Recent Advances in Operations Management Applications: Select Proceedings of CIMS 2020 (Lecture Notes in Mechanical Engineering) 9811670587, 9789811670589
This book presents the select proceedings of the International Conference on Industrial and Manufacturing Systems (CIMS
149 104 10MB
English Pages 436 [424] Year 2022
Preface
Contents
About the Editors
Risks Associated with Third-Party Logistics in Indian Restaurant Supply Chain
1 Introduction
2 Literature Review
3 Solution Methodology
4 Risks Related with 3PL in Indian Restaurant Supply Chain
5 Application to Proposed Framework
5.1 Structural Self-Interaction Matrix
5.2 Reachability Matrix
5.3 Level partitions
5.4 ISM Model Development
5.5 MICMAC Analysis
6 Result and Discussion
7 Conclusion
References
Improvement of Climate-Smart Agriculture System Based on Obstacles Assessment
1 Introduction
2 Literature Review
3 Research Methodology
3.1 Fuzzy DEMATEL Approach
4 Results and Discussion
5 Conclusion
References
Hurdle Appraisal for the Implementation of Circular Economy Notions in Supply Chain
1 Introduction and its Background
2 Solution Methodology
2.1 ISM-AHP Hybrid Approach
2.2 Identifying the Hurdles in the Circular Economy
3 Numerical Illustration
3.1 Establishing Contextual Relationships Among Different Hurdles
3.2 Conversion of SSIM into Initial Reachability Matrix
3.3 Level Partitioning
3.4 Developing ISM Framework
3.5 MICMAC Approach
3.6 Analytical Hierarchical Process
4 Results and Discussion
5 Conclusion
References
Designing an Air Purifier by Using Green Quality Function Deployment Methodology
1 Introduction
2 Research Background
3 Research Methods
3.1 Green Quality Function Deployment Technique
4 Case study
4.1 Implementation of GQFD for Designing an Air Purifier
5 Result and Discussion
6 Conclusions
References
Supply Chain Management Practice Constructs in SMEs: Development of Constructs and Its Implementation Issue
1 Introduction
2 Review Methodology
3 Literature Review
4 SMEs–SCMP Implementation Issues and Remedies
5 Conclusions
References
Analysis of QMS Practices Performed in ISO 9001 Certified Engineering Educational Institutes of India: An Interpretive Structural Modelling Approach
1 Introduction
2 Research Background
3 Research Methodology and Data Analysis
3.1 Structural Self-Interaction Matrix (SSIM)
3.2 Reachability matrix
3.3 Level Partitions
4 Results and Discussions
4.1 Categorization of QMS Dimensions
4.2 ISM-Based Final Model
5 Conclusion
References
A Review on Life Cycle Assessment of Various Dairy Products
1 Introduction
2 Life Cycle Assessment—LCA
2.1 LCA Methodology
2.2 Goal and Scope Identification
2.3 Life Cycle Inventory Analysis—LCI
2.4 Life Cycle Impact Assessment—LCIA
2.5 Interpretation
3 Results and Discussion
3.1 Raw Milk Production
3.2 Cheese Production
3.3 Yoghurt
4 Conclusion
4.1 Direction Towards Future Work
Appendix
Abbreviations
References
Design of Fuzzy Controller for Blood Glucose Level
1 Introduction
2 Model for Blood Glucose Insulin Sytstem
3 Controller Design
3.1 PID Controller
3.2 FLC
4 Result Discussion
5 Conclusion
References
Prioritization of Renewable Energy Alternatives by Using Analytic Hierarchy Process (AHP) Model: A Case Study of India
1 Introduction
2 Overview of Renewable Energy Resources in India
2.1 Wind Energy
2.2 Solar Energy
2.3 Hydro Energy
3 Research Methodology
3.1 Multi-Criteria Decision-Making (MCDM)
3.2 Analytic Hierarchical Process (AHP)
4 Decision Criteria Factors and Their Significance
4.1 Economic Factors (F-1)
4.2 Technical Factors (F-2)
4.3 Environmental (F-3)
5 A Real Case Study
5.1 Selection of Renewable Energy Sources by AHP
5.2 Data Analysis
6 Results and Discussion
6.1 Main Factors Results
6.2 Sub-factors Results
6.3 Discussion
7 Conclusions
References
Factors Influencing Customer Satisfaction and Belief Toward Car Business Model
1 Introduction
2 Literature Review
3 Objectives
4 Methodology
5 Results and Discussion
6 Conclusion
References
A Review on Musculoskeletal Disorders and Design of Ergonomics Aids with Relevance to Lower Back and Lumbopelvic Pain in Pregnant Women
1 Introduction
2 Methodology
3 Literature Review
4 Results and Discussions
5 Conclusions
References
Comparative Analysis of Supplier Selection Based on ARAS, COPRAS, and MOORA Methods Integrated with Fuzzy AHP in Supply Chain Management
1 Introduction
2 Theoretical Analysis of MCDM Techniques
2.1 Fuzzy Analytic Hierarchy Process (F-AHP)
2.2 Additive Ratio Assessment (ARAS)
2.3 Complex Proportional Assessment (COPRAS)
2.4 Multi-Objective Optimization Based on Ratio Analysis (MOORA)
3 Illustrative Example
3.1 Weightage Calculation of Criteria Using Fuzzy AHP
3.2 Supplier Ranking Using ARAS Method
3.3 Supplier Ranking Using COPRAS Method
3.4 Supplier Ranking Using MOORA Method
4 Result and Conclusion
References
Comparative Efficiency Measurement of Indian Hospitals Using Data Envelopment Analysis: A Proposed Model
1 Introduction
2 Literature Review
3 Research Methods
3.1 Introduction to DEA
3.2 Input/Output Variables and Data Selection
3.3 Assumptions in Variables and Data Selection
4 Result Discussion
4.1 Introduction of Data and Analysis
4.2 Relative Efficiency Score and Rank of Hospitals
4.3 Average Projection of Inputs and Outputs Variables
4.4 Average Slack Value of Inputs and Outputs
5 Conclusions
6 Limitations
7 Future Work
References
An Adoption of Blockchain Technology in Agri Food Supply Chain: An Overview
1 Introduction
2 Methodology
3 Literature Review
3.1 Current Scenario of AFSC
4 Blockchain in AFSC
4.1 Application of BT in AFSC
5 Challenges in Adoption of BT in AFSC
6 Conclusion and Direction for Future Research
References
Prediction Model for Talent Management Analysis Through e-HRM
1 Introduction
2 Proposed analysis
2.1 Analysis for Talent Management System
2.2 Data Analysis for Talent Management System
3 Experimental Results and Discussions
3.1 Talent Management Analysis Using Dataset 1
3.2 Talent Management Analysis Using Dataset 2
4 Conclusions
References
Study of Biomedical Waste Management Performance Indicators in Indian States
1 Introduction
2 Research Background
3 Definition of Indicators for BMWM
3.1 Collection Efficiency
3.2 Waste Diversion Rate
3.3 Uncollected Waste
3.4 GHGs Emissions
3.5 Population Density
3.6 Number of Workshops
4 Results and Discussion
4.1 Indicators Values for Different States
4.2 Principal Component Analysis
5 Conclusions
References
Determining Manpower Requirement for Material Handling in a Factory
1 Introduction
1.1 Background
1.2 Introduction to Material Flow
1.3 Objectives
2 Literature Review
2.1 Review on Material Flow Related Literature
2.2 Review on Material Handling Equipment Related Literature
3 Solution Methodology
3.1 Pick and Place Time for Forklift
3.2 Attach and Detach Time for Tugger
3.3 Pick and Place Time for Hand Pallet
4 Transporting Parameter Estimation
5 Modular Approach
5.1 Unloading Module
6 Material Handling Module
7 Kitting Module
8 Calculations
8.1 Unloading Module Calculation
9 Material Handling Module Calculation
10 Kitting Module Calculation
11 Results and Future Scope
11.1 Results and Conclusions
11.2 Future scope
Appendix
References
Data Analytics for NIRF Ranking of Indian Institutions to Check the Consistency and Validity of Ranking Framework
1 Introduction
2 Research Background
3 Research Methods
4 Application of Research Methods
5 Result Discussion
6 Conclusion
References
Design Feature Assessment for Fused Deposition Modeling Using Supervised Machine Learning Algorithms
1 Introduction
2 Summary of the Research
3 Experimental Methodology
3.1 Selection of Machine Learning Algorithms
3.2 Part Fabrication and Data Collection
3.3 Data Pre-processing and Visualization
3.4 Evaluation of ML Models
3.5 Predictive Analysis
4 Conclusions
References
Development of a 3D Printed Orthopaedic Cast for Wrist Fracture
1 Introduction
2 Methodology
2.1 3D Scanning of the Arm
2.2 CAD Modelling
2.3 Static FEA
2.4 3D printing
3 Results and Conclusion
References
Effect of IoT in Supply Chain Management—A Review
1 Introduction
2 Research Background
2.1 IoT Technologies
3 Methodology
4 Detailed Analysis of Selected Papers
5 Discussions
6 Conclusion
References
Toolpath Generation of a Human Anatomical Shape for Double-Sided Incremental Forming
1 Introduction
1.1 Basic Process of ISF
1.2 Classification of Incremental Sheet Forming (ISF)
1.3 Double Sided Incremental Sheet Forming (DSIF)
1.4 Advantages and Limitations of DSIF
1.5 Issues with DSIF
2 Research Background
2.1 CAD Model (STL Format) Generation Techniques
2.2 Available Toolpath Strategies for Incremental Sheet Forming
3 Research Methodology
3.1 Generation of STL File for Required Human Anatomy
3.2 Slicing and Toolpath Generation
4 Result and Discussion
5 Conclusion
References
Optimization of Cutting Forces in Dry Turning Process Using Taguchi and Grey Relational Analysis
1 Introduction
2 Research Background
3 Research Methodology
3.1 Design of Experiments
3.2 Taguchi’s Method
3.3 Grey Relational Analysis
4 Experimental Work
4.1 Material
4.2 Present Problem
5 Experimental Data and Analysis
6 Results and Discussion
6.1 Optimal Parameter Setting
6.2 Optimal Parameter Combination
6.3 Most Effective Factor
6.4 Confirmation Test
7 Conclusion
Appendix
References
Performance Evaluation of Cargo Warehouse Operations of an Indian Airline Using Discrete Event Simulation: A Case Study
1 Introduction
2 Research Background
3 Overview of the Existing Process in Warehouse
3.1 Cargo Booking
3.2 Process of Cargo Acceptance
3.3 Process of Cargo Uplifting
4 Solution Methodology
4.1 System Definition (see Footnote 1)
4.2 Input Data Collection (see Footnote 1)
4.3 Discrete Event Simulation (DES) Model Using ARENA
5 Results and Discussion (see Footnote 1)
5.1 Evaluation of Current Performance of the Warehouse
5.2 Evaluation of Proposed Scenarios
6 Conclusion (see Footnote 1)
References
Blockchain: A Makeover to Supply Chain Management
1 Introduction
2 Basics of Blockchain
3 Shortcomings of Existing Supply Chain Systems
4 Blockchain-Enabled Supply Chain Systems
4.1 Attributes of a Blockchain-Enabled Supply Chain
5 Use-Cases
6 Conclusion
References
Assessment of the Challenges Obstructing Performance of Indian Food Supply Chain Dynamics
1 Introduction and Background
2 Literature Review
3 Methodology
4 Sensitivity Analysis
5 Results and Discussions
6 Managerial Implications and Future Scope
References
Farm Mechanization Through Custom Hiring Centres in Punjab: A Survey
1 Introduction
1.1 The Role of Agricultural Mechanization in Emerging Regions
2 Research Background
3 Research Methods
3.1 Data Collection
3.2 Data Tabulation and Analysis
4 Results and Discussion
4.1 Condition of Economic Feasibility of PACS Custom Job Services
4.2 Impact of Custom Hiring Services on Farm Machines
5 Conclusions
References
Hospitals’ Selection Under Ayushman Bharat Scheme with Heuristic Search Method Using A** Algorithm
1 Introduction
2 Literature Review
3 Methodology
3.1 Selection of Hospitals Using A** Algorithm
3.2 Algorithm for the A** Search for Hospital Selection
4 Results and Discussions
5 Limitation and Future Scope
References
Role of Blockchain-Oriented Smart Contract in Supply Chain
1 Introduction
2 Research Background
3 Research Methods
4 Results
4.1 RQ 1: What are the Benefits of Blockchain-Oriented Smart Contract Supply Chain?
4.2 RQ 2: What are the Challenges that Relate to the Application of Smart Contracts in Supply Chain Management?
5 Discussions
6 Conclusion
References
Propeller Design and Optimization for Drones
1 Introduction
2 Methodology
3 Propeller Design
3.1 Selection of Airfoil
3.2 Optimization for Lift and Drag
3.3 Lift and Drag Variation Plots
4 Propeller Modelling
5 Propeller Analysis
6 Results and Discussion
7 Conclusion
References
Recommend Papers
File loading please wait...
Citation preview
Lecture Notes in Mechanical Engineering
Anish Sachdeva Pradeep Kumar O. P. Yadav Mohit Tyagi Editors
Recent Advances in Operations Management Applications Select Proceedings of CIMS 2020
Lecture Notes in Mechanical Engineering Series Editors Francisco Cavas-Martínez , Departamento de Estructuras, Universidad Politécnica de Cartagena, Cartagena, Murcia, Spain Fakher Chaari, National School of Engineers, University of Sfax, Sfax, Tunisia Francesca di Mare, Institute of Energy Technology, Ruhr-Universität Bochum, Bochum, Nordrhein-Westfalen, Germany Francesco Gherardini , Dipartimento di Ingegneria, Università di Modena e Reggio Emilia, Modena, Italy Mohamed Haddar, National School of Engineers of Sfax (ENIS), Sfax, Tunisia Vitalii Ivanov, Department of Manufacturing Engineering, Machines and Tools, Sumy State University, Sumy, Ukraine Young W. Kwon, Department of Manufacturing Engineering and Aerospace Engineering, Graduate School of Engineering and Applied Science, Monterey, CA, USA Justyna Trojanowska, Poznan University of Technology, Poznan, Poland
Lecture Notes in Mechanical Engineering (LNME) publishes the latest developments in Mechanical Engineering—quickly, informally and with high quality. Original research reported in proceedings and post-proceedings represents the core of LNME. Volumes published in LNME embrace all aspects, subfields and new challenges of mechanical engineering. Topics in the series include: • • • • • • • • • • • • • • • • •
Engineering Design Machinery and Machine Elements Mechanical Structures and Stress Analysis Automotive Engineering Engine Technology Aerospace Technology and Astronautics Nanotechnology and Microengineering Control, Robotics, Mechatronics MEMS Theoretical and Applied Mechanics Dynamical Systems, Control Fluid Mechanics Engineering Thermodynamics, Heat and Mass Transfer Manufacturing Precision Engineering, Instrumentation, Measurement Materials Engineering Tribology and Surface Technology
To submit a proposal or request further information, please contact the Springer Editor of your location: China: Ms. Ella Zhang at [email protected] India: Priya Vyas at [email protected] Rest of Asia, Australia, New Zealand: Swati Meherishi at [email protected] All other countries: Dr. Leontina Di Cecco at [email protected] To submit a proposal for a monograph, please check our Springer Tracts in Mechanical Engineering at https://link.springer.com/bookseries/11693 or contact [email protected] Indexed by SCOPUS. All books published in the series are submitted for consideration in Web of Science.
More information about this series at https://link.springer.com/bookseries/11236
Anish Sachdeva · Pradeep Kumar · O. P. Yadav · Mohit Tyagi Editors
Recent Advances in Operations Management Applications Select Proceedings of CIMS 2020
Editors Anish Sachdeva Department of Industrial and Production Engineering Dr. B. R. Ambedkar National Institute of Technology Jalandhar, India O. P. Yadav Reliability and Maintainability Engineering North Dakota State University Fargo, ND, USA
Pradeep Kumar Department of Mechanical Engineering Indian Institute of Technology-Roorkee Uttarakhand, India Mohit Tyagi Department of Industrial and Production Engineering Dr. B. R. Ambedkar National Institute of Technology Jalandhar, India
ISSN 2195-4356 ISSN 2195-4364 (electronic) Lecture Notes in Mechanical Engineering ISBN 978-981-16-7058-9 ISBN 978-981-16-7059-6 (eBook) https://doi.org/10.1007/978-981-16-7059-6 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore
Preface
This special issue of book consists of selected papers from the International Conference on Industrial and Manufacturing Systems (CIMS-2020). It is a detailed exposition of latest emerging trends in operations management techniques that have become inevitable for the organisations survival in the present era of global competition. On one side, the expectations of the customers related to performance, quality, service, etc. are ever increasing and at the same time, there is a pressure to lower price tags on the organisations due to stringent competition. The organisations have to strive hard for their survival and growth. Consequently, the importance of applications of operations management has been continuously increasing for the organisations to stay relevant and sustainable. The papers in this volume have been included with a focus on detailed elucidations of contemporary developments in the field of operations management. Manuscripts on the topics covering a wide range of topics such as supply chain management, human factor engineering, multi-criteria decision making approaches, circular economy, quality function deployment, value stream mapping, data analytics, soft computing techniques, block chain, optimization of process parameters, simulation approaches in the area of operations management focusing on case studies and real applications have been published in this book. The International Conference on Industrial and Manufacturing Systems (CIMS2020), from which this special edition of the book on ‘lecture notes in Industrial Engineering’ has been derived, has been started by the Department of Industrial and Production Engineering, Dr. B. R. Ambedkar National Institute of Technology, Jalandhar, India. Around 370 research papers were presented in CIMS-2020 by the renowned academicians/researchers, noted industry representatives and the delegates on various aspects of latest issues related to industrial and manufacturing engineering. In order to explore the multi platforms in academia, second series of CIMS, i.e. 2nd International Conference on Industrial and Manufacturing Systems (CIMS2021) is going to be organised in collabration with PEC University of Technology, Chandigarh, India during October 2021. The editors would like to express our gratitude towards all the authors for contributing their valuable articles for our conference. We would like to acknowledge the reviewers for their painstaking and time consuming effort in reviewing v
vi
Preface
manuscripts and providing their thorough evaluations for improving the quality of the articles. We would also like to express our sincere gratitude towards Ms. Priya Vyas, Editor and other team members of Springer Book Series for allowing the editors to publish this special edition. Last but not the least, our heartfelt regards to our worthy Director Professor Lalit Kumar Awasthi for his wholehearted support for the conduct of the conference. Jalandhar, India Uttarakhand, India Fargo, USA Jalandhar, India
Anish Sachdeva Pradeep Kumar O. P. Yadav Mohit Tyagi
Contents
Risks Associated with Third-Party Logistics in Indian Restaurant Supply Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hritika Sharma, Saket Shanker, and Akhilesh Barve
1
Improvement of Climate-Smart Agriculture System Based on Obstacles Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shubhendu Singh, Mohit Tyagi, and Anish Sachdeva
11
Hurdle Appraisal for the Implementation of Circular Economy Notions in Supply Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ruchi Gupta, Mohit Tyagi, and R. K. Garg
21
Designing an Air Purifier by Using Green Quality Function Deployment Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kanika Prasad, Akshay Kumar, Jeet Kumar Yadav, Parwez Akhtar, and Raj Ballav Supply Chain Management Practice Constructs in SMEs: Development of Constructs and Its Implementation Issue . . . . . . . . . . . . . Rohit Kumar and Manish Gupta Analysis of QMS Practices Performed in ISO 9001 Certified Engineering Educational Institutes of India: An Interpretive Structural Modelling Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Parvesh Kumar, Sandeep Singhal, and Jimmy Kansal
37
49
61
A Review on Life Cycle Assessment of Various Dairy Products . . . . . . . . . Mukesh Kumar and Vikas Kumar Choubey
75
Design of Fuzzy Controller for Blood Glucose Level . . . . . . . . . . . . . . . . . . Vijay Kumar and Amit Kumar Singh
91
Prioritization of Renewable Energy Alternatives by Using Analytic Hierarchy Process (AHP) Model: A Case Study of India . . . . . . . . . . . . . . . 103 Sudhir Kumar Pathak, Vikram Sharma, and Sandesh S. Chougule vii
viii
Contents
Factors Influencing Customer Satisfaction and Belief Toward Car Business Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Sunil Anand and Piyush Singhal A Review on Musculoskeletal Disorders and Design of Ergonomics Aids with Relevance to Lower Back and Lumbopelvic Pain in Pregnant Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 Nikhil Yadav, M. L. Meena, G. S. Dangayach, and Yashvin Gupta Comparative Analysis of Supplier Selection Based on ARAS, COPRAS, and MOORA Methods Integrated with Fuzzy AHP in Supply Chain Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Josy George, Pushkal Badoniya, and J. Francis Xavier Comparative Efficiency Measurement of Indian Hospitals Using Data Envelopment Analysis: A Proposed Model . . . . . . . . . . . . . . . . . . . . . . 157 Dilip Kushwaha and Faisal Talib An Adoption of Blockchain Technology in Agri Food Supply Chain: An Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 Ashutosh Kumar, D. J. Ghode, and Rakesh Jain Prediction Model for Talent Management Analysis Through e-HRM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 Nasreen Nasar, Sumati Ray, Abdul Wadud, and Saiyed Umer Study of Biomedical Waste Management Performance Indicators in Indian States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Anurag Deepak, Astha Sharma, Dinesh Kumar, and Varun Sharma Determining Manpower Requirement for Material Handling in a Factory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Sreenija Kompally and V. Madhusudanan Pillai Data Analytics for NIRF Ranking of Indian Institutions to Check the Consistency and Validity of Ranking Framework . . . . . . . . . . . . . . . . . 243 Mamta Yadav, Arvind Bhardwaj, and Kapil Kumar Goyal Design Feature Assessment for Fused Deposition Modeling Using Supervised Machine Learning Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Rahul Bansal, Sukhdeep Singh Dhami, and Jatinder Madan Development of a 3D Printed Orthopaedic Cast for Wrist Fracture . . . . . 271 Mohd Ahad Islam, Mukul Shukla, and Yogesh Tripathi Effect of IoT in Supply Chain Management—A Review . . . . . . . . . . . . . . . 283 Gagandeep, Nikhilesh Singh, Abhishek Charak, Mohit Tyagi, and R. S. Walia
Contents
ix
Toolpath Generation of a Human Anatomical Shape for Double-Sided Incremental Forming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 Akshay Sahu, Prashant K. Jain, and Puneet Tandon Optimization of Cutting Forces in Dry Turning Process Using Taguchi and Grey Relational Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 Sumit Verma, Vipin Kakkar, and Hari Singh Performance Evaluation of Cargo Warehouse Operations of an Indian Airline Using Discrete Event Simulation: A Case Study . . . 335 A. Tamizhinian and V. Madhusudanan Pillai Blockchain: A Makeover to Supply Chain Management . . . . . . . . . . . . . . . 351 Justin Sunny, Kenil Shah, Prajwal Ghoradkar, Manu Jose, Malhar Shirswar, Hiran V. Nath, and V. Madhusudanan Pillai Assessment of the Challenges Obstructing Performance of Indian Food Supply Chain Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365 Janpriy Sharma, Mohit Tyagi, and Arvind Bhardwaj Farm Mechanization Through Custom Hiring Centres in Punjab: A Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 Manik Rakhra and Ramandeep Singh Hospitals’ Selection Under Ayushman Bharat Scheme with Heuristic Search Method Using A** Algorithm . . . . . . . . . . . . . . . . . . 387 Manimay Dev and Dinesh Kumar Role of Blockchain-Oriented Smart Contract in Supply Chain . . . . . . . . . 401 Abhishek Rajput, D. J. Ghode, and Rakesh Jain Propeller Design and Optimization for Drones . . . . . . . . . . . . . . . . . . . . . . . 411 Vyom Patel, Keval Nikam, Shantanu Dikshit, Manav Agarwalla, and Chinmay Zagade
About the Editors
Dr. Anish Sachdeva is a Professor in the Industrial and Production Engineering Department at Dr. B. R. Ambedkar National Institute of Technology Jalandhar, India. His areas of interest are Reliability and Maintenance Engineering, supply chain management, Optimization and Simulation of Production Systems, Quality Management. He obtained his B.Tech from Dr B. R. Ambedkar National Institute of Technology (erstwhile REC Jalandhar), M.Tech from Guru Nanak Dev Engineering College Ludhiana and Ph.D. from Indian Institute of Technology Roorkee (India). He has guided 65 M Tech and 14 PhD students. He has over 150 publications in international and national journals and proceedings of international conferences to his credit. He has organized seven international conferences at NIT Jalandhar as Organizing Secretary and more than 20 short term courses in his area of expertise. He has organized several workshops and training programs for academic institutes and companies. He has edited a number of journals of repute as guest editors as well as volumes of books reputed publishers. He also shares responsibilities of editorial board member in various international journals. Pradeep Kumar is working as a Professor in the Department of Mechanical & Industrial Engineering at Indian Institute of Technology, Roorkee, India. He obtained his B.E. (Industrial Engineering) in 1982; M.E. (Production Engineering) in 1989; and Ph.D. in Manufacturing and Production Engineering in 1994- all these degrees from University of Roorkee (Now, IIT Roorkee). He has been a visiting faculty at West Virginia University USA, Wayne State University USA, AIT Bangkok, and King Fahd University of Petroleum and Minerals, Saudi Arabia. He served the Delhi Technological University as Vice-Chancellor during 2014-15. He has published/presented more than 600 research papers in International and National Journals and proceedings of International and National Conferences. He has completed numerous consultancy projects of various organizations and sponsored research projects. He has 4 patent disclosures. His research interests include Advanced Manufacturing Processes; Microwave Joining of Metals, Metal Casting; Industrial Engineering; Supply Chain Management (SCM), Quality Engineering; and Production & Operations Management. xi
xii
About the Editors
O. P. Yadav is working as a Professor and Interim Department Chair, Industrial and Manufacturing Engineering, North Dakota State University, Fargo. He obtained his B.E. (Mechanical Engineering) from Malviya National Institute of Technology, Jaipur in 1986; M.Sc. (Industrial Engineering) from National Institute of Industrial Engineering, Bombay in 1992; and Ph.D. (Industrial and Manufacturing Engineering) from Wayne State University, Detroit (USA) in 2002. He has published more than 120 scientific papers in international journals and conferences of high repute and edited more than 15 books and proceedings. He has successfully completed 29 fully funded research & consultancy projects. His research interests include Quality and reliability engineering, production & Operations Management, supply chain (logistics), inventory modeling, lean manufacturing, quantitative Modeling, statistical analysis, fuzzy logic and neural networks. Dr. Mohit Tyagi is Assistant Professor in Department of Industrial and Production Engineering at Dr. B. R. Ambedkar National Institute of Technology Jalandhar, India. He has obtained his B. Tech (Mechanical Engineering) with HONS from UPTU Lucknow in 2008 and M. Tech. (Product Design and Development) with Gold Medal from MNNIT Allahabad in 2010. He did his Ph.D. from Indian Institute of Technology, Roorkee (India) in 2015. His areas of research are Industrial Engineering, Supply Chain Management, Corporate Social Responsibilities, Performance Measurement System, Data Science and Fuzzy Inference System. He has around 7 years of teaching and research experience. He has guided 18 PG dissertations and 15 UG projects. He is presently supervising 02 M. Tech. and 03 Ph.D. scholars. He has around 95 publications in international/national journals and proceedings of international conferences and book chapters in reputed publishers of his credit. Dr. Mohit Tyagi is reviewer of many international journal of repute like International Journal of Industrial Engineering: Theory, Application and Practices, Supply Chain Management: An International Journal, International Journal of Logistics System Management, and Journal of Manufacturing Technology Management, Information Systems, Grey Systems: Theory and Applications etc.
Risks Associated with Third-Party Logistics in Indian Restaurant Supply Chain Hritika Sharma, Saket Shanker, and Akhilesh Barve
Abstract At the present time, Indian restaurant supply chains are acquiring rapid economic growth in their business sector, with a higher productivity and expanded market. This dramatic rise in the Indian restaurant business has been encountered due to the outsourcing logistics to the third-party logistics (3PL) service provider. 3PL help the restaurant firm to focus on its core in-house activities properly, by outsourcing logistics activities such as warehousing, transportation, and delivery services. Although there are several advantages of 3PL in Indian restaurant chains, there are certain risks associated with it as well. The benefits obtained from 3PL services can be maximized by having a proper understanding of all the risks related with 3PL in Indian restaurant supply chain. In light of these facts, this study aims to emphasize on the risks associated with 3PL in Indian restaurant supply chain and to analyze the relationship among them using Interpretive Structural Modelling (ISM) technique. Keywords Third-party logistics (3PL) · Indian restaurant supply chain (IRSC) · Supply chain management (SCM) · Interpretive structural modelling (ISM)
1 Introduction Logistics outsourcing has gained immense popularity among the modern enterprises. Firms in all over the world are appointing and contracting with 3PL service providers, with the aim of increasing their profit. 3PL are responsible for outsourcing logistics operations, may be as a part or as a whole depending upon the type of contract [15]. Indian restaurant supply chains are no more untouched. Indian restaurant enterprises are continuously moving toward logistics outsourcing to a 3PL service provider. The logistics services outsourced include supply-transportation services, operational and maintenance tasks, and customer delivery operations. It consequently means that the restaurant chain will now be left only with the in-house tasks such as preparation H. Sharma · S. Shanker · A. Barve (B) Maulana Azad National Institute of Technology, Bhopal, MP, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_1
1
2
H. Sharma et al.
of food and will now be able to perform well as it has to concentrate only on its core preparation sector. Logistics activities comprises transportation, warehousing, and distribution services [11]. These furthermore incorporates receiving the raw materials from the supplier end, warehousing of goods and inventory management, transporting raw materials to the restaurant chain, and delivery of completed items from the restaurant to the end-customers. Having a command on a lot of services, any mistake or malfunctioning of the 3PL may lead toward a huge loss for the restaurant firm, or may even cause the complete restaurant chain to collapse. Thus, there are certain risks associated with the implementation of logistics outsourcing and the functioning of a 3PL in Indian restaurant supply chains [9]. The relative ease in handling the restaurant business acquired by the application of outsourcing of logistics activities to a third-party business attracts many-newly emerged restaurant chains in India. In the hope of expanding their business setup, many restaurant firms indulge in a contract with the 3PL without knowing the risks associated with them. Thus, it is very essential for a restaurant enterprise to acknowledge the various risks associated with the logistics outsourcing. The objective of this work is to throw light on different risks related with the 3PL in IRSC, and to assess them based on their interdependencies.
2 Literature Review Numerous studies have been done linking with the role of 3PL in various sectors. Goswami et al. [8] scrutinized the utilization of 3PL in the Indian automobile industry with the perspective of environmental sustainability. Similarly, Azzi et al. [2] studied the role of 3PL in medicinal sector by taking into account a case study of Italian-based healthcare network. In the same succession, Yadav et al. [18] inspected the selection of 3PL provider for an Internet of things based agricultural supply chain. Durst and Evangelista [5] investigated the knowledge management practices associated with 3PL provider. However, there exists fewer literature available on risks related with 3PL in IRSC. Henceforth, this research work particularly emphasizes on the various risks related with 3PL in Indian restaurant business.
3 Solution Methodology Interpretive Structural Modeling (ISM) is a technique involving the capability to form a hierarchical model comprising of the variables present in a problem, depending upon their interdependencies [16]. The objective of utilizing ISM technique is to figure out the relationship between existing variables in a problem set, and on the basis of these inter-relationships, form a hierarchical ISM model.
