Action

Input from the Situated Environment

Communications

Adaptation

There are three external and four internal architectural domains identified. The external domains comprise those functions needed by the agent to interact with and to act within his environment. The internal domains categorize the functions needed for the agent to act and adapt as an autonomous object. • Perception: The agent has to receive input from the situated environment and map it to his or her internal structures. • Communications: The agent needs to exchange information with other agents based on an agreed protocol. • Action: The agent need to move and act in the situated environment. • Sense making: The agent needs to base his or her decision on a fused awareness of what he or she perceives, what he or she communicates, and what he or she knows. To enable autonomy, the agent needs a sense-making domain. • Memory: The knowledge of the agent must be accessible to the agent. It may have several compartments, such as short- and long-term memory, earlier decisions, pattern, etc. • Adaption: Flexibility requires that rules and memories can be changed to reflect new recognized circumstances. • Decision-making: Situational awareness leads to a desired outcome and related steps. These steps can be decided on one at a time or in form of a complex plan. The decision-making domain enables this. When such agents are used to populate a simulation, in particular when these agents do not all support the same view but are heterogeneous regarding the way they perceive, make sense, communicate, and decide, the result is a multi-agent system. These systems have been successfully applied in many domains. In particular the characteristics of autonomy and flexibility make them of interest to engineers, as they enable to add human-like behavior to simulations (life sciences and political sciences have been among the pioneers of agent-based simulation applications) as well as system’s learning and adaption of systems to new environments (e.g., it has been discussed if systems’ functionality cannot be provided by system agents that observe and adapt continuously to new situations). Again, this is a topic of current research. Figure 10.7 summarizes the characteristics.

147

Engineering Management Handbook Figure 10.7. Agents, Environment, and Societies

Agent System Agent Characteristics

Situated Environment

Agent Society

Perception

Accessible vs. Non-Accessible

Large scale vs. few Agents

Decision Making

Deterministic vs. Stochastic

Cooperative vs. Non-cooperative

Communication

Episodic vs. Sequential

Homogeneous vs. Heterogeneous

Action

Static vs. Dynamic

Open society vs. non-open society

Discrete vs. Continuous

Similar to compiling the characteristics of agents into architectural domains, the characteristics of contributing agents, the situated environment, and the agent society can be compiled into taxonomical domains. The overview is neither complete nor exclusive, but it summarizes the core domains the engineering manager must be aware of when applying multi-agent systems.

10.3.4 Simulation and Systems Engineering Methods An emergent application domain for simulation is the support of system engineers in better understanding complex systems and system of systems. The role of simulation systems can be manifold: • Simulations are used as serious games educating managers and technicians in a secure but immersive environment. • Simulations provide the foundation for what-if-analyses of alternatives: each option can be simulated and the results can be compared using the established operations research methods. This may even lead to simulation based optimization. • Simulations allow executable architectures: system architecture artifacts are used to define simulation components that act like the specified systems. This allows to bring one or many simulated systems of a portfolio together in a common synthetic environment to test the possible solution before a system has to be build. • Simulations are used as fully controlled and adaptable test beds for systems. In particular when requirements are used to derive measures and metrics for the acceptance of a system, test cases can be directly used to specify simulations providing the necessary parameters to stimulate the system under test accordingly. • Simulations are used to be executed in parallel to the controlled system: at specified times or events, the observations are compared with the predications of the simulation. If the variance between expected and observed status becomes too big, additional control or a new decision is needed. Of particular interest is furthermore the support of systems engineering and system of systems engineering by simulation applications. The observation of emergence in such systems has been the topic of 148

Simulation

concern, as engineered systems should not expose unspecified behavior that may defeat the system’s purpose, as discussed by Osmundson et al. (2008). Emergence can be reproduced in particular in agent-based simulations. Looking into agent-based methods to better understand emergence and potentially becoming able to engineer positive emergence into systems is a topic high on current research agendas. The increasing number of autonomous systems in various application domains of interest to the engineering manager also requires a better understanding of potential and limits of simulation, as the intelligent behavior of autonomous systems is constrained by the same epistemological and computational frontiers as simulation systems. The engineering manager must know these limitations. A good overview of such limits has been compiled by Oberkampf et al. (2002). A view on simulation from the perspective of philosophy of science has been provided by Tolk (2015). To summarize this chapter on simulation it should be obvious that applying simulation in and for engineering projects is challenging and requires a solid education. It showed M&S within the supporting disciplines including probability and statistics, computer science, systems modeling and architecture, and operations research. It introduced discrete event simulation, Monte-Carlo simulation, and continuous simulation as the simulation paradigms, and it mentioned M&S methodologies and application domains. The sections of this chapter cannot replace the referenced literature—which needs to be completed with experience to lead at the end to good decisions. However, the core knowledge references in this chapter will enable engineering managers to have a good start and will prepare them with tools and methods for solving real-world problems and challenges when using simulation for and in engineering.

10.4 References Diallo, Saikou Y., Padilla, Jose J., Gore, Ross, Herencia-Zapana, Heber, and Tolk, Andreas, “Toward a formalism of modeling and simulation using model theory,” Complexity, vol. 19, no. 4, 2014, pp. 56-63. Fuller, Mark, Valacich, Joe, and George, Joey, Information Systems Project Management – A Process and Team Approach. Pearson Prentice Hall, 2008. Kelton, David, Sadowski, Randall, and Sturrock, David, Simulation with Arena, McGraw-Hill Science/ Engineering/Math, 2007. Kleijnen, Jack, An overview of the design and analysis of simulation experiments for sensitivity analysis. European Journal of Operational Research, vol. 164, no. 2, July 2004, pp. 287-300. Kossiakoff, Alexander, and Sweet, William N., Systems Engineering Principles and Practice. John Wiley & Sons, 2002. National Science Foundation (NSF) Blue Ribbon Panel, Report on Simulation-Based Engineering Science: Revolutionizing Engineering Science through Simulation, NSF Press, May 2006. NATO Research and Technology Organization (RTO), NATO Code of Best practice for C2 Assessment, Command and Control Research Program (CCRP) Press, 2002. Oberkampf, William. L., DeLand, Sharon M., Rutherford, Brian M., Diegert, Kathleen V., and Alvin, Kenneth F., “Error and uncertainty in modeling and simulation.” Reliability Engineering & System Safety, vol. 75, no. 3, 2002, pp. 333-357. Osmundson, John S., Huynh, Thomas V., and Langford, Gary O., “Emergent Behavior in Systems of Systems,” INCOSE International Symposium, vol. 18, no. 1, 2008, pp. 1557-1568. Padilla, Jose J., Diallo, Saikou Y., and Tolk, Andreas, “Do We Need M&S Science?” SCS M&S Magazine, vol. 2, no. 4, 2011, pp. 161-166. Robinson, Steward, Conceptual modelling for simulation Part I: definition and requirements. Journal of the Operational Research Society, vol. 59, 2008, pp. 278-290. Ross, Sheldon, Simulation (4th edition). Academic Press, 2006. Sokolowski, John and Banks, Catherine (Editors), Principles of Modeling and Simulation: A Multidisciplinary Approach. John Wiley & Sons, 2008. Sokolowski, John and Banks, Catherine (Editors), Modeling and Simulation Fundamentals: Theoretical Underpinnings and Practical Domains. John Wiley & Sons, 2010. Tolk, Andreas, Simulation. Engineering Management Body of Knowledge. American Society of Engineering Management, vol. 1.1, Nov. 2007. 149

Engineering Management Handbook

Tolk, Andreas, Learning Something Right from Models That Are Wrong: Epistemology of Simulation. In Concepts and Methodologies for Modeling and Simulation, Springer International Publishing, 2015, pp. 87-106. Tolk, Andreas, and Hughes, Taylor K., Systems engineering, architecture, and simulation. Modeling and Simulation-Based Systems Engineering Handbook, CRC Press, 2014, pp. 11-41. Yilmaz, Levent, and Oren, Tuncer (Editors), Agent-Directed Simulation and Systems Engineering, John Wiley & Sons, 2008. Zeigler, Bernard, Praehofer, Herbert, and Kim, Tag Gon, Theory of modeling and simulation: Integrating discrete event and continuous complex dynamic systems - second edition, Academic Press, 2000.

150

Decision Analysis

11 Decision Analysis Gregory S. Parnell University of Arkansas and Innovative Decisions Inc.

151

Engineering Management Handbook

11.1 Introduction 11.1.1 What is Decision Analysis? Decision analysis is an operations research technique for analyzing complex decisions with multiple (and usually conflicting) objectives and uncertainty. One of the founders of the field, Ronald Howard, first coined the name in 1964. Decision analysis uses the axioms of probability and utility theory and the philosophy of systems analysis (Howard, 1966). The first decision analysis book (Raiffa, 1968), used probability and a single objective, net present value. The first multiple objective decision analysis book was published in 1976 (Keeney & Raiffa). Decision analysts use probability to model their belief in the likelihood of each outcome of an event based on our state of information. They use Bayes law to update beliefs as they learn new information. In addition to the mathematical foundations of decision theory, decision analysts have adopted lessons from behavioral decision theory research about the heuristics and biases people use to reason with uncertain information and make decisions (Clemen and Reilly, 2013; Kirkwood, 1997). Decision analysts have used behavioral decision research to develop effective problem structure, uncertainty, and preference elicitation protocols.

11.1.2 Why Use Decision Analysis? Engineering managers need to make defensible decisions in a complex technology management environment when faced with complex alternatives, conflicting objectives of diverse stakeholders, and major uncertainties. Engineering managers can use decision analysis for major decisions including R&D decisions (Matheson and Matheson, 1998), and systems decisions (Parnell, Driscoll, and Henderson, 2008) throughout the system life cycle. The stakeholders include consumers, owners, users, decision-makers, developers, and maintainers. Typically, engineering managers must consider technology, safety, economic, organizational, environmental, and emotional factors. In addition, major decisions sometimes involve important political, historical, cultural, social, and other considerations. Because most EM decisions impact future products and services, uncertainty can be a major concern. These uncertainties can create technical, cost, and schedule risks for the organization.

11.1.3 When Do You Use Decision Analysis? While decision analysis principles can be used for simple decisions, decision analysis is most appropriate for the engineering managers’ most complex decisions. Engineering managers typically use decision analysis to provide fact-based decision information for two types of decisions that have major consequences to the organization. The first approach is to use decision analysis to provide information at a major program decision milestone. An example would be the decision to select a system concept for design and development. The second approach is to use decision analysis to provide information for the major annual investment (e.g., R&D portfolio and capital budgeting) decisions. An example would be the major development projects to provide the technologies for future products and services.

11.1.4 Who Uses Decision Analysis? In the past 40 years, there are been a wide variety of decision analysis applications. Few application articles are published due to proprietary information, classified information, and lack of incentives of practioners to publish. Good surveys can be found in Corner and Kirkwood (1991); Keefer, Kirkwood, and Corner (2004); Philips (2007); and Parnell (2007). Decision analysis applications, including the use of decision analysis with other operations research techniques, are published in a wide variety of professional journals. The decision analysis is usually performed by employees of the company, consultants, or a combination of employees and consultants. Decision analysis education and training are obtained by academic courses, certification courses, and professional continuing education courses. Large companies that institutionalize decision analysis, sometimes develop their own internal decision analysis training pro152

Decision Analysis

grams. Many decision analysts belong to the Society for Decision Professionals1 and the Decision Analysis Society2 of the Institute for Operations Research and Management Science.

11.2 Decision Processes 11.2.1 Challenges The organizations decisions and the implementation of the decisions ultimately determine the success of the organization. Decision analysis is a technical field that uses mathematical modeling to help make sound decisions. However, leaders and key stakeholders need to effectively and efficiently interact to determine and implement the best decisions for the organization. We can think of decision analysis as a conversation with the decision-maker and key stakeholders. The conversation is only as good as the people participating. The model structure is the topic of the conversation. The numbers are used to define the topics and reason about the relevance of the topics. A well-executed decision analysis emphasizes insight about opportunities and risks. We have to design the organizational process as well as the decision model. The process should include the: • Right people that have broad and deep knowledge • Right data and information • Right forum that is conducive to discussion and interaction • Right balance of modeling and challenging the model with intuition • Right duration to meet decision deadlines but enable information gathering and socializing the results. The Handbook of Decision Analysis provides more details of the decision analysis process. Three decision analysis processes have been successfully used in organizations: the analytical process, the decision conference process, and the dialogue decision process.

11.2.2 Analytical Process When the technical (i.e., the complexity of the alternatives and the mathematics of the model) part of the decision analysis is complex and more critical that the social (i.e., the stakeholder interaction), an analytical process may be appropriate. Some decisions involve complex, technical issues that can be assessed with models. The decision-maker provides the initial guidance, the stakeholders provide input, and the decision analysts build models, collect data, analyze alternatives, and present insights at a decision briefing attended by the decision maker and the key stakeholders.

11.2.3 Decision Conference Process When the social part of the decision is more critical than the technical, the decision conference may be appropriate. Figure 11.1 describes the activities of a decision conference3. The key idea of a decision conference is the shared awareness of the key players to discuss issues, build models to evaluate alternatives, and explore the results.

1. http://www.decisionprofessionals.com/ accessed August 22, 2015 2. http://decision-analysis.society.informs.org/ accessed August 22, 2015 3. Modified from Phillips, L. D., “Chapter 19: Decision Conferencing,” Advances in Decision Analysis – From Foundations to Applications, Edwards, W., Miles, R., and von Winterfeldt, D., Cambridge Press, 2007.

153

Engineering Management Handbook Figure 11.1. Decision Conference

Compare: Gut ⇔ Model

Key Players

Awareness of Issues Prepare - objectives - participants - announcement

Discuss Issues Build Model Explore results

Shared Understanding

Commitment

Action

11.2.4 Dialog Decision Process When the decision models are complex and stakeholder participation is critical the dialog decision process may be appropriate. The dialog decision process uses regular interaction opportunities for the decision team to “dialog” with the decision makers and key stakeholders. Figure 11.2 shows the concept of the dialog decision process4. The frame is the fundamental structure that we use to view the problem. The values are the customers’ objectives and preferences. The alternatives are options to be evaluated. The evaluation is the decision analysis and the insights. Figure 11.2. Dialog Decision Process

Decision Makers Frame

Values

Decision

Alternatives Evaluation Decision Team

4. Modified from Spetzler, C., “Building decision competency,” Advances in Decision Analysis – From Foundations to Applications, Edwards, W., Miles, R., and von Winterfeldt, D., Cambridge Press, 2007.

154

Decision Analysis

11.2.5 Advantages and Disadvantages of Decision Processes Table 11.1 summarizes the advantages and disadvantages of each of the three processes. The table also identifies the individuals performing the analytical role. The analytical process is most useful for when detailed models are required and decision-makers are confident that the analysts will not lose focus on the problems. Decision conferences are most useful when the model can be done in real time and the participation of key stakeholders in essential. The dialog decision process is most useful when the models require significant development time but key stakeholder participation is essential to commitment to implement the solutions. Table 11.1. Advantages and Disadvantages of Decision Processes

Analytical role

Analytical Approach

Decision Conference

Dialog Decision Process

Analyst team

Facilitators

Decision team/Design team

May be appropriate for well-framed problems Advantages

Detailed analytical models developed Least time demand on decision-makers and stakeholders

Disadvantages

Analysts can lose focus on the evolving views of decision-makers and stakeholders Models may become overly complex Lack of stakeholder participation and data

High confidence you are solving the right problem All participants develop common understanding and shared commitment Develop and use requisite decision models that use the essential information to distinguish between the alternatives Multiple conferences can support hierarchical decision-making Must schedule all key players for same time Time commitment of managers and stakeholders (days) If needed, complex models must be developed before the conference

High confidence you are solving the right problem Planned involvement of key decision-makers and stakeholders at major decision points More analytical models of values and uncertainty Less time demand (several short meetings)

Requires scheduling periodic meetings with key players Key stakeholder availability and data collection challenges between dialog points

11.3 Decision Elements 11.3.1 Values and Outcomes Values are what we care about in the decision-making process. Decision analysts try to identify the alternatives that have the best probability of providing outcomes that will meet our values and objectives. Keeney (1992) suggests we should start a decision-making process by identifying the values and objectives that we are trying to achieve with the decision(s). This is more difficult than it seems. Keeney’s recent research has shown that people can only identify about half of their final objectives in the initial meeting and that they may not initially identify the objectives that they will assign the most importance.

11.3.2 Uncertainty Engineering managers make decisions about products and services for future consumers and customers. There are many uncertainties involved in EM. Four of the most common are technical (how well with the technologies work), cost (how much will it cost), schedule (when will it be delivered), and market (what will be the future demand). However, sources of uncertainty can include all elements of the decision environment including safety, economic, organizational, environmental, and emotional factors. In addition, major decisions sometimes involve important political, historical, cultural, and social uncertainties.

155

Engineering Management Handbook

11.3.3 Decisions Engineering managers make decisions about the best products and services to develop and market to customers and consumers. The most difficult decisions involve decision strategies that combine many decision elements including partnerships, technologies, production plans, supply chain, marketing, and logistics.

11.4 Decision Modeling—Illustrative Product Development Example 11.4.1 Basic Influence Diagram The influence diagram (Howard and Matheson, 2005) is a useful decision modeling technique that helps us identify the decision elements and their interrelationships. We will use DPL software to illustrate our influence diagrams and decision trees5. While an influence diagram solution algorithm exists (Clemen and Reilly, 2013), we will use influence diagrams to model the decision problem and decision trees solve for the best decision and obtain insights. Other decision analysis software is also available to perform a decision analysis (Patchak, 2014). We will use an influence diagram to consider an illustrative product development example. The influence diagram in Figure 11.3 identifies the decisions (squares), uncertainties (circles) and values (rounded squares). The two sequential decisions are to develop a new product and produce a new product. The arrow denotes that we have to develop the new product before we can produce the new product. If we produce the new product, the market success is uncertain. The arrow denotes that the uncertain variable market success depends on whether we produced the product. Finally, our net present value depends on development cost, production cost, and sales (market success). We can see that the major benefit of the influence diagram is that it helps identify the key variables and their interrelationships in a picture and encourages us to develop the details needed for the analysis later. Figure 11.3. Basic Influence Diagram

Market Success

Develop New Product

Produce New Product

Net Present Value

11.4.2 Basic Decision Tree In order to decide if we should develop and produce the new product, we need to know the relevant cost and market data. Suppose our current product is expected to provide a net present value of sales of $35M. However, the product development group wants to develop a new product (development cost is $10M) that has the following potential sales: -$20M in a low market, $50M in a nominal market, a $100M in a high market. The marketing department assesses the following probabilities: 0.3 for low market, 0.4 for nominal market, and 0.3 for high market. Figure 11.4 shows a decision tree using this information.

5. DPL is decision and risk analysis software developed by Syncopation Software. http://www.syncopation.com/ accessed August 22, 2015.

156

Decision Analysis Figure 11.4. Basic Decision Tree

Develop New Product [35]

Yes

Produce New Product [34] -10

Yes

No No

Market Success [34]

Low 30% Nominal 40% High 30%

[-30] -20 [40] 50 [90] [100]

[25] 35

[35] 35

The decision tree uses the following steps. First, all the values are calculated for the end nodes of the tree (triangles) for each path through the decision tree. There are five paths in Figure 11.4. Starting at the bottom and working up, the first path is not to develop and the NPV is $35M. The second path is to develop but not produce for NPV of $25M (35 minus the development cost of 10). The third path has NPV of 90 (100-10). Likewise, the fourth and fifth paths are $40M and -$30M, respectively. The second step is to start at the end nodes and solve the decision tree using the “Average Out-Roll Back” Algorithm, which calculates expected values at uncertain nodes and selects the highest NPV at decision nodes. For example, the expected value at market success of 34 (0.3*(-30) + 0.4*(40)+0.3*(90)) and at the produce new product node, the best decision is Yes since 34 > 25. Finally, the develop new product decision is No since 35>34.

11.4.3 Basic Risk Profile Risk is the probability of a bad outcome. One of the important decision analysis displays is the NPV cumulative probability distribution, sometimes called the cumulative risk profile. Figure 11.5 shows the cumulative risk profile for our basic decision. The higher expected value alternative (No development) has significantly less risk. We can see that the no development decision has a value of $35M with probability 1.0, which means we assume no uncertainty. The Yes decision (Develop and Produce) has a probability of 0.3 of -$30M, a probability of 0.4 of $40M and a probability of 0.3 of $90M (expected value of $34M). The major risk is the 0.3 probability of losing $30M. Figure 11.5. Basic Cumulative Risk Profile

So far, this seems like an easy decision, select the highest expected value and the lowest risk. 157

Engineering Management Handbook

11.4.4 Value of a Test One of the obvious weaknesses of our analysis is that we have not considered using a development test to see if the new product design will meet the market needs. The influence diagram with the addition of the development test is shown in Figure 11.6. The new node and the three new arrows each have meaning. The arrows mean that if we develop the new product, we can do the development test; the results of the test are known before the production decision; and the development test results are used in the calculation of NPV. Figure 11.6. Influence Diagram with Test

Market Success

Develop New Product

Net Present Value

Produce New Product

Development Test

We assume the test can be a success (0.9) or fail (0.1). If the test succeeds the market success will be as modeled in section 11.4.2. However, if the test fails the NPV will be 0. Figure 11.7 shows our new decision tree with the test. Information has value only if the decision changes. The test changed the decision and increased the expected value from $35M to $36M in spite of the $1M cost of the test. The highest expected value is to develop the new product and produce the product if the test is a success and not produce if the test fails. The cost of the test is an important factor, if the test cost $3M, the expected value of the develop option would have been $34M and it would have been better to not develop the new product. If the test has been free, our expected value would have been $37M. Because our expected value without the test was $35M, we would pay up to $2M for the test. Figure 11.7. Decision Tree with Test

Develop New Product [36]

Yes

Development Test [36] -10

Produce New Product Success [37] -1 90%

No

Produce New Product Failure [24] -1 10% No

Yes

Yes No

Market Success [37]

[24] 35

Low 10% Nominal 80% High 10%

[-31] -20 [39] 50 [89] [100]

Market Success [-11] [24] 35

[35] 35

In addition to the $1M increase in expected value, the test has reduced the risk from a 0.3 probability of losing $30M to a 0.09 probability (0.9*0.1) of losing $31M. This can be seen by calculating the probability of the top path through the decision tree in Figure 11.7. Of course, it would also be easy to see in the cumulative risk profile. 158

Decision Analysis

11.4.5 Value of Imperfect Information about Market Success In the previous section, we examined the value of a test. We can also consider the value of imperfect information about the market test. In Figure 11.8, we add a market survey uncertainty node to our influence diagram and three arrows. The arrow from develop new product to market survey means the market survey is done only if we develop the new product. The arrow from market success to market survey means that the probabilities assigned the market survey outcomes depend on the outcome of market success. The arrow from market survey to produce new product means that the market survey results will be known before the produce new product decision. Figure 11.8. Influence Diagram with Test and Market Survey

Market Success

Market Survey Develop New Product

Net Present Value

Produce New Product

Development Test

Suppose we have a market survey with the following data on the left side of Figure 11.9. We are given the prior probability distribution on the market success and the likelihood distributions on the on the probability of the survey prediction given the market success outcome. The likelihood distribution is a quantitative statement about the ability of the market survey to predict the true market outcome. In the decision tree, we need the preposterior distribution (the probability of the prediction) and the posterior distribution (the probability of market success given the market survey prediction). Figure 11.9 shows the standard probability calculations on the middle and right side. We enter the prior and likelihood into DPL and the software calculates the preposterior and posterior distributions. Figure 11.9. Probability Calculations Prior

Low 0.3

Nominal 0.4

High 0.3

1.0

Likelihood

Joint probability

Preposterior

Posterior

Predict Low l Low 0.7

Low l Predict Low 0.66

0.21

Predict Nominal l Low 0.2

0.06

Predict High l Low 0.1

0.03

High l Predict Low 0.09

Predict Low l Nominal 0.2

0.08

Low l Predict Nominal 0.17

Predict Nominal l Nominal 0.6

0.24

Predict High l Nominal 0.2

0.08

High l Predict Nominal 0.17

Predict Low l High 0.1

0.03

Low l Predict High 0.09

Predict Nominal l High 0.2

0.06

Predict High l High 0.7

0.21

3.0

1.0

Predict Low 0.32

Nominal l Predict Low 0.25

Predict Normal 0.36

Predict High 0.32

Nominal l Predict Nominal 0.67

Nominal l Predict High 0.25 High l Predict High 0.66

1.0

3.0

159

Engineering Management Handbook

Figure 11.10 shows the decision tree with the test and the market survey. We have assumed that the development test and the market survey are done concurrently. The expected value has increased significantly from 36 to 49M. The market survey changes the best decision strategy. If the test is a success, we only produce the new product if the market survey predicts nominal or high. Figure 11.10. Decision Tree with Test and Market Survey Produce New Product [33] Predict Low 32% -1

Yes

No

Success 90%

Market Survey [50] -1

Produce New Product Predict Nominal [45] 36%

Yes

Develop New Product [49]

Produce New Product Predict High [74] 32%

Yes

-1 No

Produce New Product Yes [33] Predict Low No 32% -1 Market Survey Failure [33] 10% -1

Produce New Product Predict Nominal [33] -1 36% Produce New Product Predict High [33] 32% -1

No

[35] 35

[-22] -20 [48] 50 [98] [100]

Low 17% Nominal 67% High 17%

[-22] -20 [48] 50 [98] [100]

35 Low Market Success 9% [74] Nominal 25% High 66% [33]

[-22] -20 [48] 50 [98] [100]

[33] 35 Market Success [45]

-1 No

Development Test [49] Yes

Low 66% Nominal 25% High 9%

Market Success [7]

Yes No Yes No

[33]

Market Success [-2] [33] 35 Market Success [-2] [33] 35 Market Success [-2] [33] 35

Another interesting benefit of the market survey is the reduction in risk. With the market survey, the worst case risk is now -22M (versus -30M), which occurs with a probability of 0.08 (versus 0.09). We know from this analysis that the market survey costing $1M resulted in an expected value of $49M. Had the test been free, our expected value with the market survey would have been $50M. Therefore, we know that the maximum we should pay for this market survey would be $13M (50 – 37).

11.4.6 Value of Perfect Information About Market Success Market managers may want to know how much they should pay for a better market survey. Decision analysts use the expected value of perfect information to assess the maximum we would pay for any information. The value of perfect information can be calculated three ways. The first way is to change the probabilities in Figure 11.9 (and Figure 11.10) to have the survey perfectly predict the outcomes. The second way is to delete the market survey node and put the market success as the first node in the tree. The third way is to have DPL calculate the expected value of perfect information using the tree in Figure 11.10. All three ways result in an expected value of perfect information of 56. This means that given we have the test and market survey would be 7M (56 - 49).

160

Decision Analysis

In addition to an increased expected value, perfect information would significantly reduce the risk to a worst case outcome of 33M (35 – 2 for the test and survey). However, this outcome would occur with probability 0.37.

11.4.7 Value of Control Although the value of perfect information assesses the expected value of knowing the outcome of a variable, it does not change the probabilities assigned to the outcomes. As we have seen, a market survey is an example of obtaining information on a probability distribution. The value of control is another important decision analysis concept. Advertising is a common marketing tool to control the market outcome. The value of perfect control assumes we can change the variable to its best outcome. Therefore, the expected value of perfect control would tell us the maximum we would pay for marketing to make the demand be the highest outcome in our distribution. In our basic decision tree, the expected value with perfect control was $90M. Therefore, the expected value of perfect control was $55M (90 – 35). One of the principles is that the expected value of perfect control is always greater than or equal to the expected value of perfect information. Intuitively, this means that we would rather control the outcome of a variable to it best outcome than know which outcome has occurred.

11.4.8 Sensitivity Analysis Because our data is seldom perfect, it is useful to consider how sensitive we are to our assumptions. We say a variable is sensitive if a value in the range of interest would cause us to change our decision. Decision analysis software allows us to perform sensitivity analysis to one or more parameters.

11.4.9 Comparison of Influence Diagrams and Decision Trees The two major modeling tools we have used in this section are influence diagrams and decision trees. As we have noted, the same model can be solved for the same answer using both tools but the algorithms are different. The two tools are complementary. The influence diagram is very useful for modeling the problem structure and capturing the relationships between variables. A complex problem can effectively be understood with an influence diagram. The decision tree is useful for displaying the mathematics of the modeling assumptions and showing the best solution strategy on the decision tree algorithm has been completed. Large decision trees are difficult to understand without careful study.

11.5 Single Attribute Utility 11.5.1 Utility In the previous section, we used net present value as our single objective. In addition, we used the expected value to compare alternatives and the cumulative risk profile to assess the risk. If we were going to make a thousand similar decisions the expected value would be wonderful criteria since we would expect to get 1,000 times the expected value! However, some decisions are only one time decisions. An alternative approach is to use a utility function (Clemen and Reilly, 2013; Kirkwood, 1997) to capture our returns to scale and our risk preference. Two exponential utility functions (top and bottom) and a linear utility function (middle) are shown in Figure 11.11.

161

Engineering Management Handbook Figure 11.11. Three Utility Functions 1 Risk Adverse Risk Neutral Risk Seeking

0.9 0.8

Utility u(x)

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0

1

2

3

4

5

6

7

8

9

10

X

11.5.2 Risk Preference Utility functions are assessed by asking questions about lotteries, e.g., would you rather has $5M with probability 1.0 or a 50/50 chance of $10M or $0. If we would take less than $5M, say $4M, we are risk adverse (see curve in Figure 11.11) and we say that we would we pay a risk premium of $1M (expected value of $5M – the $4M) to avoid the risk. Buying life insurance is an example of typical risk adverse behavior. We pay a premium to hedge against the risk to our families. If we are indifferent, we are expected value decision makers. If we would pay more than $5M, say $6M, we are risk seeking and the risk premium is -$1M ($5M – 6M). An example of risk seeking behavior is gambling in Las Vegas. Clearly, the house adjusts all games so our risk premium is negative (the house is favored to win). Otherwise, they could not make a profit.

11.5.3 Utility with Decision Trees One of the nice features of utility is that decision trees work equally well with utility as the end nodes of tree. We can use our “average out – roll back” algorithm to calculate expected utility. Also, we can perform value of information, value of control, and sensitivity analysis exactly like we did with net present value.

11.6 Multiple Objective Decision Analysis (MODA) 11.6.1 Additive Value Model In many decision problems we have other objectives besides net present value and we choose not to convert all objectives to dollars. Multiple objective decision analysis uses many mathematical equations to evaluate alternatives. The simplest and most commonly used model is the additive value model (Kirkwood, 1997). This model uses the following equation to calculate each alternative’s value: n

where

v ( x ) = ∑ wi v i ( xi ) i =1

v(x) is the alternative’s value

162

i = 1 to n is the number of the value measure xi is the alternative’s score on the ith value measure

Decision Analysis

vi(xi) = is the single dimensional value of a score of xi wi is the weight of the ith value measure n

and

∑w i =1

i

= 1 (all weights sum to one)

The additive value model has no index for the alternatives we’re evaluating because our values do not depend on the alternative. We use the same equations to evaluate every alternative.

11.6.2 Value Functions Value functions measure returns to scale on the value measures. They have four basic shapes: linear, concave, convex, and an S-curve (Figure 11.12). The linear value function has constant returns to scale: each increment of the measure is equally valuable. The concave value function has decreasing returns to scale: each increment is worth less than the preceding increment. The convex value function has increasing returns to scale: each increment of the measure is worth more than the preceding increment. The S-curve has increasing, then decreasing, returns to scale on the measure. We have several techniques to develop value curves from subject-matter experts. Our first step is to have the experts determine the shape of the value curve: linear, concave, convex, or S-curve. Next, we use value increments to identify several points on the curve—asking experts the relative value of increments in the value measure scale. Figure 11.12. Four Types of Value Functions 1

Linear Concave Convex S-Curve

0.9 0.8

V(x) [Value]

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

x [ Value Measure]

11.6.3 Swing Weights Weights play a key role in the additive value model. MODA quantitatively assesses the trade-offs between conflicting objectives by evaluating the alternative’s contribution to the value measures (a score converted to value by single-dimensional value functions) and the importance of each value measure (weight). The

163

Engineering Management Handbook

most common mistake in MODA is assessing weights based on importance only (Parnell et al., 2013). The weights depend on the value measure scales’ importance and range. If we hold constant all other value measure ranges and reduce the range of one of the measure scales, the measure’s relative weight decreases, and the weight assigned to the others increases since the weights add to 1.0.

11.6.4 Swing Weight Matrix The swing weight matrix (Parnell et al., 2013) defines what we mean in the decision context by the importance and range of variation for the value measures. The idea of the swing weight matrix is straightforward. A measure that is very important to the decision should be weighted higher than a measure that is less important. A measure that differentiates between alternatives, that is, a measure in which value measure ranges vary widely, is weighted more than a measure that does not differentiate between alternatives. The first step is to create a matrix (Table 11.2) in which the top defines the value measure importance and the left side represents the range of value measure variation. The levels of importance and variation should be thought of as constructed scales that have sufficient clarity to allow the analyst to uniquely place every value measure in one of the cells. A measure that is very important to the decision and has a large variation in its scale would go in the upper left of the matrix (cell labeled A).6 A value measure that has low importance and has small variation in its scale goes in the lower right of the matrix (cell labeled E). Table 11.2. Elements of the Swing Weight Matrix Importance of the value measure to the decision

Range of variation of the value measures

Critical

Important

Factor to Consider

Large

A

B2

C3

Medium

B1

C2

D2

Small

C1

D1

E

Consistency Rules Because many individuals may participate in the assessment of weights, it is important to ensure consistency of the weights assigned. It is easy to understand that a very important measure with a high variation in its range (A) should be weighted more than a very important measure with a medium variation in its range (B1). It is harder to trade off the weights between a very important measure with a low variation in its range (C1) and an important measure with a high variation in its range (B2). Weights should descend in magnitude as we move on the diagonal from the top left to the bottom right of the swing weight matrix. Multiple measures can be placed in the same cell with the same or different weights. If we let the letters represent the diagonals in the matrix A, B, C, D, and E, A is the highest weighted cell, B is the next highest weighted diagonal, then C, then D, and then E. For the swing weights in the cells in Figure 7.1 to be consistent, they need to meet the following relationships: • Any measure in cell A must be weighted greater than measures in all other cells. • Any measure in cell B1 must be weighted greater than measures in cells C1, C2, D1, D2, and E. • Any measure in cell B2 must be weighted greater than measures in cells C2, C3, D1, D2, and E. • Any measure in cell C1 must be weighted greater than measures in cells D1 and E. • Any measure in cell C2 must be weighted greater than measures in cells D1, D2, and E. • Any measure in cell C3 must be weighted greater than measures in cells D2 and E. • Any measure in cell D1 must be weighted greater than measures in cell E. • Any measure in cell D2 must be weighted greater than measures in cell E. • No other strict relationships hold. 164

Decision Analysis

If we denote i to be the label of the cell in the swing weight matrix and fi to be the unnormalized swing weight of the value measures in each cell, then the following strict inequalities relationships of non-normalized swing weights must hold: fA > fi for all i in all other cells fB1 > fC1, fC2, fD1, fD2, fE fB2 > fC2, fC3, fD1, fD2, fE fC1 > fD1, fE fC2 > fD1, fD2, fE fC3 > fD2, fE fD1 > fE fD2 > fE No other specific relationships hold.