Risks Associated with Third-Party Logistics in Indian …
3
4 Risks Related with 3PL in Indian Restaurant Supply Chain The various risks are described in Table 1.
5 Application to Proposed Framework The proposed ISM technique is utilized in obtaining a relationship among the various risks related to 3PL in Indian restaurant supply chain. Table 1 Risks related with 3PL in IRSC Serial number Risk
Description
Reference
α1
Inflated partnership cost
Third-parties are appointed by restaurant enterprises in order to cut down the costs drastically. On the contrary, restaurant enterprise ends up paying even more than anticipation due to exit costs, surplus and lapse charges present in the contract. In case the 3PL is unable to manage assigned logistics operations effectively, restaurant enterprises have to bear the loss
[3]
α2
Crucial information spilling
Indian restaurant enterprises have crucial information regarding their enterprise strategy, customer details, and future plans. Sharing significant information regarding the restaurant enterprise with 3PL escalates the probability of its spilling and misuse
[10]
α3
Subservience to single 3PL provider
Indian restaurant enterprises [14] generally partner with a single 3PL in order to accomplish their logistics operations. Henceforth, if any economic harm occurs to the 3PL provider, the logistics operations which were meant to be performed by the third-party business come to a halt, leading to huge loss of the corresponding restaurant firm (continued)
4
H. Sharma et al.
Table 1 (continued) Serial number Risk
Description
Reference
α4
Partnering with an incompetent An incompetent 3PL provider [17] 3PL provider lacks the potential understanding of supply chain strategies made to maximize the supply chain surplus. Subsequently, the partnered 3PL will fail to match with the competitive strategy of restaurant firm, leading to a drastic reduction in the total supply chain surplus
α5
Dearth of governance of logistics operations
Hiring a third-party for [4] managing logistics operations eventually lead to decreased access and overseeing associated with logistics operations for the restaurant enterprise. Subsequently, the restaurant firm loses significant control over its customers, warehouses, and transportation operations
α6
Reduced customer interaction
The assigned operations to [6] third-party logistics include customer delivery. Henceforth, the corresponding restaurant firm loses direct interaction with their customers which gives rise to a communication gap
α7
Improper operations handling by 3PL
3PL may mismanage to carry out [13] the warehousing and transportation operations effectively, leading to escalated costs associated with logistics operations and decreased responsiveness of the entire restaurant supply chain
α8
Unfulfillment of anticipation
3PL provider deals with many [7] enterprises at a time in order to provide its service. Subsequently, chances are high in failing to accomplish any real time demand from a particular partnered enterprise. Henceforth, the restaurant enterprise has to compromise on its anticipation of full-time availability of partnered 3PL (continued)
Risks Associated with Third-Party Logistics in Indian …
5
Table 1 (continued) Serial number Risk
Description
α9
Inappropriate inventory management
3PL may maintain inventory at [12] an excessive or insufficient manner, leading to declination in the overall efficiency and responsiveness of the restaurant supply chain
Reference
α10
Lack of interpersonal skills in employees
If the partnered 3PL has [1] employees lacking in interpersonal skills and basic etiquettes such as politeness, trustworthiness, it will become cumbersome for the 3PL to maintain good relations with the customers and restaurant firm workers. Eventually, restaurant business will have to face the drawbacks, including unsatisfied customers and disappointed staff
5.1 Structural Self-Interaction Matrix ISM technique involves input from experts in obtaining the inter-relationships among the variables, which are as follows: V: Risk i will help achieve risk j; A: Risk j will help achieve risk i; X: Risk i and j will help achieve each other; O: Risk i and j are unrelated. The SSIM for the risks is provided in Table 2.
5.2 Reachability Matrix The SSIM is transformed into an initial reachability matrix by following the already known procedure of ISM. The final reachability matrix (Table 3) is obtained utilizing the transitivity rule, according to which “if a variable ‘A’ is related to ‘B’ and ‘B’ is related to ‘C’, then ‘A’ has to be related to ‘C’”.
6
H. Sharma et al.
Table 2 Structural self-interaction matrix α10
α9
α8
α7
α6
α5
α4
α3
α2
α1
α1
X
A
X
A
A
A
A
A
X
–
α2
X
A
X
A
A
A
A
A
–
α3
V
A
V
A
A
A
O
–
α4
V
A
V
A
A
A
–
α5
V
V
V
V
V
–
α6
V
O
V
O
–
α7
V
V
V
–
α8
X
A
–
α9
V
–
α10
–
Table 3 Final reachability matrix for risks α10
α9
α8
α7
α6
α5
α4
α3
α2
α1
α1
1
0
1
0
0
0
0
0
1
1
4
α2
1
0
1
0
0
0
0
0
1
1
4
α3
1
0
1
0
0
0
0
1
1
1
5
α4
1
0
1
0
0
0
1
0
1
1
5
α5
1
1
1
1
1
1
1
1
1
1
10
α6
1
0
1
0
1
0
1
1
1
1
7
α7
1
1
1
1
1
1
1
1
1
1
10
α8
1
0
1
0
0
0
0
0
1
1
4
α9
1
1
1
0
0
0
1
1
1
1
7 4
α10 Dependence power
1
0
1
0
0
0
0
0
1
1
10
3
10
2
3
2
5
5
10
10
Driver power
5.3 Level partitions Reachability set corresponds to the risks a particular risk may facilitate in alleviating, including itself. Antecedent set corresponds to the risks which may facilitate in alleviating it, including itself. These sets are calculated for each risk and their intersection is obtained. The risk possessing same reachability and intersection set is given level I, and the next iteration is performed excluding the risks who have already given their level. The process continues until the level of each risk is known. Final iteration has been depicted in Table 4.
Risks Associated with Third-Party Logistics in Indian … Table 4 Levels of risks
7
Serial number
Level obtained
α1, α2, α8, α10
I
α3, α4
II
α6, α9
III
α5, α7
IV
Fig. 1 ISM hierarchy model for obtained risks
5.4 ISM Model Development The level partition table provides levels of each risk. Utilizing the data from level partition tables, the ISM hierarchy model is developed, depicting the interdependencies of all the risks (Fig. 1).
5.5 MICMAC Analysis According to their dependence and driver power, the variables are divided into four clusters that are autonomous, dependent, linkage, and independent elements. Autonomous risks (Cluster I) include the risks with low driver power and high dependence power. Dependent risks (Cluster II) constitute risks with low driver power but high dependence power. Linkage risks (Cluster III) involve risks with high driving as well as dependence power. Independent risks consist of risks with weak driving
8
H. Sharma et al.
Fig. 2 MICMAC analysis
power but low dependence power. Figure 2 reveals that risks α3, α4 are autonomous, α1, α2, α8, and α10 are dependent, and α5, α6, α7, α9 are independent risks.
6 Result and Discussion Study revealed that the prime risks responsible for unsuccessful third-party and enterprise partnerships are a dearth of governance of logistics operations and improper operations handling by 3PL. These prime risks give rise to other factors such as increased costs and decreased responsiveness, which in turn decreases the overall supply chain surplus in the restaurant sector. Henceforth, sincere efforts are required to mitigate with these prime risks in order to enhance the probability of a successful partnership.
7 Conclusion Logistics outsourcing is now becoming quite prevalent among Indian restaurant enterprises. Although there are numerous benefits of outsourcing logistics to a 3PL, there are certain risks associated with them as well, as already discussed in the literature review. The benefits of a 3PL can be maximized if the risks associated with them are mitigated. Thus, acknowledgement of the associated risks is necessary to find their remedial measures, which has been portrayed through this research paper.
Risks Associated with Third-Party Logistics in Indian …
9
Various risks have been found and briefed, which is a major step in their mitigation for future. Risks free 3PL-restaurant organizations are the future of Indian restaurant business. Thus, this study is recognized as a step toward the achievement of maximum benefit for Indian restaurant supply chain network.
References 1. Asian S, Pool JK, Nazarpour A, Tabaeeian RA (2019) On the importance of service performance and customer satisfaction in third-party logistics selection: an application of Kano model. Benchmarking Int J 26(5):1550–1564. https://doi.org/10.1108/BIJ-05-2018-0121 2. Azzi A, Persona A, Sgarbossa F, Bonin M (2013) Drug inventory management and distribution: outsourcing logistics to third-party providers. Strateg Outsourcing Int J 6(1):48–64. https://doi. org/10.1108/17538291311316063 3. Berglund M, van Laarhoven P, Sharman G, Wandel S (1999) Third-party logistics: is there a future? Int J Logist Manag 10(1):59–70. https://doi.org/10.1108/09574099910805932 4. Chen Z-S, Zhang X, Govindan K, Wang X-J, Chin K-S (2021) Third-party reverse logistics provider selection: a computational semantic analysis-based multi-perspective multi-attribute decision-making approach. Exp Syst Appl 166:114051. https://doi.org/10.1016/j.eswa.2020. 114051 5. Durst S, Evangelista P (2018) Exploring knowledge management practices in third-party logistics service providers. VINE J Inf Knowl Manag Syst 48(2):162–177. https://doi.org/10.1108/ VJIKMS-05-2016-0030 6. Ekeskär A, Rudberg M (2020) Third-party logistics in construction: perspectives from suppliers and transport service providers. Prod Plan Control 1–16. https://doi.org/10.1080/09537287. 2020.1837932 7. Evangelista P, Santoro L, Thomas A (2018) Environmental sustainability in third-party logistics service providers: a systematic literature review from 2000–2016. Sustainability 10(5):1627. https://doi.org/10.3390/su10051627 8. Goswami M, De A, Habibi MKK, Daultani Y (2020) Examining freight performance of thirdparty logistics providers within the automotive industry in India: an environmental sustainability perspective. Int J Prod Res 58(24):7565–7592. https://doi.org/10.1080/00207543.2020. 1756504 9. Liu H-T, Wang W-K (2009) An integrated fuzzy approach for provider evaluation and selection in third-party logistics. Expert Syst Appl 36(3):4387–4398. https://doi.org/10.1016/j.eswa. 2008.05.030 10. Marasco A (2008) Third-party logistics: a literature review. Int J Prod Econ 113(1):127–147. https://doi.org/10.1016/j.ijpe.2007.05.017 11. Marchesini MMP, Alcântara RLC (2016) Logistics activities in supply chain business process: a conceptual framework to guide their implementation. Int J Logist Manag 27(1):6–30. https:// doi.org/10.1108/IJLM-04-2014-0068 12. Pamucar D, Chatterjee K, Zavadskas EK (2019) Assessment of third-party logistics provider using multi-criteria decision-making approach based on interval rough numbers. Comput Ind Eng 127:383–407. https://doi.org/10.1016/j.cie.2018.10.023 13. Reza S, Mubarik MS, Naghavi N, Rub Nawaz R (2020) Relationship marketing and third-party logistics: evidence from hotel industry. J Hosp Tourism Insights 3(3):371–393. https://doi.org/ 10.1108/JHTI-07-2019-0095 14. Selviaridis K, Spring M (2007) Third party logistics: a literature review and research agenda. Int J Logist Manag 18(1):125–150. https://doi.org/10.1108/09574090710748207 15. Shanker S, Sharma H, Barve A (2021) Assessment of risks associated with third-party logistics in restaurant supply chain. Benchmarking Int J (ahead-of-print). https://doi.org/10.1108/BIJ06-2020-0343
10
H. Sharma et al.
16. Thakkar J, Deshmukh SG, Gupta AD, Shankar R (2005) Selection of third-party logistics (3PL): a hybrid approach using Interpretive Structural Modeling (ISM) and Analytic Network Process (ANP). Supply Chain Forum Int J 6(1):32–46. https://doi.org/10.1080/16258312.2005. 11517137 17. Wahab SN, Bahar N, Radzi NAM (2021) An inquiry on knowledge management in thirdparty logistics companies. Int J Bus Innov Res 24(1):124. https://doi.org/10.1504/IJBIR.2021. 111977 18. Yadav S, Garg D, Luthra S (2020) Selection of third-party logistics services for internet of things-based agriculture supply chain management. Int J Logist Syst Manag 35(2):204. https:// doi.org/10.1504/IJLSM.2020.104780
Improvement of Climate-Smart Agriculture System Based on Obstacles Assessment Shubhendu Singh, Mohit Tyagi, and Anish Sachdeva
Abstract Ecology is the basis of existence on this planet therefore it becomes imperative to maintain the ecological balance. Anthropogenic activities, especially in the agriculture sector are a major cause of climate change. Around the world, the largest user of land is a farmer and therefore it becomes decisive to implement practices that are environmentally and ecologically fruitful. In this context, Climate-Smart Practises (CSA) are promoted to accomplish these objectives. However, implementation of these practises precipitates certain socio-economic challenges in the supply and demand which acts as a hurdle to its enactment. This paper examines such barriers faced in the implementation of Climate-Smart Practises in the Agriculture Sector. Eight critical barriers have been identified which are then segregated into cause and effect groups by using DEMATEL (Decision-Making Trial and Evaluation Laboratory) method. The outcome of this study will help policymakers and proponents of Climate-Smart Agriculture interventions. By working to ameliorate the effects of barriers, efficient implementation CSA practises can be ensured.
1 Introduction Agriculture covers the largest area of land in the world and consumes the largest percentage of water therefore it becomes imperative to transform the agricultural sector and effectively utilize these valuable resources. It faces extreme pressure to feed the growing population especially due to uncontrolled urbanization and climate change [1]. With the increasing cost of farms inputs, shortage of labor, the uncertain pattern of rainfall and extreme climatic events the dismay for the farmers are incessant. Climate Change possesses the greatest threat to the agriculture sector in the present time [2]. It has caused extensive damage to the Indian Subcontinent [3]. Climate Change has a severe impact on the productivity of the soil as it affects the average rainfall and temperature of that region [2]. It is important to take adaptive S. Singh · M. Tyagi (B) · A. Sachdeva Department of Industrial and Production Engineering, Dr. B. R. Ambedkar National Institute of Technology, Jalandhar, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_2
11
12
S. Singh et al.
measures against climate change to ensure food safety and livelihoods of poor [4]. Emission of Green House Gases (GHGs) are the primary reason for climate change and since India is the third-largest emitter of GHGs, in which agriculture sector alone accounts for 18% of the total national emission, therefore adoption of climate-smart practises has the potential to mitigate Climate Change on a large scale [4]. Climate-Smart Agriculture (CSA) practises have been credited for improving the agricultural production system which is resilient to risks due to Climate Change [5]. Although many climate-smart innovations have been developed their prioritization is required [1]. Interventions such as crop insurance, rainwater harvesting, and weather-based agriculture advisories are some of the most preferred technologies used by the farmers [6]. Climate-smart practises which includes residue management, crop diversification, and zero tillage have been appreciated for improving the overall productivity of the rice–wheat system [7]. Many small and medium business, especially in the rural areas are generating additional income by providing services such as zero-till multi-crop planters, Land Laser Levelers and Happy Seeder to their customers [8]. However, adoption of these climate-smart interventions has been slow due to various hurdles. Identification of such hurdles is urgently required [1]. This paper aims to identify those socio-economic barriers that are inhibiting the implementation of CSA innovations. Eight such barriers have been identified by going through paper related to CSA and by a discussion with the farmer in the Bundelkhand Region of Uttar Pradesh. It is seen that there are hurdles on both supply as well as demand sides that are restricting the implementation [9]. This research divides these hurdles into cause and effect by using DEMATEL (Decision-Making Trial and Evaluation Laboratory) method and illustrates the key barrier that needs to be overcome. Further, various measures which have been drawn out by consultation with the farmers to overcome these barriers have been discussed.
2 Literature Review Agriculture is the source of income for most of the world’s poor. When the productivity of the soil decreases due to various reasons primarily one being climate change, emigration at large scale is seen especially from a poor country [2]. Emission of Green Houses Gases (GHS) is one of the major reasons for climate change and improvement in agriculture sector practises has a lot of potentials to reduce GHGs emission [4]. It has been confirmed by trials that by incorporating CSA practises along with appropriate use of few conventional practises increased the yield of the crop by 9%, overall input cost was reduced which thereby enhanced the profitability by 25% and there was reduction of usage of irrigation water by 28% which improved water productivity by 37% compared to traditional practises [8]. A Study by Notenbaert et al. [1] conducted in East Africa shows that by improving the quality of feed for farm animals, taking care of their health and by managing grassland emissions can be
Improvement of Climate-Smart Agriculture System Based on Obstacles Assessment
13
reduced by 5–24%. Among them improving diet quality had the highest ameliorating effect on the emission reduction. Although the Climate-Smart Agriculture Practises are promoted for reducing the risk in the agriculture production system induced by climatic actions [10], they fail to achieve the expected target. Low level of adoption and problems in the scaling of CSA interventions are the two major reasons of failure for the wide acceptance of CSA technologies. In addition to this, niggardly subsidies by the government hamper the prospects of small and medium enterprises to scale up the adoption [8]. While the detrimental effects of extreme climatic events are a predominant barrier in the growth of farmer, other factors such as market accessibility and few government incentives also act as a barrier [11]. Apart from this, the cost of implementation of technology is a major deciding factor for farmers for the transition to any climatesmart adaptations [6]. Since farm sizes are continuously decreasing due to the rapidly increasing population which restricts the income generation for the farmers thereby preventing them to adopt new technological innovations. This acts as a barrier for small and medium enterprises to invest in the agriculture sector [8].
3 Research Methodology The objective of this research is to evaluate critical hurdles that are hampering the efficient implementation of climate-smart Practices. To achieve the objective, a set of key barriers have been identified by interaction with the farmers from the Bundelkhand Region of Uttar Pradesh, field experts and by going through relevant research studies in this area. Table 1 describes each hurdle. After that survey was conducted with the help of a questionnaire which was based on linguistic terms and sent to 58 experts to examine the criticality of each hurdle. Once the response of the experts has been recorded, they are then analyzed using Fuzzy DEMATEL approach.
3.1 Fuzzy DEMATEL Approach Multi-Criterion Decision-Making (MCDM) is generally used by researchers to evaluate an issue which involves many criteria [17]. Fuzzy DEMATEL is one of the widely used decision-making tools developed by Geneva in 1973 to unravel complex problems [18]. It helps in obtaining the cause and effect relationship among various criteria and then divide them into cause and effect components. Since human judgements made by using crisp values can be indistinct and fickle therefore to remove such uncertainties fuzzy concept is used with DEMATEL. Algorithm of Fuzzy DEMATEL involves five steps given below.
14
S. Singh et al.
Table 1 Description of various hurdles Hurdles
Representation
Description
Financial services
(H1)
Lack of facilities for loan and insurance impede the transition process to climate-smart ways [8]
Management practices
(H2)
Insufficient technical knowledge about climate-smart interventions restrict framer’s ability to make a transition [12]
Infrastructure
(H3)
Lack of facilities such as cold storage results in loss of food grains forcing the farmer to sell their crop at low margins [13]
Awareness programs
(H4)
Low awareness about climate-smart practises results is low adoption rates [14]
Market accessibility
(H5)
More travel time to cities hamper the adaptation of new technological innovations by the farmer [1]
Land holdings
(H6)
Declining farm sizes reduce farm income and impede their ability to invest in CSA practises [8]
Govt. schemes and incentives
(H7)
Dearth of effective policies and planning for their enactment seriously restrict the spread of Climate-smart activities [15]
Availability of labor
(H8)
Shortage of labor resources in the agriculture sector impacts crop growth, forces excessive use of fertilizers and even pressing farmers to shift to non-farming activities [14, 16]
Step1. Formulation of average direct-relation matrix ‘D’ by pairwise comparison of hurdles is performed. It is formed by taking the average of expert’s opinion which is shown in Table 2. Step 2. Since the opinion of the experts are recorded in terms of linguistic expression and may involve vagueness therefore the fuzzy set theory is used to eliminate the Table 2 Linguistic scale direct-relation matrix H1
H2
H3
H4
H5
H6
H7
H8
H1
0
3
4
2
0
3
4
0
H2
1
0
1
1
0
3
2
2
H3
1
2
0
1
1
0
1
0
H4
1
3
0
0
1
2
1
4
H5
1
1
3
2
0
1
0
3
H6
3
2
1
1
0
0
2
3
H7
4
2
1
1
0
2
0
3
H8
2
1
1
1
1
2
2
0
Improvement of Climate-Smart Agriculture System Based on Obstacles Assessment Table 3 Fuzzy linguistic scale [21]
Linguistic variable
15
Influence score Triangular fuzzy number
No influence
0
(0, 0.1, 0.3)
Very low influence
1
(0.1, 0.3, 0.5)
Low influence
2
(0.3, 0.5, 0.7)
High influence
3
(0.5, 0.7, 0.9)
Very high influence 4
(0.7, 0.9, 1.0)
human vacillation. In fuzzy set theory, a response is recorded in terms of linguistic terms which are then converted into fuzzy numbers [19]. Generally, a triangular fuzzy number is preferred to represent the response received [20]. In this study linguistic expressions that are converted into triangular fuzzy number are based on five-point fuzzy linguistic scale as shown in Table 3. Step 3. Triangular fuzzy number is then converted into crisp values by using the Best Non-Fuzzy Performance (BNP) defuzzification method. BNP for a triangular fuzzy number (a, b, c) can be obtained by applying the formula BN P = l +
(u − l) + (m − l) (1) 3
After the defuzzification of linguistic terms, a matrix is obtained which has been shown in Table 4. Step 4. After the defuzzification, the matrix is normalized so that rows and columns can be standardized to improve the accuracy of the calculation. The normalized matrix has been shown in Table 5. Step 5. The Total Relation Matrix (T) is obtained from the Normalized DirectRelation Matrix by using the equation: T = N (I − N )−1 (2) Table 4 Defuzzified direct relation matrix H1
H2
H3
H4
H5
H6
H7
H8
H1
0.133
0.700
0.866
0.500
0.133
0.700
0.866
0.133
H2
0.300
0.133
0.300
0.300
0.133
0.700
0.500
0.500
H3
0.300
0.500
0.133
0.300
0.300
0.133
0.300
0.133
H4
0.300
0.700
0.133
0.133
0.300
0.500
0.300
0.866
H5
0.300
0.300
0.700
0.500
0.133
0.300
0.133
0.700
H6
0.700
0.500
0.300
0.300
0.133
0.133
0.500
0.700
H7
0.866
0.500
0.300
0.300
0.133
0.500
0.133
0.700
H8
0.500
0.300
0.300
0.300
0.300
0.500
0.500
0.700
16
S. Singh et al.
Table 5 Normalized direct-relation matrix H1
H2
H3
H4
H5
H6
H7
H8
H1
0.0330
0.1735
0.2148
0.1239
0.0330
0.1735
0.2148
0.0330
H2
0.0743
0.0330
0.0743
0.0743
0.0330
0.1735
0.1239
0.1239
H3
0.0743
0.1239
0.0330
0.0743
0.0743
0.0330
0.0743
0.0330
H4
0.0743
0.1735
0.0330
0.0330
0.0743
0.1239
0.0743
0.2148
H5
0.0743
0.0743
0.1735
0.1239
0.0330
0.0743
0.0330
0.1735
H6
0.1735
0.1239
0.0743
0.0743
0.0330
0.0330
0.1239
0.1735
H7
0.2148
0.1239
0.0743
0.0743
0.0330
0.1239
0.0330
0.1735
H8
0.1239
0.0743
0.0743
0.0743
0.0743
0.1239
0.1239
0.0330
where ‘I’ is the Identity Matrix. Total Relation Matrix is shown in Table 6. Step 6. The sum of the rows and columns of the Total Relation Matrix ‘T’ is represented by ‘R’ and ‘C’ respectively in Table 7 by using the formula given below: T = ti j n×n i, j = 1, 2, 3 . . . , n
(3)
Table 6 Total relational matrix of the study H1
H2
H3
H4
H5
H6
H7
H8
H1 0.514301 0.656501 0.587258 0.461931 0.240589 0.640743 0.660396 0.535387 H2 0.433742 0.395311 0.361705 0.327957 0.18681
0.519244 0.463968 0.488964
H3 0.323413 0.381926 0.248524 0.261549 0.184844 0.297212 0.320504 0.304878 H4 0.463311 0.556184 0.356637 0.317157 0.243974 0.519648 0.456313 0.611002 H5 0.417148 0.436358 0.455971 0.37761
0.195486 0.425032 0.375823 0.528025
H6 0.567122 0.533935 0.413151 0.368284 0.209392 0.449847 0.521371 0.575199 H7 0.624482 0.558803 0.435396 0.385782 0.218505 0.557451 0.463053 0.595479 H8 0.472881 0.437321 0.37045
0.33161
0.22502
0.474965 0.463434 0.401597
Table 7 Calculation results of the fuzzy DEMATEL Hurdle (H)
R
C
R+C
H1
Financial services
4.297105
3.8164
8.113505
R−C
H2
Management practices
3.177703
3.95634
7.134043
−0.77864
H3
Infrastructure
2.32285
3.229093
5.551943
−0.90624
H4
Awareness programs
3.524227
2.831881
6.356108
H5
Market accessibility
3.211452
1.70462
4.916072
H6
Land holdings
3.638302
3.884142
7.522444
H7
Govt. schemes and incentives
3.83895
3.724862
7.563812
H8
Availability of labor
3.177278
4.040531
7.217809
0.480705
0.692346 1.506832 −0.24584 0.114088 −0.86325
R-C
Improvement of Climate-Smart Agriculture System Based on Obstacles Assessment
17
2 H5
1.5 1
H4 H1
0.5 0
H7 0
1
2
3
4
5
6
-0.5 H3 -1
H6
7
R+C 8
9
H2 H8
-1.5
Fig. 1 Cause and effect diagram
⎡ R=⎣
n
⎤ ti j ⎦
j=1
C=
n I =1
= [ti ]n×1 (4) n×1
ti j
= t j n×1 (5)
n×1
Step 7. Casual Diagram is formed by using the values of ‘R’ and ‘C’ where (R + C) forms the horizontal axis while (R − C) forms the vertical axis of the casual diagram. As per the value of (R + C) importance of that barrier is decided. If the value of (R + C) for a particular barrier is high then that barrier is important and vice versa. Apart from that, if the value of (R − C) for a hurdle comes positive then it is grouped into cause cluster while if it comes out to be negative it is grouped into effect group. Cause and effect diagram has been shown in Fig. 1
4 Results and Discussion Order of preference for resolving hurdles is decided based upon the value of (R + C) and (R − C) as shown in Table 7. Based on this study importance rating of all the hurdles is shown in Fig. 2 which is as follows H1 > H7 > H6 > H8 > H2 > H4 > H3 > H5. Access to financial services (H1) is a critical barrier for farmers to adapt to new interventions. The government should ensure that farmers have swift access to loan facilities with low or zero interest rate. Apart from that financial assistance
18
S. Singh et al.
R+C H8
7.21780869
H7
7.563812305
H6
7.52244437
H5
4.91607229
H4
6.356107704
H3
5.551943409
H2
7.134042995
H1
8.1135 0
1
2
3
4
5
6
7
8
9
Fig. 2 Importance rating of enablers
in the form of subsidies on fertilizers and seeds can also help farmers minimize initial incurring cost. To protect the farmer from unforeseen risk, crop and livestock insurance should be promoted by the government as well as private companies. It should also be included as one of the essential requirements to avail benefits of several government schemes. Not only will this reduce insolvency among farmers but also improve their living standard. Second most predominant hurdle is dearth of effective government schemes and incentives (H7). There are very few effective government schemes which have had an extensive impact on the lives of people living in rural areas, resulting in a feeling of despair among people. The government must resurrect the dejection among people by forming schemes and providing incentives that bring a positive change in their life. Considering farmer’s reluctance to accept new technology is natural since the period of transition from conventional to smart agricultural ways involves huge initial losses. The government should compensate it by starting various schemes which would facilitate a smoother transition. This should be looked as an investment rather than a subsidy as it would help the farmer and environment in the long run. Funds should be allocated for training and giving exposure to farmers for a smooth transition to climate-smart ways. Various steps such as providing certification for transition can help in boosting the morale of the farmers should be encouraged. Barriers which have a positive value of (R − C) fall under the cause group while those having a negative value are grouped under the effect group. Causal group hurdles have more impact when resolved as compared to effect group. According to this study, hurdles grouped under the causal group are Financial Services (H1), Awareness Programs (H4), Market inaccessibility (H5), and Government schemes and incentives (H7) while barriers such as Management practices (H2), Infrastructure (H3), Insignificant land holdings (H6), and Availability of Labor (H8) falls under effect group. Among the causal group, market inaccessibility (H5) has the highest value of (R − C) which means that it is the primary causal criterion. Market inaccessibility leaves the farmers at the mercy of the traders and middlemen thereby which hampers the growth of that particular region, therefore, the formation of Farmer Producer’s Organizations (FPO) should be encouraged which would help farmers get the right
Improvement of Climate-Smart Agriculture System Based on Obstacles Assessment
19
price for their crop. FPO’s can help in marketing the crops which would help farmers reach new markets and consumer. Apart from that FPO can also sell the produce under a local name in the nearby local market directly to their customers. Other ways include tie-up by FPO with Argo Processing industries can help farmers get guaranteed buyer and price. Involvement of all stakeholder from producer to consumer is very crucial in decision-making more so in the case of the agricultural system in which links with the market are highly complex and cross-disciplinary. FPOs can play a role in integrating all stakeholders at the initial stages to have access to pertinent information which is generally distributed among stakeholders. Apart from that FPO can invest in local warehouses and cold storage thereby reducing the losses.
5 Conclusion In order to help policymakers with their decision-making process and to reduce the impact of climate change on agriculture sector by effective implementation of various climate-smart interventions, this study identifies eight major hurdles currently faced for its enactment. To analyze the interrelationship between them and minimize fuzziness because of the subjective nature of human evaluation, Fuzzy DEMATEL approach has been used to identify key hurdles. The research findings of this study justify the fact that there are obstacles on both supply as well as the demand side, therefore, the conventional method of only supply-based smart innovations will not yield the desired result. There is a need to bring innovative restructuring into the demand side as well. Inaccessibility to the financial services and market are the major curbing factor that is shackling the progress of Climate-Smart Agriculture. They also come under the causal group, therefore, need to be checked first. The outcome of this study will help decision-makers and people working on the field to enact climate-smart interventions techniques.