Assessing Un-normalized Swing Weights Once all the value measures are placed in the cells of the matrix, we can use any swing weight technique to obtain the un-normalized weights as long as we follow the consistency rules previously cited. In assigning weights, the stakeholders need to assess their tradeoffs between level of importance and level of variation in measure scale. One approach would be to assign the measure in cell A (the upper left corner cell) an arbitrary large unnormalized swing weight, for example, 100 (fA = 100). Using the value increment approach (Kirkwood, 1997), we could assess the weight of the lowest weighted measure in cell E (the lower right corner) the appropriate swing weight, for example, 1. This means the swing weight of measure A is 100 times more than that of measure E. It is important to consider what the maximum in cell A should be. Common choices are 1000 and 100. Of course fE can be other numbers besides 1. If we use 100 and 1, we have three orders of magnitude. If we use 1000 and 1 we have four orders of magnitude. Using a value increment approach, un-normalized swing weights can be assigned to all the other value measures relative to fA by descending through the very important measures, then through the important measures, then through the less important measures.

Calculating Normalized Swing Weights We can normalize the weights for the measures to sum up to 1 using the following equation:

wi =

n

fi

∑f i =1

, i

where fi is the un-normalized swing weight assessed for the ith value measure, i = 1 to n for the number of value measures, and wi are the normalized swing weights from Equation 1. MODA examples can be found in Kirkwood, 1997, Parnell, Driscoll, & Henderson, 2011, Parnell et al., 2013, and Clemen and Reilly, 2013.

11.6.5 Multiple Objective Decision Analysis with Decision Trees One of the nice features of MODA is that decision trees work equally well with multiple objective value at the end nodes of tree. We can use our “average out – roll back” algorithm to calculate expected multiple objective value. Also, we can perform value of information, value of control, and sensitivity analysis exactly like we did with net present value and utility.

11.7 Role of Engineering Manager The engineering manager plays a key role in the decision analysis process. First, the engineering manager must determine the most appropriate decision process (Section 11.2) for the type of decision, the organi-

165

Engineering Management Handbook

zation, and the leadership styles of organizational decision-makers. Second, the engineering manager must ensure that all key stakeholders participate in the process. Third, innovative alternatives are considered that have the potential to create high value for the organization. Fourth, the engineering manager must ensure that all relevant uncertainties and risks have been considered. Finally, the engineering manager and the leadership team must ensure that the solutions are implementable. Having early stakeholder participation in the decision analysis helps to ensure the solutions are implementable.

11.8 Advanced and Other Topics This chapter covered the fundamentals of decision analysis for engineering managers. Section 11.9 and specifically Table 11.3 provides references and a mapping of all of the topics to two standard references in the field. The third column in the table lists advance material that is not covered in this chapter.

11.9 References Clemen, R., Making Hard Decisions, 2nd Edition, Duxbury Press, 1996.1 Corner J. L. and Kirkwood, C. W., “Decision Analysis Applications in the Operations Research Literature, 1970-1989,” Operations Research, vol. 39, no. 2, pp. 206-219, March-April 1991. Howard, R. A., & Abbas, A. E., Foundations of Decision Analysis. Prentice Hall, 2015. Howard, R. A. and Matheson, J. E., Editors, The Principles & Applications of Decision Analysis, 1983, Volumes I & II, Strategic Decisions Group. Keefer, D. L., Kirkwood, C. W., and Corner, J. L. Perspective on Decision Analysis Applications, 1990– 2001, Decision Analysis, vol. 1, no. 1, pp. 5-24, March 2004. Keeney, R. L. Value-Focused Thinking: A Path to Creative Decision Making. Cambridge, Massachusetts: Harvard University Press, 1992. Keeney, R. L. and Raiffa H. Decision Making with Multiple Objectives Preferences and Value Tradeoffs. New York: Wiley, 1976. Kirkwood, C. W., Strategic Decision Making: Multiobjective Decision Analysis with Spreadsheets, Belmont, California: Duxbury Press, 1997.2 Matheson, D. & Matheson, J., The Smart Organization: Creating Value Through Strategic R&D, Harvard Business School Press, 1998. Patchak, William, M., “Decision Analysis Software Survey,” OR/MS Today, October 2014 [Biannual survey of DA software] https://www.informs.org/ORMS-Today/Public-Articles/October-Volume-41-Number-5/Decision-Analysis-Software-Survey, Accessed August 22, 2015. Parnell, G. S., Driscoll, P. J., and Henderson D. L., Editors, Decision Making for Systems Engineering and Management, Wiley Series in Systems Engineering, Andrew P. Sage, Editor, Wiley & Sons Inc., 2008. Parnell, G. S., Value-Focused Thinking Using Multiple Objective Decision Analysis, Methods for Conducting Military Operational Analysis: Best Practices in Use Throughout the Department of Defense, Military Operations Research Society, Editors, A. Loerch & L. Rainey, pp. 619-656, 2007. Parnell, G. S., Bresnick, T. A., Tani, S. N., and Johnson, E. R. Handbook of Decision Analysis, John Wiley & Sons, 2013. Phillips, L. D., “Chapter 19: Decision Conferencing,” Advances in Decision Analysis – From Foundations to Applications, Edwards, W., Miles, R., and von Winterfeldt, D., Cambridge Press, 2007. Raiffa, H. Decision Analysis: Introductory Lectures on Choices Under Uncertainty, Addison-Wesley Publishing Co., 1968. Spetzler, C., “Building decision competency,” Advances in Decision Analysis – From Foundations to Applications, Edwards, W., Miles, R., and von Winterfeldt, D., Cambridge Press, 2007.

166

Decision Analysis

The following are two standard references in this field. The following table maps the topics in this chapter to these references. 1. Clemen, Robert T. and Reilly, T., Making Hard Decisions: An Introduction to Decision Analysis, 2nd Edition, Duxbury Press, 2013. 2. Kirkwood, Craig W., Strategic Decision Making: Multi-objective Decision Analysis with Spreadsheets, Duxbury Press: Belmont, CA, 1997. Table 11.3. Topics Referenced to Standard Texts in the Field References

I II A B C D E III A B C C E F G H IV A B C D V A B C D E F G H VI A B C D E

Authors Introduction Elements of Decision Analysis Decision Context Values and Consequences Decisions Uncertain Events Alternative-Focused Thinking and Value-Focused Thinking Structuring Decisions Value Hierarchies Means Objectives Influence Diagram Decision Trees Comparison of Influence Diagrams and Decision Trees Sequential Decisions Time Value of Money Cash Flows and Probabilities Decisions with Single Objectives - Value Decision Tree Example with Expected Monetary Value Influence Diagram Example with Expected Monetary Value Risk Profiles Dominance Decisions with Multiple Objectives - Value Objectives, Attributes,Value Measures, and Scores Types of Value Measures Value Functions Importance Weights and Swing Weights Additive Value Model Decisions with Uncertain Scores Decision with Uncertain Events and Uncertain Scores Mathematical Foundations of Multiattribute Value Sensitivity Analysis One Way Sensitivity Two Way Sensitivity Tornado Diagram Probability Sensitivity Weight Sensitivity

1

2

C&R X X X X X X X X X X X X X X X X X X X X X X X

K X X X X X X

X X X X X X X X X X

Advanced Material

X X X X

X X

X X X X X X X X X X X X

X

A

167

Engineering Management Handbook

VII A B C D E VIII A B C IX A B C D E F G X A B C D XI A B

168

Modeling Uncertainty Probability Basics Subjective Probability Heuristics and Biases Decomposition and Probability Assessment Art of Probability Elicitation Value of Information Expected Value without Information Value of Perfect Information Value of Imperfect Information Single Attribute Risk Preferences Risk Attitude Expected Utility, Certainty Equivalent, and Risk Premiums Utility Function Assessment Constant Risk Aversion - Exponential Utility Function Axioms of Utility Theory Paradoxes and Implications Mathematical Foundations of Single Attribute Utility Multiattribute Risk References Additive Utility Model Weight Assessment with Lotteries Utility Independence Mathematical Foundations of Multiattribute Utility Resource Allocation Benefit Cost Analysis with Multiattribute Value Project Selection with Constraints using Optimization

X X X X X X X X X X X X X X X X X X X X X X

X X X X X X

X X X X X X X X X X X X X X X X

A A A A A A A

Multi-Criteria Analysis

12 Multi-Criteria Analysis Anirban Ganguly Stevens Institute of Technology

Donald N. Merino Stevens Institute of Technology

169

Engineering Management Handbook

12.1 Introduction to Multi-Criteria Analysis 12.1.1 Background Rapid globalization and continual change in the business environment has led various engineering and business organizations to search for improved techniques to achieve higher profitability in the market. In this context, a proper “decision-making strategy” is the foundation for success or failure of an organization. A key aspect of an organization’s strategic decision-making process is how to select the “best” alternative from among a set of alternatives with often-conflicting attributes. The competing projects that are to be selected are often dependent on a variety of factors, ranging from the availability of resources to the organizational infrastructure. Approaches that consider a plethora of attributes, both economic and non-economic in nature, often prove to be the best selection. Multi-Criteria Analysis (MCA), can transform dissimilar (and often incongruent) information into a common index or value. This then provides the decision-makers a rational basis for selecting the best alternative. The purpose of this chapter is to provide engineering managers with an overview of MCA along with a detailed discussion of some of its commonly used tools and techniques. According to Canada and Sullivan (1988), increased competitiveness cannot be achieved without an optimal mix of investment in people, technology and “bricks and mortar.” In this context, having a knowledge about MCA could greatly aid decision-makers to arrive at the optimal mix of trade-offs between attributes.

12.1.2 Overview of Multi-Criteria Analysis Multi-Criteria Analysis (MCA) can be defined as “the study of methods and procedures by which concerns about multiple conflicting criteria can be formally incorporated into the management planning process” (International Society on Multiple Criteria Decision Making). MCA first appeared as a decision making tool in the sixties (Evalsed, 2003) and is a structured approach used to facilitate the selection of a preferred alternative by evaluating each of the alternatives from a set of well-defined criteria. The set of criteria on which the selection of the alternatives is based should be measurable, even if the measurement is performed on a simple nominal scale. The measured outcome of the set of criteria (and its subsequent conversion to a single composite value) provides the basis for comparison of choice and subsequently facilitates the selection of the “best alternative.” The business operations of most organizations consist of decision-making that is made up of a series of choices, with the final decision being dictated by the result of analyzing each of the choices. Hence, one of the main purposes of MCA is to structure a set of important criteria that form the backbone of the decision for choosing a preferred alternative. Furthermore, in a democratic business environment, which consists of a group of decision-makers (or stakeholders), there often exist a variety of opinions with regard to the choice of the best alternative. In this context, MCA serves as a widely accepted tool because it considers the subjective opinions of all the decision-makers on each particular criterion and in turn uses the result to arrive at the “best alternative.” MCA is used in a wide variety of decision-making processes, from assessing investment decisions in alternative manufacturing processes to evaluating and selecting any hi-tech engineering project (Ganguly & Merino, 2006, 2007). Other applications of MCA include decision-making in a variety of fields like evaluating a new manufacturing system, software allocations, technology management, telecommunication management and operations research to name a few.

12.1.3 Relevance of MCA to Engineering Management With the blurring of boundaries between technology and management, engineering must redefine its role to survive in the modern day business environment (Kotnour and Farr, 2005). As a result, an in-depth knowledge of decision-making techniques using Multi-Criteria approaches could serve as a very useful tool for an engineering manager. The techniques highlighted in this chapter are simple and yet highly purposeful for facilitating any successful decision-making process. As more EM students (and consequently engineering managers) are employed by industries spanning all areas of manufacturing and services, the need for awareness of Multi-Criteria decision-making techniques like Multi-Attribute Analysis and 170

Multi-Criteria Analysis

Analytic Hierarchy Process (AHP) increases. The insight of the techniques gained from this chapter can aid the engineering managers’ knowledge about techniques and practical applications of decision-making. This, in turn, could lead to a more effective decision-making process and hence a better selection of EM projects. Additionally, using non-economic decision-making tools like MCA in tandem with economic tools like sensitivity analysis and after-tax analysis could provide any engineering manager with the knowledge of how to combine economic as well as non-economic tools to arrive at a more accurate decision regarding the choice of an engineering project. The proper selection of any engineering/manufacturing project is a complicated process that requires detailed analysis before committing huge capital investments. Hence, a proper selection among competing alternatives is a key to effectively managing engineering/technology projects. This is a typical problem for engineering managers. This chapter illustrates how to use various techniques to choose among alternative engineering projects, and, in turn, increase the profitability of the organization.

12.2 Analytic Hierarchy Process 12.2.1 Overview of Analytic Hierarchy Process Policy makers at all levels of decision-making often use multiple criteria, which, through various tradeoffs, illustrate the advantages as well as the disadvantages of various policy options under the condition of risk and uncertainty (Saaty, 1994). One of the most common approaches to the practice of “multiple criteria decision making” has been Analytical Hierarchy Process (AHP) developed by Thomas L. Saaty (1980). AHP is a decision-making process that is defined as “an approach to decision making that involves structuring multiple choice criteria into a hierarchy, assessing the relative importance of these criteria, comparing alternatives for each criterion, and determining an overall ranking of the alternatives” (Decision Support Systems Resources Glossary). AHP considers multiple criteria simultaneously, dissects a decision choice problem in various levels of hierarchy, in turn aggregate the individual solution at various levels of the hierarchy into a consolidated structure, which the determine the best alternative in the process (Roper-Lowe and Sharp, 1990).

12.2.2 The AHP Process The process of AHP starts with the construction of a hierarchy that describes the problem that is about to be tackled. While constructing the hierarchy, the overall objective (which Saaty calls “the focus”) of the project is always placed right at the top of the hierarchical tree and the main attributes a level below it. The sub-attributes are placed on the subsequent levels of hierarchy and the final level consists of the alternatives among which the selection is to be made. A simple three-level AHP hierarchy is presented in Figure 12.1. Although the level of hierarchies can be extended further, depending on the degree of details that the decision-maker wants to choose.

Selecting the Attributes The “old” right-hand rule of industrial engineering says that fewer attributes are preferred because they represent the vital few—as opposed to the trivial many. The “right hand” suggests that five attributes is an appropriate number. Selecting attributes should follow the KISS (Keep It Simple Stupid) principle.

Choosing Among the Alternatives After constructing the hierarchy, the next step is to choose between the attributes through a series of pair-wise comparisons where each attribute of that particular hierarchical level is compared with its sibling with respect to their relative importance to each other. The pair-wise comparisons are made relative to the importance, likelihood, desirability and so on and are mostly based on a numeric scale. The pair-wise comparisons are denoted in terms of the relative importance of an attribute with respect to the final alternative decisions being compared. Table 12.1 shows the 9-point AHP scale with an explanation of each of the scale levels. 171

Engineering Management Handbook Figure 12.1. A Simple Three-Level AHP Model Objective / Goal / Focus

Attribute 1

Attribute 3

Attribute 2

Alternative B

Alternative A

Table 12.1. Scale for Pair-wise Comparison Using AHP1 Relative Intensity

Definition

Explanation

1

Equally Preferred

The two attributes in question (i and j) are of equal importance

3

A Little More Preferred

One variable is a little more important than the other

5

Moderately Preferred

One variable is much more important than the other

7

Highly Preferred

One variable is very much more important than the other

9

Extremely Preferred

One variable is extremely more important than the other

Reciprocal (1/3, 1/5, 1/7, 1/9)

If attribute i has one of the above numbers assigned to it when compared with attribute j, then j has the value 1/number assigned to it when compared with i. More formally if nij = x then nji = 1/x.

Determining the Weights Next, we need to derive the weights of the lowest level of attributes through a series of pair-wise comparisons where each attribute of that particular hierarchical level is compared with its sibling. In this context, it should be noted that any pair-wise judgment could be represented in the form of a matrix where the cells denote the relationship between a particular pair-wise judgments. Each judgment represents the dominance of an element in the column over an element in the row (Saaty, 1994), based on the scale provided in Table 12.1. This is shown in Table 12.2. Table 12.2. Pair-Wise Comparison Between Attributes Attribute 1

Attribute 2

Attribute 3

Attribute 4

Attribute 5

Attribute 1

1

3

5

5

9

Attribute 2

1/3

1

3

9

7

Attribute 3

1/5

1/3

1

5

9

Attribute 4

1/5

1/9

1/5

1

7

Attribute 5

1/9

1/7

1/9

1/7

1

Table 12.2 exhibits a hypothetical pair-wise comparison table among five attributes. In this example, 172

Multi-Criteria Analysis

because Attribute 1 is “a little more” preferred to Attribute 2, cell i12 is “3” and consequently, its reciprocal, that is, cell i21, is “1/3.” Furthermore, for a set of “n” elements (or attributes) the number of pair-wise comparisons should be “n (n-1)/2”. Saaty (1994) provides a rationale for this. According to him, because there are n 1’s on the diagonal for comparing the attributes with themselves and of the remaining judgments, half are reciprocals. Therefore, there are “(n2- n)/ 2”, i.e., n (n-1)/2 judgments for a set of “n” attributes. Thus, for a set of five attributes, the number of pair-wise judgments should be 10 (i.e., “(5*4)/2 = 10”).

Weighting the Attributes After the comparisons are made, they are converted into a numeric scale and are entered into a matrix. The resulting data is normalized in order to make the matrix column stochastic in nature and the Eigenvectors are subsequently derived from the matrix. Tables 12.3 and 12.4 illustrate this point. Table 12.3. Pair-Wise Comparison Between Attributes with Totals Attribute 1

Attribute 2

Attribute 3

Attribute 4

Attribute 5

Attribute 1

1

3

5

5

9

Attribute 2

1/3

1

3

9

7

Attribute 3

1/5

1/3

1

5

9

Attribute 4

1/5

1/9

1/5

1

7

Attribute 5

1/9

1/7

1/9

1/7

1

TOTAL

1.84

4.59

9.31

20.14

33

Table 12.4. Normalized Values of Pair-Wise Comparison Between Attributes Attribute 1

Attribute 2

Attribute 3

Attribute 4

Attribute 5

Attribute 1

0.542

0.654

0.537

0.248

0.273

Attribute 2

0.181

0.218

0.322

0.447

0.212

Attribute 3

0.108

0.073

0.107

0.248

0.273

Attribute 4

0.108

0.024

0.021

0.050

0.212

Attribute 5

0.060

0.031

0.012

0.007

0.030

TOTAL

1.000

1.000

1.000

1.000

1.000

Composite Score After the pair-wise comparison has been completed, the results are combined into a composite score, which shows how well each of the alternatives to be chosen fits the overall objective (focus) of the decision-making process. Finally, the last step of the AHP is that of making the actual decision based on the overall values of the alternatives in question.

Summary of AHP 1. 2. 3. 4. 5. 6. 7.

Choose the goal or objective to be evaluated Choose the attributes (KISS) Establish a rating system for pair-wise comparisons Use the rating system to choose among the attributes Normalize the attributes to determine their weights Chose the alternatives Use the same process as in Steps 3 and 4 for each attribute/alternate combo 173

Engineering Management Handbook

8. Summarize and evaluate the ratings 9. Choose the alternate with the highest score

Are the Choices Consistent? The process discussion of AHP will conclude with an overview of the “Consistency Ratio (CR).” Consistency Ratio in AHP can be defined as an indicator of the level of consistency in the judgments of the respondents. For example, it is often seen that the respondent to any pair-wise judgment survey might prefer A to B, B to C, but C to A. The consistency ratio essentially denotes the level of inconsistency of the respondents. Any consistency ratio of 0.10 and under is acceptable (Lang and Merino, 1993; Saaty 1994) to the decision-makers and if the judgments are perfectly consistent, the consistency ratio is zero. However, an inconsistency ratio of greater than 0.10 does not necessarily denote an error in the multiple criteria analysis using AHP. It is often seen (Ganguly and Merino, 2007) that the pair-wise comparison among the attributes selected was not transitive in nature. For example, the relative importance of attribute “A” being greater than attribute “B” and the relative importance of attribute “B” being greater than attribute “C” does not necessarily signify that attribute “A” will be preferred to attribute “C.” Saaty (2001) provides a rationale behind this in the sense that evaluators often make tradeoffs that violate transitivity but, overall, are accurate in their judgments because they take into account the relative importance of the criteria themselves. There are times when an evaluator cannot make a clear decision because the tradeoffs among several activities come out to be the same (Saaty, 2001) and are not related to some other pair-wise judgment. This is the primary reason why the final AHP judgment values are not revised in many situations in spite of a high CR and the overall AHP value based on the pair-wise judgment value is used as the final decision-making criteria. For further insights into Consistency Ratio (CR) and Consistency Index (CI), readers are advised to refer to sections 7.4 and 7.5 in Saaty (2001) and Appendix 28A in Lang and Merino (1993).

Distributive vs. Ideal Mode AHP has two options of performing an operation—the distributive mode and the ideal mode (Dolan, 2000; Liberatore and Nydick, 2003). The distributive mode is followed when the objective of the analysis is to prioritize a set of options in order to determine how to distribute something among them while the ideal mode is used when the basic objective of the analyses is to identify the best set of options. Generally, the ideal mode of synthesis is used more frequently than the distributive mode of synthesis.

12.2.3 Advantages and Limitations of AHP as a Multi-Criteria Tool The popularity of AHP stems from its ability to translate subjective judgments by the decision-makers into numerical values. AHP is an intuitive technique that not only integrates subjective judgment with mathematical data, but also facilitates the participation of the entire pool of decision-makers in the decision-making process. Additionally, AHP’s simple, intuitive nature allows the participants to complete the process with relative ease, thereby arriving quickly at a conclusion regarding the selection of an alternative from a multiple set. However, in spite of its wide popularity, AHP is not devoid of limitations. The two major limitations of AHP that have been cited are its implicit assumption of transitivity among the pairwise judgments and the problem of Rank Reversal. AHP implicitly assumes a logical transitivity among pair-wise judgments, which is not always the case in a real-life decision-making process. Thus, the value of the Consistency Ratio (CR) in AHP often ends up being higher than the desired value (i.e., ≤ 0.10), reducing the authenticity of the analysis in the process. However, this is not true because the pair-wise judgments of the decision-maker are not always transitive in nature. An example of the loss of transitivity on judgment and the subsequent explanation is provided in the previous section. Another major limitation of AHP is the issue of Rank Reversal. Rank Reversal can be explained as the situation where an alternative chosen as the best out of a set fails to be chosen when another, perhaps unimportant and irrelevant, alternative is excluded from (Perez et al., 2002) or included in the set. The issue of rank reversal has been raised by the practitioners of Utility Theory as one of the major drawbacks of AHP (Saaty, 1994). How174

Multi-Criteria Analysis

ever, Saaty (1994) argues that using the “ideal mode” of AHP synthesis, rather than the distributive mode, would aid any decision-maker to preserve the original rank.

12.2.4 Conclusion We can conclude from the above that an AHP analysis aids the decision-maker in gaining valuable insight into his or her values and the relative merits of the available decision options. AHP is a structured method that can elicit more information from target respondents (usually experts or decision-makers) (Cheng and Li, 2001). According to Dolan (2000), AHP reduces a complex problem into smaller, simple parts and thus makes solving not only easier, but also much less error prone. Additionally, AHP provides a method of applying multiple viewpoints into a decision-making process in an explicit and unbiased manner— thereby making the decision-making process not only very practical, but also very complete. Several available techniques of sensitivity analysis allow how changes in the pair-wise comparisons of the criteria weights might affect the result. Finally, AHP, which is a multiple criteria decision-making process, has all the potential for overcoming many of the cognitive and practical problems associated with most other decision-making models. AHP also ensures that all-important considerations, even the ones that are very unique, are addressed while selecting the best alternative (Saaty, 1990).

12.3 Analytic Network Process 12.3.1 Overview of Analytic Network Process In spite of its wide popularity, one of the drawbacks of AHP is that it is primarily based on a one-way hierarchical structure. In the case of Analytic Network Process (ANP), all the attributes in the decision-making process can be related in any possible way—thereby incorporating a subjective judgment not only within a particular level of hierarchy, but also among various levels (referred to as clusters in ANP) of the hierarchy. This type of networked approach is suitable for complex decision-making scenarios, where interrelations among all the elements at every level of hierarchy is considered as a part of the final analysis—a situation that every engineering manager is faced with at some point in their careers.

ANP Structure ANP takes into consideration the inner as well as outer dependence among the elements (Saaty, 2001). Although inner dependence signifies the relationship between the attributes of any particular level of hierarchy (or cluster, in the case of ANP), the outer dependence indicates the relationship among various levels. This relationship is illustrated by a Supermatrix, which is a two-dimensional matrix that represents the relationship among all the clusters available in the ANP analysis and is column stochastic in nature, which is a matrix whose columns sums up to unity (Saaty, 2001). The overall decision regarding the choice of the final alternative in ANP is based on the Supermatrix.

12.3.2 The ANP Process A typical ANP decision-making process comprises of the following steps.

Step 1: Structuring the Problem ANP starts by laying down the problem in the form of an interdependent network of elements. A typical ANP network comprises all the clusters and the sub-clusters, arranged in a networked fashion, that might be beneficial in achieving the overall object of the decision-making process. For example, let us suppose that a car buyer has to decide among selling three brands of car, Brand A, B and C. In order to arrive at a purchasing decision, the buyer has to consider a set of monetary as well as non-monetary attributes. A basic ANP network for this problem is shown in Figure 12.2. Figure 12.2. Simple ANP Network for a Decision-Making Process 175

Engineering Management Handbook

• Brand A • Brand B • Brand C Alternatives (Car Brand)

• Initial Cost • Maintenance Cost

Monetary Attributes

Non-Monetary • MPG • Look of the Car Attributes

As seen in Figure 12.2, the basic ANP structure, unlike AHP, consists of a network where every node is connected with each other.

Step 2: Determining the Prioritized Weights of the Elements (or attributes) The next stage of an ANP process consists of performing a pair-wise comparison among the various elements of a cluster, the influence of the clusters on each other as well as the overall objective of the process. This analysis follows the same process and the 9-point scale used in AHP. However, in the case of ANP, the pair-wise comparisons are made between the sub-elements with respect to their parent elements. The resulting values are then normalized to add up to one. For example, in the context of the car purchase example, a possible pair-wise comparison would be to assess the relative priorities of the different brands of the car, with respect to the maintenance cost. This is shown in Table 12.5. Table 12.5. Pair-Wise Comparison Between Brands with respect to Maintenance Cost Brand A

Brand B

Brand C

Mean Normalized Weights

Brand A

1

5

3

0.6234

Brand B

1/5

1

3

0.2390

Brand C

1/3

1/3

1

0.1376

TOTAL

1.533

6

7

1.0000

Step 3: Creating a Supermatrix Once the pair-wise comparisons have been done and their values normalized, the next stage is to arrange the values in the form of a Supermatrix—a matrix that contains the normalized values of the pair wise comparisons among the different levels of hierarchy—or as they are called in ANP, clusters. The Supermatrix, in other words, is a matrix that includes the pair-wise comparison value of a sub-element with respect to its parent element. Additionally, a zero entry in a particular cell in the Supermatrix represents the absence of any relationship between the corresponding row and the column elements. Table 12.6 provides a sample Supermatrix.

176

Multi-Criteria Analysis Table 12.6. ANP Supermatrix—Un-weighted

– Brand A Brand B Brand C I – Cost M – Cost MPG Look

Brand A 0 0 0 0.6000 0.4000 0.3524 0.6476

Brand B 0 0 0 0.5000 0.5000 0.4500 0.5500

Brand C 0 0 0 0.3200 0.5800 0.3333 0.6667

I – Cost 0.6843 0.2276 0.0881 0 0 0 0

M – Cost 0.6234 0.2390 0.1376 0 0 0 0

MPG 0.6063 0.3328 0.0609 0.4443 0.5557 0 0

Look 0.3326 0.2993 0.3682 0.4200 0.4800 0 0

Table 12.6 depicts a simple ANP Supermatrix. As seen from the matrix, each cell in the Supermatrix denotes the relationship of a sub-element with respect to a parent element, with the sub-elements presented in rows and the parent elements in columns. Thus, the intersection cell of the fifth column and the first row shows the normalized value of Brand A with respect to the Maintenance Cost (Table 12.5) and so on. The “zero” values denotes the absence of relationship between the row and the column element. The next stage consists of normalizing the Supermatrix in order to make it column stochastic in nature. The new Super matrix is called Weighted Supermatrix and is provided in Table 12.7. Table 12.7. ANP Supermatrix—Weighted

– Brand A Brand B Brand C I – Cost M – Cost MPG Look

Brand A 0 0 0 0.3297 0.2198 0.1936 0.2569

Brand B 0 0 0 0.2500 0.2500 0.2250 0.2750

Brand C 0 0 0 0.1684 0.3053 0.1754 0.3509

I – Cost 0.6843 0.2276 0.0881 0 0 0 0

M – Cost 0.6234 0.2390 0.1376 0 0 0 0

MPG 0.3032 0.1664 0.0305 0.2222 0.2779 0 0

Look 0.1750 0.1575 0.1938 0.2210 0.2526 0 0

Step 4: Creating the Limit Supermatrix Once the Supermatrix has been determined, the next step is to raise it to a large power at which point the weight stabilizes. In other words, the matrix has to be raised to a sufficiently high power where all the values of a single row have to be the same. This has to be done through a number of iterations. The new matrix is called Limit Supermatrix and has the same form as the weighted Supermatrix, but all the columns of the Limit Supermatrix are the same (Saaty, 2001). Normalizing the Limit Supermatrix will result in the final prioritized weights of all the attributes and the sub-attributes from which the overall value of the set of alternatives can be determined. This is shown in Table 12.8. Table 12.8. ANP Limit Supermatrix

– Brand A Brand B Brand C I – Cost M – Cost MPG Look

Brand A 0.270 0.113 0.064 0.175 0.163 0.089 0.123

Brand B 0.270 0.113 0.064 0.175 0.163 0.089 0.123

Brand C 0.270 0.113 0.064 0.175 0.163 0.089 0.123

I – Cost 0.270 0.113 0.064 0.175 0.163 0.089 0.123

M – Cost 0.270 0.113 0.064 0.175 0.163 0.089 0.123

MPG 0.270 0.113 0.064 0.175 0.163 0.089 0.123

Look 0.270 0.113 0.064 0.175 0.163 0.089 0.123 177

Engineering Management Handbook

The weighted Supermatrix converges until the third decimal place when raised to the 18th power. Although this might sound very difficult, there are mathematical software programs (Scientific Notebook®, Mathematica®, Matlab®, etc.) that solve these problems with considerable ease.

Step 5: Determining the Final Alternative The final selection of the preferred alternative is determined by synthesizing the Limit Supermatrix. The values in the Limit Supermatrix are then normalized according to the clusters and the alternative with the highest value is considered as the preferred one. This is shown in Table 12.9. Table 12.9. Final Alternative Values based on ANP Alternative

Value

Rank

Brand A

0.270

1

Brand B

0.113

2

Brand C

0.064

3

12.3.4 Benefits and Limitations of ANP as a Multi-Criteria Tool Like AHP, the popularity of ANP stems from its ability to translate subjective judgments by the decision-makers into numerical values. ANP, like AHP, is also an intuitive technique that not only integrates subjective judgment with mathematical data, but also facilitates the participation of the entire pool of decision-makers in the decision-making process. However, in addition to having the same benefits as AHP, ANP makes a decision-making process more rational by incorporating non-linear interdependence between the different hierarchical levels of the analysis. Including interdependent relationships in any ANP model has allowed ANP to gain rapid popularity over its unidirectional hierarchical counterpart, i.e., AHP. The procedure of ANP is especially beneficial in complex engineering projects, where an engineering manager has to consider a number of interrelated factors in order to determine the viability of a particular project. Due to its complexities and the tedious nature of the survey, ANP is yet to gain popularity among the industry practitioners. For any project that is simplistic in nature, AHP is often preferred over ANP due to its simplicity. Furthermore, the process of ANP does require some complex calculation (for example, the determination of Limit Supermatrix), that often deter any practitioner to indulge in the process. However, software like Superdecisions® (www.superdecisions.com) developed by Thomas L. Saaty can prove to be useful in solving any decision-making problem using ANP.

12.3.3 Conclusion One of the fundamental advantages of ANP lies in the fact that it stretches well beyond the traditional boundaries of linear decision-making tactics. In a world that is guided by complex decision-making processes, the business and engineering managers often have to indulge in decision-making processes that can be far more well represented through a non-linear multidirectional networked approach. This type of an approach enables the decision-makers to consider consequences that not only affects the alternatives, but also derive its importance from the alternatives themselves—thereby resulting in a networked framework for decision-making. The networked structure used in ANP allows the users to identify, classify, and arrange all the factors that influence the outcome of a decision (Saaty and Vargas, 2006), enabling, in the process, the decision-maker to arrive at more robust and practical solutions regarding the project selection and evaluation.

178

Multi-Criteria Analysis

12.4 Multi-Attribute Analysis (MAA) 12.4.1 Overview As most engineering managers would agree, the selection of any manufacturing / capital project stretches beyond the economic considerations. The economic justification of projects still plays the single most important role in its choice, as more and more organizations are trying to arrive at an efficient mix of economic as well as non-economic justifications in the final selection of an engineering project. Multi-attribute analysis (also addressed as multi-objective analysis, multi-factor analysis, etc.), though different techniques, addresses this concern and enlists possible ways to include monetary considerations in the decision-making process that are often intangible in nature and are very difficult to quantify directly. It is very common to find that a project, which was considered economically viable, lose its importance once the non-economic criteria are added to the decision-making process. This is especially true in government and public-sector projects, where the benefits derived from the projects (along with other peripheral non-economic consideration) are of primary importance when it comes to the final selection of the preferred alternative among a set of many. The remainder of the chapter will provide the reader with a discussion of some of the most commonly used techniques of MAA.