References 1. Notenbaert A, Pfeifer C, Silvestri S, Herrero M (2017) Targeting, out-scaling and prioritising climate-smart interventions in agricultural systems: lessons from applying a generic framework to the livestock sector in sub-Saharan Africa. Agric Syst 151:153–162. https://doi.org/10.1016/ j.agsy.2016.05.017 2. Falco C, Galeotti M, Olper A (2018) Climate change and migration: is agriculture the main channel? IEFE Working Papers, 59:101995. https://doi.org/10.1016/j.gloenvcha.2019.101995 3. Srinivasarao C, Rao KV, Gopinath KA, Prasad YG, Arunachalam A, Ramana DBV, … Mohapatra T (2019) Agriculture contingency plans for managing weather aberrations and extreme climatic events: development, implementation and impacts in India. In Advances in Agronomy (1st ed.). https://doi.org/10.1016/bs.agron.2019.08.002 4. Sapkota TB, Vetter SH, Jat ML, Sirohi S, Shirsath PB, Singh R, … Stirling CM (2019) Costeffective opportunities for climate change mitigation in Indian agriculture. Sci Tot Environ 655:1342–1354.https://doi.org/10.1016/j.scitotenv.2018.11.225
20
S. Singh et al.
5. Tong Q, Swallow B, Zhang L, Zhang J (2019) The roles of risk aversion and climate-smart agriculture in climate risk management: evidence from rice production in the Jianghan Plain, China. Clim Risk Manag 100199. https://doi.org/10.1016/j.crm.2019.100199 6. Khatri-Chhetri A, Aggarwal PK, Joshi PK, Vyas S (2017) Farmers’ prioritization of climatesmart agriculture (CSA) technologies. Agric Syst 151:184–191. https://doi.org/10.1016/j.agsy. 2016.10.005 7. Jat HS, Datta A, Choudhary M, Sharma PC, Yadav AK, Choudhary V, … McDonald A (2019). Climate Smart Agriculture practices improve soil organic carbon pools, biological properties and crop productivity in cereal-based systems of North-West India. Catena 181(May):104059. https://doi.org/10.1016/j.catena.2019.05.005 8. Groot AE, Bolt JS, Jat HS, Jat ML, Kumar M, Agarwal T, Blok V (2019) Business models of SMEs as a mechanism for scaling climate smart technologies: the case of Punjab, India. J Clean Prod 210:1109–1119. https://doi.org/10.1016/j.jclepro.2018.11.054 9. Levidow L, Zaccaria D, Maia R, Vivas E, Todorovic M, Scardigno A (2014) Improving water-efficient irrigation: prospects and difficulties of innovative practices. Agric Water Manag 146:84–94. https://doi.org/10.1016/j.agwat.2014.07.012 10. Olayide OE, Tetteh IK, Popoola L (2016) Differential impacts of rainfall and irrigation on agricultural production in Nigeria: any lessons for climate-smart agriculture? Agric Water Manag 178:30–36. https://doi.org/10.1016/j.agwat.2016.08.034 11. Taylor M, Bhasme S (2020) Geoforum Between deficit rains and surplus populations: the political ecology of a climate-resilient village in South India. Geoforum 0035(January):1–10. https://doi.org/10.1016/j.geoforum.2020.01.007 12. Peterson CA (2014) Local-level appraisal of benefits and barriers affecting adoption of climatesmart agricultural practices: Curití, Colombia. CCAFS-Technical Report, p 36 13. Khapayi M, Celliers PR (2016) Factors limiting and preventing emerging farmers to progress to commercial agricultural farming in the King William’s Town area of the Eastern Cape Province, South Africa. S Afr J Agric Extension, 44(1):25–41 14. Khatri-Chhetri A, Pant A, Aggarwal PK, Vasireddy VV, Yadav A (2019) Stakeholders prioritization of climate-smart agriculture interventions: Evaluation of a framework. Agric Syst 174(March):23–31. https://doi.org/10.1016/j.agsy.2019.03.002 15. Rohila AK, Shehrawat P, Malik J (2018) Awareness, constraints and prospects of climate smart agricultural practices (CSAP). J Agrometeorol 20:167–171. https://doi.org/10.17501/ icoaf.2017.3104 16. Gunabhagya A, Joshi AT, Patil SS, Maraddi GN (2017) Agricultural labour shortage: an abysmal to agriculture in North Eastern Karnataka. Econ Aff 62(4):589. https://doi.org/10. 5958/0976-4666.2017.00071.7 17. Muhammad MN, Cavus N (2017) Fuzzy DEMATEL method for identifying LMS evaluation criteria. Proc Comput Sci 120:742–749. https://doi.org/10.1016/j.procs.2017.11.304 18. Shieh J-I, Wu H-H, Huang K-K (2010) A DEMATEL method in identifying key success factors of hospital service quality. Knowl Based Syst 23(3):277–282 19. Tyagi M, Kumar P, Kumar D (2018) Assessment of CSR based supply chain performance system using an integrated fuzzy AHP-TOPSIS approach. Int J Log Res Appl 21(4):378–406. https://doi.org/10.1080/13675567.2017.1422707 20. Chang YH, Yeh CH, Wang SY (2007) A survey and optimization-based evaluation of development strategies for the air cargo industry. Int J Prod Econ 106(2):550–562. https://doi.org/ 10.1016/j.ijpe.2006.06.016 21. Deng Q, Liu X, Liao H (2015) Identifying critical factors in the eco-efficiency of remanufacturing based on the fuzzy DEMATEL method. Sustainability (Switzerland) 7(11):15527– 15547. https://doi.org/10.3390/su71115527
Hurdle Appraisal for the Implementation of Circular Economy Notions in Supply Chain Ruchi Gupta, Mohit Tyagi, and R. K. Garg
Abstract Circular economy (CE) is a very wide concept, and it is spreading worldwide. With time as the population is increasing so is the pressure on limited natural resources. So, it is required to implement CE at every stage in a supply chain to make it sustainable. Every organization needs a proper balance between customer satisfaction and environmental health to cope up with future challenges. The CE is rapidly gaining momentum among business spearheads and officials in Europe, China, and other developed countries while its capability has been ignored by the developing countries so far. However, India, one of the developing countries is making an effort to crack climate-resilient development, and national legislators are investing in CE. For implementing CE in any supply chain, it is necessary to see every aspect whether it is environmental or economical. Every organization needs an ideal supply chain performance system (SCPS) to keep track of its resources and manage them effectively. An attempt has been made to make the supply chain more flexible and sustainable by identifying and assessing its hurdles. Several critical hurdles that are an obstacle for the implementation of CE in many developing countries are found out with the support of literature available, academician, and field experts. Then Interpretive structural modeling (ISM) has been applied to recognize the relationship among the hurdles. A set of directly and indirectly related hurdles are structured into a comprehensive systematic model and the matrix of cross-impact multiplication applied to classification (MICMAC) analysis has been done to categorize these hurdles into four groups, viz., autonomous, linkage, independent, and dependent. After identifying hurdles falling into the linkage cluster, prioritization has been done by using the analytical hierarchy process (AHP). Prioritization of the hurdles will help the decision-maker to decide the available criteria (viz. hurdles). Keywords Circular economy · Supply chain performance system · ISM · MICMAC · AHP
R. Gupta · M. Tyagi (B) · R. K. Garg Department of Industrial and Production Engineering, Dr. B. R. Ambedkar National Institute of Technology Jalandhar, Punjab, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_3
21
22
R. Gupta et al.
1 Introduction and its Background Supply chain practices encompass the practices starting from the procurement of entities to product delivery [27]. Companies must consider sustainability and a closure cycle with the transition from linear to CE. The CE stresses the minimum consumption of resources and the conservation of the environment,companies, therefore, plan to implement the green supply chain. CE comprises a linear sequence of ‘take-makeconsume’ that can be defined as the classic flow of conventional business models that has been changed into a modern sequence of ‘take-make-consume-dispose’ [1]. The CE stresses the least employment of resources as well as the preservation of environment; thus, businesses consider the introduction of green supply chains. The conventional supply chain creates pollution, causes environmental issues, and does not consider people and the environment. As a result of strict guidelines, a high competition level, and public pressure, the companies require environmental concerns to be included in their strategy and practices as an environmental corporate tactic [29, 30]. The CE is a philosophy that seeks to construct a new socioeconomic paradigm in the long run [13]. This strategy seeks to alter socioeconomic structures by separating economic development from natural resource diminution and environmental deterioration [15]. In production and consumption, the popular and famous concept ‘3R framework’ refers to reduction, reusing, and recycling. It calls for the ratio of recycled materials to be increased, raw materials and wastes to be reused, and resources and energy used to be reduced entirely. Based on the analysis done by Kirchherr et al. [11], in the conceptualization of CE, a core normative idea is that CE’s valid objectives are environmental sustainability, economic prosperity, and social equity and in scholarship and practice should be treated accordingly. An economic organization like CE swaps the ‘end-of-life’ idea in production/distribution and consumption processes with 3R. CE’s general idea is to maximize the use of resources to support sustainable development. CE addresses sustainability by promoting a nature-based economy in which waste can once again be utilized as secondary resources [18], which means that circular economy offers a generalized substitute to the conventional supply chain today. Nowadays managing a sustainable supply chain is of major concern [26]. The CE suggests the idea of sustainability which means a world where minimum consumption of finite resources, elimination of the waste, reduction of harmful emissions, and increase in revenue of the company without sacrificing quality product to the customer.
Hurdle Appraisal for the Implementation of Circular Economy …
23
2 Solution Methodology For the assessment of hurdles in the implementation of CE notion in SCPS, a set of hurdles with the help of a literature survey and expert opinion is recorded. Then ISM method has been applied to find out driving power and dependence power. Further MICMAC analysis has been done to categorize these hurdles into four clusters. Multicriteria decision-making (MCDM) technique AHP has been employed to prioritize the hurdles that appeared in the linkage quadrant.
2.1 ISM-AHP Hybrid Approach ISM is a well-known technique for defining relationships between objects, variables, and factors that characterize a complex problem or issue. ISM establishes a hierarchical relationship between the process/operations structures. ISM assists in the understanding and visualization of a system’s structure in the sense of hurdles. It aids in determining the driving and dependence power of hurdles. ISM is useful only to identify the interdependence among the hurdles, but prioritization of the key hurdles is difficult for a decision-maker so AHP has been employed. The detailed implementation steps for this hybrid technique are illustrated in Fig. 1.
2.2 Identifying the Hurdles in the Circular Economy Various hurdles in CE has been identified based on literature survey and consultation with academic and industrial experts. There are twelve hurdles identified which are shown in Table 1.
3 Numerical Illustration After identifying various hurdles of CE, a structural self-interaction matrix (SSIM) has been formed as shown in Table 2 to establish a contextual relationship among different hurdles. Transformation of SSIM into initial reachability matrix, level partitioning, constructing ISM framework and MICMAC analysis has been conducted for establishing relationship among considered hurdles. This study mainly focuses on key hurdles that fall in linkage clusters as they can influence the supply chain performance in context to CE. Therefore, prioritization of those hurdles has been done by using a multi decision-making approach viz AHP.
24
R. Gupta et al.
Identify the hurdles in circular economy
Establishing contextual relationship among different hurdles
Developing SSIM (Structural SelfInteraction Matrix)
Literature survey on circular economy
Obtaining expert opinion
Developing reachability matrix
Prepare reachability matrix with transitivity concept Partitioning the obtained matrix into different levels
Is there any conceptual inconsistency?
Prepare directed graph (digraph) by elimination of transitive links and using contextual relationship
Yes
No Develop ISM model from diagraph
MICMAC analysis
Prioritization of hurdles belong to linkage cluster using AHP
Fig. 1 Flow diagram for preparing ISM-AHP hybrid model
3.1 Establishing Contextual Relationships Among Different Hurdles SSIM has been developed with the help of ‘VAXO’ and here VAXO implies: • If i → j, i.e., hurdle ‘j’ is achieved through hurdle ‘i’, then it will be represented as ‘V’.
Hurdle Appraisal for the Implementation of Circular Economy …
25
Table 1 List of hurdles S.no
Variable
Hurdles
References
1
H1
Difficulty in waste or scrap collection
[2, 4, 17]
2
H2
Lack of advance technology e.g., Sorting technology
[3, 5, 6, 12, 16, 19]
3
H3
High cost of waste treatment
[10]
4
H4
Lack of safe methods for waste treatment
[5]
5
H5
Lack of industrial symbiosis
[5, 9, 20, 25]
6
H6
Increase health and safety hazardous [10]
7
H7
Complexity in reuse, remanufacturing, recycling
[7, 8]
8
H8
Difficulty in adopting the solid waste policy
[14, 25]
9
H9
Inadequate financial resources
[4, 19, 23]
10
H10
Lack of information integration in the system
[4, 19, 22, 23]
11
H11
Lack of trained professionals in environmental management (EM)
[4, 19, 21, 28]
12
H12
Lack of staff training for repair, recycling, and remanufacturing the product
The Logistics Handbook A Practical Guide for the Supply Chain Management of Health Commodities [9]
Table 2 Structural self-interaction matrix Variable
H1
H2
H3
H4
H5
H6
H7
H8
H9
H10
H11
H1 H2
V
H3
V
A
H4
V
O
A
H5
V
O
A
A
H6
V
A
O
A
O
H7
V
A
V
O
A
O
H8
X
A
A
A
O
V
H9
V
V
O
V
V
O
O
V
H10
V
O
O
O
V
O
O
V
A
H11
O
O
V
V
O
V
O
V
A
O
H12
V
A
V
A
O
O
X
V
A
O
A
O
H12
26
R. Gupta et al.
• If j → i, i.e., hurdle ‘i’ is achieved through hurdle ‘j’, then it will be represented as ‘A’. • If i ↔ j, i.e., both the hurdles ‘i’ and ‘j’ are achieved with the help of each other then it will be represented as ‘X’. • If i × j, both the hurdles ‘i’ and ‘j’ are not related to each other then it will be represented as ‘O’.
3.2 Conversion of SSIM into Initial Reachability Matrix The reachability matrix has been formed by replacing VAXO with its corresponding values in the SSIM matrix for hurdles. So, the numerical value of V is 1, A is 0, X is 1 and O is 0 as shown in Table 3. Initial reachability matrix is transformed into final reachability matrix as shown in Table 4 by considering transitivity property that is if A → B and B → C, then A → C. Example: In Table 4, H1 → H8, H8 → H6 therefore, H1 → H6 (Table 5).
3.3 Level Partitioning Level partitioning attributes each hurdle to the level and is carried out as follows: First, reachability and antecedent sets are established for each hurdle. The hurdle itself, as well as other hurdles that it can assist in achieving, create the reachability set for hurdles. The hurdle’s antecedent set consists of the hurdle itself as well as any other hurdles that may aid in its achievement. The intersection set will be built in the second Table 3 Initial reachability matrix Variable
H1
H2
H3
H4
H5
H6
H7
H8
H9
H10
H11
H12
H1
1
0
0
0
0
0
0
1
0
0
0
0
H2
1
1
1
0
0
1
1
1
0
0
0
1
H3
1
0
1
1
1
0
0
1
0
0
0
0
H4
1
0
0
1
1
1
0
1
0
0
0
1
H5
1
0
0
0
1
0
1
0
0
0
0
0
H6
1
0
0
0
0
1
0
0
0
0
0
0
H7
1
0
1
0
0
0
1
1
0
0
0
1
H8
1
0
0
0
0
1
0
1
0
0
0
0
H9
1
1
0
1
1
0
0
1
1
1
1
1
H10
1
0
0
0
1
0
0
1
0
1
0
0
H11
0
0
1
1
0
1
0
1
0
0
1
0
H12
1
0
1
0
0
0
1
1
0
0
0
1
Hurdle Appraisal for the Implementation of Circular Economy …
27
Table 4 Final reachability matrix Variable
H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 Driving power
H1
1
0
0
0
0
1
0
1
0
0
0
0
3
H2
1
1
1
1
1
1
1
1
0
0
0
1
9
H3
1
0
1
1
1
1
1
1
0
0
0
1
8
H4
1
0
1
1
1
1
1
1
0
0
0
1
8
H5
1
0
1
0
1
0
1
1
0
0
0
1
6
H6
1
0
0
0
0
1
0
1
0
0
0
0
3
H7
1
0
1
1
1
1
1
1
0
0
0
1
8
H8
1
0
0
0
0
1
0
1
0
0
0
0
3
H9
1
1
1
1
1
1
1
1
1
1
1
1
12
H10
1
0
0
0
1
1
1
1
0
1
0
0
6
H11
1
0
1
1
1
1
0
1
0
0
1
1
8
H12
1
0
1
1
1
1
1
1
0
0
0
1
8
2
8
7
9
11
8
12
1
2
2
8
Dependence 12 power
Table 5 Level 1 partitioning Variable
Reachability set
Antecedent set
Intersection set
Level
H1
1,6,8
1,2,3,4,5,6,7,8,9,10,11,12
1,6,8
1
H2
1,2,3,4,5,6,7,8,12
2,9
2
H3
1,3,4,5,6,7,8,12
2,3,4,5,7,9,11,12
3,4,5,7,12
H4
1,3,4,5,6,7,8,12
2,3,4,7,9,11,12
3,4,7,12
H5
1,3,5,7,8,12
2,3,4,5,7,9,10,11,12
3,5,7,12
H6
1,8,6
1,2,3,4,6,7,8,9,10,11,12
1,8,6
H7
1,3,4,5,6,7,8,12
2,3,4,5,7,9,10,12
3,4,5,7,12
H8
1,6,8
1,2,3,4,5,6,7,8,9,10,11,12
1,6,8
H9
1,2,3,4,5,6,7,8,9,10,11,12
9
9
H10
1,5,6,7,8,10
9,10
10
H11
1,3,4,5,6,8,11,12
9,11
11
H12
1,3,4,5,6,7,8,12
2,3,4,5,7,9,11,12
3,4,5,7,12
1 1
phase by considering the reachability and antecedent sets. The hurdles with the same reachability and interaction set will be assigned ‘level 1’. Hurdles, difficulty in waste or scrap collection (H1), increase health and safety hazardous (H6), and difficulty in adopting solid waste policy (H8) are labeled as 1. The hurdles assigned to ‘level 1’ will be removed from the rest of the sets, and the process will be repeated until all the hurdles have been assigned levels. Six levels have been assigned to all considered hurdles as shown in Table 6.
28 Table 6 CE hurdles level partitioning
R. Gupta et al. Variable
Driving power
Dependence power
Level
H1
3
12
1
H6
3
11
1
H8
3
12
1
H3
8
8
2
H7
8
8
2
H12
8
8
2
H5
6
9
3
H4
8
7
4
H10
6
2
4
H2
9
2
5
H11
8
2
5
H9
12
1
6
3.4 Developing ISM Framework The resulting outcome graph is termed as a diagraph. The diagraph is eventually translated into an ISM model after the transmissivities are removed. Figure 2 depicts ISM framework which consists of six levels from level 1 to level 6. Difficulty in waste or scrap collection (H1), increase health and safety hazardous (H6), and difficulty in adopting solid waste policy (H8) come under level 1 which shows high dependence power. Inadequate financial resources (H9) come under level 6 which shows high driving power.
3.5 MICMAC Approach The MICMAC approach helped us to understand how hurdles behave in SCPS under CE notion. The MICMAC matrix, the driving power, and dependence power are kept on Y-axis and X-axis, respectively, is shown in Fig. 3. Using their driving and dependence power, the hurdles have been divided into 4 clusters: driving, dependence, linkage, and autonomous hurdle. This study mainly focuses on linkage clusters as key hurdles which have high driving power and dependence power.
Hurdle Appraisal for the Implementation of Circular Economy …
29
Difficulty in waste or scrap collection (H1)
Increase health and safety hazardous (H6)
Difficulty in adopting solid waste policy (H8)
High cost of waste treatment (H3)
Complexity in reuse, remanufacturing, recycling (H7)
Lack of staff training for repair, recycling, and remanufacturing the product (H12)
Lack of industrial symbiosis (H5)
Level 4
Lack of trained professionals in environmental management (H11)
Lack of advance technology e.g., Sorting technology (H2)
Level 2
Level 3
Lack of information integration in the system (H10)
Lack of safe methods for waste treatment (H4)
Level 1
Level 5
Level 6
Inadequate financial resources (H9)
Driving Power
Fig. 2 ISM model framework depicting the relationship among hurdles for CE 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Linkage
Independent
H9
H4
H2 H11
H3,H7,H12
H10
H5
Dependent H1,H8
Autonomous
H6
0
1
2
3
4
5
6
7
8
Dependence Power Fig. 3 MICMAC analysis
9
10
11
12
13
30
R. Gupta et al.
Table 7 Saaty’s scale (adapted from Saaty [24])
Intensity of influence
Definition
1
Equally influential
3
Moderately influential
5
Strongly influential
7
Very strongly influential
9
Extremely influential
3.6 Analytical Hierarchical Process AHP [24] is one among the various MCDM approaches which is used for the majority of the decision-making problems because of its easier implementation. It is employed for solving the numerous intricate problems related to making a decision among the available criteria. AHP comprises arranging the decision-making problems into different hierarchical levels such as a core objective/goal of the problem, key criteria, sub-criteria, and substitutes, trailed by a pairwise comparison at each level. After defining the complete framework, experts compare each criterion against the others by means of Saaty’s nine-point scale described in Table 7. Hurdles appeared in linkage cluster, viz., H3, H4, H5, H7, H12 are considered as key hurdles but to select one most significant hurdle among them is very difficult for making decision, so prioritization of those hurdles has been done by using AHP. Steps of AHP are as follows: 1. 2.
Developing a hierarchy for the decision (Fig. 4) Deriving Weights for the Criteria.
All the criteria cannot be given equal importance. So, there is a requirement to find out the relative importance of the hurdles among themselves by comparing pairwise. The relative importance of every individual criterion with respect to the other is done
Goal
Hurdle assesment
High cost of waste treatment (H3)
Lack of safe methods for waste treatment (H4)
Lack of industrial symbiosis (H5
Complexity in reuse, remanufacturing, recycling (H7)
Fig. 4 Decision hierarchy for hurdle assessment
Lack of staff training for repair, recycling, and remanufacturing the product (H12)
Criteria
Hurdle Appraisal for the Implementation of Circular Economy …
31
Table 8 Pairwise comparison matrix Variable
H3
H4
H5
H7
H12
H3
1
3
1/3
6
5
H4
1/3
1
1/5
6
4
H5
3
5
1
7
6
H7
1/6
1/6
1/7
1
1/3
H12
1/5
1/4
1/6
3
1
using Saaty’s scale (Table 7) based on the experts’ opinion is shown in Table 8 and Table 9. Normalized pairwise comparison matrix has been found by dividing each element by sum of its corresponding column as shown in Table 10. Method of computing weights is as follows: HW = λmax W. Here ‘H’ is pairwise comparison matrix, ‘λmax ’ is principal eigenvalue that is the largest eigenvalue and ‘W’ is normalized eigenvector that is criteria weight corresponding to λmax . Criteria weights (W) corresponding to each hurdle are found by averaging across the rows. Table 9 Pairwise comparison matrix (in decimal form) Variable
H3
H4
H5
H7
H12
H3
1.00
3.00
0.33
6.00
5.00
H4
0.33
1.00
0.20
6.00
4.00
H5
3.00
5.00
1.00
7.00
6.00
H7
0.17
0.17
0.14
1.00
0.33
H12
0.20
0.25
0.17
3.00
1.00
Sum
4.70
9.42
1.84
23.00
16.33
Table 10 Normalized pairwise comparison matrix Variable
H3
H4
H5
H7
H12
Criteria weights (W)
H3
0.212
0.318
0.179
0.26
0.306
0.255
H4
0.07
0.106
0.108
0.26
0.244
0.157
H5
0.638
0.53
0.543
0.304
0.367
0.476
H7
0.036
0.018
0.076
0.043
0.02
0.038
H12
0.042
0.026
0.092
0.13
0.061
0.07
32
R. Gupta et al.
Table 11 Weighted sum value, Criteria weights and their ratio Variable
H3
H4
H5
H7
H12
Weighted sum value (P)
Criteria weight (W)
P/W
H3
0.255
0.471
0.157
0.228
0.35
1.461
0.255
5.729
H4
0.084
0.157
0.095
0.228
0.28
0.844
0.157
5.375
H5
0.765
0.785
0.476
0.266
0.42
2.712
0.476
5.697
H7
0.043
0.026
0.066
0.038
0.023
0.196
0.038
5.157
H12
0.051
0.039
0.08
0.114
0.07
0.354
0.07
5.057
Table 12 Random index (adapted from Saaty [24]) n
1
2
3
4
5
6
7
8
9
RI
0
0
0.58
0.90
1.12
1.24
1.32
1.41
1.45
Individual cell element of normalized comparison matrix has been multiplied by their corresponding criteria weight as shown in Table 11. Weighted sum value (P) has been obtained by summation across the row. n P/W = 5.403, here matrix size (n) = 5 Principal Eigenvalue, λmax = i=1n λmax −n Consistency Index (CI) = n−1 = 0.1, Index Consistency Ratio (CR) = Consistency = 0.089, here RI = 1.12 corresponding Random Index n = 5 as shown in Table 12. The calculation of CR is required for assessing consistency of experts’ opinion matrix. For a consistency level to be accepted, the value of CR must be less than 0.10. In the present study, the value of CR has been found out to be 0.089 which clearly shows that the matrix made by experts’ opinion is consistent.
4 Results and Discussion The CE hurdles have been identified from the literature survey and assessment data collected from industrial experts and academicians were kept into ISM to observe contextual relationships among them. MICMAC analysis has been done to categorize the hurdles into different clusters according to their corresponding driving and dependence power. Findings of this study give us four clusters. First quadrant represents an autonomous cluster. The hurdles appeared in this have minimal driving power and dependence, also have little impact on organization. No hurdle appeared in this quadrant. Second quadrant represents dependent cluster with minimum driving power and maximum dependence. Difficulty in waste or scrap collection (H1), increase health and safety hazardous (H6), and difficulty in adopting solid waste policy (H8) hurdles lie in second quadrant. Third quadrant shows linkage cluster. High cost of waste
Hurdle Appraisal for the Implementation of Circular Economy …
33
treatment (H3), Lack of safe methods for waste treatment (H4), Lack of industrial symbiosis (H5), Complexity in reuse, remanufacturing, recycling (H7), and Lack of staff training for repair, recycling, and remanufacturing the product (H12) hurdles lies in third quadrant. Fourth Quadrant shows independent cluster with high driving power but low dependence power. Lack of advance technology, e.g., sorting technology (H2), inadequate financial resources (H9), lack of information integration in the system (H10), and lack of trained professionals in environmental management (H11) fall in fourth quadrant. For prioritizing, AHP has been implemented to the hurdles falling in linkage cluster, viz., H3, H4, H5, H7, and H12. The value of CR which is equals to 0.089 clearly shows that the matrix made by experts’ opinion is consistent. So, the criteria weights that were obtained will be useful for the decisionmaker to take further decisions in context to CE. From the present study 47.6% weightage should be given to lack of industrial symbiosis (H5) and followed by 25.5% weightage to high cost of waste treatment (H3).
5 Conclusion Nowadays, every supply chain wants to be green and circular. For the better implementation of CE, every SCPS needs assessment of their hurdles to make supply chain more flexible, sustainable, and healthier. Figuring up the relationship among considered hurdles and identifying the driving and dependence powers of the hurdles has been done using ISM methodology. The hurdles were iterated in six levels; level 1 including difficulty in waste or scrap collection (H1), increase health and safety hazardous (H6), and difficulty in adopting solid waste policy (H8) hurdles are least important and level 6 including inadequate financial resources (H9) hurdle is most important. The results obtained from ISM are treated as an input data in MICMAC to categorize considered hurdles into four clusters, viz., autonomous, linkage, dependent, and independent. This study focuses on the linkage cluster as it has key hurdles and, in this quadrant, both the driving and dependence powers are high. The output from MICMAC analysis has been prioritized using AHP approach. The results of AHP show that lack of industrial symbiosis (H5) should be given highest weightage trailed by high cost of waste treatment (H3). Hence, the current research contributes in providing remarkable recommendation for the researchers, examiners, and analysts to make decisions in the assessment of hurdles in context to CE.
References 1. Desousa Jabbour ABL, Jabbour CJC, Godinho Filho M, Roubaud D (2018) Industry 4.0 and the circular economy: a proposed research agenda and original roadmap for sustainable operations. Ann Oper Res 270(1):273–286. 2. EC (2018) Measuring progress towards circular economy in the European Union − key
34
R. Gupta et al.
indicators for a monitoring framework – SWD 17 Final. European Commission, Strasbourg 3. Fischer A, Pascucci S (2017) Institutional incentives in circular economy transition: the case of material use in the Dutch textile industry. J Clean Prod 155:17–32 4. Geng Y, Doberstein B (2008) Developing the circular economy in China: challenges and opportunities for achieving ‘leapfrog development’. Int J Sustain Dev World Ecol 15(3):231– 239 5. Geng Y, Fu J, Sarkis J, Xue B (2012) Towards a national circular economy indicator system in China: an evaluation and critical analysis. J Clean Prod 23(1):216–224 6. Hatayama H, Daigo I, Tahara K (2014) Tracking effective measures for closed-loop recycling of automobile steel in China. Resour Conserv Recycl 87:65–71 7. Haupt M, Vadenbo C, Hellweg S (2017) Do we have the right performance indicators for the circular economy?: insight into the Swiss waste management system. J Ind Ecol 21(3):615–627 8. Jones PT, Geysen D, Tielemans Y, Van Passel S, Pontikes Y, Blanpain B, ... Hoekstra N (2013) Enhanced landfill mining in view of multiple resource recovery: a critical review. J Clean Prod 55:45–55 9. Kalmykova Y, Sadagopan M, Rosado L (2018) Circular economy – From review of theories and practices to development of implementation tools. Resour Conserv Recycl 135:190–201 10. Kazancoglu Y, Kazancoglu I, Sagnak M (2018) A new holistic conceptual framework for green supply chain management performance assessment based on circular economy. J Clean Prod 195:1282–1299 11. Kirchherr J, Reike D, Hekkert M (2017) Conceptualizing the CE: an analysis of 114 definitions. Resour Conserv Recycl 127:221–232 12. Løvik AN, Modaresi R, Müller DB (2014) Long-term strategies for increased recycling of automotive aluminum and its alloying elements. Environ Sci Technol 48(8):4257–4265 13. Masi D, Day S, Godsell J (2017) Supply chain configurations in the CE: a systematic literature review. Sustainability 9(9):1602 14. Demattos CA, De Albuquerque TLM (2018) Enabling factors and strategies for the transition toward a circular economy (CE). Sustainability 10(12):4628 15. Murray A, Skene K, Haynes K (2017) The CE: an interdisciplinary exploration of the concept and application in a global context. J Bus Ethics 140(3):369–380 16. Nakajima K, Takeda O, Miki T, Matsubae K, Nakamura S, Nagasaka T (2010) Thermodynamic analysis of contamination by alloying elements in aluminum recycling. Environ Sci Technol 44(14):5594–5600 17. Niero M, Olsen SI (2016) Circular economy: to be or not to be in a closed product loop? a life cycle assessment of aluminium cans with inclusion of alloying elements. Resour Conserv Recycl 114:18–31 18. Nubholz JL (2017) Circular business models: defining a concept and framing an emerging research field. Sustainability 9(10):1810 19. Ormazabal M, Prieto-Sandoval V, Jaca C, Santos J (2016) An overview of the circular economy among SMEs in the Basque country: a multiple case study. J Ind Eng Manag (JIEM) 9(5):1047– 1058 20. Ormazabal M, Prieto-Sandoval V, Puga-Leal R, Jaca C (2018) Circular economy in Spanish SMEs: challenges and opportunities. J Clean Prod 185:157–167 21. Preston F (2012) A global redesign? Shaping the circular economy 22. Ritzén S, Sandström GÖ (2017) Barriers to the Circular Economy – integration of perspectives and domains. Proc CIRP 64:7–12 23. Rizos V, Behrens A, Van der Gaast W, Hofman E, Ioannou A, Kafyeke T, ... Topi C (2016) Implementation of circular economy business models by small and medium-sized enterprises (SMEs): barriers and enablers. Sustainability 8(11):1212 24. Saaty TL (1980) Analytical hierarchy process. McGraw Hill Company, NY, USA 25. Sehnem S, Jabbour CJC, Pereira SCF, de Sousa Jabbour ABL (2019) Improving sustainable supply chains performance through operational excellence: circular economy approach. Resour Conserv Recycl 149:236–248
Hurdle Appraisal for the Implementation of Circular Economy …
35
26. Sharma J, Tyagi M, Bhardwaj A (2020) Parametric review of food supply chain performance implications under different aspects. J Adv Manag Res. https://doi.org/10.1108/JAMR-102019-0193 27. Sharma J, Tyagi M, Bhardwaj A (2021) Exploration of COVID-19 impact on the dimensions of food safety and security: a perspective of societal issues with relief measures. J Agribus Dev Emerg Econ. https://doi.org/10.1108/JADEE-09-2020-0194 28. Shi H, Peng SZ, Liu Y, Zhong P (2008) Barriers to the implementation of cleaner production in Chinese SMEs: government, industry and expert stakeholders’ perspectives. J Clean Prod 16(7):842–852 29. Vanalle RM, Ganga GMD, Godinho Filho M, Lucato WC (2017) Green supply chain management: an investigation of pressures, practices, and performance within the Brazilian automotive supply chain. J Clean Prod 151:250–259 30. Zhu Q, Sarkis J, Lai KH (2008) Confirmation of a measurement model for green supply chain management practices implementation. Int J Prod Econ 111(2):261–273
Designing an Air Purifier by Using Green Quality Function Deployment Methodology Kanika Prasad, Akshay Kumar, Jeet Kumar Yadav, Parwez Akhtar, and Raj Ballav
Abstract An air purifier is a device that removes impurities suspended in air and improves the quality of air. It is commonly used as a medical aid by patients suffering from allergy and asthma. In general, it reduces the particulate matter in air, which may cause respiratory health issues. This work is aimed at systematic implementation of Green Quality Function Deployment (GQFD) for designing air purifiers. GQFD is a product-oriented quality management technique based on customer feedback or survey, which imparts operational improvements in the product for addressing the issue of sustainability. The establishment of balance between the utility and sustainability in terms of impact of device on natural ecosystem is the major criterion for GQFD technique. In the process of determining the design criteria for new air purifier, GQFD provides the most critical parameter and functions from survey, which contains consumers’ perspective, and then these parameters are employed to determine the technical characteristics of the product. Subsequently, the house of quality matrix depicting the strength of the relation between customers’ needs and technical specifications is established while using these identified parameters. The dynamic nature of product development and its vast application is ideal for applying this technique. Finally, the result of GQFD is applied to the product to infer the modification in reference to customers’ feedback. Keywords Sustainability · Quality function deployment (QFD) · Green quality function deployment (GQFD) · House of quality (HOQ) · Air purifier · Voice of customer (VOC)
K. Prasad (B) · A. Kumar · J. K. Yadav · P. Akhtar · R. Ballav Department of Production and Industrial Engineering, National Institute of Technology, Jamshedpur, India e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_4
37
38
K. Prasad et al.
1 Introduction Air purifier is an emerging product in the world market due to the increase in pollution at an alarming rate, and it has the potential to attract the market. These days, air purifiers are extensively used in industries also, to get rid of impurities from air. They are designed to not only cleanse the air but also remove all different kinds of pollutants that may comprise of foul smell, dust particles, smoke and pet hackles. However, the by-products produced by it and the materials utilized for its manufacturing are dangerous for environment. The design of a sustainable product is focused on developing a product that can maintain an equilibrium and reduce the impacts on environment and society around the world. Therefore, this work aims at finding the feasibility of product development with green design as the primary focus, so that the environmental requirements may also be taken into consideration. Quality function deployment (QFD) is the most appropriate methodology for product development, when the customers are well understood, and their requirements need to be incorporated in the product. Since the main objective of this technique is to incorporate voice of customers (VOCs), it can very effectively be used throughout the entire lifecycle of product. Also, the methodology is based on feedback from customers, which can easily deploy the VOC into the product. Green manufacturing needs to be practiced in current times so as to bring attention toward environmental impact and resource consumption, in addition to quality, time, and cost. Green Quality Function Deployment (GQFD) is a technique that can address the issue of sustainability efficiently, which is of prime importance in today’s environmental scenario. The objective of this methodology is to minimize resource utilization and reduce the load on environmental load. Therefore, in the present work, GQFD is utilized for designing air purifiers in order to find the balance between utility and sustainability. The objective of this work is to develop and design the concepts for the next generation of air purifiers based on a customer-centered approach. The solution aims to meet the customers’ demands, be socially, economically and environmentally sustainable.