12.4.2 The Multi-Attribute Analysis (MAA) Process The typical MAA process consists of the following stages.

Step 1: Selecting the Attributes The starting point of any MAA consists of selecting the set of attributes, which will be evaluated as part of the final decision-making process. As stated earlier, most of the attributes selected are generally non-monetary in nature. The selection of the attributes can be made based on group discussions, interviews, surveys, experts’ solicitations, among others. Although the list of attributes influencing a particular project can be endless, it is always advisable to select a set on critical ones that have the greatest influence on the selection process (i.e., select vital few from trivial many) and only if the results warrant, add more to the list (Lang and Merino, 1993).

Step 2: Selecting the Measurement Scale The list if attributes identified in the previous stage have to be evaluated on a scale. Therefore, selecting the measurement scale represents the second stage of the process. The selection scale, like the attributes, can be decided based on discussion or subjective judgment. In many cases, the scale is simply the metric on which the measurement is made. Nevertheless, the general practice is to use the same measurement scale for all the attributes (for example, using a scale of 1 – 5 to rank attributes) in order to avoid any ambiguity in the ranking process. Once the measurement scale is selected, the attributes are ranked following the selected scale.

Step 3: Weighting the Attributes Once the attributes are ranked, the next step is to assign weights to the attributes. Normally, the range of the weights varies from 0 to 1, with 0 being the least preferred attribute and 1 being the most preferred. The weighting of the attributes can be accomplished using two different techniques. First, the relative importance of the attributes can be assigned from discussions with the stakeholders and decision-makers so that the most important attribute is assigned the highest weight and so on. The other method would be to assign the weights based on specific tradeoffs. However, unless important otherwise, it is advisable to assign equal weights to the attributes under the assumption that the entire set of identified attributes has approximately the same importance in affecting a decision-making scenario. Once the weights have been assigned, the next stage is to normalize the weights. Normalization is done to reduce data redundancy—although this is unnecessary if the attributes are assigned equal weights. Normalizations enable the scores to be converted into a standard scale that adds up to 1.00 and therefore

179

Engineering Management Handbook

provides the decision-maker with a true reflection of the attributes’ relative weights. The normalized values of the weights serve as a cornerstone for determining the overall rank of alternatives.

Step 4: Determining the Final Rank of the Alternatives The final values of the attributes are determined by multiplying the normalized weights to the rank assigned to each of the attributes (Step 2). This is called the Weighted Additive Model and is by far the technique most widely used for final ranking of a set of attributes. The attribute with the highest overall value is considered the most important followed by the one with the second highest value and so on. The additive function is:

Ai = ∑ ni = 1 Xi * Yij



(12.1)

where, Ai = The final value of alternative i Xi = The normalized weights of the attributes and Yij = Rating of attribute i for alternative j. In context of Equation 12.1, the alternative with the highest value of “A” should be the preferred one. If the decision is based on the viewpoint of several decision-makers, the standard procedure it to determine the mean of the final overall value for any particular alternative.

12.4.3 Utility Theory (Utility Analysis) Utility, as defined by the American Heritage Dictionary of the English Language, is “the quality or condition of being useful.” Therefore, Utility Theory can be stated as an attempt to infer subjective value, or utility, from choices. The concept of utility analysis has emerged as a useful tool for the selection of different projects. As a result, the perceived utility for a selected project serve as a very important consideration for selection among a set of competing projects. One of the main reasons for the huge success of utility analysis is that it affords a rational method of project selection, thus avoiding many of the fundamental logical difficulties of many widely used alternative approaches (Roth, Field, and Clarke, 1994). As an illustration of the concept of utility, let us cite a simple hypothetical scenario. Suppose an engineering manager has been given the responsibility to select a data network that would streamline the flow of information in the organization. One of the primary attribute to be considered in this case should be the reliability of data network. The utility derived from getting 99% reliability will result in a higher perceived utility than 97% reliability, all other things remaining constant. So, in this case, it will be U 99% > U 97% for the engineering manager (Lang and Merino, 1993). However, the decision-maker might become indifferent among the projects if the utility falls below a certain level. For example, for any reliability level of less that 85%, the decision-maker becomes indifferent to the projects and might start looking for other options (or base the selection of other attributes). At the point of indifference onward, the utility curve becomes asymptotic to the horizontal axis. This will be discussed further in the following section.

Stages in Utility Analysis Utility Theory (Analysis) is a multi-attribute decision-making tool that determines the expected utility for a project, based on the possible values of its attributes. Utility Analysis is comprised of the following stages.

Step 1: Selecting the Attributes Utility Analysis, like any other multi-attribute decision-making technique, starts with the selection of a set of critical attributes whose expected utility is to be determined. The expected utility of the attributes will 180

Multi-Criteria Analysis

subsequently be combined to develop the overall value of the utility function. The selection of the attributes can be based on group discussions, interviews, surveys, and experts’ opinions, among others.

Step 2: Determining the Utility Function and its Shape Utility Function can be defined as a mathematical function that expresses the preferences of decision-making attributes with respect to expected returns and risks associated with the decision. The shape of the utility function is called Utility Curve. For any decision-making attributes (say, reliability as cited in the previous section), there is a most favorable (best) outcome and a least favorable (worst) outcome. The best outcome is generally assigned a value of 1 and the worst a value of 0. The shape of the utility curve depends on the attribute and the decision-maker (Lang and Merino, 1993). For example, let us consider the attribute “reliability.” A reliability of 95% and more has a reading of 9, 90%-95%, a reading of 0.8, 88% - 90%, a reading of 0.5 and 85% - 88%, a reading of 0.2 and anything below a 85%, a reading of 0.1. Then the shape of the utility function (Utility Curve) will assume the one shown in Figure 12.3. Figure 12.3. The Shape of a Hypothetical Utility Function Utility

1 0.8

0

85%

95% 100%

Reliability

As seen in Figure 12.3, the perceived utility of the attribute starts rising only after it passes the 85% mark. A 100% reliability assumes a utility value of 1 (the best outcome) while any value below the 85% level assumes a value of 0 (worst outcome). However, it should be mentioned here that the Utility function shown in Figure 11.4 is just one of many shapes it can assume. It can be either linear or non-linear in nature depending on the attribute and the scale assigned by the decision-makers. For example, the shape of the utility function for “cost” is generally linear and negatively sloped in nature, i.e., the higher the cost, the lower the utility while that of the profit function is positively sloped. In addition, the same decision-making project can comprise of multiple utility function depending on how many attributes the decision is evaluated upon.

Step 3: Determining the Overall Utility of the Alternatives Once the utility functions are derived, the set of values is then replaced in the function in order to determine the utility value of the attribute. A composite utility function consisting of all the attribute utility value forms the overall utility value of the alternative, on which the final decision is based.

12.4.4 Conclusion The utility and value analysis allows a decision-maker to take into account a set of subjective judgments and convert them into a numerical scale. This technique is highly useful when it comes to solving decision-mak181

Engineering Management Handbook

ing problems with multiple objectives, as it effectively incorporates the perceived utilities derived from a variety of attributes, and combines them to form a composite overall utility value. The approach is also useful when money is not the sole criterion but other non-economic criteria are used in tandem with monitory aspects in the overall decision-making process. Utility theory brings structure and method to what is often an ad-hoc decision-making process, providing the decision-maker with a more rational and concrete decision.

12.5 References Blank, Leland and Tarquin, Anthony, Engineering Economy, McGraw-Hill, 2005. Canada, John R., and Sullivan, William G., Economic and Multiattribute Evaluation of Advanced Manufacturing Systems, Prentice Hall, 1988. Cheng, Eddie W. L., and Heng Li, “Information priority-setting for better resource allocation using analytic hierarchy process (AHP),” Information Management & Computer Security, vol. 9, no. 2-3, 2001, pp. 61-70. Dolan, James G., “Involving patients in decisions regarding preventive health interventions using the analytic hierarchy process,” Health Expectations, vol. 3, no. 1, March 2000, pp. 37-45. Environmental Resources Management (ERM), “Multi Criteria Analysis,” Anex 4D,.4 for Strategic Planning Guide for Municipal Solid Waste Management by Wilson, Whiteman & Tormin, Accessed online on 7th July 2008 from http://www.worldbank.org/urban/solid_wm/erm/Annexes/US%20Sizes/ New%20Annex%204D.4.pdf Evaluating Socio Economic development (Evalsed), “Multicriteria Analysis,” Sourcebook 2: Methods and Techniques (2003), Available online at http://ec.europa.eu/regional_policy/sources/docgener/evaluation/evalsed/downloads/sb2_multicriteria_analysis.doc Ganguly, Anirban and Merino,Donald N., “Economic and Non-Economic Analysis of Emerging Microwave Technology,” 27th American Society of Engineering Management (ASEM) National Conference Proceedings, Huntsville, Alabama, October 2006. Ganguly, Anirban and Merino, Donald N., “Applying Analytical Hierarchy Processing in Selection among Alternative Chemical Process,” 28th American Society of Engineering Management (ASEM) National Conference Proceedings, Chattanooga, Tennessee, November 2007. Keeney, Ralph L., “An Illustrated Procedure for Assessing Multiattributed Utility Functions,” Sloan Management Review, vol. 14, no. 1, Fall 1972, pp. 37-50. Kotnour, Timothy and Farr, John V., “Engineering Management: Past Present and Future,” Engineering Management Journal, vol. 17, no. 1, March 2005, pp. 15-26. Lang, Hans J. and Merino, Donald N., The Selection Process for Capital Projects, John Wiley & Sons Inc., 1993. Liberatore, Matthew J. and Robert L. Nydick, Decision Technology: Modeling, Software, and Applications, John Wiley & Sons, Inc., 2003. Roper-Lowe, G. C. and Sharp, J. A., “The Analytical Hierarchy Process and its Application to an Information Technology Decision,” Journal of Operational Research Society, vol. 41, no. 1, 1990, pp. 49-59. Roth, Richard, Field, Frank, and Clarke, Joel P., “Material Selection and Multi-Attribute Utility Analysis,” Journal of Computer Aided Material Design, vol. 1, no. 3, October 1994, pp. 325-342. Saaty, Thomas L., Analytic Hierarchy Process, New York: The McGraw-Hill Companies, 1980. Saaty, Thomas L., “How to Make Decisions: The Analytic Hierarchy Process,” European Journal of Operation Research, vol. 48, no. 1, Sept. 1990, pp. 9-26. Saaty, Thomas L., “How to Make a Decision: The Analytic Hierarchy Process,” Interfaces, vol. 24, no. 6, November / December 1994, pp. 19-43. Saaty, Thomas L., Multicriteria Decision Making—The Analytical Hierarchy Process: Planning, Priority Setting, Resource Allocation, RWS Publications, 2001. Saaty, Thomas L., Decision Making with Independence and Feedback: The Analytic Network Process, Pittsburgh: RWS Publications, 2001.

182

Engineering Informatics – State of the Art and Future Trends



13

Engineering Informatics – State of the Art and Future Trends Li Da Xu Old Dominion University

183

Engineering Management Handbook

13.1 Introduction Engineering informatics is an emerging engineering discipline combining information technology or informatics with a variety of engineering disciplines. It is an interdisciplinary scientific subject focusing on the application of advanced computing, information, and communication technologies to a variety of engineering disciplines. Computer-aided design (CAD), computer-aided engineering (CAE), computer-aided manufacturing (CAM) are the terms that have appeared over the last four decades in the area of computing technology in engineering. Computing technology has had significant impacts on a variety of engineering disciplines. Meanwhile, computing technology in engineering has also continuously promoted the advances in computing technology. In this evolution process, computing technology, computational methods, and a variety of engineering disciplines have increasingly intertwined as the development of the theory and practice in both disciplines (computing technology and engineering) influences each other. Since 1990, the need for a scientific subject called engineering informatics has been recognized, although the subject has not yet been formally recognized as a scientific and engineering discipline. The following are excerpted from reports from either the National Science Foundation or the National Academies: “The structuring of design information and data integration are critical requirements for data sharing between designers separated physically and in time, as well as between companies, vendors and customers. Standards do not yet exist for modeling many engineering and organizational parameters that are essential for design specification and analysis, nor are there standards for structuring rational for decisions and design procedures used” (National Research Council, 1991). “Data communication in a heterogeneous system, validation, and consistency of data, representation of textual and geometrical data, …, analytical models of manufacturing processes are all risky areas of research, requiring multiyear, cooperative efforts. Solutions to these problems are needed…” (National Research Council, 1995). “Interdisciplinary collaborations will be especially important for implementing comprehensive processes that can integrate the design of mechanical systems with the design of electrical systems and software. Successful collaborations, however, will first require overcoming incompatibilities between emerging technologies and the existing technological infrastructure and organizational cultures” (National Science Foundation, 2004). “For many organizations, a fundamental change in the engineering culture will be necessary to take advantage of breakthroughs in advanced computing, human-machine interactions, virtual reality, computational intelligence, and knowledge-based engineering…” (National Academy of Engineering, 2005). In 2008, Subrahmanian and Rachuri first proposed to use the term “engineering informatics” to cover the theory and practice in which computing technology and engineering are interfacing (Subrahmanian and Rachuri, 2008). “Informatics, with origins in the German word Informatik referring to automated information processing, has evolved to its current broad definition. The rise of the term informatics can be attributed to the breadth of disciplines that are now accepted and envisioned as contributing to the field of computing and information sciences. A common definition of informatics adopted by many departments/schools of informatics comes from the University of Edinburgh: “the study of the structure, behavior, and interactions of natural and artificial computational systems that store, process and communicate information.” Informatics includes the science of information, the practice of information processing, and the engineering of information systems” (Subrahmanian and Rachuri, 2008). Informatics has an engineering aspect, which addresses the engineering and operation of information processing systems that compute, store, communicate, and visualize information (Broy, 2006). Subrahmanian and Rachuri (2008) further indicated that the history of computing technology and engineering shows a trend of increasing sophistication in the type of engineering problems being solved. Early CAD was primarily based on computational algorithms and models. Then came the engineering use of artificial intelligence (AI), driven by theories of cognitive science and computational models of cognition. More recently, models of collaboration, and the acquisition and representation of collective knowledge have been introduced. Following this trend, engineering informatics can be defined as “the study 184

Engineering Informatics – State of the Art and Future Trends

of use of information and the design of information structures that facilitate the practice of engineering and of designed artifacts that embody and embed information technology and science to achieve social, economic and environmental goals” (Subrahmanian and Rachuri, 2008). Subrahmanian and Rachuri identified several strands of concepts that support the proposing of engineering informatics as a distinct discipline that interfacing engineering and informatics (Subrahmanian and Rachuri, 2008). As computer scientists or engineers cannot solve engineering informatics problems in the context of engineering systems alone, engineering informatics is an interdisciplinary and collaborative effort. In other words, the lack of required backgrounds among computer scientists in engineering and engineers in computing technology has led to develop a new interdisciplinary subject—engineering informatics. Engineering informatics is an interdisciplinary subject. For example, constructing an embedded software system for engineering purpose requires interdisciplinary efforts in mechanics, the domain, software, hardware, human-machine interfaces, and other disciplines. Engineering informatics is to use the knowledge from both informatics and engineering for forming engineering informatics knowledge framework and base. Similar movements have been seen in individual engineering disciplines. In the construction engineering discipline, initially, several names have been used for the interdisciplinary field related to both construction engineering and computing technology such as “computer integrated construction,” “computing in civil engineering,” and “information technology in construction.” The most commonly used terms are “information technology in construction” or “construction IT.” They were coined in the middle 1990s (Turk, 2006). According to Turk (2006), “years later more sober voices claim that many of the problems in the construction industry, that could have been solved by information technology, are not solved, however not only due to technical issues. It seems appropriate, therefore, to remove the word technology and leave just ‘construction informatics’ (CI), construction taken in the broadest sense of the word to include building, civil engineering, and structural engineering, AEC (architecture, engineering, construction) and other disciplines…” (Turk, 2006). As informatics is applicable in multiple engineering disciplines or span multiple engineering disciplines, as such, the term “engineering informatics” was proposed, coined, and started to be used. It is natural that the informatics for a specific engineering subject start expanding to cover a variety of engineering disciplines, and eventually, a more general term called engineering informatics was proposed and coined. Engineering informatics is considered as a distinct discipline, at the interface between engineering and informatics, in the same vein as bioinformatics and medical informatics (Subrahmanian and Rachuri, 2008). Subrahmanian and Rachuri proposed their view of the field of engineering informatics (for fully represent the original contents, Figure 1 was reproduced from (Subrahmanian and Rachuri, 2008). In Figure 13.1, the inner set of circles marked as informatics covers the fundamental activities associated with informatics in general. The next circle, denoted by Product and Process, identifies the multilevel, multi-scale modeling activities of products and processes. The role that informatics plays in engineering products and processes has been becoming significant in past decades. The outer circles show the inputs to engineering informatics from a number of disciplines that provide the domain knowledge and methods and tools.

185

Engineering Management Handbook Figure 13.1. The Scope of Engineering Informatics Proposed

186

Although Subrahmanian and Rachuri proposed their view of the field of engineering informatics, the scope of engineering informatics, can be further refined. As indicated by Broy (2006), software and systems engineering is the key for constructing information processing systems. In particular, software and systems engineering addresses issues such as requirements engineering and reliability engineering (Broy, 2006). Regli (2007) indicated that, in the information technology in engineering, although there have been great strides made by academic and commercial entities in the past decades, the fundamental problems of information integration remain the same. In 2008, Subrahmanian and Rachuri indicate the numerous incompatibilities in information exchange and coordination. The delays that occurred in Airbus 380 and Boeing 787 are examples of the problems of this nature (Subrahmanian and Rachuri, 2008). The information integration within or across industrial sectors is still a dream. Regli and other researchers have indicated the key technological issue of engineering informatics is “the apparent lack of fundamental progress in areas of information integration” (Regli, 2007). Before the need for engineering informatics was formally presented in 2007 and term “engineering informatics” was coined in 2007 and 2008 (Subrahmanian and Rachuri, 2008; Regli, 2007), a scientific and engineering discipline called Industrial Information Integration Engineering was formerly proposed and recognized by international organizations International Federation for Information Processing (IFIP) and the Industrial Information Integration Engineering (IEEE) in 2005. In June 2005, at a meeting of the IFIP Technical Committee for Information Systems (TC8) held at Guimarães, Portugal, the committee members intensively discussed and formally recognized the important role played by industrial information integration and the innovative and unique characteristics of IIIE as a scientific sub-discipline (Roode, 2005; Raffai, 2007). IIIE is a set of foundational concepts and techniques that facilitate the industrial information integration process; specifically speaking, IIIE comprises methods for solving complex problems when developing IT infrastructure for industrial sectors, especially in the aspect of information integration. It was decided at this meeting that the IFIP First International Conference on Research and Practical Issues of Enterprise Information Systems (CONFENIS, 2006) would be held in Vienna, Austria. In August 2006, at the IFIP 2006 World Computer Congress held in Santiago, Chile, the IFIP TC8 WG8.9 Enterprise Information Systems was established. In 2007, the Enterprise Information Systems Technical Committee was established within the IEEE SMC Society. To further respond to the needs of both academicians and practitioners for communicating and publishing their research outcomes, the science and engineering journal entitled Enterprise Information Systems, was launched in 2007 (Figure 13.2). In 2016, the science and engineering journal entitled Journal of Industrial Information Integration, exclusively devoting itself to the topics of IIIE, will be launched (Elsevier, 2016).

Engineering Informatics – State of the Art and Future Trends Figure 13.2. IIIE Discipline History

The concept of IIIE emphasizes multiple aspects, including one of the major aspects that completely overlaps with the scope of engineering informatics: engineering information integration. This chapter is focused on one of the major aspects of IIIE that completely overlaps with the scope of engineering informatics: engineering information integration. The objective of this article is to introduce to the communities of engineering and engineering informatics the current development and future opportunities that exist in engineering information integration, but it is by no means meant to be exhaustive. In Section II, we briefly discuss the relationship between engineering integration and engineering information integration. Section III describes major techniques or technologies in engineering information integration applicable to engineering informatics, while Section IV concludes this paper.

13.2 Overview of Engineering Information Integration 13.2.1 IIIE-A New Discipline of Industrial Information Integration Broadly speaking, IIIE is a set of foundation concepts and techniques that facilitate the industrial information integration process; specifically speaking, IIIE comprises methods for solving complex problems in developing information technology infrastructure for industrial sectors, especially in the aspect of information integration (Xu, 2015). IIIE has been proposed and studied through identifying its theoretical foundation, body of knowledge, frameworks, theories, and models at multiple levels. The key research questions addressed include: (1) what is the scientific foundation that will provide IIIE with the disciplinary support at the levels of frameworks, theories, and models? (2) And, at each level of IIIE (i.e., frameworks, theories, and models/techniques), how can real-world problem solving support be provided? IIIE is an interdisciplinary discipline with the typical characteristics of giant and complex system. According to the subsystems that make up a system, the number of subsystems involved, and the degree of complexity involved with the subsystems, the overall system can be categorized either as a simple system or as a giant system. If a system is made up of a huge number of subsystems, the system is referred to as a giant system. In addition, if a system has a numerous subsystems and layers and if the relationships among the subsystems and layers are complicated, the system is referred to as a complex giant system. As an interdisciplinary discipline, IIIE interacts with scientific disciplines such as mathematics, computer science, and almost every engineering discipline among the twelve engineering disciplines defined by the National Academy of Engineering in the U.S. The National Academy of Engineering is organized into 12 sections, each representing a broad engineering category. IIIE interacts with almost every one of them in separate layers. In terms of scientific and engineering methods, at the methodolog187

Engineering Management Handbook

ical layer, IIIE interacts with computer science and engineering, industrial systems engineering, information systems engineering, and interdisciplinary engineering. In terms of developing and implementing enterprise systems in different industrial sectors, at the application layer, IIIE interacts with aerospace engineering, bioengineering, civil engineering, energy engineering, communication engineering, material engineering, and earth resources engineering. In addition to the scientific and engineering disciplines, IIIE also interacts with management and social sciences. For example, any effective engineering process relies on effective management. As a result, the perspectives for the workflows that are commonly modeled and represented include managerial perspective. Based on the definition of management defined (Xu and Xu, 2011), in a broad sense, management is the most comprehensive science that covers all the disciplines. Judging from these, IIIE is defined as a complex giant system that can advance and integrate the concepts, theory, and methods in each relevant discipline and opens up a new discipline for the industry information integration purposes, which is characterized by its interdisciplinary nature. Figure 13.3 shows IIIE at the top level; relevant scientific, engineering, management, and social science disciplines at the second level; and application engineering fields at the third level. At the fourth level and the levels below, many relevant frameworks, theories, and models can be listed. Figure 13.3. Discipline Structure of IIIE

Figure 13.3 can be huge in size in order to cover all of the details involved. For example, enterprise interoperability is involved with frameworks such as the ATHENA Interoperability Framework, Business Interoperability Parameters, the CEN/ISSS eBusiness Roadmap, C4 Interoperability Framework (C4IF), the IDEAS Interoperability Framework, the European Interoperability Framework, Levels of Conceptual Interoperability, Levels of Information System Interoperability (LISI) C4ISR, NATO C3 Technical Architecture (NC3TA), and the Organizational Interoperability Maturity Model.

13.2.2 Engineering Integration In today’s global competition atmosphere, industrial systems including engineering systems need to be constantly and smoothly reengineered in order to allow them to respond to the fluctuating market and to 188

Engineering Informatics – State of the Art and Future Trends

technological evolution. In 1980s, traditionally, MRP II systems interface with engineering design systems to receive BOM and routing information. However, the interface is not always advanced, as it is unable to pass critical information back to the engineering design system. In 2000, engineering integration became one of the main components of enterprise systems (Langenwalter, 2000). Figure 13.4 shows the relationship between engineering integration, manufacturing integration, customer integration, and enterprise integration. Figure 13.4. The Relationship Between Engineering Integration, Manufacturing Integration, Customer Integration, and Enterprise Integration

In general, about 90% of a product’s cost is determined during its design cycle; its quality characteristics are also determined during the product design stage. In a typical product development process (such as in plastic injection mold design), the design information flow may not be well supported by the existing systems. If associative relations among engineering features were not available through the system, data consistency and design changes would be difficult to manage. At different stages of a product’s life cycle, from its requirement specifications to its conceptual design to its more detailed structure design and finally to its production, engineering knowledge must be integrated. A complete integration includes the design process, product data management, integration with customers, integration with suppliers, integration with the rest of the organization, and project management. The ways in which the engineering division integrates with the rest of divisions in an enterprise have been intensively researched. According to Kulvatunyou and Wysk (2000), integration can be classified into three types: 1. Data-oriented integration, which integrates CAD, CAPP, CAM, and CIM; 2. Structure-oriented integration, which is an implementation of team-oriented concepts, such as the use of a simultaneous engineering team, a concurrent engineering team, and an integrated product and process development team; 3. Procedure-oriented integration, which refers to concurrent engineering-enabling technologies include QFD, the Taguchi method, axiomatic design, and design for manufacturing and assembly. In concurrent engineering, all of the engineering processes should be supported by integrated computer-aided tools, and should be based on a consistent set of data with different application views. Such applications include conceptual design, structural design, detailed design, design analysis for certain specific engineering aspects, computer-aided process planning (CAPP), and computer-aided manufacturing (CAM) tool path generation, etc. However, this desirable scenario has not been fully realized due to the interoperability limitations of different software packages. 189

Engineering Management Handbook

A concurrent design process consists of many design activities that are interrelated with each other. Concurrent design has become increasingly important in designing complex products. When it is implemented in manufacturing enterprise systems along with engineering integration, it is likely to generate better design. Numerous concurrent design techniques have been developed, such as PERT (Project Evaluation and Review Technique), ISM (Interpretative Structure Modeling), DSM (Design Structure Matrix), Petri nets, and polychromatic sets. Each of these methods has some weakness. For example, PERT is useful for the design processes in which activities have a clear sequential relationship. However, it is inflexible and therefore unable to include feedback information and the iterative characteristics of the concurrent design. Using the adjacent matrix, ISM and DSM can apply partitioning algorithms and other algorithms in the concurrent design process. Although the Petri net is suitable for modeling concurrent processes, it does not have sufficient capacity to represent data flow or handle computational complexity. UML is a graphical and visual modeling language. Integrating UML with polychromatic sets provides a powerful tool for modeling and analyzing concurrent design processes. UML has been applied in concurrent design such that a UML model of concurrent design process has been developed and mapped into a polychromatic sets contour matrix model. Using this novel modeling and analysis method for a concurrent design process based on UML and polychromatic sets, the concurrent design process can be modeled formally and analyzed quantitatively, and the major factors that affect the concurrent design process can be considered. In the CAD/CAM field, the comprehensive design of dimensional and geometric tolerances for mechanical products using computers is called Computer Aided Tolerancing (CAT). This is a focal point of research in CAD/CAM. In the process of product design and manufacturing, the tolerance values of a mechanical part are closely related to its manufacturing process, which not only influences the quality of product but also affects the manufacturing cost. So far, considerable research has been conducted on CAT analysis and synthesis, tolerance information modeling and representation, concurrent tolerance design, dynamic tolerance control, and tolerance information verification. The research covers: (1) the concept for determining the geometric shape and the dimensional and geometric tolerance of a part using a computer. Based on this, designed dimensions and tolerances of the part with a geometric shape can be described using mathematical formulae; (2) the method to control the tolerance of design and manufacturing using computerized dimension chain; (3) the theory of tolerance, which defines the concept of tolerance according to the offset values of the real entity of a part and provides a theoretical basis for its CAT design; (4) the concept of virtual boundary requirements (VBRs), which describe tolerance and conditional tolerance; (5) TTRS (Topologically and Technologically Related Surfaces) theory, which establishes the important theoretical foundation for dimensional tolerance and geometric modeling in the CAD system; (6) and the theory based on wavelet and fractal technology with application in designing the tolerance. With the continuous development of CAT technology, a number of tolerance models have been proposed, such as attribute models, parametric models, kinematic, and DOF models. In attribute models, a tolerance can be directly stored as an attribute of either geometric entities or metric relations. Offset models can obtain the maximal and minimal object volumes by offsetting the object by corresponding amounts on either side of the nominal boundary. However, they cannot distinguish the interactions of different tolerance types. Parametric models represent tolerances as ± variations of dimensional or shape parameters. In current CAD systems, the modeling method for parametric models has been widely applied. Kinematic models use vector additions to analyze tolerances. A kinematic link is used between a tolerance zone and its datum features. TTRS models have many similarities to DOF models. With the development of three-dimensional (3D) CAD system, it has become urgent to construct a 3D dimension chain and to comprehensively design dimensional and geometric tolerances. Researchers have put forward the Variational Geometric Constraint (VGC) theory, which can effectively handle the comprehensive design of dimensional and geometric tolerance, although with some difficulties in computation. In CAD/CAM technology, the importance of CAT research has been emphasized by researchers. Consequently, CAT research is becoming more and more popular. The polychromatic sets theory has been applied in CAT. Based on the VGC theory and TTRS theory, a hierarchical reasoning model of toler190

Engineering Informatics – State of the Art and Future Trends

ance information is developed that applies the polychromatic sets theory to describe the model, optimizes computer-aided generation of tolerance types, and provides a basis for developing tolerance network and designing tolerance. The research introduces the application of the hierarchical reasoning model as well as its reasoning method, based on assembly-oriented tolerance generation. Polychromatic sets theory is a mathematical theory tool that is regarded as a promising approach for many applications. Due to the idea and the theory that were developed, namely to use a standardized mathematical model to simulate different objects, the techniques of polychromatic sets have been widely applied to areas such as product life-cycle simulation, product conceptual design, concurrent engineering, and virtual manufacturing for product modeling, process modeling, and process optimization. In recent years, engineering design has required more and more multidisciplinary design activities. Engineering designers from a number of different disciplinary areas may interact and exchange in the design process. Therefore, seamless integration and efficient processing of engineering data among numerous heterogeneous data sources plays an important role in engineering design. Hence, engineering integration is assumed to support multidisciplinary engineering design activities throughout product development cycles. The ubiquitous characteristics of data diversity, irregularity, and heterogeneity will distinctively differentiate engineering information integration from information integration in other domains in IIIE. This poses a challenge to effective engineering integration. There has been much ongoing research in this area. The topics cover: (1) the methodology for developing a virtualization-based simulation platform in support of multidisciplinary design of complex products; (2) approaches for engineering software integration and product data exchange to support interoperability among different engineering phases; (3) mathematical formulation and optimization method for engineering problems; (4) autogenetic design theory and distributed computing approaches and their applications to multidisciplinary design optimization; and (5) web services-based multidisciplinary design optimization frameworks, which provide data exchange services and integration. The research on engineering integration is becoming more prevalent now. Research has recently been conducted on the methods and models for large-scale engineering projects.

13.3 Enabling Technologies In this section, we will introduce the main enabling technologies for engineering informatics as well as IIIE, which include business process management, information integration and interoperability, enterprise architecture and enterprise application integration, and service-oriented architecture (SOA). Rapid advances in industrial information integration methods have spurred tremendous growth of a variety of techniques. These techniques include business process management, workflow management, EAI, SOA, and others. Many applications require a combination of these techniques. At present, we are at a new breakpoint in the evolution of selected enabling technologies.

13.3.1 Business Process Management Engineering design process modeling can inherit methods and approaches developed in business process management. Theiben, Hai and Marquardt (2008) introduced a methodology for modeling, improving, and implementing design processes in chemical engineering. The method inherits some methods developed in the domain of business process reengineering and workflow management (Theiben, Hai, and Marquardt, 2008). IIIE enables the integration of business processes throughout an organization with the help of Business Process Management (BPM). BPM is an approach that is focused on aligning all of the aspects of an industrial organization in order to promote process effectiveness and efficiency with the help of information technology. Through business process modeling, BPM can help industries standardize and optimize business process, increasing their agility in responding to the changing environment for competitive advantage, accomplishing business process reengineering, and realizing cost reduction. Process modeling is an interesting topic in IIIE and engineering informatics. The modeling, monitoring, and controlling of industrial processes is important, as it enables us to understand and optimize 191

Engineering Management Handbook

192

such processes. Manufacturing process modeling is a typical example. All process details in a manufacturing process that relate to the desired outputs of the process need to be understood. In general, a precise process model that relates processes is required. As such, modeling manufacturing processes is important as it enables manufacturers to understand the process and to optimize the process operation. Modeling industrial control is an example also. Such modeling draws the domain expertise of multiple disciplines/ subjects including ICT, process technology, and factory automation, and industrial communication systems. Process modeling results can be applied to process automation and factory automation. The control and predictive capability of business process management also offers useful insights into quite a few engineering fields covered in Figure 13.3. As previously mentioned, IIIE is interdisciplinary. Industrial process modeling can be listed in Figure 13.3 at a level below Level 3; as such, the industrial process control itself is a complex interdisciplinary subject. To track process-related information and the status of each instance of the process as it moves through an organization, the concept of workflow becomes important. Workflow systems have been considered as efficient tools that enable the business process management, the business process reengineering, and eventually the automation of organizational business processes. Workflow management provides increased process efficiency through improved information availability, process standardization, task assignment on an automatic basis, and process monitoring using specific management tools (i.e., WfMS). Although workflow monitoring and management spans a broad continuum, the key idea of workflow management is to track process-related information. When the first prototype of a workflow system was developed, the early idea of automation of business processes was initiated. Workflow management allows managing workflows for different types of processes, facilitating process automation and providing predictive capabilities, and it enables organizations to maintain control over their processes. Business processes and their related workflow systems have gained greater interest since the early 1990s; research about enterprise business processes and workflows has become a prominent area that attracts attention both from academia and industry. A workflow consists of a number of tasks that need to be carried out and a set of conditions that determine the order of the tasks. The Workflow Management Coalition (WMC) defines a workflow as a computerized facilitation for the automation of a business process, in whole or in part. Three types of workflows are generally recognized in literature. A production workflow is associated with routine processes, and is characterized by a fixed definition of tasks and an order of execution. An ad hoc workflow is associated with non-routine processes, which could result in a novel situation. In an administrative workflow, cases follow a well-defined procedure, but alternative routing of a case is possible. Compared with the other two types, production workflows correspond to critical business processes and possess high potential to add value to the organization. Hence, the administrative workflow is usually the focus of most studies on workflow modeling. Workflow management has been considered to be an efficient way of monitoring, controlling, and optimizing business processes through information technology support and is playing an important role in improving an organization’s performance through the automation of its business processes. Process modeling is not only expected to automate business processes within the organization, but also to automate inter-organizational business processes. As such, more efforts have been focused on the integration of inter-organizational systems to form inter-organizational architecture. For this purpose, it is necessary to study both intra- and inter-organizational business processes with a scientific approach. IIIE is required for addressing complex business processes taking place within and beyond the enterprise. Not only does the intra-organizational business process need to be addressed, but also so does the inter-organizational process. Today, workflow systems are increasingly applied to cooperative business domains including cooperative engineering design, and they are inter-organizational. As such, workflow management needs to be completed on an inter-organizational basis. Inter-organizational business process management also provides enterprises the opportunity to reshape their business processes beyond their organizational boundaries. A changing business environment requires an organization to dynamically and frequently adjust and integrate both its intra- and inter-organizational processes. Additional benefits of interconnecting business processes across systems and organizations include higher degrees of integration and the facilitation of the information and material flows.