2 Research Background Yogi Akao, who is also known as the father of QFD, first introduced this methodology in Japan in the year 1966. This technique has been extensively used in industries across the globe owing to its benefits. It is an efficient tool that can effectively deploy the customer requirements (CRs) into the appropriate technical characteristics (TCs) during several stages of product development and production, which comprises formulating marketing strategies, planning, product design and engineering, prototype evaluation, production process development, production, sales [7]. The primary focus of QFD method is to produce the right product right the very first time while converting the subjective customers’ requirements into measurable technical specifications [10]. In the previous decades, QFD methodology has been extensively
Designing an Air Purifier by Using Green Quality Function …
39
accepted and efficiently implemented in various domains, such as airlines [9] and highways. In recent years, this method has been successfully applied to improve the service quality of different sectors, such as textile industry [8] and education [1]. The QFD method has also been utilized by the construction industry to study and coordinate the conflicting demands of various stakeholders in order to minimize design alterations in construction projects [12]. Since QFD is customer-oriented quality improvement tool, it can effectively incorporate CRs into appropriate engineering characteristics in product design and development phase [14]. Moreover, QFD is a widely accepted management tool by service industries for maximizing the customers’ satisfaction. For example, it has been applied to improve service quality of government or support formulation of policy strategies, by converting the requirements of citizens into explicit TCs [12]. Apart from process and product development, this technique has been successfully applied in various fields of technology and management, for selection of non-traditional machining processes [4, 16], industrial robot [3], suppliers [2, 20], materials [13, 15], material handling equipment [5, 19], computer-aided design software [18], cotton fiber selection [6], CNC machining center [17] etc.
3 Research Methods 3.1 Green Quality Function Deployment Technique Since the past few decades, environmental issues have been introduced in new product development and QFD transformed into GQFD considering these environmental factors. GQFD is a powerful tool for developing environmentally friendly products. While analyzing the data and constructing the GQFD for a given product, different matrices are used, which are integrated in the form of house as shown in Fig. 1. Since the overall structure of the matrix looks like a house, it is named as House of Quality (HOQ) matrix. HOQ forms the engine of QFD methodology and is recognized as a process for product development that is motivated by customer expectations from product or service. The roots of this process lie in the strength and resources of the organization all working toward meeting those spoken and unspoken needs. QFD process is at its basic in this form. A multidisciplinary team works as a team to ascertain the customer requirements from market research and benchmarking of the competitor’s product. This is then quantified into prioritized engineering parameters to be incorporated into the design of the new product. The six basic elements of a HOQ matrix are as follows: (a)
Customers’ requirements (What’s)—Environmental, sustainability and other general requirements directly expressed by customers are documented and the relevant data are collected. Questionnaires, focus groups, customer complaints etc. are used for this purpose.
40
K. Prasad et al.
Fig. 1 House of quality matrix
(b)
(c) (d)
(e)
(f)
Technical characteristics (How’s)—It deals with how to meet the customers’ needs. These parameters are determined by a multidisciplinary team and contain technical characteristics of the analyzed product, which directly or indirectly contributes to the satisfaction of customers. Interrelationship matrix—It provides the strength of the relationship between different CRs and TCs. Matrix of correlation—It forms the roof of HOQ matrix where correlation between technical characteristics is indicated. The interplay between TCs that support each other (positive (+) or strongly positive (++) correlation) or do not support each other negative (−) or strongly negative (–) correlation) is defined. Planning matrix—The right-hand side of HOQ matrix contains planning matrix, which serves multiple purposes, which is used to ascertain the relative importance of identified VOCs and to get an idea about the customer’s perception of the benchmarked products. Product targets matrix—This matrix is used to summarize all values from the previously discussed matrices, which consist of three parts, namely, technical properties, competitive technical benchmarking and product targets.
Designing an Air Purifier by Using Green Quality Function …
41
Fig. 2 Management and planning tools [16]
Along with the HOQ matrix, various management and planning tools used in the QFD methodology are given in Fig. 2. The product development process, beginning with the determination of VOCs to the final production of the end product, may be listed in a step-by-step manner such that steps that require attention during intermediate process can be established and decisions may be taken accordingly. In the next phase, i.e. strategy phase, product policy and VOC are identified, and those customers’ needs are then interpreted into product concept. For establishing the TCs (How’s) of the product design, the VOC (What’s) serve as input. These design specifications define the process plans, which would help in the manufacturing process operations. The four phases of product development are shown in Fig. 3. The steps followed in preparing the HOQ matrix are as follows: 1.
2.
First, the environmental VOCs that represent what’s on the HOQ matrix are identified. The customers’ requirements that reflect the needs of targeted market segment are stated on the left side of the matrix. The targeted market segment includes the end-users or interest groups who influence the purchase decision of the product. The matrix is divided into smaller parts or subsystems to minimize the number of primary requirements in a matrix. A relative scale of 1–5 as described in Table 1 is employed for rating the importance of VOCs. In the next step, the technical requirements are identified for designing the products, which represent how these characteristics affect the customers’ needs.
42
K. Prasad et al.
Fig. 3 Four phases of QFD [11]
Table 1 Relative scale for importance rating of VOCs
3.
4.
Value
Interpretation
1
Not important
2
Important
3
Much more important
4
Important
5
Most important
Quantitative measurable characteristics are being translated from the qualitative requirements of the customer. The feedback is obtained from surveys, focus groups/clinic or customer meetings. While meeting the requirements of customer, the TCs are chosen in such a way that they do not constrain the designer. The design requirements should be in balance with the available expertise and the given time and cost frames of the project. Then, interrelationship matrix showing relationships between customer requirements and product’s technical characteristics is developed. The strength of the relationship between what’s and how’s is determined using a relative scale for prioritizing efforts and making trade-off decisions as shown in Table 2.
Table 2 Correlation index for strength of interrelationship matrix
Value
Interpretation
1
Very weak relation
3
Weak relation
5
Moderate relation
7
Strong relation
9
Very strong relation
Designing an Air Purifier by Using Green Quality Function …
5.
43
The priority weight for each technical requirement can be obtained from the following mathematical expression. Wj =
n
Pr i × Corr elation index
(1)
i=1
where, W j is the priority weight for jth technical requirement, n is the number of customers’ requirements, Pr i is the priority assigned to ith customer requirement and correlation index is the relative importance value for jth technical requirement with respect to ith customer requirement, obtained from the HOQ matrix. The relative weight for each TC is calculated using Eq. 2. Wj W j = n j=1
6.
7.
Wj
(2)
Subsequently, technical evaluation of previously manufactured products and benchmarked products is conducted. Then, an assessment based on the defined product requirements or TCs is performed. Other relevant information such as warranty or frequency of service and costs are also gathered. These data are taken into consideration for technical evaluation. Finally, the matrix is obtained and the product development strategy and product plans are finalized. Required actions and areas of focus are determined at the end.
4 Case study An example of air purifier design is also illustrated to demonstrate the application of the proposed methodology for designing a sustainable product. Several design issues that must be addressed while manufacturing are developed in collaboration with the green trend of product design and manufacturing. The ultimate aim of the GQFD methodology is to develop products that meet the current need without affecting the needs of future generations.
4.1 Implementation of GQFD for Designing an Air Purifier In this work, a simplified HOQ matrix is used where the matrix of correlation and planning matrices are not considered. Only the product targets matrix at base is incorporated at the base. The various steps followed while designing an air purifier using GQFD technique are described as below.
44
K. Prasad et al.
First, the VOCs need to be collected for designing the product. Consumers are asked about their expectations from the product, to estimate which feature is important and how much is its importance. Since there are a number of features to be explored, the most frequently mentioned characteristics are good bases for forming the attitude questionnaire. The information regarding VOC is collected through the questionnaire, in which the users are selected from open market. The questionnaire containing 25 questions was prepared for a sample size of 150. This feedback was taken to modify the survey and further resurveying provides a better insight to consumer requirements for an air purifier. The VOCs as obtained through this survey are ‘Quiet Operation’, ‘Good Air Quality’, ‘Cools Quickly’, ‘Easy Operation’, ‘Less Energy Consumption’, ‘High Durability’, ‘High Performance’, ‘Easy to Repair’, ‘Higher Life Cycle’, ‘Affordable Cost’, ‘Visual Appeal’, ‘Safe Operation’, ‘Easy to Process and Assemble’, ‘Easy Maintenance’, ‘Easy to Clean’, ‘Harmless to Living Environment’, ‘Easy to Transport and Retain’ and ‘Remote Control’. VOC contains different environmental aspects also such as ‘Harmless to Living Environment’ as well as customer’s requirement like ‘Cools quickly’, ‘Safe operation’ etc. The VOC is based on market survey to show the ‘customer weights’ where respondents were asked to provide the score for relative importance to each customers’ requirements. In the next step, technical characteristics that influence the design of air purifiers are determined and explained as follows: • Air Purification Rate—The rate at which the air purifier removes the unwanted particle will determine the extent of applicability of the machine. The machine with a faster purification rate will be more effective in providing more pure air. • Efficient Filter—Efficiency of any filter can be determined by its particle removal rate with fixed power input. A high-efficiency filter is an essential tool for a sustainable product. • Odor Removal—Parts per million (PPM) and other harmful impurities may be the high priority in terms of improving quality index, but from customer’s point of view, a bad odor is the first sign of inhabitable environment, Hence, odor removal naturally becomes a primary characteristic of product. • Balance (Torque)—The torque amount should be such that motor can easily start with optimum supply of power. • Weight—Heavy products always lack portability and are not easily compatible. A lightweight product is always preferable. In case of household setup, weight plays a major role in product development. • Hardness—A robust product is less prone to wear and tear of parts or product. This leads to less maintenance and longer life of product. • Number of Parts—Number of parts should be optimum to provide fast and reliable manufacturing and assemble system. Large number of parts will be prone to high wear and tear, and hence high maintenance. • Likelihood to Get Dirt—Device should be less prone to dust particles. More the device gets dirt particle on surface, more the maintenance need.
Designing an Air Purifier by Using Green Quality Function …
45
• Number of Types of Material—There should be less type of material used for manufacturing the device. More number of types of material will increase the investment and machine setups as well. • Particulate Matter Removal—The device should be able to reduce particulate matter content to large extent. • Physical Life Time—Product should be durable and should provide service for long period of time with a standard performance. • Amount of Energy Consumption—Energy consumption should be low to make a sustainable product, which should be high priority for modern devices. • Toxicity of Materials—The material used for manufacturing of product should not be toxic for human interaction or harmful to ecology. The use of toxic material will create a challenge for disposal. • Noise and Vibration Output—To make the product consumer friendly and ready for general purpose use, the device should generate as low noise and vibration as possible. High sound and vibration will restrict its uses for household. • Cost of Production—The cost of product should be in such a range that the final distributable product should have some sales margin and profitability to manufacturer. Lower the cost of production, higher the success rate for mass manufacturing. • Optimized Dimension—The product should be of appropriate dimension for general use. The product should be able to fit in as household equipment or commercial application. • Low Frequency of filter Replacement—The product should be economical to have a profitable manufacturing model. Regular filter replacement should be avoided to make the product economical. • Less Hazardous By-Products—The primary objective of this device is to reduce toxic gases and other waste from air. Hence, any by-product that can be harmful to ecology shouldn’t be produced by device. In the end, HOQ matrix is developed by considering environmental point of view and analyzing the whole product life cycle. In the first step, priority weights to the customers are that customers are assigned as per the relative scale shown in Table 1. Assignment of priority value 5 to customers’ requirements ‘Quiet Operation’, ‘Good Air Quality’, ‘Easy Operation’, ‘Less Energy Consumption’, ‘Higher Life Cycle’, ‘Affordable Cost’ and ‘Harmless to Living Environment’, indicate that these are of highest importance the customers. Similarly, importance rating 1 provided to ‘Easy to Transport and Retain’ reveals that this CR is of least importance to the customers based on survey responses and technical feasibility. Subsequently, VOCs are integrated into feasible TCs to develop the interrelationship matrix. Figure 4 depicts the strength of relationship between the Environmental VOC and TC and their rating factors After critically studying the relationship between Whats and Hows of design of an air purifier, it is observed that CR ‘Quiet Operation’ is highly correlated to TCs ‘Particulate Matter Removal’, ‘Amount of Energy Consumption’ and ‘Noise/Vibration output’; whereas ‘Efficient Filter’ and ‘Likelihood to get Dirt’ have the least correlation with this. In the same manner, it is found that ‘Harmless
46
K. Prasad et al.
to Living Environment’, which is a very important customers’ requirement from environmental point of view, has a strong correlation with ‘Air Purification Rate’, ‘Efficient filter’, ‘Hardness’, ‘Particulate Matter Removal’, ‘Toxicity of Materials’, ‘Noise/Vibration output’ and ‘Less Hazardous By-products’. Furthermore, it is very well known that economic consideration is a critical factor while designing any product. Therefore, ‘Affordable Cost’ is greatly associated with ‘Number of Type of material’ and ‘Cost of Production’. The strength of relationship between remaining CRs and TCs is established similarly with values from an appropriate scale of 1–9 as shown in Table 2. Once the interrelation matrix is completely filled, the relative weight for each technical requirement is calculated while using Eqs. 1 and 2.
5 Result and Discussion From Fig. 4, it is evident that ‘Particulate Matter Removal’, ‘Air Purification Rate’, ‘Efficient Filter’ and ‘Less Hazardous By-Products’ have relative weights 0.123, 0.122, 0.101 and 0.076, respectively. This indicates that these TCs relatively more important than the other TCs to satisfy customer requirements. It shows that the company must focus on these TCs to satisfy customer requirements while designing an air purifier. Moreover, TCs such as ‘Weight’ and ‘Low Frequency of filter Replacement’ are the factors that do not affect the design of air purifiers from customers’ point of view.
Raw Score Relative Weight
9
9
3
9
9
3
9 9 1 3
9
9
3
3 9 1 3
1 9
3 1 1
5 9
9
3 9
3
3 5 3 1 9
Less Hazardous By-Products
9 3
9
3
9
9
1 3 1 5
3 3
3
3
3
9
3 9 3
9
3
9 3
1
3 3
9
1 3 1 3
Low Frequency of filter Replacement
9 5
3
3 9
Cost of Production
9
Optimized Dimension
9 9 7
Noise,Vibration Output
3
Toxicity of Materials
Physical Life Time
1 3 1
Amount of Energy Consumption
Particulate Matter Removal
3
Liklihood To Get Dirt
3
Number of Types of Material
9
Number of Parts
1 9 5
Weight
7 9 9
Hardness
5 5 4 5 5 4 4 3 5 5 3 3 2 2 3 5 1 2
Balance (Torque)
Efficient Filter
Odor Removal
Quiet Operation Good Air Quality Cools Quickly Easy Operation Less Energy Consumption High Durability High Performance Easy to Repair Higher Life Cycle Affordable Cost Visual Appeal Safe Operation Easy to Process and Assembl Easy Maintenance Easy to Clean Harmless to Living Environmen Easy to Transport and Retain Remote Control
Air Purification Rate
Customers' Requirements
Green Quality Function Deployment
Importance rating
Technical Characteristics
9 3 5 1
9
9
9
9
3 9
3
9
9
1 3
269
223
106
27
21
96
90
60
91
271
177
138
148
161
87
39
27
168
0.122 0.101 0.048 0.012 0.010 0.044 0.041 0.027 0.041 0.123 0.080 0.063 0.067 0.073 0.040 0.018 0.012 0.076
Fig. 4 HOQ matrix for the design of an air purifier
Designing an Air Purifier by Using Green Quality Function …
47
6 Conclusions Since the pollution level is increasing at an alarming rate, so to control this there is a need for a better-quality air filter at an affordable cost considering the environmental factors during its design. This study helped in determining the important designing factors so as to obtain both lower cost alternative as well as a highperformance model. We were able to gather the requirements of consumers through questionnaire-based survey with rating factors and for identifying how these design parameters affect the customers’ requirement, the interrelationship between the VOC and TC’s was developed by giving a relative score. Finally, we obtained the optimized designing parameters by comparing the relative weight, and the important parameters obtained are ‘particulate matter removal, air purification rate, efficient filter and less hazardous by-products’. These four parameters were found to be most important during design incorporation while solving the issue of sustainability for meeting the customers’ requirements. This GQFD concept plays a very vital role in keeping a balance between any product development and environmental protection, which, in future, has the capacity for greater good. Though we haven’t used the full HOQ matrix in this paper, the results shown by methodology in this paper can act as building blocks for GQFD application for essential products.
References 1. AYTAC A, Deniz V (2005) Quality function deployment in education: a curriculum review 39(4):507–514 2. Bhattacharya A, Geraghty J, Young P (2010) Supplier selection paradigm: an integrated hierarchical QFD methodology under multiple-criteria environment. Appl Soft Comput 10(4):1013–1027 3. Bhattacharya A, Sarkar B, Mukherjee SK (2010) Integrating AHP with QFD for robot selection under requirement perspective. Int J Prod Res 43(17):3671–3685 4. Chakraborty S, Dey S (2007) QFD-based expert system for non-traditional machining processes selection. Expert Syst Appl 32(4):1208–1217 5. Chakraborty S, Prasad K (2016) A QFD-based expert system for industrial truck selection in manufacturing organizations. J Manuf Technol Manag 27(6):800–817 6. Chakraborty S, Prasad K (2018) A quality function deployment-based expert system for cotton fibre selection. J Inst Eng India Ser E 99(1):43–53 7. Dehariya PK, Verma DDS (2015) An application of green quality function deployment to designing an air conditioner. Eng Res Appl 5(7):147–152 8. Erdil A (2019) An evaluation on lifecycle of products in textile industry of Turkey through quality function deployment and Pareto analysis. Proc Comput Sci 158:735–744 9. Ghobadian A, Terry AJ (1995) How Alitalia improves service quality through quality function deployment,1 October. ISSN 0960-4529 10. Govers CPM (1996) What and how about quality function deployment (QFD). Int J Prod Econ 46–47(95):575–585 11. Hassan A, Siadat A, Dantan JY, Martin P (2010) Conceptual process planning an improvement approach using QFD, FMEA, and ABC methods. Robot Comput Integr Manuf 26(4):392–401 12. Mao Q, Li N, Peña-Mora F (2019) Quality function deployment-based framework for improving the resilience of critical infrastructure systems 26
48
K. Prasad et al.
13. Mayyas A, Shen Q, Mayyas A, Shan D, Qattawi A, Omar M (2011) Using quality function deployment and analytical hierarchy process for material selection of body-in-white. Mater Des 32(5):2771–2782 14. Ping Y, Liu R, Lin W, Liu H (2005) A new integrated approach for engineering characteristic prioritization in quality function deployment. Adv Eng Inform 45:101099 15. Prasad K, Chakraborty S (2013) A quality function deployment-based model for materials selection. Mater Des 49:525–535 16. Prasad K, Chakraborty S (2015) A decision guidance framework for non-traditional machining processes selection. Ain Shams Eng J. https://doi.org/10.1016/j.asej.2015 17. Prasad K, Chakraborty S (2015) Development of a QFD-based expert system for CNC turning centre selection. J Ind Eng Int 11(4):575–594 18. Prasad K, Chakraborty S (2016) A QFD-based decision making model for computer-aided design software selection. Int J Ind Eng Manag 7(2):49–58 19. Prasad K, Zavadskas EK, Chakraborty S (2015) A software prototype for material handling equipment selection for construction sites. Autom Constr 57:120–131 20. Rajesh G, Malliga P (2013) Supplier selection based on AHP QFD methodology. Proc Eng 64:1283–1292
Supply Chain Management Practice Constructs in SMEs: Development of Constructs and Its Implementation Issue Rohit Kumar and Manish Gupta
Abstract In today’s market, small- and medium-size manufacturing enterprises’ survival depends on their ability to produce more at less cost, less time, and fewer defects and provide the product of the right quality and quantity at the right time right place. Without establishing a relationship with supply chain partners and appropriate information flow with supply chain partners, small and medium-sized manufacturing enterprises will no longer effectively compete in today’s market. The key to this is the supply chain management practice. Supply chain management is a set of system strategies for integrating suppliers, inbound logistics, manufacturers, outbound logistics, and customers in order to improve knowledge sharing and cash flow efficiency among individuals and across the supply chain as a whole. Supply chain management practices are developed in the context of large enterprises, but less attention is paid to SMEs. The purpose of this chapter is to review the literature on factors for which supply chain management practice constructs are developed and its implementation issues and remedies in small and medium-sized manufacturing enterprises. This paper is beneficial for practitioners, policymakers, and regulatory bodies, CEO, and managers to develop an in-depth understanding of critical constructs of supply chain management practice and their impacts on supply chain performance. Keywords Supply chain management practice · Manufacturing SMEs · Digital technology adoption · SMEs key players’ role
1 Introduction Small and medium-sized enterprises (SMEs) have been known to manufacture and provide services with limited resources in India for many years. SMEs have emerged as a vibrant and diverse segment of the economy. The Indian industrial sector relies heavily on SMEs. In terms of production output, job growth, and exports, they play a critical role in the economy [12]. SMEs’ strength is their efficient use of resources, R. Kumar · M. Gupta (B) Department of Mechanical Engineering, Motilal Nehru National Institute of Technology Allahabad, Prayagraj 211004, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_6
49
50
R. Kumar and M. Gupta
greater organizational versatility, agility, higher creativity, and low investment [1]. Organizations currently work in unpredictably dynamic market environments. Organizations must implement supply chain management practices (SCMP) as one way to succeed in the volatile and challenging market environment [3]. The practice of SCM with information sharing provides a foundation for enhancing SC and organizational efficiency. SCM approaches have a positive impact on SC performance and, as a result, provide organizations a competitive advantage in terms of price, efficiency, and delivery dependability [7, 19]. In the current environment, organizations must work collaboratively with suppliers and customers to enhance SC performance. To compete in an unpredictably challenging market, one must be aware of the types of IT resources and applications that companies can employ to understand better and satisfy the customers’ requirements and demands across all segments and markets [18]. An enterprise’s obstacle is not to develop SCMP’ constructs but to successfully implement them, as the future will see a SC war [15]. For those organizations that are not aware of the issues that may occur during the implementation of SCMP, such an initiative may fail. This chapter aims to address complex problems that could occur during the implementation of SCMP. SCM is a collection of system strategies for integrating vendors, inbound logistics, distributors, outbound logistics, and consumers to optimize the information exchange and cash flow performance of individuals and the SC as a whole [13]. SCM strategies are utilized to link upstream (supply of the raw material, subcomponent and subassembly, and manufacturing of the product) and downstream (outbound logistics and distribution) value chain entities. Successful SCM requires the integration of these value chain entities to create a cooperative and collaborative environment that provides the smooth exchange of information, physical and cash flows. SMEs can no longer effectively sustain themselves in the global or local market without forming a relationship with SC partners and effective information flow with SC partners. Supplier alliances and strategic collaborations relate to collective and more mutual relationships between companies and their upstream and downstream suppliers and consumers. Many companies have taken bold steps today to form partnerships, reduce uncertainty, and strengthen control of supply and distribution networks by breaking down both inter and intra-firm barriers. These alliances are traditionally formed to strengthen each channel member’s financial and operational efficiency by reducing overall costs and inventories and increasing knowledge sharing [3, 8]. Hence, SMEs seek to form a relationship with SC partners (suppliers, manufacturers, and IT professionals) and more useful information flow among partners [9]. So at this stage, this chapter addresses the following issues: RQ1—What are the different factors/processes to compete in unpredictable business environments for which SCM practices are developed in SMEs manufacturing enterprises? RQ2—What are the different SCMP that are developed/tried to develop for these factors/processes in SMEs manufacturing enterprises? RQ3—If developed SCMP constructs are not implemented, then what are the implementation issues of constructs?
Supply Chain Management Practice Constructs in SMEs …
51
RQ4—How these implementation issues of constructs can be resolved? The remainder of this chapter is organized as follows: Sect. 2 presents review methodology, Sect. 3 presents the literature review, provides the definition and theory underlying each construct of SCM practice and its link with requisite factors/processes. Section 4 presents the SMEs–SCMP constructs and their implementation issue and remedies. Section 5 presents the conclusions and future direction of the study.
2 Review Methodology To describe and classify the necessary factors for which equivalent SCMP are created. An extensive literature review was conducted. An extensive online search was undertaken to find similar publications to gather, organize, and incorporate existing SCMP and their implementation issues. SCMP implementation problems in Indian SMEs are also listed in the collected publications. The papers gathered here show the relationship between SCMP, operational, and organizational efficiency, and the key factors that influence SCMP implementation in SMEs. The study discussed in this chapter was mainly performed through the use of different forms of literature reviews. Taking the following electronic database into consideration: Over the 2004–2018 time period, scientific papers, journals, articles, government reports, and business reports from companies that allow access to works were published by Elsevier (Science Direct), Emerald Insight, Scopus, and Springer.
3 Literature Review SMEs must focus on improving operating functions and SC efficiency in order to compete successfully in the global market. SMEs can improve their operating efficiency and SC performance by effectively adopting SC practices. Successful SCM requires effective integration of suppliers, inbound logistics, manufacturers, outbound logistics, and customers to develop a cooperative, collaborating, and dynamic environment that come up with information flow, physical flow, and cash flow within the SC. Previous studies found that many requisite factors/processes will effectively integrate suppliers, logistics activities, manufacturing activities, and customer from an extensive literature review. For these requisite factors/processes, several equivalent SCMP constructs are developed within SMEs. Many practices are developed in the context of large enterprises, but less attention is paid to SMEs. Developed practices for large enterprises are tried to implement in SMEs without realizing the difference between SMEs and large firms. The primary differences between large firms and SMEs are in the scope of information flow and products
52
R. Kumar and M. Gupta
flow, implementation of IS practice, implementation of information and communication technology (ICT), and IT’s proper use. SMEs have a small budget in comparison to large enterprises. Hence, any theory applicable to large firms and other countries’ SMEs may need to be tested in the context of Indian SMEs [20, 23]. Requisite factors and equivalent supply chain management practice’ constructs are explained in Table 1. Fourteen supply chain management practice’ constructs are developed in the context of SMEs till now. These SCMP constructs provide effective integration of suppliers, inbound logistics, manufacturers, outbound logistics, Table 1 Requisite factors and equivalent supply chain management practice constructs S. no.
Factors
Equivalent Supply chain Management Practice constructs
1.
Develop Effective partnership with supplier
1. a) b) c) d)
Strategic Supplier selection(Few suppliers /many suppliers) The suppliers provide material at a low price The suppliers provide material of high quality The Suppliers have a low lead time The Suppliers have latest information technology implemented for information, Communication and collaboration purpose with organization managers and other supply Chain partners
2.
Develop Effective partnership with customers
3.
Develop Effective Information flow among trading Partners
4.
Develop cooperative and collaborative environment for trading partners
5.
Improved Communication
6.
Closeness of SMEs owner with trading partners
7.
Enhance visibility of supply chain and trading Partners
8.
Effective Strategic planning for sustainable growth
9.
Develop Effective physical flow among trading Partners
5. Quality of Information sharing a) Real time information exchange among trading partners is complete, accurate, and reliable
10.
Effective production planning and control
6. Postponement. a) Organization delay final product assembly or manufacturing until customer order have been received b) Organization design product for modular assembly
11.
Improved trust among trading partners
12.
Develop Optimized logistics and distribution network
7. a) b) c) d)
13.
Develop environment to Work on Low Inventory level
8. Just in time Supply system and TQM
14.
Reduce warehousing, purchasing and transportation cost
9. Outsourcing/ Sub contracting/Third party logistics (3PL)
15.
Improve service level
.
10. E procurement
16.
Employee Empowerment
1.
11. Strategic Planning
2. Strategic Supplier partnership a) Suppliers take part in product development program b) Suppliers take part in continuous improvement program of product and information technology c) Suppliers solve problems jointly with organizations manager/CEO d) Suppliers take part in strategic planning and goal setting activities 3. a) b) c)
Customer Relationship Customers take part in product development program Organization knows about the future expectations of customers Organization interacts with customer to set reliability and other standards d) Organization measure and evaluate customer satisfaction
4. Level of Information sharing a) Organization share information with trading partners about changing needs b) Organization share information about inventory level, production and delivery schedule c) Organization share performances metric with trading partners
Internal Lean practice(ILP) Organization reduces set up time of machine Pushes supplier for shorter lead time Suppliers” factory/warehouses are located nearby Organization order in small lot sizes from its supplier
Supply Chain Management Practice Constructs in SMEs …
53
and customers. There are six validated practices named as: Strategic supplier selection and partnership, customer partnership, level of information sharing, quality of information sharing, internal lean practices, and postponement [14, 15]. Five of the six validated constructs have been implemented in every organization, regardless of the industry’s size or product type. Postponement, on the other hand, is not included in every organization. Postponement depends on firm’s business characteristics and the type of goods it produces, so it may not be acceptable in all circumstances. Strategic supplier selection and partnership come up with the integration of suppliers with the organization. Jabbour [10, 11] revealed by presenting four constructs from the principal component analysis. In this study, the author had suggested six constructs of supply chain management practice (supply chain integration, information sharing, customer service management, customer relationship, supplier relationship, and postponement) and got four principal factors extracted from principal component analysis. These four principal factors are SC integration for production planning and control support, information sharing about products and targeting strategies, strategic relationship with customer and supplier, and support customer order. Supply chain integration for production planning and control consists of supply chain integration constructs. Information sharing about products and targeting strategies consists of customer support management, customer relationship, and supplier relationship. Strategic relationship with customers and suppliers combines the constructs of customer and supplier relationships and supports customer order combining constructs of supply chain integration, postponement, and information sharing. The organization can operate with few suppliers or many suppliers determined by inventory level, price of material, quality of material, and lead time [15]. Literature keeps up that the owner–manager of SMEs remains active in purchasing relationships. Owner–manager chooses the reliability of the suppliers as the most critical factor when selecting suppliers. Suppliers should take part in product development programs to know about the requirement of material properties for the product. Suppliers are commonly selected based on price, quality of material, and lead time. SMEs can select a supplier based on the best criteria (price, quality, lead time, and reliability) by using the Analytical hierarchy process (AHP) [9, 11]. Customer relationship management involves activities such as fulfilling potential customer expectations, monitoring customer history, assessing and analyzing customer satisfaction, and establishing long-term relationships with customers. Information sharing covers the practices of formal or informal data flow with trading partners. Information quality covers the practices about information flow quality, completeness, reliability, and accuracy. Internal lean practice covers the elimination of waste, low inventory level, small lot sizes, and JIT delivery. Postponement covers the practice related to modular assembly, delaying of final product assembly until the new customer orders have been received. A close relationship with suppliers concerns cooperation between organization and supplier, close relationship with customers concerns cooperation between organization and customers with adequate information flow and information quality. This is the starting point to establish a successful SCM and a necessary but insufficient condition.