Engineering Informatics – State of the Art and Future Trends

Inter-organizational workflows are comprised of intra- and inter-organization workflows. Wolfert et al. (2010) defined intra- and inter- integration and process and application integration in this way: intra-organizational integration overcomes fragmentation between organizational units; inter-organizational integration integrates enterprises in the supply chain; process integration aligns tasks through coordination; and application integration aligns software systems to reach cross-system interoperability. Process integration, as mentioned earlier, is one of main types of integrations and can be either intraor inter-organizational. Due to the closed connections and transformations between process management and workflow management, an intra- and inter-organizational workflow management capability can enhance the performance of intra- and inter-organizational integration. Inter-enterprise workflow architecture supports the interoperations between independent enterprises. Meanwhile, an intra- and inter-organizational workflow management capability can also enhance information sharing at both the intra- and inter-organizational levels, eventually enabling all of the partners in the extended supply chain system to better collaborate, to optimize operations, and to gain competitive advantage. WfMS defines, manages, and executes workflows through the execution of software. WfMS has become a standard solution for managing complicated processes in many organizations since its appearance in the early 1990s. Despite a few failures associated with the introduction of WfMS, workflow technology has managed to become an indispensable part of enterprise systems. Workflow technology can be used to improve the business process and to increase performance, since the improvement can be quantified with respect to lead-time, wait time, service time, utilization of resources, etc. WfMS can be employed as a repository of valuable process knowledge and can act as a vehicle for collecting and distributing knowledge across the supply chain. WfMS can also be used as a platform for knowledge sharing and learning inter-organizationally, and allows the knowledge workers in each organization to perform creative intellectual activities. Practicing inter-organizational workflow management requires coping with technical challenges. The complex nature of business processes, particularly processes spread across multiple organizations, presents technical challenges. Most traditional workflow management systems assume one centralized enactment service, are only able to support workflows within one organization, and have problems in dealing with workflows crossing organizational boundaries. It is critical to ensure that technical problems such as inconsistency do not arise due to the lack of transparency across different organizations. Workflow research can be viewed in terms of three layers. The first layer pertains to issues about intra-organizational workflows, which link activities between the different units within one organization. The second layer corresponds to inter-organizational workflows, which cover distributed processes between different organizations, both of which comprise the inter-organizational workflow. The third layer concerns the workflows in e-business settings. Effective management of business processes relies on sophisticated workflow modeling and analysis. Among the modeling techniques, most of them have shown the capability in graphical representation and formal semantics in modeling workflows in an intra-organizational context. Currently, there is an urgent demand for translation between various models so that different workflow management systems can interoperate with each other. This could lead to methods that will enable the integration of heterogeneous models within a unified framework. The existing modeling techniques have advantages as well as disadvantages. Efforts regarding inter-organizational workflow modeling are exploring the better architectures in order to combine different organizational workflows while continuing to reconcile the differences. Some approaches have been specifically proposed for modeling inter-organizational workflows, such as the routing approach and the interaction model. Some cognitive approaches have been proposed for the dynamic routing of information; meanwhile, new languages have been proposed to handle the routing of information among organizations. In terms of evaluation, qualitative evaluation methods mainly focus on checking for structural soundness, which can usually be done through the validation and verification of workflows. Quantitative evaluation methods require the calculation of performance indices related to workflows. The existing techniques include computational simulation, the Markovian chain, and queuing theory, among others. At present, in the area of workflow management, there has been great interest in service workflow modeling and security management. SwSpec is a service workflow specification language that allows arbi-

193

Engineering Management Handbook

trary services in a workflow to formally and uniformly impose requirements. System flexibility has been considered to be a major functionality of workflow systems. More research is needed for such functionality in order to provide sufficient flexibility for coping with complex business processes. Other topics for research include the communication among multi-workflows in complicated business process, simplifying the workflow modeling process, and automating workflows, among other topics. Existing techniques in process modeling still have limitations as they attempt to address only some of the modeling aspects. For example, business process models may contain numerous elements with complex intricate interrelationships. Efforts are needed to address how to properly capture such complexities.

13.3.2 Information Integration and Interoperabilty

194

Subrahmanian and Rachuri (2008) indicated the numerous incompatibilities in information exchange and coordination. The delays that occurred in Airbus 380 and Boeing 787 are examples of the problems of this nature (Suvrahmanian and Rachuri, 2008). The information integration within or across industrial sectors is still a dream. Regli and other researchers have indicated the key technological issue of engineering informatics is “the apparent lack of fundamental progress in areas of information integration” (Regli, 2007). Although there has been several different explorations of different theories of design and manufacturing, progresses yet to be made that can provide effective methods for information integration (Broy, 2006). Today’s businesses of all sizes need to share data with suppliers, distributors, and customers. Information integration is not only significant for large-scale enterprise or for supply chain integration, but also at the microscopic level. Compressed product development cycles and lifetimes and just-in-time stocking imply that management systems must be interconnected, and the applications composing the information systems of enterprises increasingly need to work together. As such, the demand for integration has been increasing. As a consequence of such developments, enterprise systems are increasingly moving toward inter-organizational integration as the benefits of inter-organizational information sharing become obvious. An inter-organizational system is aimed at providing a higher-level system related to activities that involve the coordination of business processes (both intra- and inter-organizational) and is able to provide an integrated architecture to organizations within the supply chain. Now, more efforts have been focused on inter-organizational systems, and more and more enterprises have moved toward inter-organizational integration in order to support supply chain management. Inter-organizational systems are able to allow communication between partners in the supply chain. Integrated enterprise systems can collect valuable management information for all of the related business processes across the supply chain. By using integrated supply chain management, organizations can better predict their markets, can better innovate in response to market conditions, and can better align their operations across supply chain networks. The integration of inter-organizational systems is a complex task for most enterprises. Several frameworks have been proposed for information integration. Fox, Chionglo and Barbuceanu (1993) indicated that at the core of the supply chain management system lays a generic enterprise model. Hasselbring (2000) proposed a three-layer architecture for integrating different types of architectures. In Puschmann and Alt’s (2004) framework, the data level is considered as a separate layer. Giachetti’s (2004) framework includes a typical characterization of the different types of integration. However, as indicated by Wolfert et al. (2010), the contents of these frameworks are not comprehensive, and an overall framework of information integration has yet to be developed. The current level of engineering integration may be limited by the sophistication of the relevant technologies or by the lack of techniques, and the successful execution relies upon more sophisticated IIIE integration than what is currently available. It is expected that IIIE integration will attract more efficient and effective methods for automated engineering management in which the seamless integration of inter-organizational systems is highly expected. Among the new technologies, IoT and radio frequency identification (RFID) have attracted much attention. RFID is a contactless and low-power wireless communication technology that has application in many areas of the supply chain. The envisioned applications include information to be collected from a network of RFID sensors and IoT combined.

Engineering Informatics – State of the Art and Future Trends

13.3.3 Enterprise Architecture and Enterprise Application Integration “Interdisciplinary collaborations will be especially important for implementing comprehensive processes that can integrate the design of mechanical systems with the design of electrical systems and software. Successful collaborations, however, will first require overcoming incompatibilities between emerging technologies and the existing technological infrastructure and organizational cultures” (National Science Foundation, 2004). To industrial organizations, an enterprise can be an organization, a part of a larger enterprise, or an extended enterprise. An enterprise architecture (EA) defines the scope of the enterprise, the internal structure of the enterprise, and its relationship with the environment. As it describes the structure of an enterprise, it comprises main enterprise components such as enterprise goals, organizational structures, and business process, as well as information infrastructure. An EA is generally considered an important aid for understanding and designing an enterprise. Just as information infrastructure is a component of EA and the term enterprise as used in EA generally involves information systems employed by an industrial organization, EA is highly relevant to IIIE, since IIIE concerns information flow within the entire industrial organization. Enterprise architects use a variety of business models, conceptual tools, and analytical methods to describe the structure and dynamics of an enterprise. Artifacts are used to describe the logical organization of business processes and business functions, as well as information architecture and information flow. A collection of these artifacts is considered to be its EA. Software architecture, network architecture, and database architecture are partial components of an information architecture. An EA’s landscape is usually divided into various domains that allow enterprise architects to describe an enterprise from a number of important perspectives. One of the main domains in EA is the information domain. The important components in this domain include information architecture and data architecture. The other two domains with components that are also highly relevant to IIIE are the Application Domain and its component “interfaces between applications” and Technology Domain with its components as middleware, networking, and operating systems. Representing the architecture of an enterprise correctly and logically will improve the performance of an organization. This includes innovations about the structure of an organization, business process reengineering, and the quality and timeliness of the information flow that represents material flows. Enterprise integration has become a key issue for many enterprises looking to extend business processes through integrating and streamlining processes both internally and with partners in the supply chain. It consists of plans, methods, and tools. Typically, an enterprise has existing legacy systems that are expected to continue in service while adding or migrating to a new set of applications. Integrating data and applications is expected to be accomplished without requiring significant changes to existing applications and/or data. To address this issue, a solution that can help to achieve quality integration is referred to as Enterprise Application Integration (EAI). Originally, EAI was only focused on integrating enterprise systems with intra-organizational applications, but now it has been expanded to cover aspects of inter-organizational integration. EAI facilitates the integration of both intra- and inter-organizational systems. Major EAI-enabling technologies range from EDI to web services and XML-based process integration and provide a flexible, adaptable, and scalable EAI framework. Solutions comprise the efficient integration of diverse business processes and data across the enterprises, the interoperation and integration of intra- and inter-organizational enterprise applications, the conversion of varied data representations among involving systems, and the connection of proprietary/legacy data sources, enterprise systems, applications, processes, and workflows inter-organizationally. EAI entails integrating enterprise data sources and applications so that business data and processes can be easily shared. EAI must be able to integrate the heterogeneous applications that are created with different methods and on different platforms. The integration of enterprise applications includes the integration of data, business processes, applications, and platforms, as well as integration standards. Through creating an integrative structure, EAI connects heterogeneous data sources, systems, and applications intra- or inter-organizationally. EAI aims to not only connect the current and new system processes, but also to provide a flexible and convenient process integration mechanism. By using EAI, intra- or inter-or195

Engineering Management Handbook

ganizational systems can be integrated seamlessly to ensure that different divisions or even enterprises can cooperate to each other, even using different systems. A complete EAI offers functions such as business process integration and information integration, since the core of the EAI technology is business process management. Through the coordination of the business processes of multiple enterprise applications and the combination of software, hardware, and standards together, enterprise systems can exchange and share data seamlessly in a supply chain environment. In general, those enterprise applications that were not designed as interoperable need to be integrated on an intra- and/or inter-organizational basis. As such, legacy and newer systems are expected to be integrated to provide greater competitive advantages. The constantly changing business requirements and the need for adapting to the rapid changes in the supply chain may require help from service-oriented architecture (SOA). EAI provides the integration of both intra- and inter-organizational systems and databases and is moving toward integrating ES with both intra- and inter-organizational applications. The objective of EAI is to facilitate information exchange among business enterprises in a timely, accurate, and consistent fashion, in order to support business operations in a manner that appears to be seamless.

13.3.4 Service-oriented Architecture (SOA) Srinivasan, Lammer and Vettermann (2008) indicated the importance of SOA in engineering informatics. Their paper describes how product information sharing service was architected and implemented using SOA. SOA represents the latest trend in integrating heterogeneous systems that has great potential in engineering informatics. It has received much attention as an architecture for integrating platforms, protocols, and legacy systems, and it has been considered as a suitable paradigm that helps integration, since it is characterized by simplicity, flexibility, and adaptability. SOA represents an emerging paradigm for engineering informatics to use in order to coordinate seamlessly in the environment of heterogeneous information systems, enabling the timely sharing of information in the cooperative systems, and developing flexible large-scale software systems for engineering applications. Some example applications include the information integration based on SOA in agri-food industry (Wolfert et al., 2010), among others.

13.4 Summary and Challenges

196

Although the technologies and applications introduced in this chapter are currently not yet fully used in industry, they are expected to have great potential to play a major role in near future. Efforts focusing on blending the capabilities of existing technology and the emerging technologies are needed. With this blending, industries will be able to harness the power of current and emerging technologies to dramatically improve the performance of industrial information integration including engineering informatics by adopting new technologies. Research indicates that the successful engineering informatics practice relies more upon sophisticated technologies than those that are available now. Research also indicates that training engineers with the capacity of using engineering informatics presents a challenge to us (Subrahmanian and Rachuri, 2008). Although there has been several different explorations of different theories of design and manufacturing, progresses yet to be made that can provide effective methods for information integration (Broy, 2006). Lack of a single stakeholder is another challenge. As such, it is difficult to evaluate economic costs and benefits of information interoperability (Broy, 2006). In addition, developing universal metrics for information integration and solving “system of systems” design can also be challenging (Broy, 2006). The interdisciplinary nature of engineering informatics implies another challenge as the complexity level rising as it involves a multiplicity of informatics and a variety of engineering subjects. There are still many challenges and issues that need to be resolved in order for IIIE and engineering informatics to become more applicable. Engineering informatics involves complexity that mainly stems from their high dimensionality and complexity. In recent years, there have been significant developments in this newly emerging technology, as well as actual and potential applications; however, the development of advanced methodologies, especially formal methods and a systems approach, have to be synched with

Engineering Informatics – State of the Art and Future Trends

the rapid technological developments. For engineering informatics, there exists a gap between the level of complexity inherent and the rich set of formal methods that could potentially contribute. Despite advancements in the field of IIIE and engineering informatics, both in academia and industry, significant challenges still remain. Both IIIE and engineering informatics will continue to embrace cutting-edge technology and techniques, and will open up new applications that will impact industrial sectors. IIIE and engineering informatics can and will contribute to the success of this endeavor.

13.5 References Broy, M., “The ‘Grand Challenge’ in Informatics: Engineering Software-Intensive Systems,” IEEE Computer, October 2006, pp. 72-80. Elsevier, Journal of Industrial Information Integration, vol. 1, no. 1, 2016. Fox, M., Chionglo, J., and Barbuceanu, M., The Integrated Supply Chain Management System, University of Toronto, 1993. Giachetti, R., “A Framework to Review the Information Integration of the Enterprise,” International Journal of Production Research, vol. 42, no. 6, 2004, pp. 1147-1166. Hasselbring, W., “Information System Integration,” Communications of ACM, vol. 43, no. 4, 2000, pp. 32-38. Kulvatunyou, B., and Wysk, R., “A Functional Approach to Enterprise-based Engineering Integration,” Journal of Manufacturing Systems, vol. 19, no. 3, 2000, pp. 156-171. Langenwalter, G., Enterprise Resource Planning and Beyond. Boca Raton, FL: St. Lucie Press, 2000. National Academy of Engineering, “Educating the Engineers of 2020: Adapting Engineering Education to the New Century,” 2005, p. 10. National Research Council, “Improving Engineering Design: Designing for Competitive Advantage,” 1991, p. 55. National Research Council, “Information Technology for Manufacturing: A Research Agenda,” 1995, p. 81. National Science Foundation, “ED2030: A Strategic Plan for Engineering Design,” 2004, p. 10. Puschmann, T., and Alt, R., “Enterprise Application Integration Systems and Architecture-the Case of the Robert Bosch Group,” Journal of Enterprise Information Management, vol. 17, no. 2, 2004, pp. 105116. Raffai, M., “New Working Group in IFIP TC8 Information Systems Committee: WG8.9 Working Group on Enterprise Information Systems,” SEFBIS Journal, vol. 2, 2007, pp. 4-8. Regli, W., “The Need for a Science of Engineering Informatics,” Artificial Intelligence for Engineering Design, Analysis and Manufacturing, vol. 21, 2007, pp. 23-26. Roode, R., “IFIP General Assembly September 2005. Report from Technical Committee 8 (Information Systems),” Gaborone, Botswana, August 27, 2005. Srinivasan, V., Lammer, L., and Vettermann, S., “On Architecting and Implementing a Product Information Sharing Service,” Journal of Computing and Information Science in Engineering, vol. 8, 2008. Subrahmanian, E., and Rachuri, S., “Guest Editorial Special Issue on Engineering Informatics,” Journal of Computing and Information Science in Engineering, vol. 8, 2008. Theiben, M., Hai, R., and Marquardt, W., “Design Process Modeling in Chemical Engineering,” Journal of Computing and Information Science in Engineering, vol. 8, 2008. Turk, Z., “Construction Informatics: Definition and Ontology,” Advanced Engineering Informatics, vol. 20, 2006, pp. 187-199. Wolfert, J., et al., “Organizing Information Integration in Agri-Food-a Method based on a Service-oriented Architecture and Living Lab Approach,” Computers and Electronics in Agriculture, vol. 70, no. 2, 2010, pp. 389-405. Xu, L., Enterprise Integration and Information Architecture, New York: CRC Press, 2015. Xu, S., and Xu, L., “Management: A Scientific Discipline for Humanity,” Information Technology and Management, vol. 12, no. 2, 2011, pp. 51-54. 197

198

Basic Accounting and Finance

14 Basic Accounting and Finance Donald N. Merino Stevens Institute of Technology

199

Engineering Management Handbook

14.1 Introduction 14.1.1 Importance of Accounting to Engineers Why is accounting important to engineers? One reason was that engineers needed to know the cost and profitability of the products they design. For example, look at the ever-changing technology of your PC. What drives these changes? The continual increase in the cost performance of new chips, printed circuit boards, disc drives, etc. all result in new and improved products being offered every year. For companies and governments involved in research and engineering, Design for Cost (DFC) and Cost as an Independent Variable (Systems Engineering, Concurrent Engineering) are the most recent concepts that help define the synergy between economics and engineering. Knowledge of basic accounting and finance are essential to understand this concept.

14.1.2 Accounting and Engineering Economics Economic studies for capital project selection depend on estimates of cash flows. You do not have to be a professional accountant or estimator to make an economic analysis. A practitioner of engineering economics is expected to understand the fundamentals of both disciplines. The data on which such forecasts are based come from many sources, but the most important source is accounting records.

14.1.3 What is Accounting? Accounting in its simplest form is scorekeeping—it is observing, measuring, recording, classifying, and summarizing the financial data associated with the multitude of transactions occurring in the operation of a business. As such, the accounting function is historical, not predictive. Other non-accounting instruments are predictive, such as sales forecasts, production plans, expense budgets, and cash flows. However, they are usually based, at least in part, on solid historical accounting data. If that is the only purpose, one might ask why keep score? In reality, the main purpose of maintaining good accounting data is to have it available for use in decision-making.

14.1.4 Users of Accounting Information Many people use accounting information in order to make financial decisions. The following are some examples: • Individuals: You and I use accounting information to manage bank accounts, to make investments, and to make decisions concerning many different types of purchases. • Business: In corporations, the managers of the organization use accounting information to determine accountability for past activities and as a basis of decision-making for future events. • Investors and Creditors: These are people who may decide invest money in the organization. They use accounting information to evaluate risk and return on a prospective investment, i.e., as an aid in making the decision to invest or not. • Regulators: Examples of regulators include the Securities Exchange Commission (SEC), Internal Revenue Services (IRS), and other Tax Collectors. These organizations use accounting information to evaluate managers’ adherence to proper, prescribed procedures and their overall managerial performance.

14.2 Basic Accounting 14.2.1 Introduction

200

The following tutorial will help you get a better understanding of the concepts and principles of basic accounting. You should fully understand this tutorial before moving on to the next one. These ideas will be critical to fully grasping the information that is presented later. This section presents information on the various types of accounting entities, accounting transactions, basic accounting terminology, and financial statement terminology, including the well-known “accounting equation.” This basic section will not only help you better understand financial accounting and engineer-

Basic Accounting and Finance

ing economics, but also give you insights and knowledge to help you read and interpret financial reports, such as SEC filings and personal or corporate tax information.

14.2.2 Financial Accounting Accounting is the management discipline that deals with the financial condition and the financial performance of economic entities by recording, analyzing, and reporting the transactions in which they engage in order to assess their current performance and to forecast their future performance.

Entities Entities are the bounded systems whose financial records may be examined to determine their state of financial health. There are two types of entities, for-profit and not-for-profit. A convenient classification follows: • For-profit (business) entities • Sole proprietorships • Partnerships • Corporations • Not-for-profit entities • Private-sector organizations (usually charitable or religious) • Public-sector organizations (government) The above classification does not include you, the individual consumer, but when an accountant helps you prepare your income tax return, you are an economic entity. The capital selection process applies to both for-profit and not-for-profit entities. Business entities are found in the private sector of the economy. However, there are public sector entities that compete and function much like their private counterparts. Utility companies which are owned by national or local governments are one of many examples. For these and other not-for-profit entities, the terms “profit” and “loss” are replaced with “surplus” and “deficit” respectively. For profit economic entities include sole proprietorships, which are owned by one individual; partnerships, which are owned by two or more individuals; and corporations, which are owned by a few or many shareholders. The sole proprietorship is the most common form of business entity, but corporations are dominant in terms of revenues and profits. Table 14.1 compares the three forms for the year 2000. Table 14.1. Comparison of Types of Business Entities Business

Annual Receipts

Number  

 

(Millions)

Dollars Percentage

 

(Billions)

Percentage

Proprietorships (non-farm)

20.6

77.1%

1,139

4.94%

Partnerships

1.00

3.74%

2,316

10.04%

Corporations

5.10

19.1%

19,593

85.01%

26.70

100%

23,048

100%

 

Ref: U.S. Bureau of the Census, Statistical Abstract of the United States, 2004

201

Engineering Management Handbook

Proprietorships and partnerships are not “legal entities” in the eyes of the law. That is, they do not have an independent identity under the law. Therefore, their owners are fully responsible for their acts and obligations, and creditors can, if necessary, go beyond the assets of the business to seek the personal assets of the owners in order to satisfy their claims. Corporations, on the other hand, are “legal entities.” Their shareholders have limited liability, which means that corporations are responsible for their own acts and obligations. Creditors can rely only on corporate assets for the satisfaction of their claims, not those of shareholders, employees or members of the board of directors. Proprietorships and partnerships are generally managed by their owners; corporations are not. The shareholders elect a board of directors, which appoints executives to serve as managers. In short, ownership and management are divorced (and, as a result, the interests of the managers may conflict with those of the shareholders!). Examples of not-for-profit entities in the private sector include universities, schools, hospitals, museums, and charitable organizations. Examples are also found in the public sector, but these function under the aegis of federal, state, and local governments, which are economic entities, along with school districts, water sanitary districts, public utility and transit authorities, and all other governmental bodies subject to financial review. Thus, economic entities can be as large as a global corporation or as small as the corner newsstand. They can be an entire organization or one of its parts, and, as mentioned, they can also include you and the author of this tutorial. Next, our attention will be focused on corporations and governmental entities because these are the major disbursers of money for capital outlays.

14.2.3 Transactions A transaction is a piece of business—a sale, a purchase, a borrowing, the repayment of a loan, the payment for a service, the issuance of stock, the repurchase of stock, and so on. Transactions allow entities to function. Transactions, when properly recorded, analyzed, and reported give us the financial condition and the financial performance of economic entities.

14.2.4 Financial Condition The “financial condition” of an entity is its financial state of health or well-being at any given point in time. It is an assessment based on information reported in the firm’s financial statements. One critically important financial statement (report) is called the “balance sheet” or, more formally, “the statement of financial condition.” This statement is a “snapshot” or an instantaneous view of the balances of the firm’s accounts at the time that the statement is prepared. (Note: while the Balance Sheet represents the balance of accounts at one specific point in time, the Income Statement and Statement of Cash Flow show the sum effect of all transactions entered into by the firm over a stated period of time.) The points of time usually selected for publication of the financial statements are at the end of (1) each month, of (2) each quarter, and of (3) each year. Monthly issues of the balance sheet are primarily for use by management and might not be disseminated to the public. Quarterly and annual issues are generally published for use by those stakeholders who need to know or wish to know an entity’s current financial condition; stakeholders include lenders, suppliers, creditors, and shareholders.

14.2.5 Financial Statement Terminology Account The “account” is the basic unit for recording information of a firm’s financial database. Accounts are grouped into three basic types—assets, liabilities and owner’s equity. An account is a detailed record of the transactions affecting one or more of these three types. Under the present, widely-used system, each transaction affects at least two accounts, hence the name “double entry accounting system.” 202

Basic Accounting and Finance

Asset An asset is a resource that an entity owns or controls in order to achieve a future benefit (profit, share of market, and competitive advantage). Some examples of assets are cash, inventory, accounts receivable, equipment, furniture, land, and buildings. • Cash: This account shows the amount of money that a company holds in currency or in its bank account. • Inventory: This account reflects the cost (and possibly the quantity) of goods, held for use or intended for sale, that a company owns. Merchandise (goods intended for sale to customers) is recorded as an asset when it is purchased or when it is produced. In addition, raw materials (intended for use in production) are also considered to be inventory. The inventory account increases when inventory is purchased or produced and, conversely, decreases when the goods are sold or the materials are used. • Accounts Receivable: This account shows the amount of credit sales the firm has made, that is delivery of goods or services to customers who then say “Charge it!” or “Bill me…I’ll pay you later.” It is the amount of money that customers owe to the company and have promised to pay in the future for goods and services that they have already received. When the customers do pay, accounts receivable decreases and cash increases. • Notes Receivable: A promissory note that says that the customer is going to pay an agreed upon amount in the future. • Equipment and Furniture: The original (historical) cost of each item of equipment and furniture is entered (written into, keyed into) an asset account. This account, then, shows the original cost of individual items and the total cost (investment) for all of the items. • Land: This account records the cost of the land that is owned by a business. • Buildings: This account records the initial cost of buildings owned by the business. Some examples are factories, office buildings, distribution centers, etc. Land and buildings deliberately purchased for resale are entered into a different account called an “investment account.”

Liability A liability is an obligation or debt that is payable to a creditor. Some examples of liabilities include accounts payable and notes payable. • Accounts Payable: This account is the opposite of accounts receivable; it is the amount that the business owes to its suppliers as a result of credit purchases. • Notes Payable: This account is the opposite of notes receivable. It is a promissory note stating that the business will pay in the future for goods or services (previously) acquired on credit.

Owner’s Equity Imagine that all of the assets of a firm were sold and the cash received was used to pay all of the firm’s liabilities. The remaining cash (value) is called “owner’s equity.” Equity represents the value of the investment in a business by its owners. Following are accounts that affect owner’s equity: • Revenues: For proprietorships, partnerships and corporations, revenue is the total of prices for goods and services that customers agree to pay; revenue is earned through the sale of goods or services. Revenues increase equity. • Expenses: For proprietorships, partnerships and corporations, expenses are the cost of resources used for producing and delivering goods or services to customers, e.g., rent, salaries, electricity, gas, etc. Expenses decrease equity. • Retained Earnings: For corporations the retained earnings account holds the value of the accumulative profits and losses of the firm since its inception. These retained earnings (profits) are a major source of the firm’s investment capital. 203

Engineering Management Handbook

• •

Capital: For proprietorships, partnerships and corporations, capital is the amount of the owner’s investments over the lifetime of the business. Usually this investment takes the form of cash payments for shares of the firm’s stock. Dividends and Withdrawals: For a proprietorship or partnership, cash amounts withdrawn (formally) by the owner from the business to be used to pay personal expenses is called “withdrawals” or “draw.” After making a withdrawal, the asset cash and the owner’s equity both decrease. For a corporation, cash amounts paid to stockholders (not all corporations do this) is called “dividends.” Interestingly, corporations cannot deduct dividends paid for income tax purposes. However, as with withdrawals, dividends paid reduce both the cash account and equity.

Equity = Capital (initial and subsequent investments by owners) + Retained Earnings (through last year) + Revenue for this year Expenses for this year – Dividends/Withdrawals made this year

14.2.6 Financial Performance An entity can be in good financial health and still perform badly—although not for long—by living off what it has accumulated in the way of past profits. It can be in poor financial health and perform well; indicating that its health is improving and, if good performance continues, then it will eventually “get well.” Financial performance over a period of time—a month, a quarter, a year—is recorded, analyzed, and reported in the “income statement” and in two additional statements that together constitute the key financial reports for an economic entity—the “balance sheet” (see above), and the “statement of cash flows.” Not-for-profit entities use similar statements with somewhat different names. The income statement, for example, is often called the “statement of receipts and expenditures.” The owner’s equity statement is the “statement of fund balances.” For many such entities, it is not possible to prepare balance sheets, because there is no way of determining the value of certain of their assets. This is particularly true for federal, state, and local governments. How, for example, do you estimate the value of Yellowstone National Park?

14.2.7 Accounting Equation The fundamental equation of accounting is the following: Assets = Liabilities + Owner’s Equity

If the accounting statements do not result in the above equation being in balance, an accounting error has taken place (or a fraud in as occurred!). CPA firms catch errors like unbalanced balance sheets.

14.3 Income Statement 14.3.1 Introduction Income statements are read by managers, the SEC (Securities and Exchange Commission, a federal agency) and anyone who either owns stock or is interested in buying stock in a particular firm. The income statement can give the reader a fair idea of management’s overall performance as measured in terms of profitability revenue and revenue growth, costs, and changes in costs from year to year. The information in the income statement is the basis for important ratios that can provide additional insights into the efficiency and effectiveness with which the firm is being operated. It is important to note that the performance portrayed in an income statement can differ substantially from one firm to another, depending in large part on the industry in which the organization competes. For example, a natural resource company (e.g., Phelps-Dodge) would likely have a different percentage of its revenue going to research and development than would a pharmaceutical company (e.g., Merck). 204

Basic Accounting and Finance

14.3.2 The Income Statement When a product is sold to a customer “ready, willing, and able” to buy, the seller can immediately record the transaction as earned revenue. Because there are very few cash transactions in business these days, it is quite common for the customer to ask the seller to put the amount of the sale “on account” or to “bill me” for the price of the transactions. Therefore, even though no cash changes hands at the time of the transaction, the seller must reflect in the income statement not only the revenue (as noted above) but also the expenses incurred in earning the revenue. (This is in accord with what is called the “Matching Principle,” an important concept within the “Generally Accepted Accounting Principles” (GAAP) that govern accounting in the United States.) (Note: In the Balance Sheet the effect of this type of transaction would be as follows: recognition of the revenue would increase the value of the firm’s Equity; the customer’s promise to pay later would increase Accounts Receivable, with the two aspect of the transaction not only keeping the Balance Sheet in balance, but also exemplifies the “double entry” nature of the accounting system, because the transaction affected both of the two accounts “equity” and “accounts receivable.” In the unlikely event that a customer does pay in cash, the only difference between that and the cashless example described above is that the Cash account, not Accounts Receivable would be affected on the asset side of the Balance Sheet.) The expenses involved in producing goods or services that have been sold, matched as described above in the Income Statement with the Revenue earned, are also amounts owed to their respective suppliers. Once again, the Balance Sheet would be affected depending on how the suppliers are paid, i.e., in cash (very unusual) or later when an invoice is received. If the suppliers are paid in cash, the Balance Sheet would see a decrease in Equity by the amount of the expense incurred, and an increase in (the Liability) Accounts Payable. If paid in cash, the (Asset account) cash would decrease and Accounts Payable would not be affected. Either way, the balance sheet remains in balance and the equity accounts accurately reflect the difference between the assets and the liabilities that belong to the owners. If revenues exceed expenses during any given period, the earnings for the period are positive (that is, a Profit is earned). If the difference is substantial (or, occasionally, even if it is not), management may decide to distribute some of the earnings to the owners. These distributions, which are called “dividends” for corporations and withdrawals (“draw”) for proprietorship and partnerships, reduce the equity of the firm as well as (usually) the Cash account. The portion of the earnings that is not distributed to owners accrues to the firm’s benefit as an increase in Equity; the firm’s Retained Earnings and is a source of investment capital. The statement of income equation for any given accounting period can be expressed as follows: Revenues – Total Costs and Taxes = Earnings=Net Income

Letting R stand for revenue, C for costs and expenses and ∆E for increase in equity over the period (before any distribution to owners) gives us

R-C=∆E

The above equation is the model for statements of income. A typical example is given in Table 14.2 For our example, the revenues for the period are $800,000 (1). The total expenses before taxes are $ 710,000 (CoGS of $500,000 (2) + G&A of $200,000 (4) + interest of $10,000 (6)), and the earnings before taxes, often referred to as the net income before taxes, would be the difference, or $ 90,000 (7). The “net earnings” or income after taxes would be $75,000 (9). The total expenses of $710,000 are broken down into three categories: the cost of goods and services sold (CoGS), which total $500,000 (2); the Operating Expenses (includes General and Administrative expenses (G&A)), which total $200,000 (4); and the miscellaneous expenses, such as interest payments on borrowed funds which total $10,000 (6).

205

Engineering Management Handbook Table 14.2. A Typical Income Statement XYZ Company Income Statement for the Year Ending December 31, 200X (1) Revenues

 

$800,000

(2) Less cost of goods and services

 

$500,000

(3) Gross profit

 

$300,000

Less operating expenses

 

 

(4.1) Sales and marketing expenses

$100,000

 

(4.2) General and administrative expenses

$100,000

 

(4) Total operating expenses

 

$200,000

(5) Income before taxes

 

$100,000

(6) Less: Miscellaneous revenue and expenses (interest)

 

$10,000

(7) Net income before income taxes (NIBT)

 

$90,000

(8) Less: Income taxes

 

$15,000

(9) Net income after income taxes (NIAT)

 

$75,000

Earnings Before (After) Income Taxes are synonymous with Net Income Before (After) Income Taxes. There may be depreciation expenses included in either or both cost of goods sold (for production facilities and equipment) and operating expenses (for administrative facilities and equipment). In general, Income Statements cover a one-year period, with the period or “fiscal year” ending at a specified date, often December 31 of the corresponding calendar year. Many firms for various reasons operate on a fiscal year that ends at a time other than December 31. Annual reports for public firms must be audited by a certified public accounting (CPA) firm, and, if approved, carry the certification of the auditors as well as of the firm’s top managers.

14.4 Balance Sheet 14.4.1 Introduction The Balance Sheet represents the information related to the fundamental accounting equation outlined in the prior chapters. Please review that section and any definitions of the accounts classified as Assets, Liabilities, and Equity before continuing with this section. Companies report this information in annual reports to shareholders and to the SEC, as this is a document that is required to be published annually. The balance sheet will always be in balance unless there is an accounting error (or fraud).

14.4.2 The Balance Sheet The financial condition of an entity is given by its assets (what it owns) and its liabilities (what it owes). The difference between the two is ‘’owner’s equity,” which is also referred to as “net worth,” “net assets” or just “equity”. As we have seen, the Balance Sheet summarizes and exhibits the account balances of the firm in the categories: Assets, Liabilities and Equity. The Balance Sheet is sometimes called the “Statement of Financial Condition.” An important distinction is made between current assets and long-term assets, and current liabilities and long-term liabilities. Assets, Liabilities, and Owner’s Equity are tied together by the fundamental equation of accounting, which is as basic to accounting as the law of conservation of energy is to the natural sciences. The equation is:



206

Assets - Liabilities = Owner Equity

Basic Accounting and Finance

Or, as it is usually written, Rewritten in equation form:

Assets= Liabilities + Owner Equity

A= L + E

Where A = Assets, L = Liabilities, and E = Owners’ Equity You can see and touch tangible assets. You can see and touch the pieces of paper that document liabilities. However, you can’t touch equity, because it is nothing more than the difference that brings balance sheets into balance. (That occurs when the two sides of the fundamental equation are, in fact, equal.) Consider the house that you might have bought for $200,000 with a $175,000 mortgage. The house is an asset; the mortgage is a liability. Your equity is the difference: $25,000. You can’t see, hear, smell, or touch this difference, but you can see the house and you can touch the mortgage note in your desk drawer. Assets may be broken down into three categories, current, fixed and other. Current assets consist of cash and items that can be quickly (usually within a year) converted into cash. Fixed assets (often called “non-current” or “long term”) consist of land (property, which cannot be depreciated), plant (buildings) and equipment (which are depreciable, if owned). Other assets include such intangibles as patents, copyrights and any other asset that is not classified as current or fixed. Sometimes these intangible assets are considered long term because they have long expected useful lives. Usually assets are listed beginning with the most liquid asset at the top and the least liquid at the bottom. Therefore, current assets would precede fixed assets. Marketable securities such as U.S. Treasury bills or certificates of deposit represent very liquid and short-term investments. Hence, they are frequently viewed as a form of cash. Accounts receivable represent the total money owed to the firm by its customers. Inventories include raw materials, work in process (partially finished goods), and finished goods held by the firm. All of these would be considered to be current assets. All assets are entered into the financial data base at their actual (“historical”) cost, which includes transportation and installation costs, if applicable. The term “Net Fixed” means that the value displayed is the difference between total fixed asset costs and the accumulated depreciation related to those assets. The net value of fixed assets is called their “Book Value.” Liabilities are broken down into two major categories, current and long-term (non-current). Current liabilities are amounts due in one year or less. Long-term liabilities are due more than one year into the future. Like assets, the liabilities and equity accounts are listed on the balance sheet from short-term to long-term. Current liabilities include Accounts Payable (amounts owed for credit purchases by the firm), Notes Payable, outstanding short-term loans (typically from commercial banks) and accruals, amounts not yet paid, but owed for which a bill may not yet have been received. (Examples of accruals include taxes due to the government and wages due to employees.) Long-term debt represents that part of any debt for which payment is not due in the next twelve months. One important Balance Sheet term you should be familiar with is “Working Capital.” This is technically the difference between the values of the current assets and the current liabilities. Estimates of the investment required for (the “infusion of ”) working capital enter into the capital project selection process. A General Ledger is a summary of all of the organization’s accounts. An adjusted Trial Balance is prepared from the General Ledger. It is used to organize the information from the general ledger to create the Balance Sheet and Income Statement. Equity is usually broken down into at least two major accounts. The first, paid-in capital (also called Capital, Common Stock) is that portion of the difference between the assets and liabilities that was contributed by owners both initially and whenever additional capital was needed. The second, retained earnings, is that portion of the difference between assets and liabilities coming from Net Income, earned from the production and sale of goods and services, that is, from earnings that were retained in the business and not distributed as dividends to shareholders or as withdrawals to sole proprietors or partners. Managers strive to make the difference between the assets and liabilities at the end of an accounting period larger than it was at the beginning of the accounting period by increasing retained earnings 207

Engineering Management Handbook

by earning profits for the firm. When achieved, this means that the value of the firm for the owners has increased. We can formulate this objective very simply, as follows: Let A1, L1 and E1 be the assets, liabilities, and equity at the beginning of the period and A2, L2 and E2, the assets, liabilities and equity at the end of the period. By the balance sheet equation,

A1 –L1 = E1 and A2 – L2 = E2

Subtracting the first equation from the second gives (A2-A1) – (L2 –L1) = (E2 – E1) or ∆ A - ∆ L = ∆ E

If ∆ E is positive, management has succeeded in increasing the difference between assets and liabilities over the time period under study. The assets total $ 550,000 (9) and the liabilities $200,000(14). The difference of $350,000((9)-(14)) is the equity. As shown in Table 14.3, ASEM LLC had the following items on its December 31, 200X balance sheet: Table 14.3. Assets, Liabilities, and Equity for ASEM LLC  

Dates 200X $73,000

31-Dec

Notes payable

$33,000

31-Dec

$175,000

31-Dec

Accounts receivable, net

$55,600

31-Dec

Non-depreciable assets

$196,000

31-Dec

$19,500

31-Dec

$180,000

1-Jan

Income taxes payable

$23,000

31-Dec

Inventories

$24,000

31-Dec

Long-term debt

Deferred income tax liability Accumulated retained earnings

Prepaid expenses

$9,000

31-Dec

$35,600

1-Jan

$418,000

31-Dec

Accounts payable

$33,700

31-Dec

Goodwill, patents and trademarks

$12,300

31-Dec

Accumulated Net Worth Property, plant and equipment, at initial cost

Short-term Debt

$21,200

31-Dec

$178,000

31-Dec

Retained Earnings

$19,900

31-Dec

Additional paid-in capital

$69,000

31-Dec

Accumulated depreciation

208

 

Cash and equivalents

Basic Accounting and Finance

Table 14.4 contains the Balance Sheet for ASEM LLC, as of Dec 31, 200X. Table 14.4. Balance Sheet for ASEM LLC BALANCE SHEET (as of Dec. 31, 200X) ASSETS Current Assets (= 1 yr) $196,000

 

 

Long-term debt

$175,000

Deferred income tax liability

$19,500

Property, plant and equipment, at initial cost

$418,000

 

 

(Less) Accumulated depreciation

$178,000

 

 

$240,000

 

 

Net depreciable asset, at book value  

 

Total Fixed Assets

$436,000

 

 

Intangible Assets

 

Goodwill, patents and trademarks

Total Long Term Liabilities

$194,500

Total Liabilities

$305,400

 

  STOCKHOLDER’S EQUITY

$12,300

Accumulated retained earnings

$180,000

 

 

(Add) Retained earnings carried over from Income Statement

 

 

Accumulated Net Worth

$35,600

 

 

Additional paid-in capital

$69,000

Total Intangible Assets   Total Assets

$12,300   $609,900

 

$19,900

 

Total Stockholder’s Equity

$304,500

Total Liabilities & Stockholder’s Equity

$609,900

14.5 Stockholder’s (Owner’s) Equity 14.5.1 Introduction The stockholder’s equity was chosen to be expanded upon because it is often the most confusing aspect of the balance sheet and the basic accounting equation. Assets and liabilities are more easily understood, whereas stockholder’s equity is a more abstract concept. It is always important to note that if the accountant knows the value of assets and liabilities, he or she can easily calculate stockholder’s equity.

209

Engineering Management Handbook

14.5.2 Stockholder’s Equity This tutorial is an extension of section 4—the Balance Sheet. While there are multiple sources of equity, the two primary sources of equity are: • The amount provided directly by equity investors, which is called the total Paid-In Capital. • The amount retained from profits (or earnings)—that is, the amount of net income that has not been paid to owners in the form of dividends—which is called Retained Earnings. Creditors can sue the entity if the amounts due to them are not paid. Owners (Equity Investors) have only a residual claim; that is, if the entity is dissolved, they are entitled to whatever is left after the liabilities have been paid. Therefore, liabilities are the primary claim against the assets and equity is the secondary claim. We can describe the right-hand side of the balance sheet in two distinct, but correct ways: 1. As the amount of funds supplied by creditors and owners 2. As the claims of these parties against the firm’s assets The equity section is often labeled “Shareholder’s Equity” or “Owner’s Equity.” Equity consists of capital obtained from sources that are not liabilities. Table 14.3 shows the two sources of equity capital: 1. $275,000, which is labeled “Paid-in Capital”; and 2. $75,000, which is labeled “Retained Earnings” Table 14.5. Two Sources of Equity Capital for Shareholder’s Equity Dec. 31, 200X Paid-in Capital*

$ 275,000

Retained Earnings**

$ 75,000

Total Stockholder’s Equity

$ 350,000

Assume that a company’s retained earnings in the fiscal year 200X + 1 is $100,000 (Paid-In Capital is held constant). Then, the Stockholder’s equity for the following year is shown in Table 14.6: Table 14.6. Stockholder’s Equity Based Upon Two Sources of Equity Capital Dec. 31, 200X + 1 Paid-in Capital*

$ 275,000

Retained Earnings**

$ 175,000

Total Stockholder’s Equity

$ 450,000

* Paid-in capital amount comes from previous year’s balance sheet. ** Retained earnings are computed as follows Retained Earnings for 200X + 1 = Retained Earnings for 200X + Net Income after Taxes (less dividends, if any) for 200X + 1.

14.5.3 Paid-In Capital Paid-In Capital is the amount of capital supplied by equity investors. They own the corresponding equity (shares of stock representing the value to the owners of the firm). The details of how this item is reported depend on the type of organization. XYZ Company is a corporation, and its owners receive shares of common stock as evidence of their ownership. Therefore, they are called Shareholders (or Stockholders). Individual shareholders may sell their stock to another person, but this has no effect on the balance sheet of the corporation. The market price of shares of Microsoft changes practically every day, but the 210

Basic Accounting and Finance

amount of Paid-In Capital reported on the Microsoft balance sheet is not affected by these changes. This is consistent with the entity concept: that is, transactions between individual shareholders do not affect the Balance Sheet of the entity (the economic entity “Microsoft” is distinct from the economic entities “individual stockholders”).

14.5.4 Retained Earnings The other equity item, $75,000, shows the amount of equity that has been earned by the profitable operations of the company and that has been retained in the entity; hence the name, Retained Earnings. Retained Earnings represents those amounts that have been retained in the entity after some part (or none, for some firms) of the company’s earnings (i.e., profits) has been paid to shareholders in the form of dividends. In the form of an equation: Retained Earnings = Earnings – Dividends

Retained Earnings are additions to equity that have accumulated since the entity began, not those of a single year. The amount of Retained Earnings shows the amount of capital generated by operating activities and retained in the entity. It is important to note that retained earnings are not cash. Cash is an asset.

14.5.5 Example of Retained Earnings ASEM Corp. had the following items on its December 31, 200X Stockholder’s Equity Statement as shown in Table 14.7. Table 14.7. ASEM Corporations Stockholder’s Equity Statement on December 31, 200X  

 

Dates 200X

Accumulated Retained Earnings

$273,500

1-Jan

Retained Earnings

($29,600)

Jan 1 - Dec 31

Accumulated Net Worth

$320,000

1-Jan

Additional Paid-In Capital

$71,000

31-Dec

Using the given data, prepare the Stockholder’s Equity Statement for ASEM Corp., as of December 31, 200X is shown in Table 14.8. Table 14.8. Stockholders’ Equity Statement for ASEM Corp as of December 31, 200X TOTAL STOCKHOLDER’S EQUITY (as of Dec. 31, 200X) Accumulated retained earnings

$273,500

(Add) Retained earnings carried over from Income Statement

($29,600)

Accumulated retained earnings

$243,900

Accumulated net worth

$320,000

Additional paid-in capital

$71,000

Total Stockholder’s Equity

$634,900

14.6 Cash Flow Statement 14.6.1 Introduction You have already become familiar with the Income Statement and Balance Sheet. The Cash Flow Statement is the last of the major financial statements that would be taught in an introductory course in accounting. 211

Engineering Management Handbook

The Cash Flow Statement essentially converts the accrual basis of accounting that is used to create the Income Statement and Balance Sheet into a cash basis. Although the accrual style is helpful in analyzing revenues and expenses, organizations also find it useful to have an understanding about the amount of cash the organization has at its disposal.

14.6.2 The Cash Flow Statement Financial management probably spends more time recording, analyzing, and forecasting cash flow than any other financial matter. The reporting is summarized in a statement of cash flows (cash flow statement) in which the analysis starts at the beginning of the period Table 14.9 shows an example. The difference, which is either positive or negative (or possibly zero), is analyzed by breaking down the cash flow into three parts: • Cash flow from operating activities, that is, running the business on a daily basis (i.e., cash received from selling a product/service and paying the related expenses) • Cash flow from investing activities (i.e., cash flows related to buying and selling the facilities in which the firm operates and from purchasing/selling businesses—mergers and acquisitions) • Cash flow from financing activities (i.e., cash related to taking out and repaying loans, buying and selling stocks/bonds and paying dividends). • Or economic studies on project selection, estimates of future cash inflows and outflows from operating activities are prepared with the help of pro forma income statements (forecasts of income). Table 14.9 contains an example of cash flow reporting. Table 14.9. Cash Flow Statement

XYZ Company Statement of Cash Flows, 200X Cash Flow from Operating Activities Net Income………………………………………………………………$75,000 Adjustments: Depreciation Expense… …………………$100,000 Changes in working capital accounts: Decrease in accounts receivable……..…$20,000 Increase in Inventory……………………..$(40,000)* Decrease in accounts payable…………..$(30,000)* Increase in accrued wages………..…… $40,000 Change in working capital……………………..$(10,000)* Total adjustments to net income……………................................$90,000 Total cash flow from operations………………………….…….…..$165,000 Cash Flow from Investing Activities Purchase of plant………………………………$(60,000)* Total cash flow from Investing Activities....................................$(60,000)* Cash Flow from Financing Activities Issuance of long-term debt………………………. $10,000 Dividends paid…………………………………… $(12,000)* Total cash flow from Investing Activities.....................................$(2,000)* Net increase in cash and cash equivalents……….………………$103,000 * Parentheses indicate decreases in cash Note: Cash Flow statements can be prepared using two different methods, each of which determines the same, correct ending balance of cash; the methods are the direct method and the indirect method. Differences in the two methods lie only in the

212

Basic Accounting and Finance Operating Activities section. The description below uses the indirect method to prepare the Cash Flow statement.

The first section of the cash flow statement (Table 9.4) reports how much cash was generated by the operating activities of the period; that is, from the day-to-day activities that bring cash in from customers and pay cash out to employees and suppliers. To do this, we must first convert net income—the bottom line of the net income statement—from an accrual basis to a cash basis. “Cash flow from operating activities” is the difference between operating cash inflows and operating cash outflows. The second part of the cash flow statement reports cash flows from investing activities: acquisition of new fixed assets and cash inflows from sale of existing assets. The acquisition amount may not be an immediate net decrease in cash because the payment of cash may have been partially or completely offset by borrowing an equal amount (loans). Nevertheless, whatever amount of cash was paid is recorded as a cash outflow, and the amount of the borrowing is recorded separately as a financing activity. Companies may obtain cash by issuing debt securities, such as bonds or stock. These are called financing activities. Cash flows from financing activities include cash receipts or disbursements from one or all of the following: the sale of stock by a corporation to provide paid-in capital, entering into a long-term loan, repaying the loan, and distributing dividends or drawings. However, Interest payments on borrowed funds are not treated as financial activities but as operating activities. In the selection process, we usually start with the assumption that the first cost and the working capital (the funds needed to “set up shop” before cash flows in from sales) for a new venture are supplied by equity financing, that is, by investors rather than creditors. If the results are favorable, we then examine a mixture of equity and creditor financing or even consider leasing to conserve cash. The three groups of activities that affect cash flow—operating, financing, and investing—are all involved in the cash flow patterns to help analyze capital investment opportunities. The above discussion shows the cash flow statement for a company.

14.6.3 Example of Cash Flow Statement Table 14.10 and 14.11 contains for Merino Realty had the following items on its December 31, 200X Income Statement and the associated Cash Flow Statement. Note that the income tax bracket for the company is 35%. Table 14.10. Revenue and Expenses for Merino Realty  



 

Dates 200X

Revenues

$334,000

Jan 1 - Dec 31

Interest expense

$14,600

Jan 1 - Dec 31

Cost of sales (Cost of Goods Sold)

$197,400

Jan 1 - Dec 31

Administrative Salary Expense

$23,400

Jan 1 - Dec 31

Insurance Expense

$12,300

Jan 1 - Dec 31

Depreciation Expense

$27,700

Jan 1 - Dec 31

Dividends paid

$3,200

Jan 1 - Dec 31

Interest income

$6,500

Jan 1 - Dec 31

Selling and administrative expenses

$58,000

Jan 1 - Dec 31

The income tax bracket for the company is 35%.

213

Engineering Management Handbook

Using previously given information, Table 14.11 is the Cash Flow from the Income Statement. Table 14.11. Cash Flow from Merino Reality Income Statement CASH FLOW FROM INCOME STATEMENT (Jan. 1 - Dec. 31, 200X) Revenues

$334,000

(Less) Cost of sales (Cost of Goods Sold)

$197,400

Gross Margin

$136,600

(Less) Selling and administrative expenses

$58,000

(Less) Administrative Salary Expense

$23,400

Operating Income

$55,200

(Add) Interest income

$6,500

(Less) Interest expense

$14,600

(Less) Insurance Expense

$12,300

(Less) Depreciation

$27,700

Income Before Taxes

$7,100

(Less) Provision for income taxes

$2,485

Net Income

$4,615

Dividend Paid

$3,200

Net Income after Dividend Add back Depreciation Cash Flow from Income Statement

$1,415 $27,700 $29,115

14.7 Depreciation 14.7.1 Introduction Depreciation is a methodology used by organizations to distribute the cost of a capital asset over a long period of time. For example, if a company invests in an expensive super computer, the company is required by tax law in the U.S. to allocate the cost of that computer over a span of a few years, using the depreciation technique. If the company did not do this, the year in which the company bought the supercomputer would probably result in financial statements that are significantly worse than the year before. It is important to note that depreciation is considered a non-cash expense. Depreciation is often a difficult subject to grasp for students. You may need to review this tutorial two or more times before you fully grasp the concept, or you can review the textbook.

14.7.2 Depreciation Depreciation is the expense associated with allocating the cost of a capital asset (except land) over its useful life. Land, being appreciable, is not depreciated. All other capital assets that a company buys are depreciated. The portion of the first cost of an asset that is consumed through use over a period of time is called depreciation (depreciation expense).

14.7.3 Depreciation Terminology First Cost (Historical Cost) It is the original price paid for an asset, i.e., the price at which the asset was acquired, including transportation and installation. 214

Basic Accounting and Finance

Estimated Useful Life It is the length of service the business expects to get from an asset. Some define useful life as the length of time that a “prudent manager” would decide to use the asset.

Accumulated Depreciation This account is used to show the cumulative sum of all depreciation expense for an asset from the date on which the asset is acquired. The balance of this account increases over the life of the asset that is being depreciated.

Book Value It is the first cost of a depreciable asset, less its accumulated depreciation. Book Value = First Cost - Accumulated Depreciation

Depreciable Cost Depreciable cost can be defined as: Depreciable Cost = First Cost of the Asset – Estimated Salvage Value

Estimated Salvage Value It is also known as estimated residual value, or scrap value. It is the expected cash value of an asset if sold at the end of its useful life.

Plant Assets Disposal Plant assets have a useful life. When these plant assets cease to be productive, they are usually disposed of. They can be sold, exchanged or scrapped. When they are sold, a gain or loss on the transaction may be incurred. It is a gain (technically called a “depreciation recovery”) if the asset is sold for more than its book value and it is a loss if the asset is sold for less than its book value. We will explain how to calculate the after tax amounts in section 14.8.3.

14.7.4 Depreciation Methods The depreciable cost of eligible capital assets is expensed over its estimated useful life. Tax authorities allow various depreciation methods to be used, and so there are a number of methods for calculating depreciation. We will discuss (1) the straight-line method, (2) MACRS, and (3) the double-declining balance method.

Straight-Line Method Straight Line Depreciation Per Year =



First Cost - Salvage Value Useful Life in Years

An equal amount of depreciation expense is assigned to each year of the asset’s useful life. The depreciable cost is divided by the useful life in years to determine the annual depreciation expense. SL Example: A computer is purchased for $2,200 on January 2001. The salvage value of the computer is $200 and its useful life is four years. Calculate, using Straight Line method, • Depreciation expense • Accumulated depreciation at end of each year • Book value at the end of the year for each year of useful life of the asset 215

Engineering Management Handbook SL Depreciation = Cost – Salvage No.Years SL Depr. = 2200 – 200 = 500 4 SL Depr Rate = 1 = 1 = 0.25 = 25% Useful life 4

Table 14.12 presents depreciation schedule for this example. Table 14.12. SL Depreciation Example Depreciation Rate

Depreciable Cost

Depreciation Amount

1

0.25

$2,000

$500

$500

$1,700

2

0.25

$2,000

$500

$1,000

$1,200

3

0.25

$2,000

$500

$1,500

$700

4

0.25

$2,000

$500

$2,000

$200

Year

Initial Asset Cost

Accumulated Depreciation

$2,200

Asset Book Value $2,200

Thus, Annual Depreciation Amount = Depreciation Rate X Depreciable Cost = (.25 X 2000) = $500 / yr Depreciable Cost = Cost - Salvage Value = $2200 -$200 = $2000 Note that at the end of amortization, the Book Value must equal the estimated Salvage Value.

MACRS: Modified Accelerated Cost Recovery System MACRS is mandatory for the federal tax returns under Tax Reform Act of 1986. An accelerated method similar to the double-declining balance method, it allows deducting larger amounts during the first years of the asset’s life. The depreciation of a particular asset depends on the classification of that asset. MACRS defines eight property classes (3 years, 5 years, 7 years, 10 years, 15 years, 20 years, and two real property classes). These property classes and the applicable MACRS rates can be referred from Chapter 16 of Lang/Merino text. To determine the annual depreciation expense for an asset using MACRS, first locate the asset in the appropriate property class table. In the table, find the year of ownership for the asset and multiply the first cost of the asset by the percentage given in the table for that year of ownership.

MACRS Example Stevens acquired, for an installed cost of $40,000, a machine having a recovery period of five years. Using the applicable MACRS rates, the depreciation expense each year is shown in Table 14.13.

216

Basic Accounting and Finance Table 14.13. MACRS Depreciation Example Year

Cost (in $) (1)

Percentages (%) (2)

Depreciation (in $) (1 * 2 = 3)

1

40,000

20

8,000

2

40,000

32

12,800

3

40,000

19

7,600

4

40,000

12

4,800

5

40,000

12

4,800

6

40,000

5

2,000

TOTAL

100%

40,000

Double-Declining Balance Method This method is an accelerated depreciation method as MACRS is. This method calculates depreciation by allocating larger amounts to the depreciation expense account in the earlier periods of the useful life compared with the later periods. It computes annual depreciation by multiplying the asset’s book value by a constant rate, which is two times the straight-line depreciation rate. There is much more on depreciation. However, at this time you should be aware that the depreciable life of an asset (the period of time over which its cost is prorated) does not have to be, and often is not, the same as its useful life. Accelerated depreciation and inflation often bring the book value of an asset far below its market value, which is why the discussion that follows on gains and losses from the disposal of assets is important.1

14.7.5 Gains and Losses from the Disposal of Assets Gains and losses from the disposal of assets result from the differences between market value and book value. If market value exceeds book value, there is a gain; if it is less, there is a loss. Consider the following example: Fair Market Value = FMV; Book Value = BV Tax Rate = TR; FMV – BV = Taxable Gain/ Loss Taxable Gain/ Loss * TR = Taxes; After Tax Entry = FMV – Taxes

Example: (Gains and Losses)—The car bought for $30,000 is sold for a cash payment of $20,000 at the end of two years, at which time its book value is $18,000 ($30,000 less two years of accumulated depreciation at $6,000 per year). Its market value on the day of sales is therefore $2,000 more than its book value. The tax rate is 40%. FMV-BV = Taxable Capital Gains: $20,000 – $18,000 = $2,000 Taxes: $2,000 * 0.4 = $800 After Tax Cash Flow: $20,000 - $800 = $19,200

This completes our overview of accounting. The concepts presented here can be understood by one more example given later in the tutorial. We hope you have found this discipline a more conceptual and stimulating subject than its image as “bookkeeping” usually conveys.

14.8 After Tax Analysis 14.8.1 Introduction The section is related to prior chapters on the Income Statement, Balance Sheet, Cash Flow Statement and Depreciation. You may wish to review those topics before continuing with this section. 217

Engineering Management Handbook

The tax code is very complicated for individuals and corporation in the United States. You probably, at some point in your life, have filled a tax return with the Internal Revenue Service (IRS). Organizations have an even more daunting task with more complicated rules.

14.8.2 After Tax Analysis and Cash Flow The taxes paid and tax related items like depreciation and investment tax credits have a significant impact on the economics of all aspects of corporate activity, but particularly on Capital Investment. Economic decision-making is based on Cash Flows (not accounting profit). If you take a course in Corporate/Managerial Finance you will find that corporations are managed on the Cash Flow. In Section 14.6 (Cash Flow) you saw that Cash Flows can be generated as part of: • Operating Activities • Capital/Finance Activities Operating cash flows can be determined from the Income Statement. Note that an operating cash flow is Net Income after Tax plus depreciation. Remember that depreciation is a non-cash (accrual) expense. Operating cash flows are periodic over a number of years. Capital Cash Flows can be determined from the Balance Sheet. There are a number of activities that impact the Capital Cash Flows. Change in inventory levels, financing activities and capital expenditure are some of these activities. For engineering economics we are interested in investments for depreciable (plant, equipment, etc.) and non-depreciable (land) capital. For capital projects we are also interested in loans (to finance capital) and the after tax sale/disposal of capital at the end of the project. In the following sections we will provide the calculations and format for • Net Cash Flow from Operating Activity • Net Cash Flow from Capital Activity as well as the After Tax Cash flows for • Depreciation • Loans/Borrowings • Salvage/Disposal of Capital Items

14.8.3 After Tax Cash Flow from Depreciation Charges Section 14.7 described various depreciation methods and how to calculate them. Note that depreciation is a non-cash (accrual) entry and as such is added back to the Net Income after tax to yield the cash flow from operating activities.

14.8.4 After Tax Cash Flow from Investment Tax Credit From time to time government/agencies (U.S., State and sometime local) give incentives for business to make capital Investments. This is usually during a recession or natural disaster (like Hurricane Katrina) and is designed to increase capital spending. Why Capital Spending? The concept is that for every dollar spent on capital goods, the Gross National Product (GNP and GDP) will increase many fold (6 to 8 times). In Macro-economics this is called the “Multiplier Effect.” The Investment Tax Credit (ITC) allows a company to subtract some part of the Capital purchase price from the Income Tax it pays. For example, if the ITC is 10% and the company makes a $200,000 capital investment then the company can deduct $20,000 from its taxes (usually in year 0 or 1). It is assumed that the company will keep the Capital asset in service for a certain number of years. If it does not, it may have to give back some of the ITC. This is called re-capture. You need to seek Tax Counsel on this matter because the rules are complex. ITC for solar panels or energy saving items like Hybrids cars are examples that apply to consumers. 218

Basic Accounting and Finance

14.8.5 After Tax Cash Flows from Loans The after tax analysis with loan repayment includes are the steps in Table 14.13 plus interest expenses. Interest is subtracted from the net cash flow before income taxes are calculated since interest is tax deductible, requiring it to be eliminated from the taxable income. Interest charges benefit the organization because it reduces the taxable income and, hence, the income taxes. The principal payments are being subtracted in Table 14.5 since principal is NOT tax deductible. When dealing with loans, we need to understand the difference between interest and principal. Principal is the actual amount you have borrowed or the remaining, unpaid balance owed to the lender. Interest is most easily described as the fee charged by the lender for lending you the principal. It is usually expressed in as annual percentage of the principal. Interest is generally tax deductible, meaning it REDUCES taxable income. Table 14.14 provides an example on how to “amortize” a loan. The example assumes that the interest rate is 10% and the loan is repaid in four years. The amount of the loan is $80,000. Annual payment is calculated using the time value of money equation:

A = P (A/P, I, N) = $80,000(A/P, 10, 4) A = $80,000 x .3155 = $25,240/year



Table 14.14. Loan Balance / Amortization Table Year

Beginning-of- year balance

Annual payment*

Annual Interest

Annual Principal Repayment

End-of-Year Balance

(1)

(2)

(3)

(4)

(5)

(6)

10% of (2)

(3) - (4)

(2) – (5)

01

$80,000

$25,240

$8,000

$17,240

$62,760

02

$62,760

$25,240

$6,276

$18,964

$43,796

03

$43,796

$25,240

$4,380

$20,860

$22,936

04

$22,936

$25,240

$2,294

$22,936

0

Total

80,000

14.9 Accounting Process 14.9.1 Introduction The accounting process is governed by laws to prevent fraud, insider trading, and unfair practices. This tutorial does not introduce these laws; however, you should note that if there is any doubt as to the ethics of the accounting process in your organization, you should consult references available on the subject. Most references guide you to practice methods introduced below.

14.9.2 The Accounting Process The accounting process is a systematic and logical methodology for processing a myriad of transactions to produce the financial statements on which managers, bankers, creditors, suppliers, and shareholders rely. We highlight this process below but suggest that you also go to one or more of the suggested references provided by your professor.

14.9.3 Double Entry Accounting Double entry system of accounting means that each business transaction influences at least two accounts. The accounting equation needs to balance, so if there is a change on one side of the equation, some item(s) on the other side need to be changed to maintain the balance. 219

Engineering Management Handbook

The process begins with the journal or daily record of transactions. We know that every transaction must be recorded in at least two accounts; otherwise, balance sheets would not balance. One of these accounts is debited (entered on the left or “debit” side of the T-account), and the other is credited (entered on the right of “credit” side of the T-account). The words debit and credit refer to the left and right side of the T-account respectively for the purposes below. Their abbreviations are dr. (debit) and cr. (credit). Each “simple” journal entry (a simple entry is one that affects just two accounts) is therefore recorded as shown in Table 14.15. Table 14.15. T-account Method of Recording Debits and Credits Debit Name of first account

Credit

$ XXXXXXX

Name of second account

$ XXXXXXX

Note that accounting requires that the sum of all the debits pertaining to a transaction must equal the sum of all the credit for those same transactions. Not shown, but part of the journal record of the transaction, are the date, the amounts to be “posted” in the affected accounts and, if needed, a brief description of the transaction. Table 14.16 shows how assets, liabilities, and equity should be recorded. Table 14.16. Location of Typical Items That are Posted Increase in assets

Debits

Decrease in assets

Credits

Increase in liabilities

Credits

Decrease in liabilities

Debits

Increase in equity

Credits

Decrease in equity

Debits

14.10 Financial and Managerial Accounting 141.10.1 Introduction This section covers cost-volume-profit (CVP) analysis. CVP analysis is the study of effects of output volume on revenue (sales), expenses (costs) and net income (net profit). The most basic CVP analysis computes the monthly break-even point in number of units and in dollar sales. To apply this analysis, some simplifications must be assumed. The costs have to be simplified into fixed and variable costs. Total Sales Revenue - Total Costs = Profits Sales - Costs = Profits Sales = Costs + Profits Sales = Fixed Costs + Variable Costs + Profits

Therefore, the CVP equation can be written as: Units sold * Selling Price per Unit = Fixed Costs + (Variable Cost per Unit * Units Sold) + Profits

14.10.2 Breakeven Analysis The breakeven point is the point where the sales revenue equals the total costs and is demonstrated in Figure 14.1. The breakeven point is the point where the sales revenue equals the total costs. 220

Basic Accounting and Finance Figure 14.1. Concept of Breakeven Analysis Units

Total Revenue Total Cost Profit

Breakeven Sales (In Units)

Variable Cost

Fixed Cost Loss

Dollars

As you can see, the net income is zero at the breakeven point.

14.10.3 Contribution Margin It is the amount that covers fixed costs first and then provides profits for the period. When the contribution margin is higher than the fixed costs we have a profit, and when the contribution margin is lower than the fixed costs, we have a loss. If the contribution margin is equal to the fixed costs, the profit is zero and this point is called breakeven point. Contribution Margin = Sales Revenue - Variable Costs

14.10.4 Contribution Margin Ratio Contribution Margin Ratio is the proportion of each sales dollar available to cover fixed costs and provide profits. It is the ratio of contribution margin to sales.

Contribution Margin Ratio = Contribution Margin Sales

14.10.5 Breakeven Sales in Dollars We get this equation in terms of the contribution margin ratio. Sales = Fixed Costs + Variable Costs + Profits Sales - Variable Costs = Fixed Costs + Profits

This equation can be rewritten in terms of the Contribution Margin if the equation is divided by Sales on both sides. Contribution Margin Ratio = Contribution Margin Sales Hence, Contribution Margin Ratio = Fixed Costs + Profits Sales

To calculate the breakeven sales, the profits are set at zero, and it gives us the following equation:

Contribution Margin Ratio =

Fixed Costs Breakeven Sales 221

Engineering Management Handbook

From this equation we get the Breakeven Sales,

Breakeven Sales =



Fixed Cost Contribution Margin Ratio

Breakeven Sales in Units = Fixed Costs (Unit Selling Price – Variable Unit Cost)

14.10.6 Target Net Profit To achieve the target net profit, the minimum amount of units that must be sold can be calculated as follows: Selling Price per Unit * Units Sold = Fixed Costs + (Variable Cost per Unit*Units Sold) + Target Net Profit Units * (Selling price per unit - Variable Cost per Unit) = Fixed costs + Target Net Profit Units * Unit Contribution Margin = Fixed Costs + Target Net Profit Units to Earn Target Net Profit = (Fixed Costs + Target Net Profit) / Unit Contribution Margin

14.10.7 Sales to Achieve Target Return on Sales Return on Sales, also known as Income percentage of revenue, helps you determine if the organization is generating enough returns on the sales effort. Following is the formula used to calculate Return on Sales: Return on Sales = Income / Revenue

14.10.8 Degree of Operating Leverage Operating leverage ratio is the ratio of fixed to variable costs. A company has less leverage when the fixed costs are lower than the variable costs. When the proportion of fixed costs in relation to variable costs is high, the operating leverage is high, and the profits are more sensitive to changes in sales volume. The degree is a measure of how changes in sales volume affect the profits. The lower the costs are for manufacturing a unit, the higher the contribution margin per unit. This is why it is important to manage a good cost structure.