54
R. Kumar and M. Gupta
The next level requires coordination and collaboration between buyers and suppliers [2]. This includes specified workflow, physical flow, sharing information through Electronic Data Interchange (EDI) and the internet and joint planning and other mechanisms that facilitate the just in time (JIT) system and TQM in the companies. Many approaches (e-procurement, strategic planning, JIT supply system, and TQM) have been developed to ensure cooperation and collaboration with suppliers. The JIT supply system and TQM practices are a series of activities designed to achieve high volume production while maintaining raw materials, work in progress, and finished product inventories to a minimal level. Strategic planning is a process focusing on manufacturing processes, technical innovation, financial consideration, and market penetration. The organization integrates strategies in each of these areas to produce and sell high-quality products at a low cost. Electronic Procurement (Eprocurement) is a virtual purchasing application that also enhances the visibility of data by leveraging supplier negotiation. It allows organizations to control their supplier hence reducing purchasing costs. Very often an E-procurement tool also interfaces with Enterprises Resources Planning (ERP) to automate many purchasing and payment tasks services. Outsourcing and subcontracting are related to a type of service of activities by an external party to accomplish related functions that are desired to be rendered or managed by the purchasing organizations. Third-party logistics (3PL) is a type of service of multiple activities by an external party to accomplish related functions that are desired to be rendered or managed by the purchasing organizations. The previous research stated that implementation of these SCM practices (strategic supplier selection and partnership, customer relationship, level of information sharing, quality of information sharing, internal lean practice, postponement, JIT supply system, TQM, Strategic Planning, E-procurement, outsourcing, subcontracting, Third-party logistics (3PL), holding stocks) leads to higher level of operational performance, and this argument can be generalized to all manufacturing firms regardless of their size [2, 15]. To achieve agility and improve operational efficiency, most SMEs incorporate these SCMP constructs in their SC. SMEs can enhance their operational efficiency by reducing lead times in production, inventory levels, better resource planning, and more effective costing as explained in Table 1. Sixteen requisite factors to compete in the unpredictable business environment require successful implementation of 14 SCMP as explained. These SCMP constructs, however, have not been successfully implemented in Indian SMEs. There are a number of key players/enablers that mainly influence the successful implementation of these constructs in Indian SMEs. These key players are responsible for restricting SCMP successful implementation in Indian SMEs. These key players and their roles are explained in Table 2. Top management is the first enabler on the list. The value of SC practices and their effect on SC and organizational success must be understood by top management. Top management can provide a well-thought-out plan for what to implement and how to implement it, and the information sharing practices should be implemented within SMEs for effective information flow among SC partners. Issues related to the adoption of SC are explained in Table 2.
Supply Chain Management Practice Constructs in SMEs …
55
Table 2 Supply chain management practice constructs implementation issue S. no. 1.
Factors/ SCMP constructs Develop Effective partnership with supplier.
2.
Develop Effective partnership with customers.
3.
Develop Effective Information flow among trading Partners.
4.
Develop Cooperative and collaborative Environment for trading partners.
5.
Improved Communication
6.
Closeness of SMEs owner with trading partners.
7.
Enhance visibility of supply chain and trading Partners. Effective Strategic planning for sustainable growth Develop Effective physical flow among trading Partners.
8. 9.
10.
1. Top management Support.
2. Suppliers/customers support
3.SMEs culture .
4.SMEs structure and governance 5. Information Sharing practice
Effective production planning and control
11.
Improved trust among trading partners
12.
Improve service level
13.
Develop Optimized logistics and distribution network Employee Empowerment.
14.
SMEs’ Key players/Enablers
15.
Develop environment to Work on Low Inventory level
16.
Reduce warehousing, purchasing and transportation cost
. 6. Information and Communication Technology.
Issue Don’t have any strategic planning for implementing SCMP in SMEs Not able to shape the supply chain decisions based on the past experience and educational background Not able to realize where IS practice and ICT having greatest impact on supply Chain Not able to conduct training and development programs for Employee Do not take part in new product development programs and support inventory management functions Do not invest on implementation of IS practice and ICT to improve their supply chain transparency and visibility Do not facilitates technology to know about degree of information transparency, knowledge sharing, and collaboration Not able to create a new culture that enables employee to adapt and learn constantly about SCMP and technology. Do not facilitate high level cross department collaboration. Skills shortage for successful implementation of Information sharing practice Resources shortage Resistance to change from employees Integration issue with supplier, customer and existing systems. Financial constraints. Skills shortage for successful implementation of Information and communication technology. Lack of knowledge of ICT Resources shortage Resistance to change from employees. Integration issue with supplier, customer and existing systems. Financial constraints. Lack of government support and policies. Security and privacy issues
4 SMEs–SCMP Implementation Issues and Remedies This section seeks issues and enablers related to the implementation of SCMP in Indian SMEs. Implementation of SCMP is highly dependent on organizations’ inter and intra linkage. In SMEs, the size and budget constraints of SMEs restrict the
56
R. Kumar and M. Gupta
adoption of information sharing (IS) practice, ICT, and employee training. Implementation of IS practice and ICT can enhance the SC’s visibility and develop effective and efficient information flow among trading partners and provide an integrated SC. The pillars of SC integration are cooperation, collaboration, information sharing, trust, partnership, shared technology, and a fundamental shift away from managing individual functional processes to managing integrated chains of processes. The integration of logistics with other functional areas will help the organization to realize the full potential of its value-added activities and hence, to gain a significant competitive advantage [22]. The growth of SMEs is highly dependent on financial condition, industrial factors such as level of demand and the intensity of competition, and internal factors such as leadership and managerial skills of the owner, skills of the employees, organization culture, organization structure, and governance (Oswald & Kleinemeier, 2016). Implementation of IS practice, ICT; and top management support system can develop effective SC integration. But there are some issues related to IS practice, ICT, and top management support that are mentioned in Table 2. Top management should be aware of the most significant impact of SC practice on organizational performance. Top management must pay attention to strategic planning for the successful implementation of SCMP. The strategy is a prerequisite for implementing SCM in any firm. The strategy is crucial since it gives direction to any implementing program. Strategy as issue: SCM is a set of system approaches utilized to effectively integrate suppliers, inbound logistics, manufacturers, outbound logistics, and customers for improving the information flow and cash flow performance of the individuals and the supply chain as a whole. Developing a strategy for SCM becomes a complicated task in itself. The major issue for top management at this stage is to develop an appropriate strategy for the successful implementation of SCMP. Development of an appropriate strategy Before developing an appropriate strategy, top management should assess their own organization and SC capabilities [23]. This is a prerequisite step because a firm’s SCM strategy would depend on its organization and SC capabilities. A model for selfassessment based on organization structure and governance, organization culture, IS practice, ICT, usage of IT, transportation, and warehousing system can be used. The strategy must focus on four key elements: people (employees, supplier, customer, IT team, and 3PL), organization (culture, leadership, structure, and governance), systems, and technological processes (IS practice, ICT, digital technology). Type of product, product life cycle, type of industry, and dynamic environment mainly affect the formulation process of SCM strategy. Top management should have these four factors in mind while assessing the SC capabilities and formulating an appropriate strategy [21, 23]. Table 2 discusses how to incorporate SCM activities due to supplier/customer support, SME culture, and SME structure and governance. Adoption issues of IS practice and ICT: Implementation of IS practice and ICT can develop effective SC integration. IS practice and ICT are also crucial for inbound logistics, outbound logistics, and reverse logistics. The information system is crucial for achieving effective logistics activities. The information system must be
Supply Chain Management Practice Constructs in SMEs …
57
responsive in order to anticipate and accommodate operational change and customer demand. The primary concern of top management at this point is to consider the effect of IS and ICT on the practical implementation of SCM activities, and the secondary concern is whether IS and ICT should be incorporated within SMEs. Since SMEs have a smaller budget than large companies, top management should examine whether SMEs require IS Practice and ICT [6]. Handling Issues Information sharing practices such as enterprise resources planning (ERP), material requirements planning (MRP), manufacturing resources planning (MRPII), and electronic data interchange (EDI), and RFID have a positive impact on operational performance [5]. Before implementing the IS practice and ICT, top management should know where IS practice and ICT having the greatest impact on SCMP. Top management should aware about the implementation barriers of IS practice and ICT in SMEs. Many challenges come in the way of a successful implementation of informationsharing practices. The critical hurdle to SMEs implementing information-sharing is a lack of competence and tools for effectively sharing the information about the inventory levels and production schedule. These barriers are mainly categorized into five categories namely: organization barriers (organization infrastructure, organization structure and governance, employee skills), security and privacy barriers, technology foundation-related barriers, SME budget barriers, and government support and policy [2, 17, 18]. Implementing information and communication technology creates major improvements in business processes and supply chain operations. As a result, SMEs must concentrate on enhancing their skills in terms of employee training and growth and knowledge management systems. Management commitment and support are needed for ICT adoption [16]. Top management support and dedication toward implementing IS and ICT for the betterment of the SC operations. Top management should understand: • About the need of urgency of IS and ICT in SMEs. • Where is the change occurring, and where does the implementation of IS and ICT have the most significant impact on SC operations? • What is the focus of IS and ICT initiatives? (to improve customer experience and engagement, improve decision making, improve innovation, improve engagement and collaboration with employees, suppliers, and business partners). Supply chain partners should be prepared to adopt IS and ICT for better and realtime information exchange on inventory status and delivery schedules. Adopting IS, and ICT in SMEs needs investment, and suppliers to SMEs are generally small fragmented players that carry less motivation for investment.
58
R. Kumar and M. Gupta
5 Conclusions This chapter’s most significant impact will be on the academicians who intend to investigate SCMP’ construct implementation in SMEs. This research chapter presents some insight generated from the facts for developing SCMP constructs and their implementation issue. • Five SCM strategies are incorporated to establish a successful relationship with suppliers and consumers. Strategic supplier selection (few vs. many suppliers), strategic supplier partnership, customer relationship, level of information sharing, and quality of information sharing are the five strategies. Hence, two requisite factors to compete in the unpredictable business environment: successful relationships with suppliers and consumers require successful implementation of these five SCMP. • Suppliers and customers must engage in the product development program and the performance improvement program of information technology use in order to have effective relationships with them. • Indian SMEs need to share information with SC partners about changing needs, inventory level, production, and delivery schedules to have effective relationships with suppliers and customers. Real-time information exchange among SC partners needs to be complete, accurate, and reliable. • Sixteen requisite factors to compete in the unpredictable business environment require the successful implementation of 14 SCMP. • SMEs have developed 14 SCMP (Strategic supplier selection and partnership, customer relationship, level of information sharing, quality of information sharing, internal lean practice, postponement, JIT supply system, TQM, Strategic Planning, E-procurement, outsourcing, subcontracting, Third-party logistics (3PL), holding stocks) to improve their performance in the competitive market. These 14 developed practices positively impact SC performance in terms of SC flexibility, SC integration, SC efficiency, and responsiveness to customers. • After successfully implementing these 14 constructs, SMEs (B2C) can establish better relationships with suppliers, customers, and other partners. On the other hand. SMEs (B2B) can establish better relationships with original equipment manufacturers (OEM) or large enterprises, hence improving their learning curve. • After successfully implementing these 14 constructs, SMEs can improve their supply chain performance in terms of supply chain integration, flexibility, and efficiency. • Information sharing practices such as enterprise resources planning (ERP), material requirements planning (MRP), manufacturing resources planning (MRPII), and electronic data interchange (EDI), and RFID have a positive impact on operational performance. • The primary concern of top management is to consider the effect of IS and ICT on the practical implementation of supply chain management activities, and the secondary concern is whether IS and ICT should be incorporated within SMEs.
Supply Chain Management Practice Constructs in SMEs …
• •
•
• • •
59
Since SMEs have a smaller budget than large companies, top management should examine whether SMEs require IS Practice and ICT. Formulating strategic planning, implementing IS practice and ICT, and managerial support is a prerequisite step for successfully implementing supply chain management practice’ constructs in SMEs. The strategy must focus on four key elements: people (employees, supplier, customer, IT team, and 3PL), organization (culture, leadership, structure, and governance), systems, and technological processes (IS practice, ICT, digital technology). Top management should aware about the implementation barriers of IS practice and ICT in SMEs. These barriers are mainly categorized into five categories, namely: organization barriers (organization infrastructure, organization structure and governance, employee skills), security and privacy barriers, technology foundation-related barriers, SME budget barriers, and government support and policy. Top management support and vision toward the implementation of information sharing practice and ICT helps SMEs to adopt IS and ICT at the right time. The government can raise the awareness of ICT by organizing a conference, publishing papers, and hosting a workshop on the topic. An ICT-based SC in SMEs can instantaneously collect information about product flow, financial flow, and production system schedule and efficiently access these data to all the SC partners. An ICT-based SC in SMEs can record and analyze the customer and distributor history in real-time and prognosticate customers’ future demand. Hence, by formulating strategic planning, implementing IS practice (EDI, ERP, MRP, MRPII, RFID), and ICT help SMEs to successfully and efficiently implement SCMP and smoothen their information flow, cash flow, and physical flow by improving their SC efficiency and effectiveness.
References 1. Ali A, Husain F (2014) MSME’S IN INDIA: problems, solutions and prospectus in present scenario. Int J Eng Manag Sci 5(2):109–115 2. Bayraktar E, Demirbag M, Koh SL, Tatoglu E, Zaim H (2009) A causal analysis of the impact of information systems and supply chain management practices on operational performance: evidence from manufacturing SMEs in Turkey. Int J Prod Econ 122(1):133–149 3. Chong AYL, Chan FTS, Ooi KB, Sim JJ (2011) Can Malaysian firms improve organizational/innovation performance via SCM? Ind Manag Data Syst 111(3):410–431 4. Cuthbertson R, Piotrowicz W (2008) Supply chain best practices–identification and categorisation of measures and benefits. Int J Product Perform Manag 57(5):389–404 5. Fasanghari M (2008) Assessing the impact of information technology on supply chain management. In: 2008 International symposium on electronic commerce and security. IEEE, pp 726–730 6. Gandhi NS, Thanki SJ, Thakkar JJ (2018) Ranking of drivers for integrated lean-green manufacturing for Indian manufacturing SMEs. J Clean Prod 171:675–689
60
R. Kumar and M. Gupta
7. Gawankar SA, Kamble S, Raut R (2017) An investigation of the relationship between supply chain management practices (SCMP) on supply chain performance measurement (SCPM) of Indian retail chain using SEM. Benchmarking Int J 24(1):257–295 8. Gunasekaran A, Patel C, McGaughey RE (2004) A framework for supply chain performance measurement. Int J Prod Econ 87(3):333–347 9. Hong J, Zhang Y, Ding M (2018) Sustainable supply chain management practices, supply chain dynamic capabilities, and enterprise performance. J Clean Prod 172:3508–3519 10. Jabbour ABL, Alves AG, Viana ABN, Jabbour CJC (2011) Factors affecting the adoption of supply chain management practices: evidence from the Brazilian electro-electronic sector. IIMB Manag Rev 23(4):208–222 11. Jabbour AB, Viana ABN, Jabbour CJC (2011) Measuring supply chain management practices. Meas Bus Excell 15(2):18–31 12. Khanna R (2018) Status of MSMEs in India: a detailed study. J Appl Manag Jidnyasa 10(2):1–14 13. Koh SL, Demirbag M, Bayraktar E, Tatoglu E, Zaim S (2007) The impact of supply chain management practices on performance of SMEs. Ind Manag Data Syst 107(1):103–124. https:// doi.org/10.1108/02635570710719089 14. Li S, Rao SS, Ragu-Nathan TS, Ragu-Nathan B (2005) Development and validation of a measurement instrument for studying supply chain management practices. J Oper Manag 23(6):618–641 15. Li S, Ragu-Nathan B, Ragu-Nathan TS, Rao SS (2006) The impact of supply chain management practices on competitive advantage and organizational performance. Omega 34(2):107–124 16. Luthra S, Mangla SK (2018) Evaluating challenges to Industry 4.0 initiatives for supply chain sustainability in emerging economies. Process Saf Environ Prot 117:168–179. https://doi.org/ 10.1016/j.psep.2018.04.018 17. Novais L, Maqueira JM, Ortiz-Bas Á (2019) A systematic literature review of cloud computing use in supply chain integration. Comput Ind Eng 129:296–314 18. Oswald G, Kleinemeier M (2017) Shaping the digital enterprise: trends and use cases in digital innovation and transformation.https://doi.org/10.1007/978-3-319-40967-2 19. Srinivasan M, Mukherjee D, Gaur AS (2011) Buyer–supplier partnership quality and supply chain performance: moderating role of risks, and environmental uncertainty. Eur Manag J Els 29(4):260–271. https://doi.org/10.1016/j.emj.2011.02.004 20. Thakkar J, Kanda A, Deshmukh SG (2008) Supply chain management in SMEs: development of constructs and propositions. Asia Pac J Mark Logist 20(1):97–131. https://doi.org/10.1108/ 13555850810844896 21. Thakkar J, Kanda A, Deshmukh SG (2009) Supply chain performance measurement framework for small and medium scale enterprises. Benchmarking Int J 16(5):702–723 22. Thakkar J, Kanda A, Deshmukh SG (2012) Supply chain issues in Indian manufacturing SMEs: insights from six case studies. J Manuf Technol Manag 23(5):634–664 23. Varma S, Wadhwa S, Deshmukh SG (2006) Implementing supply chain management in a firm: issues and remedies. Asia Pac J Mark Logist 18(3):223–243. https://doi.org/10.1108/135558 50610675670
Analysis of QMS Practices Performed in ISO 9001 Certified Engineering Educational Institutes of India: An Interpretive Structural Modelling Approach Parvesh Kumar, Sandeep Singhal, and Jimmy Kansal Abstract This study was carried out to investigate and explore the ISO 9001-based QMS dimensions, analyze interrelationships and their combined impact on engineering education in India using interpretive structural modelling (ISM) methodology. In this paper, 20 quality dimensions/elements are identified through Delphi technique to construct an ISM-based model. This approach helps in establishing the hierarchy of these quality dimensions, which will assist to enhance the quality of engineering education (EE). The objective of this article is to frame a blueprint of QMS implementation and evaluation of QMS dimensions for continuous quality improvement of engineering education (EE). The results give a feasible solution that the dimensions such as quality mission and vision (QM&V) statement and top management commitment and leadership (TMCL) are the main driving powers for establishing an effective QMS in engineering institutes. The limitation of the study is that the proposed ISM model has not been tested and validated statistically. Therefore, in the future, the model needs to be tested and validated statistically using different structural approaches. Keywords Quality management systems (QMS) · Interpretive structural modelling (ISM) · Engineering education (EE) · Quality dimensions · Engineering institutions (EI)
1 Introduction Today’s economies are dominated by service industries; In general, India’s GDP has a higher percentage of services than the manufacturing sector. The accelerated increase in this zone comes mostly from the pressures of liberalization and privatization. For
P. Kumar · S. Singhal (B) Department of Mechanical Engineering, National Institute of Technology, Kurukshetra, India J. Kansal Snow & Avalanche Study Establishment, DRDO, Chandigarh, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_7
61
62
P. Kumar et al.
instance, the education, health care, tourism, telecom, transport and information technology sector have experienced stronger growth over the last decade [6]. In specific, the engineering sector in India is undergoing a revolution with the exposure of a completely modern category of education providers, comprising private organizations, distance learning providers, self-financed government institutions, overseas education providers, etc. In today’s trade, merely meeting or surpassing previous performance won’t give the extent of improvement necessary to stay competitive [6]. Despite some attractive empirical studies in the field of engineering education (EE), the critical dimensions of quality have not been fully described. To enhance the quality of education, professional and educational institutions should adopt a course of action that accredited institutions around the world can follow [8]. The autonomy to formulate strategies, make decisions, develop a curriculum and implement rigorous accreditation concepts to modify teaching–learning into a university is an important criterion [1]. The syllabus propound in most of the engineering institutions in India does not meet the global criterion. Hence, the difficulty arises from the absence of comprehensive and existing QM models in engineering institutions (EI) [9]. The limited number of state-funded technical institutes in India and the withdrawal of government funds in this area have led to the development of private institutions. As a result, the stakeholders involved have a broad radius of options to select from [5]. As an effect, it is necessary to acquire changes in educational procedures to thrive a well education system. In such a vital domain, the implementation of a QMS appears to be one of the strongest tools to address not only market provocations but also stakeholders. In this context, this article attempts to study the implementation of a quality management system (QMS) in the EE sector in the Indian outline, with an approach to assess its relationship with performance.
2 Research Background Quality improvement techniques need to be applied in all service-oriented organizations/firms. These techniques play a vital role in fulfilling the demands of the users of the service. Like other service-oriented organizations/firms, the educational institutions also aspire to confront the wishes of the users [6]. In the present time’s exigent environment, the engineering institutes must provide quality education with appurtenant knowledge along with viable skills and that too within the limited or restricted budget. In this situation, the features of QMS fit effectively as they inculcate the longing for continuous improvements, such as self-improvement, work improvement, society improvement, etc. [7]. The implementation of QMS is mainly deep-rooted in the industries, but in modern times, there has been a strong usage to adopt the practices of QMS in educational institutions as well. The first step in implementing QMS in an institute is to adopt a framework that corresponds to its ultimate goal of continuous quality improvement.
Analysis of QMS Practices Performed in ISO 9001 Certified …
63
Six core values that form the backbone of the QMS are leadership/administration, educational management, HRM, data management, customer satisfaction; a system approach to management was identified. [4] Identified some “quality indicators” and organized the importance of quality indicators for evaluating the performance of technical institutes in India. Author examined the relationship between the organizations’ performance and the dimensions of quality management. They featured the five critical factors, i.e. knowledge management, leadership, human resource management, customer satisfaction, top management commitment, which show there is a notable correlation between the key elements of QM and the performance of institute [3]. Singh examined the ten critical factors of quality management and their correlation in engineering education (EE) and emphasized that the most critical dimension for building a productive QMS in engineering institutes (EI) is the commitment of the top management and quality vision [11]. Researchers vigorously recommended that a blend of these quality dimensions to be managed for future analysis on QMS. The current research strives to merge quality management dimensions proposed by different authors for conferring a QMS execution model for engineering educational institutes (EEI). Twenty leading quality concepts were secured using Delphi technique (Table 1). It incorporates brainstorming with a team of 40 professionals, which includes researchers, professors, scholars, academicians and stakeholders. However, the authors have suggested the critical dimensions for a constructive QMS execution in the field of engineering institutes (EI), but they were failed to construct a structure for its successful execution. The author recommended that to enhance the QMS fulfilment, quality management system beliefs should be prioritized [2]. The interpretive structural modelling methodology helps in quantitatively prioritizing strategic problems in quality evaluation and also offers a perception of correlation that enables to find out the driver and dependent powers. That is why the interpretive structural modelling–ISM methodology is applied to set the hierarchy of selected quality dimensions to enhance the quality of engineering education.
3 Research Methodology and Data Analysis The logic of interpretive structural modelling approach selected for this research will be discussed in this section. The first step is to select the quality management dimensions pertinent to the research problem. In the second step, a relationship and dependency are identified between the selected quality parameters/dimensions. After identifying the relationship and dependency, a structural self-interaction matrix— SSIM is generated depending upon the collation of the QMS dimensions/parameters. Next, the reachability matrix (RM) is developed with the help of structural selfinteraction matrix, and transitivity test is done. After completing the transitivity test, the well-structured matrix model is generated. In last, separation of the constituents is completed, and final structural model is developed. Steps involved in interpretive structural modelling methodology are as follows [7]:
64
P. Kumar et al.
Table 1 Quality management dimensions for engineering education
Serial number
Quality management system dimensions
1
Top management commitment and leadership (TMCL)
2
Policy and strategy (PS)
3
People management (PM)
4
Institutional resource management (IRM)
5
Human resource management (HRM)
6
System approach to management (SAM)
7
Information management system (IMS)
8
Employee involvement (EI)
9
Team work (TW)
10
Academic culture (AC)
11
Service culture (SC)
12
Continuous improvement (CI)
13
Customer satisfaction (CS)
14
Industry institution partnership (IIP)
15
Innovative academic philosophy (IAP)
16
Society satisfaction (SS)
17
Quality mission and vision statement (QM&V)
18
People results (PR)
19
Infrastructure management (IM)
20
Institute results (IR)
S-1 Recognizing the quality dimensions relevant to the research problem. Delphi approach will be used for this purpose [10]. S-2 Developing a specific relation among the various QMS dimensions/parameters for dependency check. S-3 Establishing a structural self-interaction matrix of the QMS dimensions, which will describe the set-wise interrelationship among the quality dimensions. S-4 Establishing a reachability matrix (RM) using structural self-interaction matrix and transitivity test will be done. Now with the help of RM, an interpretive structural modelling (ISM) initial diagraph is formed. S-5 Subdividing the RM into ranking of levels in accordance to reachability and antecedent set for each quality dimension. S-6 Final digraph will be formed using reachability matrix (RM) after separating the transitivity relatedness. S-7 Now develops an interpretive structural modelling (ISM) model using final digraph. S-8 Finally, analyze the variability and modify the ISM model if required.
Analysis of QMS Practices Performed in ISO 9001 Certified …
65
3.1 Structural Self-Interaction Matrix (SSIM) The interpretive structural modelling (ISM) approach proposed the use of specialist beliefs based on management methods in the establishment of a logical relationship between quality dimensions/parameters. The specialist team determined the nature of logical relationship between the 20 QMS dimensions in this study (Table 2). Symbols used to designate the relationship between the dimensions/parameters are as follows: • • • •
V Dimension i will help to achieve dimension j; (V: i → j) A Dimension j will help to achieve dimension i; (A: i ← j) X Dimensions i and j will help to achieve each other; ( X: i ↔ j) and O Dimensions i and j are unrelated; (O: i↨j).
Table 2 Structural self-interaction matrix (SSIM) Dimensions 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1
TMCL
X V V V V V V V V V
V
V
V
V
V
V
A
V
V
V
2
PS
A X V V A V V V V V
A
V
V
V
V
V
A
V
V
V
3
PM
A A X A A A A A A A
A
V
A
A
A
A
A
V
A
A
4
IRM
A A V X A V V V V V
A
V
V
V
O
V
A
V
V
V
5
HRM
A A V V X V V V V V
A
V
V
V
V
V
A
V
V
V
6
SAM
A A V V A X V V V V
A
V
V
V
V
V
A
V
V
V
7
IMS
A A V V A A X V V V
A
V
V
V
V
V
A
V
V
V
8
CI
A A A A A A A X A A
A
V
A
A
O
V
A
O
O
V
9
V
TW
A A V V A V V V X V
A
V
V
V
A
V
A
V
V
10 AC
A A V A A A A A A X
A
V
A
A
A
V
A
V
V
V
11 SC
A V V V V V V V V V
X
V
V
V
V
V
A
V
V
V
12 CI
A A A A A A A A A A
A
X
A
A
A
A
A
A
A
A
13 CS
A A A A A A A A A A
A
V
X
A
A
A
A
V
V
V
14 IIP
A A A A A A A A A A
A
V
A
X
A
V
A
V
V
V
15 IAP
A A V O A V V V V V
A
V
V
V
X
V
A
V
V
V
16 SS
A A A A A A A O A A
A
V
A
A
A
X
A
O
O
A
17 QMVS
V V V V V V V V V V
V
V
V
V
V
V
X
V
V
V
18 PR
A A A A A A A A A A
A
V
A
A
O
V
A
X
O
V
19 IM
A A A A A A O O A A
A
V
A
A
O
A
A
O
X
V
20 IR
A A A A A A A A A A
A
V
A
A
A
A
A
V
V
X
66
P. Kumar et al.
3.2 Reachability matrix Structural self-interaction matrix arrangement (V, A, X, O) is converted into a binary matrix (1 s and 0 s), referred to as the initial reachability matrix (IRM). The basic rules for the conversion of 1 s and 0 s are as follows: • If V is entered within the (i, j) of structural self-interaction matrix SSIM, then the (i, j) entry within the reachability matrix RM becomes one (1) and the (j, i) entry becomes zero (0). • If A is entered within the (i, j) of structural self-interaction matrix SSIM, then the (i, j) entry within the reachability matrix RM becomes zero (0) and the (j, i) entry becomes one (1). • If X is entered within the (i, j) of structural self-interaction matrix SSIM, then the (i, j) entry within the reachability matrix RM becomes one (1) and the (j, i) entry also becomes (1). • If O is entered within the (i, j) of structural self-interaction matrix SSIM, then the (i, j) entry within the reachability matrix RM becomes zero (0) and the (j, i) entry also becomes zero (0). In this case, there is no transitivity so the initial reachability matrix—IRM is used to establish the final reachability matrix—FRM (Table 3). Driving power—DRP can be defined as collective value of the each critical parameter evaluated horizontally, which can help others to achieve it. Dependence power— DPP can be defined as the collective value of the each critical parameter evaluated vertically, which can help itself but not others. From Table 3, it is depicted that the quality dimensions are levelled based on the driving power—DRP. The values of the dependent power—DPP evaluated in respect of 20 criterions are used to construct ISM digraph.
3.3 Level Partitions The reachability and antecedent set for each block is established with the help of final reachability matrix—RM. The antecedent set found in this section comprises of the quality dimension itself and the dimension that it may help achieve. Next, the intersection set is developed for all quality dimensions. After locating the peaklevel quality dimension in the ISM hierarchy, it will be eliminated from the set of remaining dimensions. From Table 4, it is observed that continuous improvement— CI (dimension12) is at level I. Similarly, the levels of all the remaining quality dimensions/parameters are identified. These levels and rankings will help in the construction of interpretive structural modelling—ISM-based final model.
1
TMCL
PS
PM
IRM
HRM
SAM
IMS
CI
TW
AC
SC
CI
CS
IIP
IAP
SS
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
cx
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
1
2
Table 3 Reachability matrix
0
1
0
0
0
1
1
1
0
1
1
1
1
1
1
1
3
0
0
0
0
0
1
0
1
0
1
1
1
1
0
1
1
4
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
1
5
0
1
0
0
0
1
0
1
0
0
1
1
1
0
1
1
6
0
1
0
0
0
1
0
1
0
1
1
1
1
0
1
1
7
0
1
0
0
0
1
0
1
1
1
1
1
1
0
1
1
8
0
1
0
0
0
1
0
1
0
1
1
1
1
0
1
1
9
0
1
0
0
0
1
1
1
0
1
1
1
1
0
1
1
10
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
11
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
12
0
1
0
1
0
1
0
1
0
1
1
1
1
0
1
1
13
0
1
1
0
0
1
0
1
0
1
1
1
1
0
1
1
14
0
1
0
0
0
1
0
0
0
1
1
1
0
0
1
1
15
1
1
1
0
0
1
1
1
1
1
1
1
1
0
1
1
16
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
17
0
1
1
1
0
1
1
1
0
1
1
1
1
1
1
1
18
0
1
1
1
0
1
1
1
0
1
1
1
1
0
1
1
19
0
1
1
1
0
1
1
1
1
1
1
1
1
0
1
1
20
(continued)
2
13
6
5
1
18
7
14
4
14
15
16
14
3
16
19
Driving Power (DRP)
Analysis of QMS Practices Performed in ISO 9001 Certified … 67
PR
IM
IR
18
19
20
Dependence power (DPP)
1
QMVS
17
2
0
0
0
1
cx
Table 3 (continued)
4
0
0
0
1
2
12
0
0
0
1
3
9
0
0
0
1
4
4
0
0
0
1
5
9
0
0
0
1
6
10
0
0
0
1
7
11
0
0
0
1
8
10
0
0
0
1
9
10
0
0
0
1
10
3
0
0
0
1
11
20
1
1
1
1
12
11
0
0
0
1
13
11
0
0
0
1
14
8
0
0
0
1
15
14
0
0
1
1
16
1
0
0
0
1
17
16
1
0
1
1
18
15
1
1
0
1
19
17
1
1
1
1
20
4
3
3
20
Driving Power (DRP)
68 P. Kumar et al.
1, 17
1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20
2, 3, 4, 6, 7, 8, 9, 10, 12, 13, 14, 15, 16, 18, 19, 20
3, 12, 18
3, 4, 6, 7, 8, 9, 10, 12, 13, 14, 16, 18, 19, 20
3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 15, 16, 18, 19, 20
3, 4, 6, 7, 8, 9, 10, 12, 13, 14, 15, 16, 18, 19, 20
3, 4, 7, 8, 9, 10, 12, 13, 14, 15, 16, 18, 19, 20
8, 12, 16, 20
1
2
3
4
5
6
7
8
2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 19, 20
12
12, 13, 18, 19, 20
12, 14, 16, 18, 19, 20
3, 6, 7, 8, 9, 12, 13, 14, 15, 16, 18, 19, 20
12, 16
1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 17 20
12, 18, 20
12, 19, 20
12, 18, 19, 20
11
12
13
14
15
16
17
18
19
20
10
9
4, 7, 9, 10
8
4, 7, 9, 15
4, 6, 9, 15
5
4, 6, 7, 9
3
2
1, 17
1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 17, 18, 19, 20
1, 2, 4, 5, 6, 7, 9, 10, 11, 13, 14, 15, 17, 19, 20
1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 13, 14, 15, 17, 18, 20
1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 14, 15, 16, 17
1, 2, 5, 6, 7, 11, 15, 17
1, 2, 4, 5, 6, 7, 9, 11, 14, 15, 17
1, 2, 4, 5, 6, 7, 9, 11, 13, 15, 17
18, 19, 20
19, 20
18, 20
17
16
15
14
13
II
IV
III
XIV
V
X
VII
VII
I
XII
VIII
VIII
VII
VIII
IX
XI
IX
VI
XI
XIII
Intersection Level
1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 12 20
1, 10, 17
1, 2, 4, 5, 6, 7, 9, 10, 11, 17
10
1, 2, 4, 5, 6, 7, 9, 10, 11, 15, 17
3, 4, 6, 7, 8, 9, 10, 12, 13, 14, 16, 18, 19, 20
3, 9, 12, 16, 18, 19, 20
9
1, 2, 4, 5, 6, 7, 8, 9, 11, 15, 17
1, 2, 4, 5, 6, 7, 9, 11, 15, 17
1, 2, 4, 5, 6, 9, 11, 15, 17
1, 5, 11, 17
1, 2, 4, 5, 6, 7, 9, 11, 17
1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 15, 17
1, 2, 11, 17
Antecedent set
Dimensions Reachability set
Table 4 Iteration matrix
Analysis of QMS Practices Performed in ISO 9001 Certified … 69
70
P. Kumar et al.
4 Results and Discussions The present study shows that the QMS dimensions (dimension 17, dim.1, dim. 11, dim. 6, dim. 9, dim. 2, dim. 5, dim. 7, dim. 15 and dim 4) having sturdy driving powers (DRP) are in the bottom and belongs to cluster IV, which drives the other quality dimensions towards quality. The quality dimensions (dim. 14, dim. 13, dim. 8, dim. 20, dim. 3, dim. 19, dim. 18, dim. 16 and dim. 12) belong to cluster II, having less value of DRP and can simply be controlled by the quality dimensions of strong DRP. Hence, these quality dimensions were able to find a higher place in the ISM digraph. The dimension 10 (AC) is autonomous so require a different path from dimension 17 (QMVS) to reach dimension 13 (CS) for the achievement of institutional goals.
4.1 Categorization of QMS Dimensions In this section, four clusters (C-1, C-2, C-3 and C-4) are framed to categorize the 20 QMS dimensions explained earlier. It is observed from Fig. 1 that the four clusters are divided on the basis of driving power (DRP) and dependence power (DPP) of the QMS dimensions.
Fig. 1 Clusters of driving and dependence power
Analysis of QMS Practices Performed in ISO 9001 Certified …
71
Fig. 2 Interpretive structural modelling—ISM-based digraph
Cluster 1: It contains the dimensions of weaker DRP and DPP, known as autonomous dimensions. In this case, dimension 10 (AC) is observed as autonomous dimension and is shown separately in the ISM final digraph (Fig. 2). Cluster 2: This cluster is having dimensions with weak DRP and strong DPP and consists of dependence dimensions. In this case, dimensions 3, 8, 12, 13, 14, 16, 18, 19, 20 are dependence dimensions and are placed at the peak position of the ISM final model. Cluster 3: The dimensions of this section are known as linkage dimensions having strong DRP and strong DPP. The dimensions of this cluster are totally interdependent on each other so are unstable. In the present case, no linkage dimension is observed. Cluster 4: This section contains the most critical quality dimensions, which are known as driving dimensions. Because of the ability of strong DRP and weak DPP, they contain independent dimensions, which drive the whole model. In this case, dimensions 1, 2, 4, 5, 6, 7, 9, 11, 15, 17 are observed as independent dimensions.
72
P. Kumar et al.
4.2 ISM-Based Final Model It is observed from Fig. 3 that the quality mission and vision statement (QMVS) is ranked level XIV and is the base of the ISM model, which drives the other quality dimensions such as top management commitment and leadership (TMCL), policy and strategy (PS), HRM, IMS. Consequently, TMCL supports and drives system approach to management (SAM) and Team Work (TW). The TW drives IR and IR drives back to reach customer satisfaction CS. Furthermore, the IMS has a direct capacity to drive IAP and IRM and EI. It is clear from the interpretive structural modelling-based final model that the QMVS along with TMCL has the ultimate and strong DRP to achieve quality directly through PM, IIP, SC, PS, HRM, SS, PR, EI, IM. CI can be sustained through the process of AC and CS.
Fig. 3 ISM-based final model
Analysis of QMS Practices Performed in ISO 9001 Certified …
73
5 Conclusion In this study, a model based on interpretive structural modelling has been developed, which helps to determine the sequence of operations that should be performed to upgrade the quality of engineering education (EE). Delphi approach is used to recognize the 20 quality dimensions for the successful implementation of QMS in EE. It is observed that for the enhancement of quality, the dimensions that have a higher value of driving power should be focused more than the dependence power. The quality mission and vision statement (QMVS) (dimension-17) has the highest value of driving power for an institute and hence it forms the basis for the implementation of QMS. The results indicate that the dimensions like society satisfaction (dimension-16), people results (dimension-18), employee involvement (dimension8) and infrastructure management (dimensions-19) are dependent on many other dimensions. These quality dimensions represent a critical result for top management (TM). Therefore, TM must pay special attention to work with these QMS parameters. Finally, quality mission and vision statement (QMVS) dimension-17, top management commitment and leadership (TMCL) dimension-1, people management (PM) dimension-3, team work (TW) dimension-9, system approach to management (SAM) dimension-6, service culture (SC) dimension-11, policy and strategy (PS) dimension-2, human resource management (HRM) dimension-5, information management system (IMS) dimension-7, innovative academic philosophy (IAP) dimension-15 and institutional resource management (IRM) dimension-4 are driver variables, which indicates that the practical viewpoint and ordered work environment in the institution/organization can help in attaining quality goals. The study shows that apart from the industries and other service organizations, interpretive structural modelling (ISM) methodology is applicable to QMS of engineering education (EE).
References 1. Das K (2019) Integrating lean, green, and resilience criteria in a sustainable food supply chain planning model. Int J Math Eng Manag Sci (IJMEMS) 4(2):259–275 2. Kansal J, Singhal S, Khurana A (2017) Design of an instrument to measure the performance of ISO 9001 based quality management systems in knowledge based organizations. J Sci Ind Res (JSIR) 76(12):767–770 3. Kansal J, Singhal S (2017) Application and validation of DMAIC six sigma tool for enhancing customer satisfaction in a government R & D organization. Int J Qual Res 11(4):931–944. https://doi.org/10.18421/IJQR11.04-13 4. Leiber T (2018) Impact evaluation of quality management in higher education: a contribution to sustainable quality development in knowledge societies. Eur J High Educ 8(3):235–248. https://doi.org/10.1080/21568235.2018.1474775 5. Mahadeven R, Shivaprakash NC, Bose SK (2013) Quality assessment of technical education in Indian Engineering Institutions. IEEE Glob Eng Educ Conf EDUCON 1:973–977. https:// doi.org/10.1109/EduCon.2013.6530225
74
P. Kumar et al.
6. Mehta N, Verma P, Seth N (2013) Total quality management implementation in engineering education in India: an interpretive structural modelling approach. Total Qual Manag Bus Excell 25(1–2):124–140. https://doi.org/10.1080/14783363.2013.791113 7. Mehta N, Verma P, Seth N, Shrivastava N (2014) Identification of TQM criterions for engineering education using Delphi technique. Int J Intell Enterp 2(4):325–352. https://doi.org/10. 1504/IJIE.2014.069075 8. Pal Pandi A, Rajendra Sethupathi PV, Jeyathilagar D (2016) The IEQMS model for augmenting quality in engineering institutions—an interpretive structural modelling approach. Total Qual Manag Bus Excell 27(3–4):292–308. https://doi.org/10.1080/14783363.2014.978647 9. Papic L, Garcia AC (2017) Significant factors of the successful lean six-sigma implementation. Int J Math Eng Manag Sci (IJMEMS) 2(2):85–109 10. Purwihartuti K, Zusnita WO (2016) Quality management systems and performance of organization. Int J Econ Commer Manag IV(11):598–611 11. Singh C, Sareen K (2006) Effectiveness of ISO 9000 standards in Indian educational institutions: a survey. Int J Serv Technol Manag 7(4):403–415
A Review on Life Cycle Assessment of Various Dairy Products Mukesh Kumar and Vikas Kumar Choubey
Abstract Life cycle Assessment is an organised tool, defined by ISO 14040: 2006, generally used to assess the environmental impact throughout product life cycle (from beginning to recycle or disposal). Globally 107 kg per capita per year milk is being consumed by the world population. In this paper a snapshot has been presented for environmental LCA in dairy industry. Dairy industry has been responsible for the emission of 4% of total global anthropogenic greenhouse gases. India and Australia are the most GHG producing country in post firm gate processing. Methane gas is the most dominating factor for the Greenhouse gas emission by dairy products. Most variation on post firm gate has been occurred due to the use of various packaging materials, transportation, different processing method in industries, and energy consumption. Cheese is the most studied dairy product. Fertilisers use, production of agriculture products and manure management are the main drivers at pre firm or milk production emission. Mostly researchers 10 out of 15 were used Global warming potential as impact assessment. Eco-invent was the most utilised database along with Impact 2002+ as the most utilised Impact assessment tool. To reduce carbon footprint, use of renewable energy (as availability) and energy efficient equipment’s are the most common recommendations. Keywords Carbon footprint · Life cycle assessment · Dairy products
1 Introduction Increasing population of the world needs food products to survive. Food industry has been changed over the last decade due to this the environmental burden on climate change, water quality, loss of biodiversity, quality of soil and many more are directly linked through increased population growth [29, 34, 35]. Milk is one of the most consuming energy and protein rich food product. According to the international dairy federation 107 kg per capita per year milk consumed by world population. In M. Kumar · V. K. Choubey (B) Department of Mechanical Engineering, National Institute of Technology, Patna, Patna, India e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_8
75
76
M. Kumar and V. K. Choubey
2015 total milk production was more than 800 million tonnes [16, 19], and dairy sector grows rapidly at an average growth rate of 1.8% per year and estimated as it will be double in 2050 that of 2020 [30, 41]. When entire chain of milk production and dairy production is taken into study then around 18% of anthropogenic emissions of GHGs is generated by the livestock sector [15]. Livestock is directly related to water consumption and carbon emission. Dairy sector was responsible for the emission of CO2 eq. of approximately 1069 million tonnes in which 1328 million tonne contributed by milk production and remaining by meat [31]. Dairy products were the second most of GHG producing industry after meat production and contribute approximately 4% of total GHG, 10% of global EP and 6% AP [6, 15, 21, 31, 39]. The average global emissions at the firm gate are estimated to be 2.4 and regional variation from 1.3 to 7.5 CO2 eq. per kg of FPCM. In study of whole dairy supply chain milk production was the highest GHG contributing area. But from study it shows that the variation of GHG till firm gate is similar to all the reason it varies on post firm gate [15]. In account of dairy industry, due to excess release of methane gas dairy firm is major contributors of GHG emission [31]. Life cycle assessment—LCA is a tool that has been used for finding environmental changes due to product production and consumption and their potential impact on human life [6, 20, 26, 39]. According to IDF LCA is most effective tool used to evaluate environmental impact by the product. LCA is a tool that tests various damage categories (Human health, Ecosystem, Resource depletion) by evaluating various midpoint impact categories. Product is being studied from raw material to recycle or final disposal of the product. Various Environmental impacts are tested at different scenarios in LCA in reporting phase. The major studied impact categories are Carbon footprint (Greenhouse gas emission) measured in the form of equivalent CO2 per kg of product (Fig. 1). GHG Contributing Factor CO2
21
Nitrous oxide
10
27
Methane
38
52
0%
10%
20%
52
30%
40%
Developed country
50%
60%
70%
80%
90%
Developing country
Fig. 1 Greenhouse gas contributing factors in developed vs developing country
100%
A Review on Life Cycle Assessment of Various Dairy Products
77
2 Life Cycle Assessment—LCA Due to higher environmental impact of milk and other milk product (Dairy) assessment of milk production is necessary. Emission from milk production at firm gate is approximately equal in all the reason, but post firm gate emission such as (transportation of milk, production of dairy products, waste water treatment, packaging material consumption, energy consumption, and chemical use) varies from industry to industry due to different processes [45]. Life cycle assessment is a tool suggested and defined by ISO standards 14,040 and 14,044—ISO, 2006. LCA is a systematic tool that has been used for the analysis of environmental load of product from beginning to recycle phase of product. LCA tests product at all the phase of production and consumption of product and calculate environmental load. LCA also interprets the damage potential from the emission of the product [38].
2.1 LCA Methodology Concept of LCA started in the 1960s and there was some effort made to develop LCA methodology in 1970s. It gains attention in the field of environmental science since 1990s [1]. LCA applied in four steps namely 1. Goal and scope identification, 2. Life cycle inventory (LCI), 3. Life cycle Impact Assessment (LCIA) and 4. Interpretation. Detail framework of LCA has been shown below in Fig. 2. Purpose of LCA is to identify the environmental impact of product throughout whole life cycle, controlling the emission through identifying reason of highest impact within product life and comparison of emission from two or more same product and process [38].
Goal and Scope identifecation
Life cycle Inventory LCI
Life cycle Impact Assessment - LCIA
Interpretation
Setting Functional Unit
Data collection
Setting impact category
Uncertainety Analysis
System Boundary Identifecation
Calculation models
Impact calculation
Sensitivity analysis
Fig. 2 Life cycle assessment framework modelled from [26, 27]
78
M. Kumar and V. K. Choubey
2.2 Goal and Scope Identification Setting goal and Defining Scope is the most important phase in LCA because all the analysis is based on the initial selection. In the Goal and scope definition phase firstly find the purpose at which LCA is to be applied, setting of system boundary, defining the functional unit (FU) [26]. System boundary in LCA is generally designated by the Input–Output process flow diagram. In dairy production study there are two main system boundaries are identified Cradle to firm gate and second firm gate to retailor. In cradle to firm gate all the upstream data are considered such as (milk production, Manure management, grazing, firming, livestock production etc.). In post firm gate system boundary all the processes incorporates into production, packaging, transportation, waste management, etc. are accounted [40]. Functional unit in LCA is used to normalise the inventory data. Generally, in cradle to firm type LCA 1 kg fat and protein corrected milk or 1 kg energy corrected milk is used. In post firm analysis of LCA generally 1 kg of final product is used [11].
2.3 Life Cycle Inventory Analysis—LCI LCI is the time consuming and more work intensive process compare to all the phases of LCI. Data collection has been done in this phase made thus it take more time. If industry has good database then more time to be saved. There are several guidelines about data collection was given in ISO 14040–2006, but for dairy industry the International Dairy Federation (IDF) publishes guidelines about data collection module and questionnaire format. In this phase all the Input–output and emission related data has been collected and normalised it in according tom functional unit. In this phase all the emissions, inputs (raw material, livestock, fertilisers, energy, chemical, packaging material, Etc.) are shown in a tabulated form [4]. There are generally two different LCI modelling approach are used Consequential life cycle assessment modelling and attribution life cycle assessment modelling [10].
2.4 Life Cycle Impact Assessment—LCIA LCIA is the third step in LCA according to ISO and there are various impact assessment tools are available in the market some are available in paid version some are nonpaid. All the impact assessment are differing according to their calculation methods and their midpoints and endpoints of result. LCIA purpose has to convert the elementary flows from the LCI into their possible contributions on the environmental impacts that has been considered in the LCIA [24, 31]. Life cycle inventory results examined on various midpoint impact categories such as global warming potential, Acidification potential, Eutrophication potential, Mineral use, Land occupation, etc. and also
A Review on Life Cycle Assessment of Various Dairy Products Inventory results
Common Midpoint
79 Damage (Endpoint)
Global warming potential Human health Ozone Depletion Inventory Data CO2, N2O, PO4, CFC……Inputs etc.
Eutrophication
Photochemical oxidant formation
Environment
Acidification Potential
Human toxicity potential Use of Natural Mineral use
Resources
Land use
Fig. 3 General LC impact model [41]
their long term impacts are also calculated through endpoint impact such as Human health, Ecosystem and resource depletion [37]. A general life cycle impact model is illustrated below in Fig. 3.
2.5 Interpretation This is the fourth and last phase of LCA in this phase results obtained from LCI and LCIA are tested for their completeness of data, sensitivity and uncertainty [45]. The sensitivity test of LCI results on various categories such as variation in result due to use of different allocation procedure such as mass allocation, chemical allocation, economic allocation etc. [8]. In sensitivity analysis of results selected parameters are individually increased by 10% and parameters are analysed.
3 Results and Discussion Snapshot of the study has been shown in Table 1. In the structural table various criteria has been used to categorised i.e., researchers and their studied country, the product
80
M. Kumar and V. K. Choubey
Table 1 Structural snapshot of literature Author
Country
Impact categoriesa
System boundary
[31]
India
AP, PM, CED. Gate to gate EQ, EC, EP, FEP, GWP, GHG, HT, MEP, OLD,PO, PS,TET,RE
[44]
China
GWP, AP, EP, EC
Survey
[14]
Italy
CC, OLD, HT, PM, EQ, PO, AP, TET, FEP, MEP, FD, LC, WC
Cradle-to-distribution centre-gate
[33]
India
[45]
Functional unit Product studied 1 kg of final product at customer
Milk, curd, paneer, ice-cream, butter, ghee, lassi
1 kg of fat and protein corrected milk (FPCM), 10 g dry matter of packaged cheese
Milk, cheese
AP, PM, CED. Gate to gate EQ, EC, EP, FEP, GWP, GHG, HT, MEP, OLD,PO, PS,TET,RE
1 kg of final product at customer
Toned milk, double toned milk, ghee, butter, curd, sambaram, cream, ice cream, zip—up, ice cream candy
Ireland
CED, GWP
Gate to gate
1 kg solid in processed product
Dairy products like pasteurized milk UHT milk, yoghurt, cream, butter, cheese
[2]
Italy
CC, OLD
Cradle to industry gate 1 ton of Cheese and protein in the whey different w/pcs
[36]
Italy
AD, GWP,OD, HT, FD, MEP, TET, PO, AP, EP
Gate to gate
123 g of mozzarella cheese made from 1 L of cow milk
Cheese and whey
[32]
Brazil
GWP, AP, EP, OLD, PS,HTP
Cradle to firm gate
1 kg energy corrected milk
Milk production
[11]
Serbia
GWP, AP, EP, OLD, PS,HTP
Gate to gate
1 kg of final product at customer
Pasteurized milk UHT milk, yoghurt, cream, butter, cheese (continued)
A Review on Life Cycle Assessment of Various Dairy Products
81
Table 1 (continued) Author
Country
Impact categoriesa
System boundary
Functional unit Product studied
[28]
USA
ET, AD, CC, FEP, HT, MEP, PO
Cradle-to-grave
One ton of product produced
Cheese and whey
[23]
Portugal
AD, AP, EP,GWP, OD, LC, PO, ECD
Cradle to grave
1 kg of final product at customer
Yoghurt
[12]
Romania
CC, EP, AP, Cradle to gate HT, ET, LC,AD
One kg of dairy product
Milk, sour cream, natural yoghurt, curd, butter, cream cheese, fresh cheese, soft cheese, whey
[4]
Peru
GWP, AP, EP
Cradle to firm gate
1 kg energy corrected milk (ECM)
Milk
[7]
Portugal
AD, GWP, PO, AP, EP
Cradle to firm gate
1 tonne of raw Milk milk ready to be delivered at the farm gate.
GWP, EP, EC, PO
Cradle to grave
1 kg of final product at customer
[5]
a Note:
Milk, cheese, yoghurt
Abbreviations are provided in Appendix
used for the case study, Impact categories used, system boundary and functional unit. Fifteen research paper is included in the table for their analysis from different country and number of papers are included for their help and support. Selection of paper for analysis is done on the basis of research paper which published in a reputed journal, main focus of research paper is on milk and dairy products based on LCA. According to study Cheese was the most studied dairy product followed by Pasteurised milk, Yoghurt, cream, whey and curd. Global warming potential, Acidification potential, Eutrophication—Potential, Photochemical oxidation, ozone layer depletion was the most utilised impact category in dairy product analysis. Graphical analysis of dairy product study and impact assessment are shown below in Figs. 4 and 5, respectively (Table 2). There are the three types of system boundary generally followed by the researchers in LCA study of milk and dairy product. Gate to gate analysis is the most common system boundary used to analyse the LCA of Dairy products which are shown in Table 3. Gate to gate system boundary is used because the variation of carbon emission and the various impact on the environment are comparatively same [22, 25, 33]. In the analysis of functional unit “one kg of Final product” is most commonly used when system boundary is Gate to Gate analysis. That indicates when only dairy product
M. Kumar and V. K. Choubey Number of researcher study
82 10 8 6 4 2 0
Dairy products
Fig. 4 Product category analysis
Number of researchers use
10 8 6 4 2 0
Life cycle impact categories
Fig. 5 Impact assessment analysis
is taken into consideration then One kg of final product is the main Functional unit. When the study focusses on Milk production then 1kh of FPCM and 1 kg of ECM are used as a functional unit.
3.1 Raw Milk Production Raw milk Is produced in the dairy firm and having various activity involved that affect the environment. In this paper some of the main environmental affecting activity are shown in Table 3 with some critical impact categories. GWP is the most utilised factor in milk production after that AP, EP and Land use are the Critical impact that is assessed [3, 9, 43]. The two important Activity feed production, manure management having a higher amount of impact on the environment, in this paper these two activities
Unit
Kg CO2 eq
Kg SO2 eq
Kg PO4 eq
Kg R11 eq
Kg C2 H4 eq
Impact category
Global warming potential
Acidification potential
Eutrophication potential
Ozone layer depletion
Photochemical smog
(1.66–2.1)E−3
(5–6.1)E−5
0.32–0.413
0.69–0.89
–
–
0.0011
–
7.8
(1.01–1.31)E−3
(3.24–4.3)E−5
0.0203 - 0.027
0.043–0.057
3.5–4.5
[11]
6.7–9.47
Cream
[11]
[12]
Cheese
Table 2 Environmental impact analysis of various dairy product
3.24E−3
4.24E−6
3.3E−3
0.0144
3.37
[33]
(5.89–6)E−3
2E−4
–
0.2636–0.2658
20.69–21.3
[11]
Butter
1.51E-3
1.1E-4
–
0.1401
39.16
[33]
(3.8–6.92)E−3
1.01E−5
0.0065–007
0.0144–0.0195
1.42–2.63
[11]
Yoghurt
–
–
3.92E−04
3.49E−2
3.35
[33]
A Review on Life Cycle Assessment of Various Dairy Products 83
Unit
Kg CO2
Kg SO2 eq
Kg PO4 eq
MJ/Kg
Impact category
Global warming potential
Acidification potential
Eutrophication potential
Energy consumption
1.55–3.32
0.00092–0.00201
0.00573–0.0124
0.242–0.499
–
–
–
0.33–0.53
0.00057–0.00107
0.0033–0.006
0.109–0.238
[44]
–
–
–
0.11–0.18
[42]
Manure management
[44]
[42]
Feed production
Table 3 Environmental impact analysis of milk
1.68–3.344
0.00192–0.00334
0.0114–0.02007
0.857–1.528
[44]
Total 1.44–2.04
[42]
3.73–6.73
0.00588–0.0133
0.01298–0.0273
0.95–1.88
[43]
84 M. Kumar and V. K. Choubey
A Review on Life Cycle Assessment of Various Dairy Products
85
are detail analysed in Table 2. Total GWP in milk production is between 0.857–1.528 [44], 1.44–2.04 [42] and 0.95–1.88 [43] and variation is between 0.857 and 2.014 kg CO2 eq. Average emissions in the land use category are estimated at 0.09 kg CO2 eq. per kg of FPCM [15]. At the milk production unit or dairy firm Methane is the most contributing factors for the GHS around 50% of the total GHG is generated due to extraction of methane on the manure management activity. Around 35–38% of GHG contributor is NO2 gas formed due to use for fertilisers and also in manure management remaining GHG emission due to emission of CO2 gas [13, 15.
3.2 Cheese Production Because Cheese was the most studied dairy product as shown in Fig. 4 and also verified through [11, 41]. Cheese has been detail analysed and their various environmental impact has been shown in Table 2. In Table 2 data from two researcher has been compared with five critical impact category listed. Djekic et al. [11] analysed 7 dairy plant in Serbia and shows GHG emission from Cheese production is 6.7–9.47 kg CO2 eq. and acidification potential has been 0.69–0.89 kg SO2 eq. id dairy cheese production Raw milk production, raw milk Transportation and milk processing are the main three point where GHG is emitted. According to [17] GHG emission has been 7.203 kg CO2 included with 6.709 kg CO2 in RMP and 0.03 kg CO2 in RMT and 0.464 kg CO2 in milk processing. Finnegan et al. [18] shows there is a lot of variation in the GHG emission per Kg CO2 emission is between 0.46 and 1.92 because of large variation on the processing methods and Cumulative energy demand.
3.3 Yoghurt Yoghurt is a dairy product studied by the numerous author. In this paper comparison between two authors has been tabulated. According to [23] yoghurt production main contribution to environmental impact is production of raw milk and concentrated milk powder with remarkable contributions of 62% of GWP, 91% Acidification and 92% of Eutrophication. In the above Table 2 Global warming potential has been 1.42–2.63 kg CO2 eq. [11] and 3.35 kg CO2 eq. [12].
4 Conclusion A detailed literature review has been presented through this paper, studies focussing on the environmental life cycle assessment on milk and dairy products. Raw milk and various dairy products (Cheese, yoghurt, butter, and cream) are studied in detail.
86
M. Kumar and V. K. Choubey
A total of 15 research paper are fully analysed and listed in tabulated form and more than 35 research paper has been used to write this paper. In the whole study Cheese was the most common studied dairy product (8 out of 15), followed by raw milk as well as processed milk, cream, yoghurt whey. GHG emission from butter production is between 20 and 36 kg CO2 eq., while in production of cheese it was 6.7–9.47 kg CO2 eq. Global warming potential (10 out of 15) was the most common used impact category followed by acidification potential, Eutrophication, ozone layer depletion etc. ecoinvent database was the leading database in life cycle assessment while IMPACT 2002+ was most utilised impact method.
4.1 Direction Towards Future Work After studying several research articles was clearly visible that the variation in GHG emission for several dairy products was varying plant to plant. Different processing plants follow different operations for transforming raw milk to dairy products. So it is recommended that every dairy plant needs to evaluate and significant improvement is required to minimise GHG emission.