14.11 Advanced and Other Topics This chapter covered the fundamentals of accounting, basic financial accounting, and some of the advanced topics in cost costing and cost estimation. Section 14.13 provides a mapping of all of the topics to two standard references in the field. The topics not covered in this chapter are as follows: I. A. II. A. B. C. III. G. H. I. J. IV. A. B. C.

222

Basic Accounting—Fundamentals Asset & Inventory: LIFO; FIFO Basic Financial Accounting Financial Ratios Capital Structure of Firm Stocks, Bonds and Financial Instruments Advanced Cost Accounting—Fundamentals Activity Based Costing (ABC) Flexible & Master Budgets Performance Assessments Ethical Considerations—SoX Advanced Cost Estimation Statistical Cost Estimation Use of Cost Indices and Cost Factors Design for Cost / Affordability / Target Costing

Basic Accounting and Finance

14.12 Summary One chapter can hardly cover the basics of accounting and finance other than a cursory presentation of the subject matter. Yet, few subjects are more important in modern business than the understanding of the finances of a corporation or an individual project. Engineering managers must understand the finances of a business to provide value.

14.13 References The following are two standard references in this field. The following table maps the topics in this chapter to these references. Merino, Donald N., Accounting for Engineers, Engineering Management Body of Knowledge, American Society of Engineering Management, vol 1.1, Nov. 2007. Riggs, Henry E., Financial and Cost Analysis for Engineering and Technology Management, John Wiley & Sons, Inc., 1994, ISBN: 0-471-57415-5; ISBN (13): 978-0471574156. Easton, Peter, Halsey, Robert, McAnally, Mary, and Hartgraves, Al, Financial & Managerial Accounting for MBAs, 1st edition, Cambridge Business Publishers, 2008, ISBN: 0-9787279-1-6.

Topics Referenced to Standard Texts in the Field References Authors I. A B. C. D. E. F. G. H. I. II. A. B. C. III. A. B. C. D. E. F. G. H. I. J. IV. A. B. C.

Basic Accounting—Fundamentals Debits and Credits Assets, Liabilities Income, Expenses Equity Definitions Income Statement Balance Sheet Cash Flow Statement Integration of Income Statement; Balance Sheet and Cash Flow Statement Asset & Inventory Basic Financial Accounting Financial Ratios Capital Structure of Firm Stocks, Bonds and Financial Instruments Advanced Cost Accounting—Fundamentals Fixed and Variable Costs Break-even Analysis Job, Process and Standard Costs Direct and Indirect Costs Cost of Goods Sold; Overhead Costs Contribution Analysis—Profit and Loss Calculations Activity Based Costing Flexible & Master Budgets Performance Assessments Ethical Considerations—SoX Advanced Cost Estimation Statistical Cost Estimation Use of Cost Indices and Cost Factors Design for Cost / Affordability / Target Costing

1 Riggs

2 Easton, ..

3 3 3, 4 2 4 3 11

1 1 1,2 1-3 1-3 1-3 1-3 2, 3

1, 6

6

10 10 10 part

4–6 7–9 12, 13

13 13 17 16 16 18 16 12 10 part Readings

15 16 18 19 19 20 19 20, 21 22 Readings

NA NA NA

15 16 23

223

224

Engineering Economics

15 Engineering Economics Donald N. Merino Stevens Institute of Technology

225

Engineering Management Handbook

15.1 Capital Expenditures 15.1.1 Importance of Capital How efficiently a country, industry, company, or individual uses capital decides whether they prosper economically and whether they survive. Generally, capital investments increase the wealth of the investors. In the case of a country this would be the Gross Domestic Product (GDP). This increase is called a multiplier effect. However not all capital investments yield that multiplier and in some cases the capital spent may not increase GDP. Entities need to choose wisely. The “right” investments in equipment or education can yield great returns.

15.1.2 Capital Selection Process In order to allocate capital efficiently a rationale selection process is needed. The Strategic Plan of an entity that contains capital expenditures usually drives this process. These capital expenditures can be for plant and equipment, software development, hardware manufacture or any activity is required to meet the plan.

15.1.3 Minimum Attractive Rate of Return The minimum acceptable rate of return (MARR) for investments that investors choose including allowance or risk.

15.2 Mathematics of Finance 15.2.1 Rates of Return Money has a time value. The best example is lottery prize payments. You win a Megabucks prize of USD $1,000,000. Based on the rules and regulations, the prize was to be paid in 20 equal annual payments starting this year. However, you wanted to receive the prize as a lump sum payment. Different lottery financing companies are ready to offer you various amounts as a lump sum payment to buy the prize of USD $1,000,000 that is going to payout in 20 years. The reason the dollar amounts vary is due to the fact each company has a different time value and money (% / year). Financing companies are ready to give up current purchasing power for future purchasing power. The offer of each company is different, which means that giving up current purchasing power has different values for different companies. For example, one financing company might be willing to pay USD $573,495 [A(P/A, i, N) where A = 1,000,000 / 20, i = 6% and N = 20] while another one might pay only USD $490,905 [A(P/A, i, N) where A = 1,000,000 / 20, i = 8% and N = 20]. We can calculate rate of return (the rate that creditors and investors will accept for giving up a current purchasing power for future purchasing power), from current and future purchasing power of money by using engineering economics calculation methods.

15.2.2 Simple and Compound Interest They are two kinds of interest: (1) simple interest and (2) compound interest. The basic difference between these two is that with simple interest, you do not earn interest on interest, but with compound interest, you do.

Simple Interest You earn interest only on the principle, which is the money that was initially invested. For example, if you invested USD $1,000 (P) with 10% (i) simple interest for five years (N), you would earn USD 100, 10% of USD 1,000, the first year, USD 100, 10% of USD 1,000, the second year, and so on for as long as your money remained in the bank. In five years, you will receive USD $500 as an interest and your USD $1,000 as the principal. Therefore, your future bank balance from leaving your money at bank for 5 years will be USD $1,500 (F). As you can see the interest on your money is P x i for each period and the total interest amount is P x i x N. 226

Engineering Economics

The equation for simple interest for any combination of F, P, i, and N is; F = P (1 + i x N) where F= Future Value; P = Present Value; I = interest rate; N = no. of years For the example above: The future balance (value) F = $1,000 (1 + 0.10 X 5) = $1,000 X 1.5 = $ 1,500

Compound Interest You earn interest on the principle, which is the money that was initially invested and on the interest that you earn in previous periods. For example, if you invested USD $1,000 (P) with 10% (i) compound interest, compounding yearly, for 5 years (N), your earnings on this USD $1,000 would be: At the end of the year five, you will receive USD $610.51 as an interest and your USD $1,000 as the principal. Therefore, your future bank balance from leaving you money at bank for 5 years will be USD $1,610.51 (F). As you can see the total interest amount on your money is P x ((1+ i)n- 1) The equation for compound interest for any combination of F, P, i, and N is; F = P (1 + i) ^ N For the example above: The future balance (value) F = $1,000 (1 + 0.10)5 = $1,000 X 1.61051 = $ 1,610.51

Comparison of Simple Interest and Compound Interest If we analyze the results of interest calculations for sections 15.2.2.1 and 15.2.2.2, the total return on the investment was USD $110.51 more for compound interest due to the fact that we earned interest on interest. Now, we know the effect of interest method. What about the effect of formula components and length of compounding period? Let us analyze them one by one for both simple and compound interest methods. We are using the example in sections 15.2.2.1 and 15.2.2.2 to make the comparisons. We will change only one component each time to analyze the effect of that component. Principal (P) Let us assume that the available fund to be invested is doubled to USD $2,000 from USD $1,000 for both simple and compound interest examples above. Let us calculate the future balances (values) for both methods. Simple Interest: FNEW = $3,000 Compound Interest: FNEW = $3,221.20 We can easily notice that when we doubled the investment amount, the future balance, value, is doubled for both methods. Note: USD $1,000 is the initial principle. Interest Rate (i) Let us assume that we are still investing USD $1,000 and got better interest rate for both simple and compound interest methods. The new interest rate is 20%. Let us calculate the future balances (values) for both methods. Simple Interest: FNEW = $2,000 Compound Interest: FNEW = $2,488 In simple interest, the total interest earned doubled from USD $500 (USD $1,500 -USD $1,000) to USD $1,000 (USD $2,000 – USD $1,000) while the total interest earned increased more than double for

227

Engineering Management Handbook

compound interest from USD $610,51 (USD $1,610,51 - USD $1,000) to USD $1488 (USD $2,488 USD $1,000). Note: USD $1,000 is the initial principle. Number of Periods (N) Let us assume that we are still investing USD $1,000 with the same interest rate (10%). Moreover, we will invest money for longer term, 10 years. Let us calculate the future balances (values) for both methods. Simple Interest: FNEW = $2,000 Compound Interest: FNEW = $2,594 In simple interest, the total interest doubled from USD $500 (USD $1,500 - USD $1,000) to USD $1,000 (USD $2,000 – USD $1,000) while it increases more than double for compound interest from USD $610,51 (USD $1,610,51 - USD $1,000) to USD $1,594 (USD $2,594 - USD $1,000). Note: USD $1,000 is the initial principle. Compounding Period Compounding period is the length of the time period that elapses before interest compounds. Therefore, at the end of each compounding period you receive interest. Because simple interest does not pay interest on interest, the length of compounding period does not have any effect on calculations. Therefore, we will analyze the effect of compounding for compound interest only. We are about to invest USD $1,000 with 12% nominal interest rate, which is the periodic rate multiplied by the number of compounding periods per year. Moreover, we have four compounding options: (1) yearly compounding, (2) semiannually compounding, (3) quarterly (every three months) compounding, and (4) continuous compounding. The future balances and interest earned at the end of first and tenth year are shown in Table 15.1. Table 15.1. Demonstration of Simple vs. Compound Interest Future Value F of Present Sum P Simple vs. Compound P = $1000

Future Balance

Example: Bank

Interest

N

@ EOY 1

@ EOY 10

A

12% Simple

0

1120

2200

B

12% Compounded Annually

1

1120

3106

C

12% Compounded Quarterly

4

1126

3262

D

12% Compounded Monthly

12

1127

3300

E

12% Compounded Continuously

360

1127

3320

Future Value of F The shorter the compounding cycle the higher is its effect. The table above illustrates this phenomenon. The interest rate is 12% annually, but in case A it is applied as simple interest, in cases B through E we are shortening the compound cycle from one year to quarterly to monthly to continuously. The last column shows the balance at the end of 10 years. As you can see the balance increases as the cycle is shorten, but eventually levels off to $3320 by considering continuous compounding.

228

For Yearly Compounding We will invest USD $1,000 and receive 12% interest. USD $120 will be paid as at the end of first year. Therefore, our future balance at the end of first year is USD $1120 and USD $120 is our interest.

Engineering Economics

For Semiannually Compounding First, we need to find how much interest we will earn for compounding period. Therefore, we need to calculate our periodic rate, the rate for the specified compounding period. Because our nominal interest rate is 12% and we have two periods in a year, our periodic rate is 6%. In this case, when we invested USD $1,000 at the beginning, we will earn 6% for the first 6 months. Then, we will earn another 6% on both the initial amount and the interest earn in the first six months. Fyear= $1,123.60 Our future balance at the end of first year is USD $1123.60 and USD $123.60 is our interest. For Quarterly Compounding Let us calculate the periodic rate for a quarter. We have four quarters in a year and our yearly interest rate is 12%. Therefore, periodic rate is 3%. In this case, when we invested USD $1,000 at the beginning, we will earn 3% for the first 3 months. Then, we will earn another 3% on the balance. This will repeat until the end of year 1. Fyear = $1,125.51 From the results, it is obvious that our future balance at the end of first year is USD $1125.51 and USD $125.51 is our interest. We have learnt about the definition and calculation steps of both simple and compound interest. We know that formulas of simple and compound interest calculations include P, i, and N to determine future value (balance). Let us compare those two interest techniques. Payback is the length of time, usually expressed in years, needed to recover initial cost of a capital investment. It can also be defined as the number of years it takes for the sum of the annual net cash flows to equal zero. For a capital project, yearly cash flows can be either equal or unequal. We can segregate payback period calculations into these two classes.

15.2.3 Effective Interest Rates Periodic Rate: The periodic rate is the compounding rate of interest for a specific compounding cycle, like, daily, monthly, quarterly. If the interest is compounded daily, then it is a rate for 1 day, if it is compounded monthly then the rate is for 1 month. Nominal Rate: The nominal rate is the periodic rate scaled to an annual basis. If the rate is 1% per month, then its nominal rate is 12% / year. In other words, what is the rate per year in terms of straight extrapolation of rate of the year? To go from periodic rate to the nominal rate we must drive the rate by the number of periods that happens within the span of a year. Effective Rate: The effective rate is the equivalent of the periodic rate compounded for the number of periods per year. In the effective rate, of periodic rate is 1% per month, then at the end of the year we will get more than 12%. The 1% monthly has been compounded 12 times. So the effective rate is (1+0.01)^12 – 1. Table 15.2 illustrates the effect if compounding if the compounding cycles change. The nominal rate is 10%. The periodic rate for semiannual, quarterly or daily is calculated by dividing the nominal rate by the number if compounding in a year. The last column shows that the effective rate increases with shorter compounding cycle and levels off at 10.52% by increasing the number of cycles to infinity.

229

Engineering Management Handbook Table 15.2. Effects of Compounding Period Nominal and Effective Rates - Examples Example: Compounding

Interest Periods / Yr

Period Annual

Effective Rate

I

 

M

r

r/M

i

1

10

10

10

Semiannual

2

10

5

10.25

Quarterly

4

10

2.5

10.38

Monthly

12

10

0.833

10.47

Daily

360 / 365

10

0.002

10.51

Continuous

Infinity

10

0

10.52

Note, that as the compounding period decreases (example from yearly to daily to continuously), the effective rate increases.

15.2.4 Compounding and Discounting Compounding and discounting is done mainly to convert from current worth to future worth and vice versa. As you check these equations, we deal with four items, F, P, N, and i. So to simplify calculations, we introduce a shorthand notation to represent these relationships and is shown in Figure 15.1. Figure 15.1. Shorthand Notation Used for Engineering Economics

 Compounding - F = P (1 + i)^N - money grows into the future  Discounting

F = P (F/P, i, N)

P = F (P/F, i, N)

- F = P (1 + i)^-N - in today’s dollars Shorthand Notation

For compounding we are looking for a future sum F. This is given by multiplying the current value P by a factor that given P yields P at a given I and N... Hence, one can calculate the value of these factors for any combinations of i’s and N’s. This has been done for the factors we normally use and values are presented in a set of tables at the back of the Lang and Merino (1993). The inverse of F/P is the discounting factor and is shown by P/F. Knowing this notation, we can calculate a future sum, by multiplying the present worth by its factor that we look up at the back of the text, for the given i and N.

15.2.5 Cash Flow Patterns While companies have a wide range of capital projects to select from, the resources for these projects are constrained. Therefore, different projects are to be evaluated from various perspectives before being undertaken. We will concentrate on the economics of projects in this tutorial. When it comes to the evaluation 230

Engineering Economics

for economic purposes, we need to estimate cash flows. Cash flows are the amount of cash a company/project generates and/or uses during a period of time. It also tells us about the company’s/project’s financial health. During the life of a project, various cash flows occur. Some of cash flow streams are: • Cash flows in from sales of goods and services. • Cash flows out for expenses incurred in producing goods and services. • Cash flows out for interest payments on borrowed funds. • Cash flows out for the payment of taxes. • Cash flows in from the disposal of assets throughout the life cycle and its termination. • Cash flows out for replacing assets throughout the life cycle. • Cash flows in from borrowed funds. • Cash flows out to pay back the principal on borrowed funds.      The above list identifies almost all of the important cash flows except the “initial cost,” “working capital,” and the “imputed cash flows from benefits and disbenefits.” The initial cost is the first cost of a project and includes both depreciable and non-depreciable assets. It can be called by different names such as “investment cost,” “initial investment,” “capital investment” and “capital expenditure.” Then come the “working capital,” which are the funds needed to operate a capital facility before revenue becomes available for covering expenses. Working capital is not a depreciable item. “Imputed cash flows” are the benefits and disbenefits produced by non-profit entities. For example, in a highway project, if the new highway will reduce the travel expenses for the highway users, this is an imputed cash flow for this project. Figure 15.2 illustrates how a cash flow diagram should be drawn. Figure 15.2. Cash Flow Notation

S = Salvage Value

1 2

F

3

4

5

F’s = difference between revenue and expense Net cash flow for a period shown at the end of period

P

P = first cost plus working capital

To draw the cash flow diagram, you start from the current time, time 0. In this diagram we can see that P is the first cost plus working capital, which is what the company invests in the project and is shown downward (negative side). At the end of the first year we have to make an additional investment, F. In the first year there is not enough revenue (or no revenue) to cover the cost. From the second year we can see positive net cash flow each year and the amount increasing. At the end of the project, there is also the salvage value of the plant and equipment from the initial investment. Salvage value is an inflow, so it is shown as positive. This diagram typifies the cash flow for a simple startup business. Note that salvage value can be negative or positive. Examples of negative salvage value include cleaning up toxic water and nuclear power disbanding. 231

Engineering Management Handbook

15.2.6 Loan Programs and Personal Finance This model is the most common loan or mortgage model used. For instance, if you take a fixed rate loan for your car or your house, your payments to the bank looks like what is represented as shown in Figure 15.2. This is called amortization and is shown in Figure 15.3. That is the amount to be paid during the payment period is the same, but the proportions between the interest and principal change. Initially, more of the payment goes toward the interest and less toward the principal. However, toward the end, more of the Payments - Interest and Principal Combined payments go towardConstant the principal.

(amortization)

Figure 15.3. Constant Payments - Interest and Principle Combined (amortization)

Interest Variable 0

Principal Variable 1

2

3

4

5

P The advantage of the model is the fixed payments, which help people with relatively fixed income (like salary) to better plan and manage their budgets. Up until the end of WWII, loans were mostly interest payment only. The result was that most people could not accrue equity in their home. After WWII, with VA and FHA loans, the amortized loans (mortgages) became the norm—and with it equity for millions in their homes, firms, and businesses.

15.3 Figures of Merit (FoM) There are generally five figures of merit that are used to compare and rank alternative capital investment opportunities. They are: • Present Worth (PW) or Present Value (PV) or Net Present Value (NPV) • Future Worth (FW) or Future Value (FV) or Net Future Value (NFV) • Annual Worth (AW), or if the annual worth is negative, then Annual Cost (AC) or Equivalent Uniform Annual Cost (EUAC). • Internal Rate of Return (IRR) • Benefit Cost Analysis (BCA) The three worths—present, annual and future—are directly linked through the rate of return factors derived before. AW is FOM used in retirement or replacement problems. FW is FOM used for insurance, pension, and bond analysis. Any worth can be converted to any other worth because of equivalence. • Present Worth—most commonly used • Annual Worth—used where annual comparisons are appropriate • Future Worth—used in insurance, pensions, etc.

15.3.1 Present Worth

232

This is the most widely figure of merit and is based on the discounted cash flow. Present Worth (PW) is defined as the monetary sum that is equivalent to a future sum when the interest is compounded at a given rate. It can be also defined as the discounted value of a future sum when discounted at a given rate. It is also referred as Net Present Value (NPV), PV or NPW.

Engineering Economics

So, the Present Worth (PW) is given by, PW = A (P/A, i, N) + S (P/F, i, N) - P where: PW = Present Worth; A = Uniform Series of Annuities, i = Interest Rate per period; N = Number of periods; S = Salvage Value; P = Present Value.

Example 1: Automatic Welder Problem Data: - Cost of Automatic welder = $100,000 - Expected annual income = $40,000 - Salvage value = $10,000

- Installation cost = $15,000 - Duration of life = 5 years - Interest rate = 20%

Figure 15.4. Cash Flow Diagram for Purchasing an Automatic Welder S = $10,000

A = $40,000

0 1

i = 20%

5

$115,000

PW = = = =

A (P/A, 20, 5) + S (P/F, 20, 5) - P ($ 40,000 x 2.9906) + ($10,000 x 0.4018) – 115,000 $ 119,624 + 4,018 – 115,000 $ 8,642 for 5 years

15.3.2 Annual Worth The Annual Worth (AW) can be obtained by simply multiplying the PW that have just been calculated by the A/P factor. So, AW = PW (A/P, I, N) Also, AW can be found out by using the following formula too: AW = S + S (A/F, i, N) – P (A/P, i, N) where: AW = Annual Worth; A = Uniform Series of Annuities, i = Interest Rate per period; N = Number of periods; S = Salvage Value; P = Present Value; (A/F) = Sinking Fund Factor; (A/P) = Capital Recovery Factor. • To Convert AW to PW: Multiply by P / A factor = A (P / A) = P • To Convert AW to FW: Multiply by F / A factor = A (F / A) = F

15.3.3 Future Worth The Future Worth (FW) is the value of an asset or cash at a specified date in the future that is equivalent in value to a specified sum today. Future Worth could be obtained by multiplying the present or annual worth by the proper F/P and F/A factors, respectively. The FW can also be calculated using the following formula: FW = A (F/A, i, N) – P (F/P, i, N) + S 233

Engineering Management Handbook

where: FW = Future Worth; A = Uniform Series of Annuities; i = Interest Rate per period; N = Number of periods; S = Salvage Value; P = Present Value; (F/A) = Compound Amount Factor for Equal Payment Series; (F/P) = Compound Amount Factor for Single Payment.

15.3.4 Capital Recovery Capital Recovery of an investment is a uniform series representing the difference between the equivalent annual cost of the first cost and the equivalent annual worth of the salvage value. Capital Recovery of an investment is a uniform series representing the difference between the equivalent annual cost of the first cost and the equivalent annual worth of the Salvage Value. The generalized formula for the Capital Recovery is given as: CR = P (A/P, i, N) – S (A/F, i, N) where: CR = Capital Recovery; P = First Cost; S = Salvage Value at the end of the time period. Now, let us move on to the CR for an infinite horizon. In the case of an infinite time horizon, the A/P factor equals i and the A/F factor equals zero. This means that we do not consider salvage value in situations like this. So, in other words, for a project with infinite time horizon, S = 0. Therefore, the formula for CR over an infinite time horizon is given by: CR = Pi where: CR = Capital Recovery; P = First Cost; i = MARR

15.3.5 Capitalized Cost Capitalized Cost is defined as the sum of the first cost P of an investment plus the present worth of perpetual periodic cash disbursements and is demonstrated in Figure 15.5. Figure 15.5. Capitalized Cost      

0

1 A

to infinity

MARR = i P Now, in a situation like this, where the planning horizon is infinity, the Capitalized Cost is given by: CC = P + A/i where, CC = Capitalized Cost; P = First Cost; A = Uniform Annuity;  i = MARR. This is because, as N approaches infinity, A/P factor for any interest rate i, approaches i. P/A, which is the inverse of A/P, approaches 1/i. Therefore, we get the formula of CC given above.

234

Engineering Economics

15.3.6 Internal Rate of Return (IRR) As mentioned before the IRR is the return investors will get if the cash flow estimates in the cost studies prove true. The IRR can also be defined as the rate of return that makes the net present worth of a set of cash flow streams equal to zero. IRR is also known as the hurdle rate, or cutoff rate, or targeted rate, or the profitability index PW = A (P/A, i, N) – P + S (P/F, i, N) When calculating IRR, we set the PW equal to zero. Then using the provided information and the above equation we solve for the IRR.

A Simple IRR Problem Given that capital cost of a project is $100,000. With yearly savings of $30,000 over the service life of 5 years, calculate the IRR. First, we set the PW to zero,       P = $100,000       A = $30,000       N = 5 years       PW = A (P/A, i, N) – P + S (P/F, i, N) Remember (P/A, i, N) = ((1+i) N – 1) / (i (1+i) N) Using a calculator it is easy to solve for the IRR. Using a calculator, Goal Seek in Excel, interpretation it is easy to solve for the IRR thus producing i = 15.27%. Present Worth and IRR are often used to determine the best course of action to follow. First, the Present Worth is calculated using the MARR. If the present worth is positive, IRR is then determined to find out what rate of return the cash flow estimates might produce. The IRR methodology is also applied to incremental cash flow patterns to determine the return on incremental investments due to lower expenses or higher revenues, or both. However, there are several shortcomings when calculating IRR. The first shortcoming is “Multiple Solutions.” This occurs when the cumulative cash flow crosses the x-axis and changes signs. In cases where additional capital is invested after year 0, the added capital may cause multiple changes in signs and result in multiple solutions. The second shortcoming is the ‘Reinvestment Fallacy.” Can the funds from a particular project be invested at the IRR rate?

15.3.7 Benefit Cost Analysis (BCA) Introduction to BCA The last figure of merit to be discussed in this course is the benefit cost ratio (BCR). For projects aiming to improve the welfare of the public, BCR is often the preferred figure of merit. This is because it does not solely focus on the financial return of a project, but measures if the benefits of a project outweigh its costs. This is the figure of merit often used by government entities to justify their selection of projects from the large pool of projects placed before them.  

Cash Flows

A key difference between the BCR and other figures of merit is noted in the cash flows. Cash inflows in relation to the benefit-cost analysis consist of the benefits to the public. Whereas cash outflows in relation to the benefit-cost analysis consist of the costs to the government for providing those benefits. 235

Engineering Management Handbook

For example, let us consider the cash flows in relation to the Mississippi Levee; the cash inflows in this case include the initial capital investment and periodic maintenance costs. The cash outflows include benefits like flood control and recreation, besides protecting the economic well-being of cities like New Orleans. Cash flows in relation to the benefit-cost analysis can also be further divided into the following five groups: • Positive Benefits (B) • Disbenefits (D) • Initial Cost (I) • Cash Costs (C) • Cash Receipts (R) As mentioned earlier a project is deemed worthwhile only if the benefits derived from it exceed the costs of the project. The benefit cost ratio (BCR) can thus be expressed as either B – C or B/C. For a project to be deemed feasible; B – C > 0 or B/C > 1

Conventional and Modified BCR Formulas The conventional formula for the BCR is BCR = (B – D) / (I + (C – R)) The modified formula for the BCR is B/C = ( (B – D) – (C – R)) / I where: B = Benefits derived from the project; D = Disbenefits derived from the project; I = Initial investment in the project; C = Operating and maintenance cost involved with the project; R= Revenue earned from the project. The key difference between the two formulas is in the treatment of the term (C – R). In the conventional BCR formula the term (C – R) is part of the denominator, while in the modified BCR formula it is in the numerator, leaving Initial cost as the sole term in the denominator. In the conventional formula, all costs are taken as cash outflows while the net benefit is taken as the cash inflow. In the modified BCR formula only the initial cost is taken as a cash outflow, while the net benefit minus the O&M costs is taken as the cash inflow. It should be noted that while calculating BCR in either formula, one should ensure that all cash flows are brought back to their present worths or annualized.

Example using Conventional and Modified BCR Given the following; - Benefits = $2,000,000/yr = B - Initial Costs = $ 1,000,000/yr = I - User Fees = 0 = R

Conventional BCR B/C = (B - D) / (I + (C – R))     = 2,000,000 / (1,000,000 + 500,000)       = 1.333

Modified BCR B/C = ((B – D) – (C- R)) / I        = (2,000,000 – 500,000) / 1,000,000 236

= 1.500

- Disbenefits = 0 = D - O&M Costs = $ 500,000/yr = C

Engineering Economics

15.4 Retirements and Replacements Retirement and replacement problems are of great importance in a capitalistic economy. Retirement and replacement studies are essential for any company to remain competitive. Every machine used by a company undergoes a certain amount of wear and tear as time goes by. After a certain point, the machine needs to be replaced. Either because it is no longer suitable for use, or it is no longer economical to maintain and operate. Replacement problems are typically handled as a least cost problem. This means that the decision-makers are interested in minimizing cost. Retirement and replacement studies also include non-economic factors in the decision-making process. However, in this chapter we will deal only with the economic factors. For the retirement and replacement problems we deal with in this chapter, we will generally assume discrete cash flows and that we are dealing with before tax rate of returns.

15.4.1 Types of Retirement and Replacement Problems Type 1—Simple retirement/without replacement This is a problem that offers a very simple alternative; to keep the asset or to retire it without replacing it. Here, the net cash flows over the remaining life are discounted by the MARR and then compared to the opportunity cost of not selling the asset. If the resulting PW is less than zero the asset is retired.

Type 2—Retirement with an identical replacement In this case, the asset being considered as a replacement is similar to the asset considered for replacement. In this case we find the optimal economic life and the life of the asset. If the asset is older than the optimal economic life it is replaced. If it is not, it is kept.

Type 3—Retirement with different replacements but similar to each other In this case, the challenger assets are different from the defender. However, the challengers themselves are different.

Type 4—Retirements with different replacements not similar to each other Generalized retirement and replacement problems require knowledge of dynamic or mixed integer programming.

15.4.2 Total Costs and Economic Life The following covers the most frequent type of retirement and replacement problems—Type 2 problems. Figure 15.5 shows the optimal economic life of a machine. The total cost reaches a minimum when the Operating and Maintenance (O & M) costs equal the capital cost. The corresponding N value to this point is the optimal economic life. This is point is also considered to be the best time to replace or retire the asset.

237

Engineering Management Handbook Figure 15.6. Generalization of an R&R Problem

EUAC Total Costs O&M Costs Capital Costs N

Time

Figure 15.6 shows the optimal economic life of a machine. The total cost reaches a minimum when the Operating and Maintenance (O&M) costs equal the capital cost. The corresponding N value to this point is the optimal economic life. This is point is also considered to be the best time to replace or retire the asset.

15.4.3 Capitalized Costs There are three simple steps needed to calculate the total capital cost. 1. Calculate the annualized first cost (P) by converting P to an annuity over the project life, using the capital factor (A/P, i, N). 2. Calculate the annualized salvage value by converting the future value (F) to an annuity using the sinking fund factor (A/F, i, N) 3. Add the two capitalized costs obtained from steps 1 and 2. Standard formula used for calculating total capital costs. EUAC = P (A/P, i, N) – S (A/F, i, N)

Note: Alternative formula.

EUAC = (P-S)*(A/P, i, N) + S (i)

Example for Capitalized Costs SM Industries is looking into expanding its operations with a new facility in Rio. The first year salvage value is $8,000,000 and will decrease each year after. O&M costs are $1,500,000 and will increase each year after. Assume a life of 4 years, and initial costs of $40,000,000. Assume discrete end of year cash flows and a before-tax rate of return of 8%. Find out when this facility would have to be replaced.

Step 1 The first thing is to calculate the salvage values and O&M costs for each year as shown in Table 15.3. Table 15.3. Salvage Value and O&M Costs by Year   EOY

Salvage Value

O&M Costs

1

$8,000,000

$1,500,000

2

$6,400,000

$1,875,000

3

$6,120,000

$2,343,750

4

$4,096,000

$13,929,688

Initial Cost

238

Rio

$40,000,000

Engineering Economics

Step 2 We calculate the capitalized cost for each year as shown in Table 15.4. Table 15.4. Capitalized Costs by Year 1

3

4 = 2x3

5

6

7 = 5x6

8 = 4-7

(A/P,8%, j)

Annualized IC

Salvage (Sj)

(A/F,8%, j)

Annualized Sj

CR

0

2 Initial Cost (IC) $40,000,000

 

 

 

 

 

 

1

 

1.0800

$43,200,000

$8,000,000

1.0000

$8,000,000

$35,200,000

2

 

0.5607

$22,428,000

$6,400,000

0.4807

$3,076,480

$19,351,520

3

 

0.3880

$15,520,000

$6,120,000

0.3080

$1,884,960

$13,635,040

4

 

0.3019

$12,076,000

$4,096,000

0.2219

$908,902

$11,167,098

EOY

15.4.4 Operating and Maintenance Costs O&M costs cover the costs of running a factory, office, or a plant. It includes labor, materials, etc. O&M costs usually increase with time, but there are a few exceptions, for example, computer software. There are three steps required to calculate the annualized O&M costs. 1. Obtain the yearly cost by converting the future cost to present worth. Use the present worth factor (P/F, i, N). 2. Find the cumulative PW. 3. Find the annualized cost by converting the cumulative PW to annuity. Use the capital recovery factor (A/P, i, n). An example for EUAC O&M Costs is shown in Table 15.5. Table 15.5. EUAC Calculation for O&M Costs 1

2

3

4 = 2x3

5

6

7 = 5 x6

EOY

O&M

(P/F,8%, j)

PW of O&M

Cum. PW

(A/P,8%, j)

O&M EUAC

1

$1,500,000

0.9259

$1,388,850

$1,388,850

1.0800

$1,499,958

2

$1,875,000

0.8573

$1,607,438

$2,996,288

0.5607

$1,680,018

3

$2,343,750

0.7938

$1,860,469

$4,856,756

0.3880

$1,884,421

4

$13,929,688

0.7350

$10,238,321

$15,095,077

0.3019

$4,557,204

The total EUAC is calculated by adding the capitalized recovery costs and the O&M costs. The optimal economic life is determined by finding the year with the lowest EUAC figure. The asset in question should be replaced after this point. Table 15.6 shows the calculation for total EUC for an O&M example. Table 15.6. Total EUAC for an O&M Example EOY

CR

O&M EUAC

Total EUAC

1

$35,200,000

$1,499,958

$36,699,958

2

$19,351,520

$1,680,018

$21,031,538

3

$13,635,040

$1,884,421

$15,519,461

4

$11,167,098

$4,557,204

$15,724,301

From Table 15.6 it is seen that the lowest EUAC occurs at the end of year 3. Hence, the facility should be replaced at the end of year 3.

239

Engineering Management Handbook

15.5 Inflation Inflation can be defined as an increase in the general price level of goods and services. It can also be defined as a decrease in the purchasing power of the dollar. Inflation is an important part of life and business for both individual consumers and corporations.

15.5.1 Causes of Inflation Inflation is variable in nature and is influenced by several factors. 1. Consumer Demand: When consumer demand for a certain product or service increases, it usually leads to the increase in the price of the product or service. 2. Product Improvements: Certain product improvements cause price increases because the improved product has more or better features. An example would be switching to Color TV from a Black and White TV. 3. Technological change: Price increases are also caused by technological changes. The product using a new technology usually cost more because the new product cost reflects the cost of research and development. An example would be switching from wired networks to Bluetooth networks.