Appendix Abbreviations AP PMRI CED EC EP FD FEP GWP HT LC MEP OD PS TET WC GHG
Acidification potential Particulate matter/respiratory inorganics Cumulative energy demand Energy consumption Eutrophication potential Freshwater depletion Freshwater eutrophication potential Global warming potential Human toxicity Land competition/use Marine eutrophication potential Ozone depletion Photochemical smog Terrestrial ecotoxicity Water consumption Greenhouse gas
A Review on Life Cycle Assessment of Various Dairy Products
EQ FPCM ECM
87
Ecosystem quality Fat and protein corrected milk Energy corrected milk
References 1. Arzoumanidis I, Petti L, Raggi A, Zamagni A (2013) Life cycle assessment for the agri-food sector. Product-oriented Environmental Management Systems (POEMS). Springer, Dordrecht, pp 105–122 2. Bacenetti J, Bava L, Schievano A, Zucali M (2018) Whey protein concentrate (WPC) production: environmental impact assessment. J Food Eng 224:139–147 3. Baldini C, Gardoni D, Guarino M (2017) A critical review of the recent evolution of life cycle assessment applied to milk production. J Clean Prod 140:421–435 4. Bartl K, Gómez CA, Nemecek T (2011) Life cycle assessment of milk produced in two smallholder dairy systems in the highlands and the coast of Peru. J Clean Prod 19(13):1494–1505 5. Berlin J, Sonesson U, Tillman AM (2007) A life cycle based method to minimise environmental impact of dairy production through product sequencing. J Clean Prod 15(4):347–356 6. Burek J, Nutter DW (2020) Environmental implications of perishables storage and retailing. Renew Sustain Ener Rev 133:110070 7. Castanheira ÉG, Dias AC, Arroja L, Amaro R (2010) The environmental performance of milk production on a typical Portuguese dairy farm. Agric Syst 103(7):498–507 8. Cederberg C, Sonesson U, Henriksson M, Sund V, Davis J (2009) Greenhouse gas emissions from Swedish consumption of meat, milk and eggs 1990 and 2005. Report, 793, pp 1–97 9. Cederberg C, Mattsson B (2000) Life cycle assessment of milk production—a comparison of conventional and organic farming. J Clean Prod 8(1):49–60 10. Dalgaard R, Schmidt J, Flysjö A (2014) Generic model for calculating carbon footprint of milk using four different life cycle assessment modelling approaches. J Clean Prod 73:146–153 11. Djekic I, Miocinovic J, Tomasevic I, Smigic N, Tomic N (2014) Environmental life-cycle assessment of various dairy products. J Clean Prod 68:64–72 12. Doublet G, Jungbluth N, Stucki M, Schori S (2013) Life cycle assessment of Romanian beef and dairy products. SENSE-harmonised environmental sustainability in the European food and drink chain, seventh framework programme: project, 288974) 13. Eide MH (2002) Life cycle assessment (LCA) of industrial milk production. Int J Life Cycle Assess 7(2):115–126 14. Famiglietti J, Guerci M, Proserpio C, Ravaglia P, Motta M (2019) Development and testing of the product environmental footprint milk tool: a comprehensive LCA tool for dairy products. Sci Total Environ 648:1614–1626 15. FAO (2010) Greenhouse gas emissions from the dairy sector a life cycle assessment. In: Food and agriculture organization of the United Nations animal production and health division 16. FAO (2015) FAO statistical pocketbook 2015. Food and agriculture organization of the United Nations 17. Finnegan W, Goggins J, Clifford E, Zhan X (2017) Global warming potential associated with dairy products in the Republic of Ireland. J Clean Prod 163:262–273 18. Finnegan W, Yan M, Holden NM, Goggins J (2018) A review of environmental life cycle assessment studies examining cheese production. Int J Life Cycle Assess 23(9):1773–1787 19. Food and Agriculture Organization of the United Nations (2020) The impact of COVID-19 on food and agriculture in Asia and the Pacific and FAO’s response. Report of the ... FAO regional conference for Asia and the Pacific. 20. Gallucci T, Lagioia G, Piccinno P, Lacalamita A, Pontrandolfo A, Paiano A (2020) Environmental performance scenarios in the production of hollow glass containers for food packaging: an LCA approach. Int J Life Cycle Assess 1–14
88
M. Kumar and V. K. Choubey
21. Gerber P, Vellinga T, Opio C, Steinfeld H (2011) Productivity gains and greenhouse gas emissions intensity in dairy systems. Livest Sci 139(1–2):100–108 22. Gollnow S, Lundie S, Moore AD, McLaren J, van Buuren N, Stahle P, Christie K, Thylmann D, Rehl T (2014) Carbon footprint of milk production from dairy cows in Australia. Int Dairy J 37(1):31–38 23. González-García S, Castanheira ÉG, Dias AC, Arroja L (2013) Environmental life cycle assessment of a dairy product: the yoghurt. Int J Life Cycle Assess 18(4):796–811 24. Hauschild MZ, Huijbregts MA (2015) Introducing life cycle impact assessment. In: Life cycle impact assessment. Springer, Dordrecht, pp 1–16 25. International Dairy Federation (2016) Bulletin of the International Dairy Federation 473/2014 26. ISO (2006a) 14040: Environmental management–life cycle assessment—Principles and framework, International organization for standardization 27. ISO (2006b) ISO 14044:2006, Environmental management—Life cycle assessement— Requirements and guidelines, ISO 14044, International organization for standardization 28. Kim D, Thoma G, Nutter D, Milani F, Ulrich R, Norris G (2013) Life cycle assessment of cheese and whey production in the USA. Int J Life Cycle Assess 18(5):1019–1035 29. Kumar M, Choubey VK (2021a) Modeling the causes of post-harvest loss in the agri-food supply chain to achieve sustainable development goals: an ISM approach. In: Challenges Opportunities Circ Econ Agri-Food Sect. Springer Singapore, pp 133–149. https://doi.org/ 10.1007/978-981-16-3791-9_8 30. Kumar M, Choubey VK (2021b) Modelling the interaction among the key performance indicators of sustainable supply chain in perspective of perishable food. Int J Logist Syst Manage1(1):1. https://doi.org/10.1504/ijlsm.2021.10039305 31. Kumar M, Kumar Choubey VK, Deepak A, Gedam VV, Raut RD (2021c) Life cycle assessment (LCA) of dairy processing industry: a case study of North India. J Clean Prod 129331. https:// doi.org/10.1016/j.jclepro.2021.129331 32. de Léis CM, Cherubini E, Ruviaro CF, Da Silva VP, do Nascimento Lampert V, Spies A, Soares SR (2015). Carbon footprint of milk production in Brazil: a comparative case study. Int J Life Cycle Assess 20(1):46–60 33. Mahath CS, Kani KM, Dubey B (2019) Gate-to-gate environmental impacts of dairy processing products in Thiruvananthapuram, India. Resour Conserv Recycl 141:40–53 34. Noya I, González-García S, Berzosa J, Baucells F, Feijoo G, Moreira MT (2018) Environmental and water sustainability of milk production in Northeast Spain. Sci Total Environ 616:1317– 1329 35. OECD (2020) COVID-19 and the food and agriculture sector: issues and policy responses, OECD. 36. Palmieri N, Forleo MB, Salimei E (2017) Environmental impacts of a dairy cheese chain including whey feeding: an Italian case study. J Clean Prod 140:881–889 37. Rosenbaum RK, Hauschild MZ, Boulay AM, Fantke P, Laurent A, Núñez M, Vieira M (2018) Life cycle impact assessment. In: Life cycle assessment. Springer, Cham, pp 167–270 38. Roy P, Nei D, Orikasa T, Xu Q, Okadome H, Nakamura N, Shiina T (2009) A review of life cycle assessment (LCA) on some food products. J Food Eng 90(1):1–10 39. Rufí-Salís M, Petit-Boix A, Villalba G, Ercilla-Montserrat M, Sanjuan-Delmás D, Parada F, Arcas V, Muñoz-Liesa J, Gabarrell X (2020) Identifying eco-efficient year-round crop combinations for rooftop greenhouse agriculture. Int J Life Cycle Assess 25(3):564–576 40. Tillman AM, Ekvall T, Baumann H, Rydberg T (1994) Choice of system boundaries in life cycle assessment. J Clean Prod 2(1):21–29 41. Üçtu˘g FG (2019) The environmental life cycle assessment of dairy products. Food Eng Rev 11(2):104–121 42. Wang X, Kristensen T, Mogensen L, Knudsen MT, Wang X (2016) Greenhouse gas emissions and land use from confinement dairy farms in the Guanzhong plain of China–using a life cycle assessment approach. J Clean Prod 113:577–586 43. Wang X, Ledgard S, Luo J, Guo Y, Zhao Z, Guo L, Liu S, Zhang N, Duan X, Ma L (2018) Environmental impacts and resource use of milk production on the North China Plain, based on life cycle assessment. Sci Total Environ 625:486–495
A Review on Life Cycle Assessment of Various Dairy Products
89
44. Wang L, Setoguchi A, Oishi K, Sonoda Y, Kumagai H, Irbis C, Inamura T, Hirooka H (2019) Life cycle assessment of 36 dairy farms with by-product feeding in Southwestern China. Sci Total Environ 696:133985 45. Yan M, Holden NM (2018) Life cycle assessment of multi-product dairy processing using Irish butter and milk powders as an example. J Clean Prod 198:215–230
Design of Fuzzy Controller for Blood Glucose Level Vijay Kumar and Amit Kumar Singh
Abstract To maintain good health is the biggest challenge of this generation. And the environment we are living is not only polluted but it is hard to find good healthy food everywhere. One of the most dangerous diseases is diabetes; it is incurable and it is a major health problem. A nonlinear model of type 1 diabetes was taken into consideration, which has been implemented in MATLAB-SIMULINK environment. In this research, developing a system which monitors the glucose level and also regulates the injection of insulin rate for controlling the blood glucose. Mamdanitype fuzzy logic and PID controller is used to stabilize the blood glucose in normal range. The analysis of result is based on disturbances which are incorporated to patient model in the form of meal, delay or noise in glucose sensor. Comparison of Fuzzy controller and PID is done in both situation disturbances and without disturbances. The result of simulation provides the superiority of the proposed Controller. Keywords Bergman model · Glucose · Fuzzy logic controller · Diabetes · Closed loop
1 Introduction Majority of the population across the world suffering from diabetes and number of patients are increasing day by day. Now, it becomes the most common disease in human life. It is seen in all age group of people and became a very serious health issue in society. More than 9% of adult population whose age is between (21–89) are affected by this crucial problem, and this number is increased by 56% within 20 years [1]. This occurs because of unhealthy lifestyle and insalubrious food. Diabetes is not seen only in the developed countries and western world, but also seen in developing countries [2]. Diabetes is an incurable disease most of the people are affected by it, according to the survey which is done by WHO all around the globe says that 180 V. Kumar · A. K. Singh (B) Department of Instrumentation and Control Engineering, Dr. B. R. Ambedkar National Institute of Technology, Jalandhar, India e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_9
91
92
V. Kumar and A. K. Singh
million people is being affected. This number will be double i.e. (360) million in the upcoming future year 2030 [3]. According to international diabetes federation, every 6th second a person dies due to diabetes [4]. In India out of ten every 8 people are suffering from diabetes [5]. With increase in the number of diabetes patient we need to develop an efficient device which aim is to diagnosis the diabetes patient with better accuracy. Many scientist and researchers are engaged in developing new technique and device which helps in solving the problem of diabetic patient. Most of the researchers have been working for finding many methods that are related to the diagnosis and treatment of diabetes disease. Automatic close loop control system (ACLCS) is also known as the artificial pancreas. ACLCS is a technique that is used to design the device for diabetes patients. The functionality of APD (Artificial pancreas device) is discussed later. When the body is not properly working to maintain blood glucose in normal range cause diabetes and if the diabetes is not treated and prolonged for long time cause serious health issues such as loss of vision, kidney disorder, heart problem and other diseases. Mainly there are two kinds of diabetes i.e., Type 1 and Type 2. Type 1 diabetes is a patient in which the cell that produces the insulin are destroyed and insulin is released by β cell of pancreas is very less (below to 10% of normal) [6]. Type 1 diabetes patient depends on the external source of insulin to maintain the blood glucose. Due to the absence of much Insulin, it is unable to reduce the blood glucose at a faster rate and this is the case of high blood glucose (Hyperglycemia) above 180 mg/dl [7] and these patients require treatment by injection of insulin into the body, by doing regular exercise or insulin infusion pump. Indication of type 1 person are continual urination, tiredness, reduce of weight and hunger. Type 1 diabetes patient is 10% of diabetes population. Type 2 patient is most common, which accounts for 90–95% of the diabetic population. In this type insulin is not affect the cell of the body for increasing up taking of glucose and Type 2 people are insulin resistant and relative insulin deficient [8]. There are three types of control strategies involved in treatment of diabetes patient i.e., open loop, semi closed loop and last one is automatic control loop. In open loop control, patient inject a dose of insulin according to glucose measurement by invasive method i.e., Finger prick [8]. Usually patient performed this activity 3–4 times daily a day. This method is very painful and uncomfortable to patient and also this method is not suitable because it involves approximation of insulin delivered Challenges faced by invasive CGMS (continuous glucose monitoring system) is that there is time delay is present all time because time taken by insulin to reach the internal from insulin pump. Closed loop control also Like CGMS, insulin faced with time-delay like CGM [9]. Insulin infusion rate in the semi close loop control are adjusted with respect to the blood glucose reading. Disadvantage of this control loop is that it can provide very large sampling time. Closed loop control method is also called artificial pancreas device. It is the most effective technique to diagnosis the diabetic patient [10]. APD is replica is our natural pancreas and performs the same activity of natural pancreas and performs the same activity of natural pancreas and helps in maintaining blood glucose in desired physiological range. Advantages of APD are that it enables diabetes patient to regulate BGC (glucose level) by injecting proper Amount of insulin at right and
Design of Fuzzy Controller for Blood Glucose Level
93
without involvement of human intervention. In closed loop control, mathematical model of patient is used which describes the glucose insulin system. Parameter used in the model varies from patient to patient and variable. So, we have to design a controller which is robust to uncertainty in model parameter and disturbances in the form of meal. APD system performs three functions: sensor which continuously monitors the blood glucose, which is wireless. Insulin pump delivers insulin according to need and last one is controller which determines the amount of insulin dosage [11]. The simple diagram of APD system is in Fig. 1 [12]. Bergman et al. [13] were the first Scientists which provide physiological Model of glucose and insulin kinematics. History of minimal model of blood glucose control came from two decades after world war 2nd; the science of cybernetics was introduced by Norbert wiener at MIT. Weiner recognized the importance of the rapidly developing fields of control theory and systems analysis to problems in biology and medicine [13]. Bergman minimal model is also called single compartmental model which is nonlinear it means that our body is treated as compartment with glucose and insulin as basal concentration. It includes different variables for indicating the concentration of glucose, remote insulin and insulin in the blood plasma. Minimal model contains two models one of them tells about glucose kinetics and other tells about insulin kinetics. The model takes input in form of insulin and meal and output of model is blood glucose which is to be controlled. In the past, various control technique have been developed for blood glucose model including Linear Quadratic Regulator (LQR) control [14] and H infinity control [15]. In this paper FLC (Fuzzy logic controller) is used to regulate blood glucose, insulin and disturbances are taken as input into patient model. Result of Simulation is shown with or without disturbances on diabetic patient model and also step response of system is presented. The simulation of BGC is implemented on MATLABR2016. Fig. 1 Artificial pancreas device [12]
94
V. Kumar and A. K. Singh
2 Model for Blood Glucose Insulin Sytstem BMM (Bergman minimal model) is mostly the used model to show the relationship of glucose insulin system. Diagram of BMM is shown in Fig. 2 [16]. BMM is a first order differential equation that is nonlinear. For linearization uses some modification to design the control model under operating situation. Parameters used in model are listed in Table 1 [17]. The block diagram of close loop control and open loop control for glucose system is shown in Figs. 3 and 4, respectively. dG(t) = −P1 [G(t) − Gb ] − X(t)G(t) + D(t) dt
(1)
dX(t) = −P2 X(t) + P3 [I(t) − Ib ] dt
(2)
Fig. 2 Bergman minimal model [16]
Table 1 Description of parameters used in Bergmanmodel
The plasma glucose conc (mg/dl)
G(t)
The plasma insulin conc. (μu/ml)
I(t)
The remote compartment insulin (min−1 )
X (t)
Plasma glucose Basal value (mg/dl)
Gb
Plasma insulin Basal value (μu/ml)
Ib
System parameter
P1 , P2 , P3
Plasma insulin decay rate
(min−1 )
n
Threshold value of glucose mg/dl
h
Insulin secretion by beta cell
U
Rate of insulin infusion
U(t)
Design of Fuzzy Controller for Blood Glucose Level
95
Fig. 3 Closed loop control of glucose control [17]
Fig. 4 Open loop control for diabetes control [16]
dI(t) = −n[I(t) − Ib ] + Y[G(t) − h]+ + U(t) dt
(3)
Y[G(t) − h]+ represents internal regulatory function which is absent in diabetes function so we ignore this term. D(t) is disturbances in the form of meal which increases the blood glucose and then decreases in approximately 3–4 h. due to the action of insulin delivery. Here, the disturbance is the exponential decay function as below: D(t) = A exp−Bt
(4)
where A, B are the constant and the value are taken as A = 1.59, B = 0.05, t is in min and D(t) is the function of disturbance which has the unit of mg/dl/min [18] Blood glucose parameters differ from patient to patient and depend on weight and height of the patient. The value of basal blood glucose used in glucose model is 71 mg/dl and the value of the system parameter used p1 = 0.312300017 l/min., p2 = 0.0123 l/min. and p3 = 0.0121 l/min, plasma insulin decay rate (n) is 0.090265 l/min [18]. The system parameter is converted into hour which is shown in the result. The parameter of Bergman model given in Table 1. When the insulin goes inside or the outside then the insulin rate is directly proportional to the difference of I(t) and the Ib of plasma insulin I(t). Where Ib is known as the base level of the insulin. When the insulin level in to the plasma is greater than the base level of the insulin then the insulin goes to the insulin comportment and otherwise insulin goes outside from the insulin comportment. When the Glucose goes to the inside or the outside then the glucose rate is directly proportional to the
96
V. Kumar and A. K. Singh
G(t) and the Gb . Where Gb is known as the basal value of plasma glucose. When the glucose in the plasma is greater than the basal value of the plasma then glucose goes into the plasma compartment and otherwise glucose goes outside from the plasma compartment [19].
3 Controller Design 3.1 PID Controller PID controller is a traditional controller and also called three term controllers, it includes three terms proportional, derivative and integral which are denoted by Kp , Kd and Ki . The diagram of PID controller is depicted in Fig. 5. The system’s output due to the integral parameter depends on the past error and the output due to the derivative parameter depends on the future error. By using all these control actions i.e., PID we get a good performance control technique. PID Controller calculates the error e(t) which is difference between desire value r(t) and measured value (o/p) [20]. According to error generated controller applies a correction based on proportional, derivative and integral term. e(t) = desire value – measured value. The overall output of PID controller is U(t) = kp e(t) + kd + ki e(t) dt
(5)
3.2 FLC In the control system FLC is a powerful technique used in the design of controllers and it is an advanced process control. Fuzzy logic controllers have expert knowledge which raised demands in designing of controllers In this paper, algorithm is designed
Fig. 5 Block diagram of PID controller [20]
Design of Fuzzy Controller for Blood Glucose Level
97
Fig. 6 Block diagram of FLC [22]
for the treatment of blood glucose using FLC (fuzzy logic controller) [21]. The designed controller helps the patient to maintain the glucose concentration when the patient is incorporated into meal as a disturbance in diabetic model. To implement the FLC controller we need two inputs i.e., change of glucose e(t), the rate of change of glucose. And one output is in the form of insulin dose. FLC consists of the following process [22]. Block diagram of FLCis depicted in Fig. 6. Fuzzification: in fuzzification mapping of crisp data input into linguistic variables. The membership function (MF) is used in FLC like bell, sigmoidal, triangular, Gaussian, etc. Here we used both the inputs and outputs have three membership functions. Membership function (MF) of input and output variables are L (low), M (medium) and H (high). Here triangular membership function is chosen in design of the controller. Rule base: from the analysis of input and output fuzzy sets, the rule base consists of IF–THEN statements. The antecedent part used in rule base is connected through ‘AND’ logical operation i.e., IF the change of glucose is low (L) ‘AND’ the rate of change of glucose is low (L), THEN rate of insulin injection should be low (L). There are nine IF–THEN rule are designed which are shown in Table 2. Inference engine: Inference engine is central unit which controls the entire rule in fuzzy rule base. Defuzzification: in the process of defuzzification the output of fuzzy logic converted into crisp value it converts the linguistic values of a FLC output into a crisp value. The calculation of output is done using the centroid method (Figs. 7, 8 and 9). Table 2 Rules of fuzzy logic controller
Rale of change of glucose Change of glucose
Low
Medium
High
Low
Low
Low
Medium
Medium
Low
Medium
High
High
Medium
High
High
98
V. Kumar and A. K. Singh
Fig. 7 Input MF of change of glucose
Fig. 8 Input MF of rate of change of glucose
Fig. 9 Output MF of the of FLC
4 Result Discussion The simulation of BCG is implemented on MATLAB R2016. BERGMAN Minimal model is used as patient’s dynamics. Results are shown with disturbances and without
Design of Fuzzy Controller for Blood Glucose Level
99
disturbances. According to the rise in the level of glucose insulin is injected from remote compartment Also delay is introduced in meal disturbances which occurs at certain regular interval of time in the form of peak. According to glucose level, insulin is injected to compensate the effect of high glucose. In this paper our desired blood glucose level is 80 mg/dl. To stabilize the blood glucose level in normal range we have to develop a controller which tracks the desired glucose level in small amount of time. In this paper maximum limit of insulin injection is 10 and upper limit of glucose is 250 mg/dl and lower limit is 70 mg/dl. The step responses of Bergman model and insulin injection rate without controller are shown in Figs. 10 and 11, respectively. Patient blood glucose is far away from desire blood glucose limit which is 80 mg/dl. And also the insulin injection rate is constant throughout the time. This result is not as the desire performances therefore we incorporated controller into the system. The response of Bergman model without and with disturbances is shown in Figs. 12 and 13. Figures 12 and 13 is the comparative analysis of PID and FLC controller. Figure 12 shows the response of Bergman model without disturbances in which the blood glucose level is settles desired limit very fast in the case of FLC Controller but in case of PID controller, it settles after some time. When we introduce the disturbance as a meal in the Bergman model, Response of glucose level rises in the form of spikes as shown in Fig. 13. FLC controller rejects the disturbances in short period of time. In PID controller peak of spike is large as compared to FLC and disturbances are resisting for long time. The comparison of various parameters with the PID and the FLC is shown in Table 3.
Fig. 10 Step response of the Bergman model
100
V. Kumar and A. K. Singh
Fig. 11 Step response of insulin injection rate
Fig. 12 Comparison of PID and FLC without disturbance in Bergman model
Fig. 13 Comparison of PID and FLC with disturbance in Bergman model
Design of Fuzzy Controller for Blood Glucose Level
101
Table 3 Comparison of PID and FLC Sr.no
Condition
Parameters
1
With meal
Settling time (hr.)
2
Without smeal
PID 6.75
FLC 5.89
Steady state value
80.6
Steady state error
0.6
79.9 0.1
Settling time (hr.)
6.5
5.70
Steady state value
80.3
79.8
Steady state error
0.3
0.2
5 Conclusion In this paper, PID and controller is used to controlling of blood glucose level. Mamdani-type scheme was used in the fuzzy controller. The designed controller controlling the uncertainty in patient’s models simultaneously rejecting disturbances in the form of meal and model also tracks the desired glucose level. Insulin injection rate changes continuously according to the rise in glucose level. The designed controller injects proper amount of insulin according to need which shows the performance of FLC controller is better than PID controllers. The designed FLC controller improves the automation of insulin delivery and controlling of blood glucose in diabetes patient.
References 1. Gyuk P, Vassányi I, Kósa I (2019) Blood glucose level prediction for diabetics based on nutrition and insulin administration logs using personalized mathematical models. J Healthc Eng.https:// doi.org/10.1155/2019/8605206 2. Mohammad TM, Mina R, Kadhum Q, Mahdi S (2011) Back stepping-based-PID-controller designed for an artificial pancreas model. Al Khwarizmi Eng J 7(4):54–60 3. Kovács L (2006) Extension of the Bergman minimal model for the glucose-insulin interaction. Periodica Polytech Electr Eng 50(1–2):23–32 4. Rungta A, Nadu T (2017) A review on classification of diabetes using fuzzy logic and optimization technique. Int J Comput Intell Res 13(8):2143–2150. http://www.ripublication. com 5. Eren-Oruklu M, Cinar A, Colmekci C, Camurdan MC (2008) Self-tuning controller for regulation of glucose levels in patients with type 1 diabetes. In Proceedings of the American control conference, July, pp 819–824. https://doi.org/10.1109/ACC.2008.4586594 6. Pagkalos I, Herrero P, Toumazou C, Georgiou P (2014) Bio-inspired glucose control in diabetes based on an analogue implementation of a β-cell model. IEEE Trans Biomed Circuits Syst 8(2):186–195. https://doi.org/10.1109/TBCAS.2014.2301377 7. Yadav J, Rani A, Singh V (2016) Performance analysis of fuzzy-PID controller for blood glucose regulation in type-1 diabetic patients. J Med Syst 40(12):1–15. https://doi.org/10.1007/ s10916-016-06028. Gharghory SM, El-Dib DA, Mahmoud M (2016) Low power fuzzy control system for adjusting the blood glucose level. Proc Int Conf Microelectron ICM (December):333–336. https://doi. org/10.1109/ICM.2016.7847883
102
V. Kumar and A. K. Singh
9. Gao J, Yi P, Chi Z, Zhu T (2017) A smart medical system for dynamic closed-loop blood glucose-insulin control. Smart Health 1–2(May):18–33. https://doi.org/10.1016/j.smhl.2017. 04.001 10. Id AS, Della RF, Klickstein I, Russell J, Sorrentino F (2019) Optimal regulation of blood glucose level in Type I diabetes using insulin and glucagon, pp 1–23 11. Sivaramakrishnan N (2017) Optimal model based control for blood glucose insulin system using continuous glucose monitoring 9(4):465–469 12. Ata S, Khan ZH (2017) Model based control of artificial pancreas under meal disturbances. In 2017 international symposium on recent advances in electrical engineering, RAEE 2017, pp 1–6. https://doi.org/10.1109/RAEE.2017.8246033 13. Bergman RN (2003) The minimal model of glucose regulation: a biography. Adv Exp Med Biol 537:1–19. https://doi.org/10.1007/978-1-4419-9019-8_1 14. Sánchez ChávesI Y, Morales-Menéndez R, Martínez Chapa SO (2005) Linear quadratic control problem in biomedical engineering. Comput Aided Chem Eng 20(C):1195–1200. https://doi. org/10.1016/S1570-7946(05)80041-0 15. Chee F, Savkin AV, Fernando TL, Nahavandi S (2005) Insulin injection control for blood glucose regulation in diabetic patients 52(10):1625–1631 16. Kang H, Han K, Choi MY (2012) Mathematical model for glucose regulation in the whole-body system. Islets 4(2):84–93. https://doi.org/10.4161/isl.19505 17. Faiz-Ul-Hassan, Adil M, Khaqan A, Shuja S, Tiwana MI, Qadeer-ul-Hassan, Malik S, Riaz RA (2017) Closed loop blood glucose control in diabetics. Biomed Res (India) 28(16):7230–7236 18. Lynch SM, Wayne Bequette B (2002) Model predictive control blood glucose in type 1 diabetics using subcutaneous glucose measurement 19. Abu-Rmileh A, Garcia-Gabin W (2010) A gain-scheduling model predictive controller for blood glucose control in type 1 diabetes. IEEE Trans Biomed Eng 57(10 PART 1):2478–2484. https://doi.org/10.1109/TBME.2009.2033663 20. Sasi AYB, Elmalki MA (2013) A fuzzy controller for blood glucose-insulin system. 2013(May), 111–117 21. Sharma R (2016) Tuning of digital PID controller for blood glucose level of diabetic patient. November 2017, 2–7. https://doi.org/10.1109/RTEICT.2016.7807837 22. Yasini S, Karimpour A (2008) Active insulin infusion using fuzzy-based closed-loop control
Prioritization of Renewable Energy Alternatives by Using Analytic Hierarchy Process (AHP) Model: A Case Study of India Sudhir Kumar Pathak, Vikram Sharma, and Sandesh S. Chougule
Abstract Energy consumption is on the increase as a result of the growing population of India. Current conventional energy sources do not have an appropriate amount to meet energy demand. Economical and renewable energy alternative solutions must also be considered to satisfy this need. Renewable energies can be seen as a solution to this rising need for electricity. Although the selection of alternative renewable energy sources is a multi-criteria decision-making (MCDM) issue, and it is important to consider many competing factors. The main aim of this study is to develop an analytical hierarchy for the prioritization of three potential renewable energy resources such as solar, wind, and hydro energy in the context of India. The factors were recognized from the available literature studies and also through interactions with ten experts from academics and several industries. Then, nine factors were selected and categorized into three groups: economic factors, technical factors, and environmental factors. The analytical hierarchical method (AHP) is used to calculate the relative weights of the evaluation factors and to identify the alternatives. The final result showed that hydro energy with a priority weight percentage of (39.91%) had the highest rank followed by solar energy (34.9%) and wind energy (25.18%). So, hydro energy is India’s best alternative to renewable energy sources, supported by solar and wind energy. Keywords Analytic Hierarchy Process · MCDM method · Renewable energy
1 Introduction In the past few years, the demand for renewable energy sources has been increased worldwide; in advanced nations for sustainable energy sources and in emerging nations for satisfying future energy demands. For a huge energy-demanded country like India, sources of renewable energies are of importance for economic growth, so that industrial development projects in the various sectors are not restricted by energy S. K. Pathak (B) · V. Sharma · S. S. Chougule Department of Mechanical-Mechatronics Engineering, The LNM Institute of Information Technology, Jaipur, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_10
103
104
S. K. Pathak et al.
scarcity [1]. Around 675 million population in Asia cannot access electricity, and around two billion population rely on biomass energy for cooking. India is seventh worldwide, which is trying to occupy its human resources [2]. Recently, India has established a comprehensive energy supply path. Awareness of energy usage among citizens has been encouraged to the use of solar, wind, and hydro energy. India plans to generate 175 GW of renewable electricity, consisting of 100 GW of solar energy, 60 GW of wind power, and 5 GW of small hydropower plants by 2022 [3]. To get the optimistic goals for the generation of renewable energy (175 GW) by 2022, the government needs to set up 330,000 new employment and opportunities for livelihood. Due to the feasible geographical position of India, the key renewable energies are solar, wind, hydro energy, and ocean energy. To meet the current energy demand from various sources are now in a phase where industrial exploration is necessary. Even so, the prioritization of the best renewable energy alternatives cannot be purely based on technical, economic, and environmental factors that must be considered [4]. This study aims to prioritize the application of renewable energy alternatives in India. The main and sub-criteria factors are identified through a comprehensive review of existing literature and the opinion of experts. Also, ranked these renewable energy alternatives based on these factors as per their severity by assigning the weights using the decision-making AHP method. The previous literature suggests that the identification of the best alternative in renewable energy is feasible, but the study of the alternative in renewable energy sources is not carried out according to their deployment intensity. Many research studies which underline the study of alternatives for renewable energy are given below. Table 1 presents the summarized literature for MCDM methods used in various segments of renewable energy alternatives. Table 2 shows the scale of importance for pairwise comparison matrices. Table 1 Summary of MCDM methods used in renewable energy alternatives Name of author
Year
Country
Methodology
Description
Tasri et al. [5]
2014
Indonesia
Fuzzy AHP
Identification and prioritization of suitable renewable energy sources for Indonesia
Luthra et al. [6]
2015
India
AHP
Ranking the barriers to the implementation of sustainable technologies in the sense of India
Buyukozkan et al. [7]
2017
Turkey
Integrated DEMATEL, ANP, TOPSIS
Selection of the best sources of renewable energy in the sense of Turkey
Kolak and Kaya [8]
2017
Turkey
Fuzzy AHP-TOPSIS
Prioritize the renewable energy alternatives in Turkey
Prioritization of Renewable Energy Alternatives by Using … Table 2 Saatey’s scale of importance for pair-wise comparison matrices
105
The intensity of no. Verbal definition 1
Equal important
3
Moderate important
5
Strong important
7
Very strong important
9
Extremely important
2, 4, 6, 8
Intermediate values (When compromise is needed)
This paper proposed AHP for the prioritization of sustainable/renewable energy alternatives in the sense of India. An overview and summary of renewable energy resources have been covered in Sect. 2. The AHP research methodology was quickly stated in Sect. 3, accompanied by an assessment of decision criteria and their physical importance in Sect. 4. Data analysis of the AHP study for the selection of renewable energy was introduced in Sect. 5. The results are discussed in Sect. 6. Section 7 shall contain a conclusion.
2 Overview of Renewable Energy Resources in India As per the introduction, Currently, India is facing the main problem in the energy production sector. It has been recognized that renewable energies are the lone alternative solutions that can be used to confirm a safe electricity demand. In India, comprehensive research and development programs were conducted in the search of renewable energy sources. Due to geographical location, India has a strong opportunity for renewable energy development, but not many of them are currently possible [9]. The Ministry of New and Renewable Energy (MNRE), under the supervision of the Government of India, aims to build a 500 GW renewable energy capability by 2030. Today, a large part of renewable energy is untapped. According to MNRE, India is expected to have 900 GW of renewable energy from commercially exploitable sources like solar energy: 750 GW; wind power: 102 GW; bioenergy: 25 GW; and minor hydropower: 20 GW [10]. The below concerning renewable energy alternate sources have been included in this study.
2.1 Wind Energy Wind energy is the main form of renewable energy. The availability of wind in India is during the season of monsoon for around two months before and after the monsoon. Even so, wind energy can be generated with the help of wind turbines. The combination of the turbine and generator produces electricity. A 7 m/s wind speed
106
S. K. Pathak et al.
is required to produce electricity from wind energy. However, the maximum wind speed in Bangladesh is around 5–6 m/s. Although, wind mapping is still needed to be done. As per the annual report of the MNRE for 2017–2018, the predicted wind power capacity was 302,251 GW [11].