15.5.2 Types of Inflation 1. 1. Cost Push Inflation: This type of inflation is caused by increases in the cost of production (e.g., minimum wages or increased safety). With all other factors remaining the same, the higher the cost of production, the lower the amount of goods produced. At a given price level, rising wages or increasing raw material costs, result in companies downsizing labor forces and lowering production rates. 2. Demand Pull Inflation: When supply is unable to meet demand, there is a resultant increase in prices. This may happen because of a sudden spike in demand and production levels cannot be ramped up fast enough. It may also occur because producers are deliberately curtailing supplies. 3. Structural Inflation: These are price increases triggered by Cost of Living Allowances (COLASs) or other contractual objectives. 4. Energy: The impact of the increase in the price of energy on products is significant and complex and its effects on inflation merit its own category. Increased energy costs are reflected in production costs and transportation costs. The common method of measuring inflation is through the use of price indices. “Price indices for measuring price level effects are dimensionless ratios that compare prices of a specified set or combination of goods and services in a selected base period to the prices of the same or functionally equivalent set at any other period.” Increasing price indices indicate inflation, while falling price indices indicate deflation. These price indices are based on models constructed through surveys, definitions, examinations of published prices, etc. The base year is assigned a cost of 100 and following years are measured against this level. Thus, future years may have lower or higher price levels.

Consumer Price Index (CPI) This is one of the best-known price indices and is issued and complied by the U.S. government. The CPI is further subdivided into two indices. The CPI-W which is the cost of the “the market basket” of goods and services bought by wage earners. The second is the CPI-U, which is measured for all urban dwellers. Inflation can also be defined in relation to the CPI. The inflation rate is defined as the percentage change in the annual CPI.

Wholesale Price Index (WPI) Also known as the Producers Price Index (PPI), this is the cost of the “market basket” of goods and services brought by the industry. 240

Engineering Economics

GNP or GDP Deflator The GNP or GDP deflator measures the cost of all goods and services in the GNP or GDP.

Industry Specific Index These are cost indices for individual disciplines/industries. These indices take an in-depth look at an individual discipline/industry and the movement of capital and operating costs through the years. For example, the Chemical industry has the Marshall/Swift indices.

15.5.3 Using Price Indices The first step is to ensure you have selected the right price index. The second step is to use an appropriate time period. Often ratios are used to determine the amount of inflation.

Example Using the CPI Index When John joined the Clysedale Brewery in September 1980, his starting salary was $24,000. What would his equivalent starting salary be if he joined in the year 2000?   1980 2000

Starting Salary $24,000 ???

CPI 84 171

Using the F/P Future worth factor we calculate “i”. F = P(1+i)n => F/P = (1+i)n => 171/84 = (1+i)20 => i = 3.62% Thus, the equivalent starting salary would be: F = 24000 (1+0.0362)20 = $48,969

15.5.4 Inflation and MARR We will now examine the relationship between MARR and inflation. There are three elements to this relationship. ‘i’ = The MARR with inflation ‘f ’ = Rate of Inflation ‘ieq’ = The MARR without inflation (1 + ieq) = (1 + i) / (1 + f ) => ieq = (i - f ) / (1 + f ) => i = ieq + f + ieqf => I = ieq + f or ieq = i - f Thus, we can see that the i (MARR with inflation) is approximately equal to ieq (MARR) plus f (inflation). On the other hand, that ieq (MARR without inflation) is approximately equal to i (MARR with inflation) minus f (the inflation).

15.5.5 Cash Flows and Inflation Cash flow estimates over a life cycle based on constant dollars are not adjusted for price level effect. These are also know as actual dollar cash flows and keyed into a specific year. Cash flows can also be adjusted to current or real dollars. These cash flows are in the years specified. We will now look at three cases of cash flows and inflations.

241

Engineering Management Handbook

Case 1: No Inflation and Constant Dollars In this case we use ieq as the rate. The case is represented below in a cash flow figure with A denoting constant dollars and N the number of years. Case 2: With Inflation and Constant Dollars In this case we use ieq as the rate. The case is represented below in a cash flow figure with A denoting constant dollars and N the number of years. In this case before calculating the figure of merit, we ensure that all cash flows have been converted to constant dollars. Case 3: With Inflation and Current Dollars In this case we use i as the rate. The case is represented below in a cash flow figure with A denoting current dollars and N the number of years.

15.5.6 Common Problems Relating to Inflation We will focus on two common types of inflation problems. A: Inflating Constant Dollars to Current Dollars: In this case, we have been given a value in constant dollars (for example, the current rate of soft lumber) and we want to convert this figure to current dollars over the life of the project. To do so we need to inflate our constant dollar value. We apply the F/P (Future Worth Factor) to inflate constant dollars. B: Deflating Current Dollars to Constant Dollars: In this case, we have been giving a value in current dollars and we want to convert them to constant dollars. For example, we have been given the cash flows of operating expenses across the life of the project and we want to convert those cash flows to their present constant dollar value. To do so we need to deflate our current dollar value. We apply the P/F Factor to deflate current dollars.

15.6 After Tax Analysis The taxes paid and tax related items like depreciation and investment tax credits have a significant impact on the economics of all aspects of corporate activity, but particularly on Capital Investment. Economic decision-making is based on Cash Flows (not accounting profit). If you take a course in Corporate/Managerial Finance, you will find that corporations are managed on the Cash Flow. Cash Flows can be generated as part of: • Operating Activities • Capital/Finance Activities

242

Operating cash flows can be determined from the Income Statement. Note that an operating cash flow is Net Income after Tax plus depreciation. Remember that depreciation is a non-cash (accrual) expense. Operating cash flows are periodic over a number of years. Capital Cash Flows can be determined from the Balance Sheet. There are a number of activities that impact the Capital Cash Flows. Change in inventory levels, financing activities and capital expenditure are some of these activities. For engineering economics we are interested in investments for depreciable (plant, equipment, etc.) and non-depreciable (land) capital. For capital projects we are also interested in loans (to finance capital) and the after tax sale/disposal of capital at the end of the project. In the following sections we will provide the calculations and format for: • Net Cash Flow from Operating Activity • Net Cash Flow from Capital Activity As well as the After Tax Cash flows for: • Depreciation • Loans/borrowings • Salvage/disposal of capital items

Engineering Economics

15.6.1 Net Cash Flow from Operating Activities Table 15.7 is the modified (Normal) Income Statement format used in Engineering Economics for the selection of Capital Projects. Table 15.7. Net Cash Flow from Operating Activities Line

Cash Flow

1 2 3 4 5 6 7 8 9 10 11 12

Operating Revenue Cash Expenses Operating Income Before Depreciation Depreciation Operating Income Interest Expenses Pretax Net Income Income Taxes Per Year Investment Tax Credit Net Income Depreciation Net Cash Flow from Operating

Operations

1-2 3-4 5-6

7+8+9 10 + 11

Note that the format is very similar to the income statement format in Chapter 3. The format in Chapter 3 was expanded to include separate items for: • Line: 4:- Depreciation • Line: 6:- Interest Expenses from Loans • Line: 9:- Investment Tax Credit Depreciation was added back to the Net Income after Tax to yield the operating cash flow. Note that the $ amounts in this format are for N years where N = Project Life.

15.6.2 Net Cash Flow from Capital Related Line Table 15.8 is the modified balance sheet format used in Engineering Economics for the selection of Capital Projects. Table 15.8. Net Cash Flows from Capital Related Activities Line

Cash Flows

Operations

13 14a 14b 14c 15 16 17

Principal Repayment Depreciable Capital Non-Depreciable Capital Loan Proceeds Capital Gains/Losses Working Capital Net Capital Cash Flow

13+14+15+16

Note that the format of Table 15.8 is similar to the Balance Sheet and Cash Flow Statement discussed in Chapter 14 (Basic Accounting). You will note that there are other items that will impact on balance sheet cash flows. For the selection of Capital Projects these include: • Principal Repayment of Loans used to purchase capital goods • Depreciable and non depreciable capital • Loans proceeds 243

Engineering Management Handbook

• Capital gains/losses when capital is disposed of (salvage) • Working Capital You should know the definition of these items and how they differ. Consult Chapter 14 (Basic Accounting).

15.6.3 After Tax Cash Flow from Depreciation Charges Various depreciation methods is discussed earlier in the handbook. Note that depreciation is a non-cash (accrual) entry and as such is added back to the Net Income after tax to yield the cash flow from operating activities.

15.6.4 After Tax Cash Flow from Investment Tax Credit (ITC) From time to time government/agencies (U.S., State, and sometime local) give incentive for business to make capital Investments. This is usually during a recession or natural disaster (like Hurricane Katrina) and is designed to increase capital spending. Why Capital Spending? The concept is that for every dollar spent on capital goods, the Gross National Product and Gross Domestic Product (GNP and GDP) will increase many folds (6 to 8 times). In macroeconomics, this is called the “Multiplier Effect.” The Investment Tax Credit (ITC) allows a company to subtract some part of the Capital purchase price from the Income Tax it pays. For example, if the ITC is 10% and the company makes a $200,000 capital investment then the company can deduct $20,000 from its taxes (usually in year 0 or 1). It is assumed that the company will keep the Capital asset in service for a certain number of years. If it does not, it may have to give back some of the ITC. This is called re-capture. You need to seek Tax Counsel on this matter because the rules are complex. ITC for solar panels or energy saving items like Hybrids cars are examples that apply to consumers.

15.6.5 After Tax Cash Flows from Loans The after tax analysis with loan repayment includes are the steps in Table 15.1 plus interest expenses. Interest is subtracted from the net cash flow before income taxes are calculated because interest is tax deductible, requiring it to be eliminated from the taxable income. Interest charges benefit the organization because it reduces the taxable income and, hence, the income taxes. The principal payments are being subtracted in Table 15.2 because principal is NOT tax deductible. When dealing with loans, we need to understand the difference between interest and principal. Principal is the actual amount you have borrowed or the remaining, unpaid balance owed to the lender. Interest is most easily described as the fee charged by the lender for lending you the principal. It is usually expressed in as annual percentage of the principal. Interest is generally tax deductible, meaning it REDUCES taxable income. Table 15.9 provides an example on how to “amortize” a loan. The example assumes that the interest rate is 10% and the loan is repaid in 4 years. The amount of the loan is $80,000. Annual payment is calculated using the time value of money equation: A = P (A/P, I, N) = $80,000(A/P, 10, 4) A = $80,000 x .3155 A = $25,240/year

244

Engineering Economics Table 15.9. Loan Balance / Amortization Table Year

Beginning-ofYear Balance

Annual Payment*

Annual Interest

(1)

(2)

(3)

01 02 03 04 Total

$80,000 $62,760 $43,796 $22,936

$25,240 $25,240 $25,240 $25,240

(4) 10% of (2) $8,000 $6,276 $4,380 $2,294

Annual Principal Repayment (5) (3) - (4) $17,240 $18,964 $20,860 $22,936 80,000

End-of-Year Balance (6) (2) – (5) $62,760 $43,796 $22,936 0

As you are shown in Table 15.7 the interest is a periodic expense that reduces operating income before taxes. From Table 15.4 notice that the principal repayment is a capital related item (return of capital). To summarize: • Table 15.9 Interest (Column 4) goes to Table 15.7 Line 6 • Table 15.9 Principal Repayment (Column 5) goes to Table 15.84 Line 13

15.6.6 After Tax Cash Flow for Salvage/Disposal of Assets This topic was covered in Chapter 14 – Gains & Losses for the Disposal of Assets 14.4.

15.6.7 Total Cash Flow Discounted Table 15.10 adds the total periodic cash flow (Line 12 – Table 15.7) to the Capital related cash flows (Line 17 – Table 15.8). Table 15.10. Total Cash Flow (Operating and Capital) Discounted Line

Cash Flows

Operations

18 19 20 21

Total Cash Flow Discounted Factor Net Present Value Cumulative NPV

12 + 17

Line 1: is truly the “bottom line” in all Economic Analyses. This is used for choosing among capital projects, mergers and acquisitions, long-term corporate planning and in all sorts of financial decisions. Because Engineering Economics is built on the Time Value of Money we use the Discounted Cash Flow (DCF). This is line 20 and 21 (accumulated). To calculate the DCF you need to use a MARR (Minimum Attractive Rate of Return) and apply this to the project cash flows. The cumulative DCF is the Net Present Value (NPV) for the project and is a widely used Figure of Merit.

15.6.8 ATA Example Innovation Inc. is planning to start a new business in order to increase the efficiency of their consulting operations. Listed in Table 15.11 are the economic data for their proposed investment.

245

Engineering Management Handbook Table 15.11. Economic Data for the ATA Problem 1

Depreciable Capital - year 0

$2000

2

Salvage Value (FMV) (at the end of project life)-BV

$900

3

Non-depreciable Capital - year 0

$500

4

Non-depreciable Capital (returned at the end of project life)

$750

5

Expected Revenue ($ / yr.)

$9500

6

O&M Cost ($ / yr.)

$120

7

Useful life –years

8

Working Capital - year 0

$600

9

Working Capital - (returned at the end of the project life)

$600

10

Loan Proceeds - year 0

$1200

11

Interest on Loan - per year

12

Loan Period – years

2

13

Tax rate - per year

40%

14

ITC - year 0

$150

15

ITC - year 1

$0

16

MARR per year

2

10%

18%

(Note: All units are in ‘000 dollars)

This system qualifies a special two-year MACRS Depreciation (with factors 0.55 and 0.45). Assume working capital is returned in year 2. Also assume that the company has income from other projects and this system is sold at the end of year 2. ITC is $150 and being given at the end of year 0.

Solution to the ATA Example As a first step we need to calculate the interest and principal repayments for the loan which is shown in Table 15.12. Table 15.12. ATA Loan Amortization Table

Year

Beginning Balance

Annual Payment

Interest

Principal Repayment

Ending Balance

0

 

 

 

 

 

1

$1,200.00

$691.43

$120.00

$571.43

$628.57

2

$628.57

$691.43

$62.86

$628.57

$0.00

Total

 

 

$182.86

$1,200.00

 

The second step is shown in Table 15.13 in which we find the depreciation expenses and accumulated depreciation expenses (using the special MACRS rate) for this system.

246

Engineering Economics Table 15.13. Depreciation Expenses for the ATA Problem

Yr.

Initial

Depr.

Dep.

Accum.

Ending

Cost

Rates

Exp

Dep

BV

0

$2,000

 

 

 

 

1

 

55%

$1,100

$1,100

$900

2

 

45%

$900

$2,000

$0

Total

 

 

$2,000

 

 

Next we determine the after tax cash flows from salvage of depreciable and non-depreciable capital:

Depreciable: Taxes

= (FMV - BV) * TR = ($900 – 0) * 40% = $360 After tax cash flows FMV – salvage

= $900 - $360 = $540

Non - depreciable: Taxes = (FMV - BV) * TR = ($750 - $500) * 40% = $100 After tax cash flows FMV - salvage

= $750 - $100 = $650

We then calculate the Net Cash Flow from operating income as shown in Table 15.14. Table 15.14. Net Cash Flows from Operating Income  

 

0

1

2

Total

1

Operating Revenue

0

9500

9500

19000

2

Cash Expenses

0

120

120

240

3

Operating Income

0

9380

9380

18760

4

Depreciation

0

1100

900

2000

5

Operating Income

0

8280

8480

16760

6

Interest Expense

0

120

62.86

182.86

7

Pretax Net Income

0

8160

8417.14

16577.14

8

Income taxes

0

3264

3366.86

6630.86

9

Investment Tax Credit

150

0

0

150

10

Net Income AT

150

4896

5050.29

10096.29

11

Depreciation

0

1100

900

2000

12

Net C.F. from Operations

150

5996

5950.29

12096.29

247

Engineering Management Handbook Table 15.14. Net Cash Flows from Operating Income (continued) 13

Principal Repayment

 

($571.43)

($628.57)

($1,200.00)

14a

Depreciable Capital

($2,000.00)

$0.00

$540.00

($1,460.00)

14b

Non-depreciable capital

($500.00)

$0.00

$650.00

$150.00

14c

Loan Proceeds

$1,200.00

 

 

$1,200.00

15

Capital Gains/Losses

 

 

$0.00

16

Working Capital

($600.00)

$0.00

$600.00

$0.00

17

Net Capital Cash Flow

($1,900.00)

($571.43)

$1,161.43

($1,310.00)

18

Total Cash Flow

($1,750.00)

$5,424.57

$7,111.71

$10,786.29

19

Discount Factor @ 18%

1

0.8475

0.7182

-

20

Net Present Value

($1,750.00)

$4,597.32

$5,107.63

$7,954.96

21

Cumulative NPV

($1,750.00)

$2,847.32

$7,954.96

 

15.7 Decision Analysis 15.7.1 Types of Problems There are three classes of problems in economic decision-making. • The first is decision under certainty – where you assume that you have sufficient information on the inputs. That is, you are certain about input estimates. To explore alternatives, we use sensitivity analysis. • The second involves decisions under risk mean that you have inputs that have probabilities. These can be expressed as distributions. Risk management uses simulation techniques (example, Monte Carlo Simulation) to look at various alternatives. • Lastly, there are problems where you do not have any data or the data is random (like in stock market) and cannot be expressed as a distribution. To solve these problems, you can use certain rules or heuristics. These are known as principles of choice.

15.7.2 Choosing Among Alternatives Using ATA The most prevalent industry approach to choose among competing alternatives is the After Tax Analysis (ATA) (Lang and Merino, 1993). Figure 15.6 describes a typical ATA decision process followed by a written description of the decision-making process. Why is After Tax Analysis the standard methodology for industry? ATA is necessary because the government is a “partner” in every capital decision. Depreciation rates, tax rates, investment tax credits and capital gains taxes greatly influence the attractiveness of capital expenditures. In addition, ATA is used to determine the impact of borrowing on the project’s profitability. Borrowing at after tax interest rates that are below the project’s Internal Rate of Return (IRR) will increase the overall Rate of Return (RoR). Note that the opposite is true. These borrowing options are called Financial Leverage. Companies first look at a project’s RoR assumptions with full equity (100% of company’s cash with 0% of loans) to determine if the investment should be included in the capital expenditure budget. Financial leverage is considered for some subset of projects. Those projects tend to be ones that are required by law and are mandatory and not optional or elective. Also financial leverage is considered when there is a readily available pool of funds and/or lease options.1

1. A lease could be considered the same as a loan in certain circumstances.

248

Engineering Economics Figure 15.7. General Decision Model to Determine Economic Feasibility

- Specific Alternatives - Develoop Scenarios

1. Develop Alternative / Scenarios - Mutually Exclusive (MEAs)

2. Cost Estimation - Capital - Operating Expense

3. Benefit Estimation - Revenues - Other Savings

4. Economic Analysis - Establish Economic Critiera (MARR, FoM, Tax Rate, etc.) - Use After Tax Analysis (ATA Model) - Calculate Life Cycle Costs (EUAC)

5. Evaluate Intangibles / Non-economic - Environmental, Aesthetic, Legal, etc. - Set Goals / Criteria

6. Decision Analysis A. Conduct Economic Analysis B. Conduct Sensitivity Analysis & Determine Vital Few C. Conduct Non-Economic Analysis D. Combine A, B, C into Decision Model

No

7. Does Alternative Meet Criteria? - Economic - Non-Economic Yes 8. Final Decision

15.7.3 Decision Model Figure 15.7 illustrates a typical decision process used to choose among competing alternatives. Step 1 develops feasible alternatives that are mutually exclusive. In this case alternatives are various bus fleet sizes. In addition to alternatives, scenarios can be constructed which could combine alternatives along with other factors such as potential breakdowns or disaster scenarios. Step 2 estimates the capital and operating costs. The appendix explains the estimates for the case under consideration. Step 3 is the most critical step because of the need to estimate the benefits. Benefits can be savings in operating cost compared to a base case. This could be caused by a more fuel-efficient design than the base case. Step 4 is the economic analysis. The first part is to establish economic criteria like Minimum Attractive Rate of Return or MARR. The MARR reflects the opportunity cost for the investor’s capital. Risk plays a role because some investments may be more risky than others may. The time horizon needs to be determined. The owners need to decide to use full equity (% of loans) or some sort of financial leverage (10% to 90% of loans). Generally, most companies conduct the economic analysis using full equity and then, after choosing the most economical alternative, look at financ249

Engineering Management Handbook

ing options. Because this is an after tax analysis an applicable tax rate needs to be estimated for the chosen time horizon. The last criteria are the Figure of Merit (FoMs). Given that the ATA model is run on an Excel spreadsheet it is relatively easy to report on more than one FoM. FoMs include the Net Present Value (NPV) and the Equivalent Uniform Annual Cost or EUAC. EUAC is Life Cycle Costs (LCC). Step 5 involves evaluation of the intangibles and non-economic factors impacting the decision. There are a number of multi-attribute tools that can be used. Analytical Hierarchy Process (AHP) and Utility Analysis are two common techniques. Step 6 involves the decision process. Sensitivity analysis should be employed to determine the most sensitive attributes that impacts / influences the decision. This helps to separate the “vital few” from the “trivial many.” It is an aid in decision-making because it focuses the effort on the most important variables. Next, a decision needs to be made whether an economic or non-economic analysis is to be employed. If all the attributes can be monetized and converted into dollars then the standard ATA with a FoM such as NPV or EUAC can be used to either maximize benefits or minimize costs. However, if the downtime cannot be monetized then some form of non-economic analysis must be employed. There are at least three different process flows depending upon the downtime values and costs. This will be discussed in the next section. Step 7 involves the decision whether the analysis yields an alternative that meets the economic and/or not-economic criteria. If it does than a decision is made. If it does not, the process needs to be repeated starting with step number 1. This process needs to continue until a mutually exclusive feasible solution is reached. Step 8 is the final decision. If all the economic and non-economic criteria are met then a decision can be made to accept the alternative under review.

15.8 References Newman, Donald, Eschenbach, Ted and Lavelle, Jerome, Engineering Economic Analysis, Oxford University Press, 2004, ISBN: 0195168070; Library Number: TA177.4 N48 2004. Merino, Donald and Lang, Hans, The Selection Process for Capital Projects, John Wiley & Sons, Inc., 1993, ISBN: 0471634255; Library Number: TA177.4.L42. Eschenbach, Ted, Engineering Economy – Applying Theory to Practice, Oxford University Press, 2002, ISBN: 0195161521; Library Number: TA177.4.E833 2003.

250

Project Management’s Role in Engineering Management



16

Project Management’s Role in Engineering Management Kenneth W. McDonald United States Military Academy

251

Engineering Management Handbook

16.1 Introduction Engineering management (EM) is a unique and specialized form of management that is focused on the application of engineering principles to business practice. The EM discipline relates to management and engineering by bridging the gap between the disciplines. As stated in Chapter 1, EM has undergone tremendous adjustments to the discipline as trends in business and education have changed. This metamorphosis of EM is important to understanding how EM remains a very relevant and applicable discipline in today’s world. As different areas of engineering have emerged, EM has adjusted to offer courses in those disciplines and to produce quality EM engineers. As well, EM has ebbed and flowed with the adjustments to the management career field. If one looks the EM profession, you will see a dynamic, relevant and exciting field that embraces the changes associated with the professions of engineering and management. It is through this open embrace of change that the EM profession maintains a tremendous edge over other engineering disciplines. The EM professional becomes an expert in the field of engineering as well as management. In the engineering profession, the Accreditation Board for Engineering and Technology (ABET) accredits EM programs and in doing so, EM majors can achieve the Professional Engineering (PE) license and thus become recognized as an expert in the EM profession. Likewise, with an EM degree, EM majors can also earn the Project Management Professional (PMP) certification, which society recognizes those earning the PMP as experts in the field of project management. How is this possible? A degree in EM can produce experts in both fields of study (PE and PMP). What a tremendous recognition of the value of the EM profession. This chapter focuses on the management aspect of the EM profession. Specifically, project management. When we look at project management’s (PM) role in EM, we need to understand where PM and EM intersect in the real world. As stated before, EM is the bridging of disciplines of engineering and management (Hicks, Utley, and Westbrook, 1999). In this context, the application of engineering principles to an industrial challenge will in some way apply PM principles, tools and techniques. In industry, there are constantly projects cropping up that fall in the realm of EM. Just from perspective of the realm of EM, one can see that there is a natural tendency for engineer managers to run and participate in projects. From developing improvements to production lines, bringing new technology online, etc., the engineer manager will inevitably have to work in the realm of PM. Most everything becomes a project. It is what one could consider a natural “tool set” that most engineer managers will need to master. In Figure 16.1, PM encompasses the intersections of EM and engineering and management. It is in this realm that PM integrates directly to EM. All engineer managers will run projects sooner or later throughout their careers. Figure 16.1. Project Management Related to Engineering Management

Engineering

Engineering Management

Management

Project Management

16.2 Project Management A project is a temporary endeavor undertaken to create a unique product, service, or result (PMI, 2013). In the EM profession, projects are a major part of the profession and will be encountered over and over 252

Project Management’s Role in Engineering Management

again. A project has a definite beginning and end and therefore temporary; however, it is does not mean it is necessarily short in duration. Many projects can last for years and keep the EM professional engaged throughout. Properly moving a project to completion requires knowledge and experience in PM. By definition, PM encompasses the knowledge, skills, tools and techniques applied to tasks so that project objectives can be met (Kerzner, 2006). PM is a discipline that requires study and experience to master. Entire academic programs and books are devoted to PM and the tools and techniques used to implement proper PM. As a profession, PM is well-known through the Project Management Institute, Inc. (PMI). PMI developed the Project Management Book of Knowledge (PMBOK®) as a repository for up-to-date PM processes, tools and techniques as approved by the PM profession. PM integrates the PM processes known as initiating, planning, executing, monitoring and controlling, and closing. In Figure 16.2, these processes are laid out as they are encountered in the PM process. There is more than one way to manage a project to successful completion, and one could argue that PM is more of an art than a strict science. One enters into the PM process by starting a project. The project charter and scope are delivered during this phase. It provides the launching point from which the project begins. After the project is started, the project moves into the monitoring and controlling aspect of PM. The monitoring and controlling aspect of PM is critical. Failure to properly monitor and control a project leads directly to cost over runs for a number of reasons the biggest of which is improper PM. The monitoring and controlling processes observe project execution so that any potential problems and challenges to successful completion may be identified as early as possible. Once challenges/issues are identified, corrective action is taken to avoid schedule slippage, cost overruns, and other detrimental effects imposed by deviations from the plan. Monitoring and controlling must be performed frequently enough to allow the PM sufficient cognizance of the health of the project so that any corrective action required may be taken prior to events having an adverse impact on the project’s cost, schedule, or performance (Parnell, Driscoll, and Henderson, 2010). These processes required to track, review and regulate the progress and performance of the project and identify any areas in which changes to the plan are required and initiated the applicable changes. Making up the monitoring and controlling processes include initiating, planning, executing and closing. The initiation process requires those items for initiating the project are those required to define a new project by obtaining authorization to start the project (PMI, 2013). Figure 16.2. Project Management Process Groups (PMI, 2013)

Monitoring & Controlling Processes Planning Processes

Enter Phase/ Start Project

Initiating Processes

Closing Processes

Exit Phase/ End Project

Executing Processes

The planning process is critical for ensuring the project is completed on time, on budget and on schedule. Poor planning is the number one contributor of project failure. Planning defines scope, objectives and the course of action required to attain the project objectives. Following the planning process

253

Engineering Management Handbook

is the execution. Execution is the processes required to perform and complete the work defined in the plan. Planning and executing are represented by arrows forming a circular pattern. This pattern represents the iterative process of these two areas. The circular pattern indicates that planning and executing both impact the other and that the two should be updated continuously through the project and adjusted accordingly (PMI, 2013). Nearing the completion of the project, closing the project becomes of great importance. The closing processes are those performed to finalize any activity not completed across all aspects of the project to formally close the project. Projects can continue on for extended periods of time if this phase is overlooked. It is quite important to ensure that close-out procedures are executed with vigor similar to the beginning of the project. In some cases, the end of the project seems to be given a less than enthusiastic effort because the team is tired and ready to move on. Indeed, most team members will be released (depending on the project team structure) prior to the end of the project so that workforce becomes an issue. Finally the project enters the exit phase the project is completed (PMI, 2013). Obviously, there is more than one way to manage a project and bring it to successful completion. The key is understanding how to apply the proper PM techniques associated with the six PM processes to ensure success. PM becomes more of an art than a science.

16.2.1 Initiating Processes The initiating processes are those processes that define the project and gain authorization to start. It is not limited to starting new projects but it includes a new phase of an existing project, which is separate and distinct. For in-phase changes, the project is normally very large or complex. The initiating process may be performed at different levels (project, organizational, program, or portfolio). Likewise, the initiating process can take on different approaches depending on the company/firm. One approach that is quite successful if done correctly is stakeholder (customers, sponsors, etc.) involvement during the initiation phase. This allows for more interaction and a shared understanding of the scope, success criteria and deliverable acceptance. It can also allow for “value focused thinking” to become part of the process from the beginning. Value focused thinking is the concept by where you inculcate stakeholder values into your decision-making process. In the case of the initiation process, stakeholder values are addressed in the initiating process by including stakeholders at the beginning to help shape the scope, process and deliverables. It also ensures shared buy-in to the project management process (Parnell et al., 2010). Initiating defines and authorizes the project through a project charter. It provides the project manager the authority to execute the project using company/firm resources. A major aspect of this document is a defined start, defined project boundaries, and the establishment of formalized project record (PMI, 2013; Meredith, Mantel, and Shafer, 2015). The major inputs that assist in developing the project charter are: a. Expert judgment: Expert judgement is expertise provided from within the company/firm as well as any group or individual with specialized knowledge or training applicable to the project. b. Statement of work: The project statement of work is a detailed description of the project deliverables (products or services). The statement of work can come from within the company/firm if the project is internal or from a number of external sources if the project is from outside the organization. c. Business case: The business case describes the required information from a business perspective to determine whether or not there is enough benefit to the company/firm to do the project. There are several methods (numeric and non-numeric) for determining the level of cost vs. benefit of the project. d. Agreements: Agreements are necessary and help define the objectives of the project. Agreements can take various forms but regardless of form, the intent is to establish intent of the customer and service provider. e. Enterprise environmental factors: Simply stated, enterprise environmental factors are those aspects of the company/firm’s operating procedures as well as government and industry standards of practices. Finally, marketplace conditions also come into play here and help set the conditions from which the project will operate. f. Organizational process assets: Are those processes established by the company/firm that are standard operating procedures. This includes templates (e.g., project charters), historical records, lessons learned, after action reviews, etc. 254

Project Management’s Role in Engineering Management

Hopefully, the results of these inputs will produce an effective project document that will lay the foundation of a successful project. The project charter is a senior-level document that defines the scope of project. It formally authorizes the project and gives the project manager authority to begin the project and to employ company/firm resources toward the project. It includes, but is not limited to the following (PMI, 2013; Meredith et al., 2015): a. Purpose: A short and concise statement which includes the general goals of the project and their relationship to the company/firm’s objectives. b. Project objectives: More detailed statements of the goals, their priorities, defining success and how the project is terminated. c. Project overview: This is a higher level description of the project. A discussion of the management and technical aspects of the project is included. d. Project schedule: The project schedule is outlined to include major milestones and “phase gates.” Major tasks are listed with associated time. e. Project resources: Project budget is discussed in this section. There is detailing of capital and expense requirements. All contractual items are listed in detail. The cost monitoring the control procedures are addressed here as well. f. Stakeholders: All key stakeholders are listed with appropriate insights and analysis accompanying the list. The expected personnel requirements of the project are listed here as well. Additionally, any special personnel requirements and training are addressed. Essentially, all unique requirements associated with personnel should be addressed. g. Risk management plan: The risk management plan addresses mainly the high level risks that are anticipated against the project. The risks are spelled out in detail so that they are clearly understood to include the potential impacts to the project and company/firm. This section is quite critical and requires tremendous effort. Although the effort is placed into this section, it cannot cover all potential catastrophes. However, if done correctly, the major anticipated risks can be identified and proper mitigation measures taken to lessen their impact. h. Evaluation methods: How a project performs is a matter of properly evaluating it against standards and methods that are spelled out initially. The evaluation methods lay out a description of the procedures used during the monitoring controlling of the project. These are the basic elements of the project charter and constitute the foundation from which the project management plan is developed. It should at a minimum set the high-level boundaries of the project. The size and detail of the project charter depends on the size and complexity of the project.

16.2.2 Planning Processes The planning process is critical to setting the conditions for overall success of the project. Indeed, inadequate planning is the greatest contributor to most project failures. Poor planning makes achieving project schedule, cost and performance objectives almost impossible. In the project planning process, the project management plan is the deliverable. The project management plan allows for project scope and objectives to be clearly defined and established. There are several techniques and approaches to assist in this planning effort. The major inputs that assist in developing the project management plan are: a. Project charter: As previously described, the project charter is the foundational document that sets the project up for success. b. Outputs from other processes: These represent a number of different outputs that can be used to assist in developing the project management plan. c. Enterprise environmental factors: As stated previously enterprise environmental factors are those aspects of the company/firm’s operating procedures as well as government and industry standards of practices. d. Organizational process assets: Processes established by the company/firm that are standard operating procedures. This includes templates (e.g., project charters), historical records, lessons learned, after action reviews, etc. 255

Engineering Management Handbook

The first step in the planning process is analyzing the preliminary project charter and project management processes in order to develop a project management plan. The project management plan uses all the necessary subordinate plans and integrates them into a cohesive effort directed toward achieving the project goals. The project management plan encompasses the activities needed to identify, define, combine, unify and coordinate the various processes to successfully accomplish the project. The subordinate plans include, but are not limited to, (Kerzner, 2006; PMI, 2013; Meredith et al., 2015; Parnell et al., 2010): • Scope management plan • Requirements management plan • Schedule management plan • Cost management plan • Quality management plan • Process improvement plan • Human resource management plan • Communication management plan • Risk management plan • Procurement management plan • Stakeholder management plan These subordinate plans include project management techniques that allow the specific plans to be implemented, monitored and controlled. For example, the project schedule management plan is used to ensure that the project schedule is followed and maintained to ensure the project time and performance objectives are achieved. The schedule management plan is a major input to the planning schedule management, defining activities, estimation of activity resources and durations and schedule development. These plans become integral pieces to an overall complex planning process and require special attention in order to achieve overall project success. For the purposes of this chapter, the specific management plans listed will not be detailed. Suffice it to say, the project management plan takes tremendous effort to pull off correctly; however, it is effort well spent if there are minimal instances of “changes” to the project further into the project schedule. Changes that occur further into the schedule can have a tremendous impact the overall cost of a project (PMI, 2013). There are a number of project management methods and techniques that assist in developing the project management plan. Understanding and mastery of these techniques are critical to the success of the project manager and the project. Table 16.1 lists a number of the project management techniques. For example, the work breakdown structure (WBS) is an effective technique to assist in the project scoping process. The WBS is prepared to help determine the tasks required to complete a specific project. The WBS is not limited to one particular format or structure. What is important is the process of breaking down larger tasks into smaller tasks. This becomes a hierarchical process – starting with an overall project objective followed by successive smaller tasks until all tasks are identified. A WBS can appear as a tree diagram (Figure 16.3) with level one tasks directly below the overall project objective followed by level two tasks. In the case of a university cafeteria improvement project example, Figure 16.3 illustrates a simplistic WBS (Meredith et al., 2015).

256

Project Management’s Role in Engineering Management Table 16.1. Project Management Tools and Techniques a. Linear responsibility charts b. Scheduling - Schedule milestone list - Activity sequencing - Activity resource estimating - Activity duration estimating c. Project configuration management Requirements d. Order and magnitude cost estimate e. Product acceptance criteria f. Resource allocation - Critical path method - Resource loading - Resource leveling - Constrained resource scheduling g. Staffing management plan

h. i. j. k.

Earned value analysis Schedule baseline Cost baseline Quality baseline - Quality assurance - Quality control l. Risk - Risk identification - Qualitative risk analysis - Quantitative risk analysis - Risk response m. Critical path method n. Value engineering o. Stakeholder communication plan p. Document control register q. Change control system

Figure 16.3. Work Breakdown Structure University Cafeteria Operations Inprovement Project

Logistics

Food Preparation

Food Delivery to Customers

Clean Up Operations

Defrost Food Prep Food

Cook Food

The overall objective of this project is to increase the efficiency of the university cafeteria operations. Under the level one task #2—food preparation—the three listed subtasks include defrost food, prepare food and cook food. This logical breakdown allows planners working the project scope management plan to accurately identify the requirements associated with every task supporting the objective of improving cafeteria operations. Although this example may seem simplistic, the WBS is a highly effective tool in supporting the project scope management plan. Other effective tools and techniques identified during the planning process to include, but are not limited to, those identified in Table 16.1 (PMI, 2013; Kerzner, 2006; Meredith et al., 2015; van Gigch, 1991; Forsberg, Mooz and Cotterman, 2000; Smith and Reinertsen, 1998; Parnell et al., 2010). Following work with the WBS the linear responsibility chart (LRC) is used in conjunction with the WBS. When larger tasks are broken down to the basic tasks, a LRC can take those tasks and assign proper personnel responsibility. The LRC shows the critical interfaces between tasks and individuals and highlights areas that require special management attention (Meredith et al., 2015). Such a chart is illustrated in Figure 16.4, which takes the cafeteria WBS in Figure 16.3 and assigns a number of tasks to different individuals and teams. Note: a number of implied tasks are not listed on the WBS and include project plan and budget. 257

Engineering Management Handbook

Program Manager

Lead Planner

Planning Team

Engineering Manager

Engineering Team

Operations Manager

Operations Team

Logistics Manager

Logistics Team

Lead Budget Analyst

Budget Team

Resource Manager

Resource Team

Establish Project Plan

5

6

2,4 1,2

1

3

4

3

4

3

4

3

4

3

4

Establish Project Budget

5

6

2,4

4

4

3

4

3

4

3

4

1,2

3

3

4

Plan Efficient Cafeteria Layout

5

6

1,2

3

4

1,2

1

3

4

3

4

3

4

3

4

• Identify Planning Factors

5

6

6,2

3

3

4

3

4

3

4

3

4

3

4

• Determine Cost Factors

6

6

3

3

3

3

3

3

3

3

1,2

3

3

3

Plan Food Distribution Layout

6

6,2

3

4

1,2

3

3

3

4

4

4

4

3

3

Plan Resource Requirements

5

6,2

3

3

4

4

3

3

3

4

4

4

1,2

3

• Identify Material Requirements

5

6

3

4

4

4

3

3

3

4

4

4

1,2

3

• Identify Inventory Requirements

5

6

3

4

4

4

3

3

3

4

4

4

1,2

3

Acquire Kitchen Equipment

6

6,2

3

4

3

3

3

3

1,2

3

Plan Personnel Training

5

6,2

3

4

3

4

1,2

3

Project Manager

Senior Vice President for Programs

Figure 16.4. Linear Responsibility Chart

3

3

3

3

1 - Responsible 4 - Consultation Possible 2 - Supervision 5 - Must be Notified 3 - Consultation Mandatory 6 - Formal Approval

258

The associated generic managers and teams are included to demonstrate the complexity of the overall organization and the interfaces between departments. As you move from left to right in the chart, you will notice relationships with particular individuals and teams. The project management plan is important because it is the foundation for executing the project. Indeed, formal approval is most likely to occur all the way through the chain of command from the project manager, through the program manager, to senior vice president for programs. The LRC illustrates the relationships between the lead planner, his or her team and the other departments. For planning purposes, it is important to consult each department because they have valuable information that the planner uses. At a minimum, the lead planner needs to consult with each department manager. As would be expected, a 3 is used to identify a mandatory consultation requirement on the part of the lead planner. You can argue that the lead planner has a mandatory requirement to consult with the team as well but may not be how the organization is set up. In this case, it is assumed that the department managers are the “gatekeepers” to their departments and that the lead planner needs to consult with them versus going directly to the department team (Parnell et al., 2010). Solid work must be part of planning process to ensure success of the project. Many projects fail, regardless of size, when left to planners or individuals who have limited practical experience. Therefore, experienced project managers must be part of the planning team for the plan to be a success. Their expert advice brings a level of practical experience that equates to time and money savings when the final project management plan moves to execution. As Figure 16.2 illustrates, the planning process is an iterative process that includes the executing and monitoring and controlling processes. As execution begins, there are inevitable changes (information updates, challenges with resource allocation, scheduling delays, value engineering ...), which are identified in either the executing or monitoring and controlling processes. Once identified, the change/information is fed back into the planning phase to allow for project management plan updating. No one can predict the changes that occur and therefore the close integration of the

Project Management’s Role in Engineering Management

planning, execution and monitoring and controlling processes is essential for success. The project manager must constantly seek this type of feedback in order to adjust the plan and execute accordingly. The only certainty is change so the project manager must be ready to adjust (Parnell, 2010).

16.2.3 Executing Processes The executing processes consist of those processes that are executed to complete the scope of work identified in the project management plan. Very concisely, the executing process involves tremendous coordination of personnel and resources, stakeholder expectation management, and executing all directed and implied tasks associated with the project management plan. It is complete project integration management. The executing processes integrates the management areas of quality assurance, human resources, communications, procurement and stakeholder. The executing process requires the PM team to perform a myriad of actions to execute the project management plan. As discussed, the project management plan is made up of several specific management plans that utilize several tools and techniques to execute the project. The PM team is required to orchestrate the integration of these plans. Additionally, the PM team must track the deliverables (products, results and/or services) from each of the subordinate plans. Of particular importance is the communications plan. The distribution of accurate information allows the PM team to track progress and status of the project. Another important task is managing stakeholder expectations. Stakeholder’s involvement during this process is troublesome at times. The project manager should always be aware of what information is passed along to stakeholders. Projects do not go well all the time but, if executed correctly, they do become successful. Stakeholders, not unexpectedly, react to downturns and poor performance that naturally occur during the project’s lifecycle. If the project manager the information passed to stakeholders is not handled correctly it becomes problematic. Additional tasks during the executing process include but are not limited to (PMI, 2015): • Direct and manage project work • Perform quality assurance • Acquire project team • Develop project team • Manage project team • Manage communications • Conduct procurements It is in the best interest of the project manager and the company/firm to ensure information that affects the project management plan be updated as quickly as possible to ensure appropriate corrective/ improvement action is implemented in a timely manner. Most deliverables take the form of tangible items such as a road, building, etc., but intangible deliverables such as training, information, etc. are also provided. The executing process is complex and requires integration of many areas (PMI, 2015).

16.2.4 Monitoring and Controlling Processes The monitoring and controlling processes consists of those processes whose function is to track and manage the progress and performance of the project. In controlling the project, data is gathered, which requires a change to the project, and changes are made to the project plan according. The monitoring and controlling process is an iterative process, which ebbs and flows accordingly. In Figure 16.2, the monitoring and controlling process encompasses all the other processes and rightly so. It is through these processes that adjustments are made throughout the project in response to heartbeat of the project. Like a flowing stream, when faced with an obstacle, adjusts its flow but continues its journey until it reaches its final destination. It never stops but continues to progress and adjusts accordingly. The fundamental purpose of the monitoring and controlling process is to monitor the other processes so that effective control measures be directed to keep the project performing to expected performance expectations, on time, and below cost (Meredith et al., 2015).

259

Engineering Management Handbook

One of the first steps to proper monitoring is identifying those essential elements requiring control. The first item for consideration for a project manager is control of performance, time and cost. The project manager must establish clear boundaries for control and identify the level of importance for each category. It is safe to say that the boundaries and level of importance are not the same for each project and are driven by the project’s overall project charter (Meredith et al., 2015; Fisk et al., 2004; Palmer, 2006). Continuous monitoring in each of the subordinate plans allows the PM team to keep current with the changing dynamics of the project and to register the project’s health through the prism of performance, time and cost. In Figure 16.5, the three main project concerns are scope, time and cost. A project manager is always concerned with these three and they are referred to as the golden triangle. Figure 16.5 demonstrates how continuous monitoring of a project motivates confidence for a project manager that the project is on-track or that it requires action to restore it to the proper status. Figure 16.5 is a three-dimensional “snapshot” of the project status at a specific time and compares the cost and required results. The project manager can see where the actual project is in relation to the “performance target.” For example, if the actual results right of the performance target, the project manager can see that for the given performance, the cost is too high, which means that the project is exceeding the budgeted amount for a given value. Therefore, corrective action by the project manager is required. Figure 16.5. Project Performance Target = Scope, Time, and Cost

260

The actions of the project manager are varied depending on his/her level of authority. The project manager can either, increase resources (money, equipment or personnel) or shorten the time. An example of shortening time is called “crashing the schedule.” Crashing the schedule is a technique used in the critical path method to bring a project back on schedule if a particular task is going over the scheduled time allowed. This technique requires placing resources against the task in order to reduce its duration. The obvious outcome of using this technique is increasing the cost of the project. The project manager and the team understands the importance of these monitoring and controlling techniques and will ensure analysis of the results provide an accurate picture of the project’s status (Parnell et al., 2010). There is also earned value (EV), which is also a method for monitoring a project. EV analysis is a commonly used method for measuring the overall performance of a project (PMI, 2013, Meredith et al., 2015, Parnell et al., 2010). It is a comparative analysis of the projected budget, actual costs and the work accomplished (value). Figure 16.6 illustrates an EV graph representing the facility layout portion of the university cafeteria facility renovation in Figure 16.3. In this example the EV is lagging behind the budgeted amount and the actual expenditures. The project manager will conclude that the project is behind schedule. The project cost to-date exceeds the value accumulated for the work performed.

Project Management’s Role in Engineering Management Figure 16.6. Earned Value Graph

The chart also tells another story beyond a summary of goals being achieved or not. The contractor is surging as indicated by the stair stepping pattern of the EV line. This pattern normally indicates challenges with the contractor. At the 25 February date, the contractor is stagnant (EV does not increase) as expected with most project start dates. The project manager responds with an increase in actual budget expenditure. Cash flow is extremely important for the contractor as well as the project manager. Both have expectations (project manager—value for the money expected and the contractor—money necessary to create value) and these expectations must be balanced by both parties. Even though the EV is below what the project manager would want, there are situations where paying the contractor ahead of the EV is the best course-of-action for project success. This occurs most often when a good working relationship is established between the project manager and contractor. In this case, continuing to work with the contractor to keep progress going instead of, for example, dismissing him from the project and pursuing litigation is probably the best course of action for the PM in this case. The data past the 11 March date shows that the contractor responded with an upsurge in the EV. However, on 25 March, the contractor slows down again and EV begins to plateau. In this instance, the PM reacted differently and lowers the payment to the contractor in order to get him to respond. Of note is near the end of the project, there is great gain in EV for the overall project. This is a very typical pattern for a renovation project. Engineers have a tendency to estimate a concave shape over the project, indicating optimism in terms of how quickly value (functionality, performance) can be developed. In reality, the EV curve assumes a convex shape because of rework imposed by test failures, delayed schedules and response surges in activity, and cost-conserving measures put in-place to mitigate the threat of running out of budget before the expected (or required) value has been delivered. Although it is not what a project manager would like to see, this is not uncommon for a project that was behind schedule as depicted in this example (Parnell et al., 2010; Meredith et al., 2015). The monitoring and controlling process uses (PMI, 2013): a. Expert judgment: Use of the judgment of the project manager and the PM team. They interpret the information provided and develop appropriate actions. b. Analytical techniques: These techniques help forecast budget and value outcomes. These include but are not limited to: • Regression analysis • Grouping methods 261

Engineering Management Handbook

• Causal analysis • Root cause analysis • Forecasting methods • Failure mode and effect analysis • Fault tree analysis • Decision tree analysis • Reserve analysis • Trend analysis • Earned value management • Variance analysis c. Project management information system: This is part of enterprise environmental factors and encompasses a host of information systems and are limited only by the company/firms commitment to these systems. d. Meetings: Meetings are critical if done efficiently and effectively. Effort must be placed into proper meeting management. If not done correctly time and resources are wasted. The outputs from the monitoring and controlling process establish guidance for change to the project. A number of products are produced from the tools and techniques mentioned earlier. These include, but are not limited to, (PMI, 2013): e. Change requests: Change requests are the result of the comparison of planned results and actual results. Change requests can impact project scope (expand, adjust, and/or reduce). Impact from change requests must be followed through by the project manager and PM team. Attention to the secondary and tertiary effects of any change cannot be understated. Therefore, it is understood that policies and procedures should be developed by the company/firm. Several changes can occur and listed below are a few more common ones (PMI, 2013). • Corrective action • Preventive action • Defect repair f. Work performance reports: Work performance reports are essentially documentation of actual work performed. It is important that a physical or electronic copy of the work performed be cataloged and stored for the duration of the project. Although not mentioned earlier, document control is critical here. In most companies/firms, document control is a full-time job, requiring a certain skill level. g. Project management plan updates: These updates are focused on all the project management plans used to keep the project tracking on time, on budget and at performance. Each of these management plans are listed (PMI, 2013): • Scope management plan • Requirements management plan • Schedule management plan • Cost management plan • Quality management plan • Scope baseline • Schedule baseline • Cost baseline h. Project documents update: This is the process of updating all project documents required to manage the project. These documents include, but are not limited to (PMI, 2013): • Schedules • Cost forecasts • Work performance reports • Field changes • Challenge logs 262

Project Management’s Role in Engineering Management

16.2.5 Closing Processes The closing process is the most difficult of the process groups for a number of reasons. A good project manager will ensure that closeout procedures are closely adhered to so that the project can be properly turned over to the client and that anything required legally is completed correctly. The closing process involves all the necessary administrative and contractual closing procedures to ensure proper project closeout. This process includes all the subordinate project plans as well as any phases (complex/large projects) that are associated with the project. The administrative closing procedures are those procedures and project relationships dealing with different aspects of the project (PMI, 2013). The largest portion of the administrative work required includes documentation control and archiving. This process is important just in case there is possible legal action taken against the company/ firm. All contracts start out with good intentions on the part of both parties, but there are lawsuits brought against a company/firm by a contractor for numerous reasons. Good document control plan allows the project manager and the PM team too quickly and efficiently respond to any legal request. The contract closure procedure includes those activities needed to complete contractual obligations. These include, but are not limited to, (PMI, 2013): • Formal acceptance by the customer • Post-project review • Record impacts • Document all lessons learned • Apply appropriate updates (organizations, billing, operational procedures, etc.) • Archive all relevant documentation • Close out procurement activities and termination of all applicable agreements • Perform team members’ assessments • Release project resources One important process is ensuring proper contractor documentation completion. Near the end of most projects, the daily progress reports will tend to get overlooked in the last few weeks, especially if a contractor is near 100% paid. If a project manager fails to hold the contractor to full compliance for contractual requirements such as progress reporting, the project manager is liable. It is imperative that the PM team ensures all aspects of close out procedures are adhered too and followed (Parnell et al., 2010).

16.3 Summary Engineering management is a tremendously relevant and important discipline. It bridges the gap between management and engineering. This bridge is supported by a number of difference disciplines. Of most import is PM. PM as it relates to EM is nestled in an arching way to assist the EM professional in being able to manage projects. Many of those skills found in EM are found in PM. Proper PM consists of initiating, planning, executing, monitoring and control and closing processes. The initiating processes are those processes that define the project and gain authorization to start. It is not limited to starting new projects but it includes a new phase of an existing project, which is separate and distinct. For in phase changes, the project is normally very large or complex. The initiating process may be performed at different levels (project, organizational, program, or portfolio). Likewise, the initiating process can take on different approaches depending on the company/firm. A major deliverable coming out of the initiating processes is the project charter. Several inputs occur during this initial process phase, which include expert judgment, statement of work, business case, agreements, enterprise environmental factors and organizational process assets. The planning process is critical to setting the conditions for overall success of the project. Poor planning makes achieving project schedule, cost and performance objectives almost impossible. In in the project planning process, the project management plan is the deliverable. The project management plan allows for project scope and objectives to be clearly defined and established. There are several techniques and approaches to assist in this planning effort. The major inputs that assist in developing the project

263

Engineering Management Handbook

management plan include the project charter, outputs from other processes, enterprise environmental factors and organizational process assets. The executing processes consist of those processes that are executed to complete the scope of work identified in the project management plan. Very concisely, the executing process involves tremendous coordination of personnel and resources, stakeholder expectation management, and executing all directed and implied tasks associated with the project management plan. It is complete project integration management. The executing processes integrates the management areas of quality assurance, human resources, communications, procurement and stakeholder. The executing process requires the PM team to perform a myriad of actions to execute the project management plan. These requirements include but are not limited to direct and manage project work, perform quality assurance, acquire project team, develop project team, manage project team, manage communications and conduct procurements. The monitoring and controlling processes consists of those processes whose function is to track and manage the progress and performance of the project. The monitoring and controlling process is an iterative process that ebbs and flows accordingly. It is through these processes that adjustments are made throughout the project in response to heartbeat of the project. The monitoring and controlling never stops but continues to progress and adjust accordingly. The fundamental purpose of the monitoring and controlling process is to monitor the other processes so that effective control measures be directed to keep the project performing to expected performance expectations, on time, and below cost. The monitoring and controlling process uses expert judgment, analytical techniques, project management information system and meetings. The closing process is the most difficult of the process groups for a number of reasons. A good project manager will ensure that closeout procedures are closely adhered to so that the project can be properly turned over to the client and that anything required legally is completed correctly. The closing process involves all the necessary administrative and contractual closing procedures to ensure proper project closeout. The closing process requires formal acceptance by the customer, post-project review, record impacts, document all lessons learned, apply appropriate updates, archive all relevant documentation, close out procurement activities and termination of all applicable agreements, perform team members’ assessments and release project resources.

16.4 References Forsberg, K., Mooz, H., and Cotterman, H., Visualizing Project Management. 2nd Ed. New York: John Wiley & Sons, Inc., 2000. Hicks, P., Utely, D., and Westbrook, J., “What are we teaching our engineering managers?” Engineering Management Journal, vol. 11, no. 1, March 1999, pp. 29-34. Kerzner, H., Project Management: A Systems Approach to Planning, Scheduling, and Controlling, 9th ed., Hoboken, NJ: John Wiley & Sons, Inc., 2006. Palmer, D., Maintenance Planning and Scheduling Handbook. 2nd ed. New York: McGraw-Hill, 2006. Parnell, G., Driscoll, P., and Henderson, D., Decision Making in Systems Engineering and Management. 2nd Ed., New York: John Wiley and Sons, 2010. PMI, A Guide to the Project Management Body of Knowledge (PMBOK Guide), 5th ed., Newtown Square, PA: Project Management Institute, Inc., 2013. Smith, P. G. and Reinertsen, D. G., Developing Products in Half the Time. 2nd ed. New York: John Wiley & Sons, Inc., 1998. van Gigch, J. P., System Design Modeling and Metamodeling. New York: Plenum Press, 1991.

264

Systems Engineering

17 Systems Engineering Robert Cloutier Stevens Institute of Technology

Mary Bone

Stevens Institute of Technology

265

Engineering Management Handbook

17.1 Introduction 17.1.1 What is Systems Engineering? Although the discipline called systems engineering (SE) may seem self-descriptive, the concept is deceptively difficult to define. One must begin with an understanding of what constitutes a system. The International Council on Systems Engineering (INCOSE), which is the largest professional organization dedicated to the practice and advancement of SE, defines a system as “a combination of interacting elements organized to achieve one or more stated purposes” (INCOSE, 2011). Possibly a more descriptive definition was provided by Rechtin (1991) who said a system is “a set of different elements so connected or related as to perform a unique function not performable by the elements alone.” Rechtin went on to explain that one could disassemble an automobile (which can be defined as a system that has the function of transportation) and spread the parts out on the ground. While all the parts are there, no single part is able to transport a person from one place to another. It takes the assemblage of those parts, in a specific manner, for the system to be able to provide the unique function of transportation not found in any individual part, nor in a partial collection of the parts—it takes the whole set. Therefore, if a system is a collection of parts, which when assembled in a specific manner, are able to perform a purpose, then SE must be the practice of designing such a system. However, going back to our automobile, many different engineering disciplines are required for that design to be accomplished. Expertise in internal combustion engines is required, as is in automobile body design, human controls, and even electrical wiring. Knowledge from all of these domains (and many others) is necessary to design this system called an automobile. INCOSE (2004) defined SE as “an interdisciplinary approach and means to enable the realization of successful systems.” However, this definition was been expanded by INCOSE in Version 3.2.2 of their handbook (INCOSE, 2011): Systems Engineering is an interdisciplinary approach and means to enable the realization of successful systems. It focuses on defining customer needs and required functionality early in the development cycle, documenting requirements, and then proceeding with design synthesis and system validation while considering the complete problem. Systems Engineering considers both the business and the technical needs of all customers with the goal of providing a quality product that meets the user needs. It becomes clear from this definition that SE is an approach to creating a system that will satisfy the desired functionality goals of the system. The approach includes defining the system, designing the system, and validating the system—ensuring the right system was built. SE is a collection of tasks that are necessary to provide a quality system to a customer. Figure 17.1 represents a mind map of key SE concepts. The major branches are perspectives, management, technical and focus. The branches flowing out of those key concepts further describe those concepts. These key concepts will be further discussed throughout this chapter. Figure 17.1. Mind Map of SE Key Concepts (Cloutier, Baldwin, and Bone, 2015)

266

Systems Engineering

Stevens Institute of Technology has put forth the following definition for SE: Systems engineering is the translation of a need/deficiency into a system architecture through the iterative process of functional analysis, allocation, implementation, optimization, test, and evaluation. It incorporates all technical parameters to assure compatibility between physical and functional interfaces, hardware and software interfaces, in a manner that optimizes system definition and design; and the integration of performance, manufacturing, reliability, maintainability, supportability, global flexibility, scaleability, upgradeability and other specialties into the overall engineering effort. In this chapter we will also look at the steps necessary to provide such a quality system, and what processes are involved when someone performs SE. But first, let us look into the history of SE to better understand how the discipline has arrived at where it is today.

17.1.2 Why Did Systems Engineering Originate? While some will state “systems engineering is a new engineering discipline”, that is actually incorrect. Although SE is relatively newer when compared to classical engineering, such as mechanical and civil, it has existed as a discipline since before World War II. Later, NASA was in need of an approach to meet President Kennedy’s challenge—take a man to the moon and back by the end of the decade. Figure 17.2. U.S. Army Corporal Sounding Rocket

SE began to emerge in a number of companies in the late 1940s and early 1950s that were attempting to solve complex problems. AT&T Bell Labs needed an approach to provide in transmission designs, as well as a way to provide consistent communications with Western Electric; JPL was developing the U.S. Army Corporal sounding rocket in 1945 as shown in Figure 17.2 and found they lacked any consistency in the engineering processes, and that the rocket was very unreliable. Both efforts needed a process to transition new ideas from research concepts into engineering designs. Later, NASA was in need of an approach to meet President Kennedy’s challenge – take a man to the moon and back by the end of the decade. The common denominator in each of these needs was an interdisciplinary approach to integrate complex ideas and concepts into a system that would allow a complex idea to become a reality. Arthur Hall wrote an early textbook detailing the practice of SE at that time. Figure 17.3 is the cover of Hall’s book published in 1962 . He noted in the Preface that “the growing recognition of the need for SE over the past decade has been attended by the need for philosophical foundations.” He stated that his 267

Engineering Management Handbook

book was intended “to increase awareness and understanding of SE as a process, and to sharpen definitions and approaches to the main recurring problems of the process-problem definition, goal setting, systems synthesis, systems analysis and choice among alternative systems.” Figure 17.3. Hall’s Book on Systems Engineering, Published in 1962

17.1.3 The Systems Engineering Lifecycle Everything has a lifecycle. Animals are conceived, born, and mature into adults, enter old age, and then pass away. The same can be said about a system. It is conceived by the customer, an engineer, or both; the idea is further developed into a concept. Some preliminary design is conducted to ensure the system is feasible, and if it is determined to continue, detail design and development ensues. The product is produced and used, and at some time, the system outlives its usefulness and is retired. This is the systems lifecycle. There are many different lifecycles defined by various experts for SE, and the lifecycle used on a project is normally dictated by the industry or customer. Figure 17.4 shows some typical lifecycles. Figure 17.4. Common Systems Engineering Lifecycles in Use Today Typical High-Tech Commercial Systems Integrator Study Period User Requirements Definition Phase

Concept Definition Phase

Implementation Period

System Specification Phase

Acq Prep Phase

Typical High-Tech Commercial Manufacturer Study Period Product Requirements Phase

Product Definition Phase

Product Development Phase

Source Select. Phase

Operations Period

Verification Phase

Development Phase

Deployment Phase

Implementation Period Internal Test Phase

Engr Model Phase

Operations and Maintenance Phase

Deactivation Phase

Operations Period

External Test Phase

Full-Scale Production Phase

Manufacturing, Sales, and Support Phase

Deactivation Phase

ISO/IEC 15288 Utilization Stage

Production Stage

Development Stage

Concept Stage

Retirement Stage

Support Stage

US Department of Defense (DoD) 5000.2 A

B

Pre-systems Acquisition

C

FOC

IOC

Systems Acquisition

System Development and Demonstration

Concept and Technology Development

Sustainment

Production and Deployment

Operations and Support (including Disposal)

US Department of Energy (DoE) Project Planning Period Pre-Project

Typical Decision Gates

New Initiative Approval

Preconceptual Planning

Concept Approval

Conceptual Design

Project Execution Preliminary Design

Development Approval

Adapted from: INCOSE Systems Engineering Handbook, v.3.1

268

Final Design

Construction

Production Approval

Mission Acceptance

Operations Approval

Operations

Deactivation Approval

Systems Engineering

Although each of the lifecycles accomplishes the same work, the order or naming conventions used may differ. In Figure 17.4, one of the lifecycles is that documented by ISO/IEC 15288 Systems Engineering— System Life Cycle Processes (ISO/IEC 15288, 2015). The remainder of this chapter will use that lifecycle to discuss SE. There are a number of supporting processes that also occur, and are shown in Figure 17.5. Figure 17.5. ISO/IEC 15288 SE Processes – Gray indicates changes in 2015 revision

SE is concerned with all aspects of the systems lifecycle, and must consider each part of the lifecycle during the design and development of a new system. As can be seen by studying Figure 17.5, many of these processes are very specialized, and can take some time to master. This is an important aspect of SE. Although a systems engineer must be aware of, and knowledgeable about each process, on large and/or complex projects, the execution of each process may be conducted by a systems engineer that has become an expert in that particular process. The purpose of this chapter is not to explore each process, but to introduce the reader to the more significant processes and let the reader explore the remaining processes by reading ISO/IEC 15288. This chapter will further explore the following phases that can be mapped to 15288: • Stakeholder Requirements Definition • Requirements Analysis 269

Engineering Management Handbook

• • • • • • • •

Architecture Design Implementation Integration Verification Transition Validation Operation and Maintenance Disposal

17.1.4 The Role of Systems Modeling and Systems Simulation A topic that does not fit into the processes listed in ISO/IEC 15288: 2015, and many times is described by the tool being used is system modeling and simulation. On the surface, modeling seems to be a straightforward concept. A model is a facsimile of the item of interest. When one sends a fax of a memo using a fax machine, the original is copied, and a facsimile of the original is transmitted. Alternatively, one may build a model of a volcano out of clay for a science project. By no means is the model a volcano—it is a representation of a volcano. Models need not be exact—they only need to emphasize that part of the item of interest that is to be examined. While the model of a volcano may look like the original, it does not have to display, nor could it, the extreme temperatures of the flowing lava and the life threatening nature of the gases expelled. In the SE domain, modeling and simulation are words that elicit different mental models, based on the training and experience of any given systems engineer. To some, it will mean dynamic modeling or simulation of the system to predict or understand system performance. To others, it may evoke the thought of cost modeling to apportion cost and to ensure the system in built within a specified budget. And to others it may mean using the use of the Object Management Group’s (OMG, 2008) Systems Modeling Language (SysML, 2008). There is no agreement to the proper terminology, so the proposed terminology used here is that preferred by the author. It is beyond the scope of this chapter to go into more depth of this topic. If the reader is interested in more detail, the INCOSE Model Based SE (MBSE) working group, and the OMG website would be good starting places.

17.2 Stakeholder Requirements Definition This phase is sometimes referred to as requirements elicitation. Stakeholders are anyone that will be affected in anyway by the system of interest, that is, the system being designed. The stakeholders may be active stakeholders, or passive stakeholders. Active stakeholders are individuals, things, or other systems that play an interactive role during the operation of the system of interest. One easily overlooked active stakeholder might be the environment—rain, snow, and ice may play a very acute role with an automobile, and therefore might be considered an active stakeholder. Passive stakeholders are individuals, things, or other systems, protocols, procedures, etc., which can influence the success of the system. For instance, a regulatory body such as the Department of Transportation is a passive stakeholder in the design of a new automobile. During stakeholder requirements definition, the goal is to capture the functionality and attributes of the system of interest, in the words of the individual stakeholders. Functionality is what the system of interest does—go fast, hold five passengers, etc. The attributes of the system of interest is that it is safe, that it fits on the roads, etc. It is important to ask probing questions, and fully understand the customer needs. A recommendation is to capture the stakeholder requirements using a noun-verb form. For instance: The driver will be able to listen to the radio while driving the car. The noun is the “driver,” and the verb is “listen.” 270

Systems Engineering

17.2.1 Use Cases and Scenarios One method that is helpful in eliciting requirements from the stakeholders is to develop use cases or scenarios. This is done by working with the stakeholder to capture graphically or textually a description of how the system of interest will be used. In all probability, there will be a number of different scenarios that must be captured. For instance, there may be scenarios for normal operations, degraded operations, maintenance operations, and failed operations to name a few. Other scenarios may include daytime operations, night operations, foul weather operations, etc. The more of these conditions that are understood early, the probability of a more complete set of requirements increases. Figure 17.6 is a sample operational use case for a hybrid SUV. Figure 17.6. Sample Hybrid SUV Operational Use Case uc HSUVUseCases [Operational Use Cases]

HybridSUV

Start the v ehicle

«extend»

Driv e the v ehicle

Driv er

«include»

Accelerate

«include»

«include»

Park

«include»

Steer

Brake

Adapted from: Sparx Enterprise Architect tool, http://www.sparxsystems.com/

17.2.2 Performance Criteria During the requirements elicitation, it is important to understand “soft” words used by the stakeholders. Word like: better, lower than, improved, and improved reliability are difficult to design and verify. All efforts should be made during this phase to reach some quantifiable agreement to what each term means. Examples of this might include: • 25% improved reliability over current performance • 5% better gas mileage than existing models • Whites are 10% whiter, as measured using an optical spectrometer

17.2.3 Inputs and Outputs Another important aspect to understand during requirements elicitation is what interfaces exist between the system of interest and those systems or entities that come in contact with the system of interest. While it is a simple example to understand what the inputs for an automobile will be: fuel, cargo, passengers, security, etc.—these all may have some influence on the design. Same is true for outputs—emissions and horsepower for instance. Table 17.1 shows a simple input/output matrix with examples of the type of data that might be useful for typical inputs and outputs. 271

Engineering Management Handbook Table 17.1. Input/Output Matrix INPUTS Intended

OUTPUTS Unintended

Desired

Undesired

Signal

Pulse shape, data rate, signal to noise ration

Electrical noise

Data rate, accuracy

Error rates, false alarm rates

Electrical

Nominal voltage

Surge voltages and timing

Voltage, current, frequency, stability

Electromagnetic interference, electrical shock

Mechanical

Activation force

Shock and vibration

Movement, resistance

Acoustic noise levels

Environmental

Normal temperature range

Temperature and humidity extremes

Particle density, air flow

Heat, effluents

17.2.4 Conclusions The key during this phase is to listen to the customers, and gather as many requirements as you can. Try to understand what the customer is asking for, and what they are not saying—capture everything possible using dedicated note takers, audio, video—anything that will help recall the conversations. It is also important to understand that mistakes, including poorly communicated or wrong information, at this phase can lead to expensive and time consuming fixes later. This is best demonstrated by the diagram shown in Figure 17.7. It shows that the cost to extract defects goes up exponentially the later in the program they are discovered. Figure 17.7. Input/Output Matrix

272

Systems Engineering

17.3 Requirements Analysis During requirements analysis the systems engineer will sort, assess, evaluate, combine, and expand on the stakeholder requirements for the purpose of better understanding the system of interest in the context of the performance and system constraints expressed by the stakeholders. System requirements are normally expressed in the form of “shall” statements, and express some capability the system is expected to perform. A well-formed system requirement should contain a single measurable capability to avoid confusion or conflict. Examples of well-formed requirements might be: R1 The door shall allow people to pass between the living room and the kitchen R1.1 The door shall comply with the 2006 International Residential Code® New Jersey Edition R1.2 The door shall permit people to see between rooms R1.3 The door shall be equipped with a lock It is also likely that during the interviews and discussions with the stakeholders, conflicting requirements were discovered—one stakeholder wants it to be ecologically responsible, and the other wants it to have very high horsepower. The systems engineer will work to resolve or balance these two apparently conflicting requirements. It is important that the systems engineer create a requirements traceability matrix to track the source of each requirement and how each requirement will be verified to ensure it is satisfied. This becomes important as the system evolves—sometimes requirements must be modified, or even deleted. One must know the source of the requirement to gain approval/concurrence with the change or deletion. Table 17.2 contains a requirements traceability matrix that is used to track requirements. Table 17.2. Requirements Traceability Matrix Requirement

Source

Acceptance Criteria

Test Type/Test Method

C1

Stakeholder B

Horsepower >= 250

Test

C2

Stakeholder A

Emissions