2.2 Solar Energy Taking into consideration the geographical location, experts note that India has excellent potential for solar energy. The maximum intensity of daily solar insolation in India ranges from 5.0 kWh /m2 in winter to 8.36 kWh /m2 in summer. The duration from February to June offers a good amount of insolation. India has received fairly good sunshine between September and October. In the short winter season between November and January, then during the peak rainy season during July and August, there is less insolation in the country. The overall scenario of the potential of solar energy in India is, therefore, pretty good for power generation applications. According to a report of MNRE, India generates 5,000 trillion kWh of solar power every year [12].
2.3 Hydro Energy India has one of the largest river irrigation networks with thousands of dams worldwide. In the northern ranges of the Himalayas, there are several rivers and streams with regular flows. Since small projects in the field of hydropower will have an alternative to the remote hilly areas where the extension of the grid system is relatively uneconomic to meet the demand. The capacity of small hydropower projects in India is about 19.749 GW till 2018 [13].
3 Research Methodology The aim of the study is to identify the best renewable energy alternatives for electricity generation in India. To fulfil this purpose, we propose the AHP method to compute the weights of the assessment criteria and rank these renewable energy alternatives. In this study, a survey is completed for various main criteria and sub-criteria factors with the help of relevant experts. In this survey, ten relevant experts engaged in the selection process of various factors which are considered in this study for the selection of renewable energy alternatives. Due to limited time, only ten experts are engaged in this survey. Such experts included four senior research fellows, four university professors, two renewable energy Institute panel members in India. Due to limited time, only ten experts are engaged in this selection process.
Prioritization of Renewable Energy Alternatives by Using …
107
3.1 Multi-Criteria Decision-Making (MCDM) The MCDM approach is used by researchers to make effective decisions. By using statistical data, the MCDM attempts to solve the problem. It helps to create choices between alternatives concerning particular objectives and measurable criteria. This approach is mostly used in research fields to choose the option for the problem, which includes claims and several targets [14].
3.2 Analytic Hierarchical Process (AHP) Thomas L. Saaty explored the AHP model in the 1980s. It is very helpful for the decision-maker to fix the issue and rate the best solution to find the right decision [15]. One of the key issues facing organizations today is the need to select the best and consistent solutions in ways that preserve strategic consistency. AHP is an important mathematical model to support decision theory. During this study, AHP is used to select the best renewable energy resources for the production of electricity throughout India. To accomplish this goal, three main parameters have been laid out to allow unpredictable decisions to policymakers and identified as economic, technical, and environmental. Also, sub-criteria were established for the main criteria and used for certain decisions on the prioritization of renewable energy sources. The important steps involved in this methodology are as follows: • • • • •
Create the hierarchical structure Make a pair-wise comparative matrix and start comparing hierarchical factors Derive the matrix of priority Calculate the predilection accuracy Find the detailed priorities of the alternative.
Table 2 shows Saatey’s scale of importance for pair-wise comparison matrices. Since a pair-wise analogy can be very arbitrary. AHP uses a consistency check for comparisons, Eqs. 1 and 2 shows how the consistency index (CI) and consistency ratio (CR) calculated, respectively. CI =
(λ¯ max − n) n−1
(1)
The consistency ratio is calculated as CR =
CI RI
(2)
Table 3 represents the value of the random index (RI) for matrices of order (N) 1–10 attained by approximating random indices with the help of a sample size of 500 [16].
108
S. K. Pathak et al.
Table 3 The values of random consistency index n
1
2
3
4
5
6
7
8
9
10
RI
0
0
0.58
0.9
1.12
1.24
1.32
1.41
1.45
1.49
If the value of consistency ratio (CR) is less than or equal to 0.1, it suggests that the estimation within the matrix is acceptable or displays a good level of consistency. But if the value of CR is greater than the acceptable value, inconsistency has occurred in the matrix then the assessment process should be reviewed, reconsidered, and improved.
4 Decision Criteria Factors and Their Significance After completing the extensive literature review and survey with the help of relevant ten experts, following main criteria: Economic factors (F-1), Technical factors (F2), Environmental factors (F-3), and sub-criteria factors: Investment cost (F-11), Operation and maintenance cost (F-12), Job creation (F-13), Plant/infrastructure design (F-21), Technological maturity (F-22), Risk and reliability (F-23), Pollutant emission (F-31), Impact on the ecosystem (F-32), the Land requirement (F-33) were selected. These factors affect the selection of suitable renewable energy alternatives. Table 4 shows these factors with their codes. Three main criteria and nine sub-criteria factors for the selection of renewable energy alternatives have been finalized following a review of the previous literature and consultation with professionals in this field. These criteria and alternatives are presented in the form of a hierarchical structure (shown in Fig. 1). A precise summary of all factors is provided below to indicate their impact on the final objective of this study. Table 4 Classification of main and sub-criteria factors for selection of renewable energy Main criteria
Main criteria code
Sub-criteria
Sub-criteria code
References
Economic factors
(F-1)
Investment cost
F-11
[17]
Operation and maintenance cost
F-12
[8]
Technical factors
(F-2)
Environmental factors
(F-3)
Job creation
F-13
[18]
Plant/Infrastructure design
F-21
[19]
Technological maturity
F-22
[20]
Risk and reliability
F-23
[21]
Pollutant emission
F-31
[22]
Impact on ecosystem F-32
[12]
Land requirement
[23]
F-33
Prioritization of Renewable Energy Alternatives by Using …
109
Fig. 1 Hierarchical structure of factors affecting renewable energy alternatives
4.1 Economic Factors (F-1) Economic factors include the accessibility to the funds and conditions which emerge from the contextual expectations of a precise population group [24]. Economic criteria provide for the integration of the benefits and related costs in the execution of the project as per the parameters set out in the following sub-criteria. • Investment cost (F-11): It includes the total infrastructure and material costs. Based on these factors we can select the particular renewable energy resource. • Operational and maintenance cost (F-12): Corrective maintenance and preventive costs are included in this segment. • Job creation (F-13): Number of regional jobs created to install, maintain and repair these renewable energy technologies.
4.2 Technical Factors (F-2) This consideration requires various technical and operating criteria for the implementation as well as during the running period of these renewable energy technology systems [25]. It relates to the following sub-criteria: • Plant/Infrastructure design (F-21): Usually, the plant contains a variety of equipment. This equipment can be easily incorporated or not, the working of the plant is dependent on technical factors. • Technological maturity (F-22): Conversion efficiency of primary energy into electricity and the level of sustainable energy growth. • Risk and reliability (F-23): Risks that are exposed to financing because of fluctuations in the reportable exchange trend. A system’s ability to perform under design conditions but also to support failures.
110
S. K. Pathak et al.
4.3 Environmental (F-3) The environmental criteria are taken into account to the environmental effects for the growth of the renewable energy project, according to the framework set out in the preceding sub-criteria: • Pollutant emission (F-31): Pollution of greenhouse gasses emitted by the sustainable energy system to be implemented. • Impact on the ecosystem (F-32): Production of waste that affects the environment and human health. • Land requirement (F-33): Land and water services needed for the implementation of a renewable energy system.
5 A Real Case Study This segment offers a case report on alternative sources of renewable energy in India. The most frequently used economic, technical, and environmental factors and the participation of renewable energy alternatives: wind energy (E-1), solar energy (E-2), and hydro energy (E-3) were involved in this analysis.
5.1 Selection of Renewable Energy Sources by AHP While in the case of prioritizing main parameters, sub-criteria factors, and renewable energy alternatives, the below-mentioned group of experts participating in the survey method for the selection process. The questionnaire was constructed using the proposed AHP and was answered by ten experts. With each expert, 13 comparative matrixes were obtained as follows: three main and nine sub-criteria factors with a set of renewable energy alternatives. Each expert has been offered a certain weight and the analysis of all actions was carried out using the AHP method. [5].
5.2 Data Analysis In the case of prioritization of the factors, a comparative matrix (see Table 5) was obtained by solving the process. The target matrix and the consistency ratio for the parameters have been derived from Table 5. To obtain the priorities of sub-factors and renewable energy alternatives, the same procedure was applied. Tables 6, 7 and 8 shows the pair-wise comparative matrices for sub-criteria factors. Such pair-wise matrices were eventually resolved to obtain the results described and discussed in the next section.
Prioritization of Renewable Energy Alternatives by Using …
111
Table 5 Pair-wise comparative matrix for the 1st level criteria Main factors
Economic
Technical
Environmental
Criteria Weight
Economic
1
2
3
0.5390
Technical
1/2
1
2
0.2973
Environmental
1/3
1/2
1
0.1638
Maximum Eigenvalue = 3.0092; Consistency index = 0.004604; Consistency ratio = 0.007938
Table 6 Pair-wise comparative matrix for the 2nd level criteria for economic factors (F-1) Sub-factors
F-11
F-12
F-13
Criteria weight
F-11
1
3
2
0.5390
F-12
1/3
1
1/2
0.1638
F-13
0.5
2
1
0.2973
Maximum Eigenvalue = 3.0183; Consistency index = 0.009154; Consistency ratio = 0.01578
Table 7 Pair-wise comparative matrix for the 2nd level criteria for technical factors (F-1) Sub-factors
F-21
F-22
F-23
Criteria weight
F-21
1
1
2
0.3873
F-22
1
1
3
0.4429
F-23
1/2
1/3
1
0.1698
Maximum Eigenvalue = 3.0092; Consistency index = 0.004604; Consistency ratio = 0.007938
Table 8 Pair-wise comparative matrix for the 2nd level criteria for environmental factors (F-1) Sub-factors
F-31
F-32
F-33
Criteria weight
F-31
1
1
1/2
0.2409
F-32
1
1
1/3
0.2106
F-33
2
3
1
0.5485
Maximum Eigenvalue = 3.0183; Consistency index = 0.00915; Consistency ratio = 0.01578
Tables 9, 10, 11, 12, 13, 14, 15, 16 and 17 have set out the performance scores of renewable energy sources for the relevant sub-criteria. Here, renewable energy Table 9 Pair-wise comparative matrix for investment cost factor (F-11) Alternatives
Wind Energy
Solar energy
Hydro energy
criteria weight
Wind energy
1
1/5
1/3
0.1095
Solar energy
5
1
2
0.5812
Hydro energy
3
1/2
1
0.3091
Maximum Eigenvalue = 3.0036; Consistency index = 0.001848; Consistency ratio = 0.003186
112
S. K. Pathak et al.
Table 10 Pair-wise comparison matrix for operation and maintenance factor (F-12) Alternatives
Wind Energy
Solar energy
Hydro energy
Criteria weight
Wind energy
1
2
2
0.4905
Solar energy
1/2
1
1/2
0.1976
Hydro energy
1/2
2
1
0.3119
Maximum Eigenvalue = 3.0537; Consistency index = 0.02687; Consistency ratio = 0.04632 Table 11 Pair-wise evaluation matrix for job creation factor (F-13) Alternatives
Wind Energy
Solar energy
Hydro energy
Criteria weight
Wind energy Solar energy
1
2
1/2
0.3119
1/2
1
1/2
Hydro energy
0.1976
2
2
1
0.4905
Maximum Eigenvalue = 3.0537; Consistency index = 0.02688; Consistency ratio = 0.04634 Table 12 Pair-wise evaluation matrix for plant/infrastructure design factor (F-21) Alternatives
Wind energy
Solar energy
Hydro energy
Criteria weight
Wind energy
1
1/2
1/5
0.1222
Solar energy
2
1
1/3
0.2299
Hydro energy
5
3
1
0.6479
Maximum Eigenvalue = 3.0036; Consistency index = 0.001848; Consistency ratio = 0.003187 Table 13 Pair-wise comparison matrix for technological maturity factor (F-22) Alternatives
Wind energy
Solar energy
Hydro energy
Criteria weight
Wind energy Solar energy
1
2
3
0.5390
1/2
1
2
Hydro energy
0.2973
1/3
1/2
1
0.1638
Maximum Eigenvalue = 3.0092; Consistency index = 0.004604; Consistency ratio = 0.00793 Table 14 Pair-wise evaluation matrix for risk and reliability factor (F-23) Alternatives
Wind Energy
Solar energy
Hydro energy
Criteria weight
Wind energy
1
2
1/3
0.2395
Solar energy
1/2
1
1/4
0.1373
Hydro energy
3
4
1
0.6232
Maximum Eigenvalue = 3.0183; Consistency index = 0.009186; Consistency ratio = 0.01580
Prioritization of Renewable Energy Alternatives by Using …
113
Table 15 Pair-wise comparison matrix for pollutant emission factor (F-31) Alternatives
Wind Energy
Solar energy
Hydro energy
Criteria weight
Wind energy
1
1/3
1/2
0.1593
Solar energy
3
1
3
0.5889
Hydro energy
2
1/3
1
0.2519
Maximum Eigenvalue = 3.0539; Consistency index = 0.02694; Consistency ratio = 0.04646
Table 16 Pair-wise evaluation matrix for impact on ecosystem factor (F-32) Alternatives
Wind Energy
Solar energy
Hydro energy
Criteria weight
Wind energy
1
1
1/2
0.25
Solar energy
1
1
1/2
0.25
Hydro energy
2
2
1
0.50
Maximum Eigenvalue = 3; Consistency index = 0; Consistency ratio = 0
Table 17 Pair-wise evaluation matrix for land requirement factor (F-33) Alternatives
Wind Energy
Solar energy
Hydro energy
Criteria weight
Wind energy
1
1/2
1/3
0.1638
Solar energy
2
1
1/2
0.2973
Hydro energy
3
2
1
0.5390
Maximum Eigenvalue = 3.0092; Consistency index = 0.004604; Consistency ratio = 0.007939
resources are represented by (E-1), (E-2), and (E-3). After that these renewable energy alternatives transformed into pair-wise comparison matrixes and then resolved. Table 18 summarizes the findings obtained for the three renewable energy forms of criteria studied for the local priorities of the alternatives.
6 Results and Discussion The AHP analysis of renewable energy alternatives ranking framework can be structured as a three-phase hierarchy and presented in this section.
6.1 Main Factors Results The priority weight of the main criteria resulting from this calculation is shown graphically in Fig. 2. It indicates that the performance of the economic factor (F-1)
114
S. K. Pathak et al.
Table 18 Overall priority matrix for selection of suitable renewable energy technology (final result) Main criteria
Priority weight
Sub-criteria code
Priority weight
Wind energy
Solar energy
Hydro energy
Economic (F-1)
0.5390
F-11
0.5390
0.1096
0.5813
0.3092
F-12
0.1638
0.4905
0.1976
0.3119
F-13
0.2973
0.3119
0.1976
0.4905
F-21
0.3873
0.1222
0.2299
0.6479
F-22
0.4429
0.5390
0.2973
0.1638
F-23
0.1698
0.2395
0.1373
0.6232
F-31
0.2409
0.1593
0.5889
0.2519
F-32
0.2106
0.2500
0.2500
0.5000
F-33
0.5485
0.1638
0.2973
0.5390
Overall priority weight of renewable energy alternatives
0.2518
0.3490
0.3991
Ranking of renewable energy alternatives
3
2
1
Technical (F-2)
Environmental (F-3)
0.2973
0.1638
Fig. 2 Priority weight for the main criteria
has a maximum priority weight of “0.5389” followed by the technical factor (F2) with “0.2972”. The environmental factor (F-3) has the lowest priority weight of “0.1637”.
6.2 Sub-factors Results The global weights of the sub-factors are calculated by multiplying the local weight of each sub-factor by the local weight of the main factor of its type. If the global weights of the factor have been met, they can be organized in a hierarchy by assigning
Prioritization of Renewable Energy Alternatives by Using …
115
Fig. 3 Priority weight for the sub-criteria factors
a special rank to each according to the importance obtained. Figure 3 reveals that maintenance costs are the most important sub-criteria dependent on the main criteria for the selection of renewable energy resources. The final ranking of sub-barriers is: (F-11) > (F-13) > (F-22) > (F-21) > (F-33) > (F-12) > (F-23) > (F-31) > (F-32). It was concluded from Fig. 3 that investment cost (F-11) got the highest priority weight of “0.29” among all sub-factors, followed by the job creation (F-13) with a global priority weight of “0.1602”. The third and fourth positions among all subfactors are the technical maturity (F-22) & plant/infrastructure design (F-21) with respective priority weights of “0.1316” and “0.1151” The last five sub-factors in this list with marginal priority weights are land requirement (F-33) with “0.08983”, operation and maintenance cost (F12) with “0.08827”, risk and reliability (F-23) with “0.05049”, pollutant emission (F-31) with “0.03946” and impact on ecosystem (F-32) with “0.03449”.
6.3 Discussion In this paper, a benchmarking method was introduced to take tough decisions to eliminate the challenges associated with future energy demand. Understanding technical and economic challenges leads to valuable lessons in the creation of renewable energy policies and institutions in developed countries like India. The managers are aware to take decisions on the most successful and profitable use of renewable energy alternatives in India. Also, India will play a major role in the fight against climate change as a future global leader in the production and implementation of clean energy. The positive effect of green energy on the atmosphere and community
116
S. K. Pathak et al.
Fig. 4 Final weight percentage of renewable energy alternatives
is important. This will help renewable energy managers to concentrate on regions with unique capabilities and adopt steps to address the weakness possibility. The government should advocate the abolition of inappropriate, incompatible, and insufficient policies that support fossil fuels and technology, and refuse to consider the benefits of green energy technical, environmental, and economic benefits. In the sense of global warming, the topic of climate change and energy scarcity, India has identified an immediate need to prepare and adopt policies to increase its share in renewable energy, especially in hydro energy. Figure 4 presents the total priority weight percentage of three renewable energy alternatives. In the view of researchers, hydro energy is the best suitable renewable energy alternative with a weight percentage of (39.91%). The second is solar power at (34.9%) and wind energy is at last with (25.18%). Hydropower, followed by solar and wind, is the most suitable kind of renewable energy supply in India.
7 Conclusions The present study is conducted to prioritizing renewable energy alternatives according to their relevance in the Indian context. To begin with, several factors were recognized from the previous studies, which were then shortlisted by consulting with professionals in this field. These factors categorized into three main criteria and nine sub-criteria followed by three renewable energy alternatives. After that AHP methodology is used to form the pair-wise comparison matrices and assign the weight to these factors and alternatives. After calculation, the results showed that the economic category with the weight percentage of (53.90%) was important in the main criteria accompanied by technical (29.73%) and environmental (16.38%).
Prioritization of Renewable Energy Alternatives by Using …
117
For all of the alternatives evaluated, hydro energy was of the greatest importance with an overall weight percentage of (39.91%). The second was solar power (34.90%) followed by wind energy (25.18%). Such studies have demonstrated the significance of these renewable energy forms for energy planning in remote/rural parts of India. In terms of market capitalization and the industrial revolution, the rating of clean energy identified in this work helps investors to define the priorities of investment in the renewable energy sector. To the change from fossil fuels to renewable energy, this research needs to be done to establish a strategy to increase the contribution of renewable energy in India. This study opens up new horizons to resolve the issues of future energy demand in the context of India. So let us all make a serious and collaborative effort to set a precedent and lead by example by introducing renewable energy innovations across the nation to fulfill the dream of “Clean India, Green India”.
References 1. Shihan M (2019) Investigation E. Selection of renewable energy sources using analytic hierarchy process 2. Kamalapur GD, Udaykumar RY (2012) Electrical power and energy systems rural electrification in India and feasibility of photovoltaic solar home systems. Int J Electr Power Energy Syst 33:594–599. https://doi.org/10.1016/j.ijepes.2010.12.014 3. Mainali B, Silveira S (2012) Energy for sustainable development renewable energy markets in rural electrification: country case Nepal. Energy Sustain Dev 16:168–178. https://doi.org/10. 1016/j.esd.2012.03.001 4. Sindhu S, Nehra V, Luthra S (2017) Investigation of feasibility study of solar farms deployment using hybrid AHP-TOPSIS analysis: case study of India. Renew Sustain Energy Rev 73:496– 511. https://doi.org/10.1016/j.rser.2017.01.135 5. Tasri A, Susilawati A (2014) Selection among renewable energy alternatives based on a fuzzy analytic hierarchy process in Indonesia. Sustain Energy Technol Assess 7:34–44. https://doi. org/10.1016/j.seta.2014.02.008 6. Luthra S, Kumar S, Garg D, Haleem A (2015) Barriers to renewable/sustainable energy technologies adoption: Indian perspective. Renew Sustain Energy Rev 41:762–776. https://doi.org/ 10.1016/j.rser.2014.08.077 7. Büyüközkan, Gülçin SG (2017) Accepted Manuscript. Energy. https://doi.org/10.1016/j.ene rgy.2017.01.137. 8. Çolak M, Kaya ˙I (2017) Prioritization of renewable energy alternatives by using an integrated fuzzy MCDM model : a real case application for Turkey 80:840–53.https://doi.org/10.1016/j. rser.2017.05.194 9. Bazar B (2003) Selection of renewable energy sources using analytic hierarchy process 267–276 10. Nautiyal H, Varun (2012) Progress in renewable energy under clean development mechanism in India. Renew Sustain Energy Rev16:2913–9. https://doi.org/10.1016/j.rser.2012.02.008 11. Kumar A, Kaushik N, Sharma S, Mishra S (2010) Renewable energy in India: current status and future potentials.https://doi.org/10.1016/j.rser.2010.04.003 12. Mahtta R, Joshi PK, Kumar A (2014) Solar power potential mapping in India using remote sensing inputs and environmental parameters. Renew Energy 71:255–262. https://doi.org/10. 1016/j.renene.2014.05.037 13. Bhattacharya SC, Jana C (2009) Renewable energy in India: historical developments and prospects. Energy 34:981–991. https://doi.org/10.1016/j.energy.2008.10.017
118
S. K. Pathak et al.
14. Ishizaka A, Labib A (2011) Review of the main developments in the analytic hierarchy process 38:14336–14345 15. Saaty TL (2008) Decision making with the analytic hierarchy process 1 16. Pressures S, Resources W, Interests V, Values C. The Analytic Hierarchy Process ( AHP ) for decision making by Thomas Saaty decision making involves setting priorities and the AHP is the methodology for doing most decision problems are multicriteria maximize profits satisfy customer demands Maximize emp n.d.: 1–69 17. Punia Sindhu S, Nehra V, Luthra S (2016) Recognition and prioritization of challenges in growth of solar energy using analytical hierarchy process: Indian outlook. Energy 100:332–348. https:// doi.org/10.1016/j.energy.2016.01.091 18. Ikram M, Sroufe R, Zhang Q (2020) Prioritizing and overcoming barriers to integrated management system (IMS) implementation using AHP and G-TOPSIS. J Clean Prod 254:120–121. https://doi.org/10.1016/j.jclepro.2020.120121 19. Kahraman C, Cebi S (2009) A comparative analysis for multiattribute selection among renewable energy alternatives using fuzzy axiomatic design and fuzzy analytic hierarchy process. Energy 34:1603–1616. https://doi.org/10.1016/j.energy.2009.07.008 20. Rapal BKAL, Sumabat AKR, Lopez NSA (2017) Analytic hierarchy process for multi-criteria site selection of utility-scale solar and wind projects. Chem Eng Trans 61:1255–1260. https:// doi.org/10.3303/CET1761207 21. Ligus M, Peternek P (2018) Determination of most suitable low-emission energy technologies development in Poland using integrated fuzzy AHP-TOPSIS method. Energy Proc 153:101– 106. https://doi.org/10.1016/j.egypro.2018.10.046 22. Aya˘g Z (2017) Evaluating solar power plant location alternatives through a multiple-criteria decision making method 878–84. https://doi.org/10.20472/iac.2017.029.005 23. Ghimire LP, Kim Y (2018) An analysis on barriers to renewable energy development in. Renew Energy. https://doi.org/10.1016/j.renene.2018.06.011 24. Algarín CR, Llanos AP, Castro AO (2017) An analytic hierarchy process based approach for evaluating renewable energy sources 7:38–47 25. Yusoff S, Kardooni R (2012) Barriers and challenges for developing RE policy in Malaysia 28:6–10
Factors Influencing Customer Satisfaction and Belief Toward Car Business Model Sunil Anand and Piyush Singhal
Abstract In today’s competitive business scenario, the pivot in supply chain is consumer satisfaction. Apart from this some other factor like the environmental issue plays an important role in making the business model sustainable, which leads to an agile nature of the manufacturer. Consumer being important entity of supply chain become part of this continuously changing innovative environment. These will also affect consumer satisfaction level during the use of the product and at the time of use end. The purpose of this study is to focus on these factors and access the level of satisfaction and belief of consumer regarding Indian car business model. For this a research tool is developed, surveys of existing car customers were carried out and the data collected is analyzed with the help of exploratory factor analysis. Four latent factors considered important named impact of agility, unorganized secondary market, environmental issue, and eco-leasing, which can be related to customer satisfaction and belief. On the basis of factors evaluated and business model suggested for Indian car market, a sustainable business model can be implemented, which can be tested for the present challenging scenario of automobile sector. Study focuses on long term leasing model with concept of circular economy for Indian car market. Keywords Agility · Eco-leasing · Secondary market · Sustainability
1 Introduction The Indian car market deals mainly with conventional business model based on selling of product. More business models like car leasing, carpooling or sharing, etc. are also emerging in Indian market but with less popularity in India. The present selling business model needs to be re-evaluated in terms of sustainability and efforts
S. Anand (B) Symbiosis University of Applied Sciences, Indore, M.P 453112, India P. Singhal GLA University, Mathura, U.P. 281406, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_11
119
120
S. Anand and P. Singhal
are made in this sense [7]. The reason being benefits in terms of environment and society and at the same time for firms also, sustainability is the solution [30]. In order to achieve this, many efforts are being made, and the most prominent steps involved various legislative by government authorities regarding emissions, set up of various private–public partnership for monitoring, supply chain partners focused on continuously changing product and process leading to an agile environment, emission labeling, fast new product development, etc. All these continuously changing elements have an impact on the existing customer using the product. In this study, an attempt is made to study the various factors which influence customer satisfaction and belief toward the present business model. At the same time perception of the customer about the new style of business model making its foot into the market also needs to be checked. With the aid of literature review specific concerned questions which relate to the impact on the satisfaction of customers about present business model scenario and perception about the leasing business model has been developed. Data collection is done through Google forms send to the various car owners through emails. The data collected is then analyzed through principles of exploratory factor analysis and factors are established, which could be considered as the pivot in feature studies.
2 Literature Review The fast launch of new models as per the consumer demand and requirement of the car market as per the consumer demand and some time to stimulate the demand leads to agile system [3, 7]. For the automobile industry, customers may resist innovation because it conflicts their beliefs and it threatens to create changes in their routines [26, 28]. The life cycle of passenger car is getting reduced due to frequent introduction of more products and models into the market [22, 23]. The small product life cycle threatens the end-user about obsolescence and dissatisfaction on up-gradation in a small period of time. Agility may also forces customer to enter into the secondary car market. This market is dominated over 90 percent by the unorganized sector. The limitation of the unorganized secondary car market is reduced goodwill and involvement of brokers [29]. Secondary car market in India is dominated by the unorganized sector in which there is no consumer protection [5]. In second hand car market, there is little information about the quality of the car and thus buyer always find himself as cheated [6]. The uncertainty about the used car market increases with the various legislative which imposed ban on the cars of a particular age. To understand this issue, study is required to be done on government and dealer outlets role regarding environmental issues like NGT legislative and efforts of dealers in creating awareness regarding pollution control. The problem of emission can be tackled by assessment of effect in actual than what suggested officially by adaptation of various regulation and legislation [4, 9, 15]. Proper inspection and maintenance system is the most important factor to control the emissions under Indian perception needs to be formulated [21, 32]. Thus review suggests checking the customer view on
Factors Influencing Customer Satisfaction and Belief Toward Car …
121
legislative and role of government and dealers regarding awareness on Environmental issues. This study also involves understanding the view of customer regarding the traditional business model and alternate business model. In traditional business model, vehicle manufacturer is the main player, directing both supplier and the distribution system. In this model the main source of profit is the sales of vehicle [32]. Vehicle producers earns their profits mainly from the cars sale and the spare parts sales, but they do not focus on earnings associated with the use of the vehicle. The automotive industry is a leader in the development of new products, processes, and organizational solutions. The sale strategy will drastically change by 2035, and the main source of revenue and profits will be the holistic intermodal mobile solutions rather than traditional sales, after-sales, and financial services. The industry will focus on society, and the environment and traditional way of selling cars in showrooms slowly leads to end [8]. Business models, if developed and managed properly, can support sustainable business processes, products, services, and environmentally and socially beneficial forms of consumption [12]. Sustainable business model is the need of the time. It is define as the meeting needs of the present generation without compromising the ability of future generations to meet their needs [19]. A better business model which relays on the concept of ecoleasing of cars with the inclusion of PSS and circular economy is suggested for Indian car market [18]. The proposed alternative business model is based on the concept of eco-leasing, which involves a low or no down payment leasing of product with low-environmental impact. This system includes the pre-production, production, and distribution in a single phase. The car manufacturer will take care of distribution, maintenance, and return of car at the end of life, i.e., the lease period [2]. The transformation needed for converting traditional to eco-leasing model is depicted in Fig. 1.
Traditional business model
Alternative business model
Fig. 1 Transformation from traditional to alternate model [2]
122
S. Anand and P. Singhal
Leasing of automobile can reduce the cost to the customer as well improve on depreciation factor which makes the consumer worry [1, 2]. On the basis of review of literature, an effort is made to point out the research gap in the next section of study and the objectives of study are formulated on the basis of the gap identified.
3 Objectives As literature suggests, the focus should be toward service concept instead of selling concept and is the global perception also. When the maximum share of market is beard by traditional business model in Indian market, the perception of customer toward the business must be needed to understand in terms of customer satisfaction or their belief toward the running trend. To have a sustainable approach the alternate model which focus on eco-leasing must be tested in terms of existing customer view. The objective of this study is to develop and validate a scale for consumer satisfaction and belief toward present business model along with the eco-leasing model and to develop factors using exploratory factor analysis.
4 Methodology An inductive approach is useful when the information available is limited. Also an inductive approach is used to develop theory [20]. According to the objectives mentioned and on the basis of literature, the methodology used is an inductive research with exploratory factor analysis (EFA) as a tool for analysis. Exploratory factor analysis (EFA) is useful in inductive research. This involves development of new instrument’s scale [16, 27]. As suggested through literature review about the limited prior information makes the use of EFA more appropriate for the study undertaken. The methodology is a quantitative type of research which uses explorative factor analysis, and the research design used is a survey method. The technique adopted for data collection is convenience sampling. The sample size used for the study is 320. Instrument for data collection involves a structured questionnaire. The questionnaire includes a series of 25 questions separated into two sets; first set covers 6 demographic questions involving basic details about the education, age of respondent, and their car. Second, part of the questionnaire includes 17 questions. The questions were frame using 5-point Likert scale. The respondent was to rate each question on the basis, various points highlighted through literature review. The data collected was analyzed using SPSS Software. Frequency distribution of data under the category of age and education is given in Fig. 2. Exploratory factor analysis is used for investigating the relationships for a large number of variables and to find way for summarization of information in a smaller set of factors [13, 14]. It also evaluate relationships among the factors. For EFA,
Factors Influencing Customer Satisfaction and Belief Toward Car …
Education
UG
PG
Ph.D. 16%
36%
123
Age 20-30
31-40
51-60
>60
41-50
10% 3% 12% 26%
48%
49%
Fig. 2 Frequency distribution
Table 1 KMO and Bartlett’s test
Kaiser–Meyer–Olkin measure of sampling adequacy 0.83 Bartlett’s Test of Sphericity: Approx. Chi-Square
1936.47
Df
136
Sig
0.00
the Kaiser-MeyerOlkin Measure (KMO) of sampling adequacy must be above 0.50 [17]. For Kaiser–Meyer–Olkin (KMO) test of sampling adequacy and Bartlett’s test of Sphericity see Table 1. EFA tries to uncover complex patterns by exploring the dataset and testing predictions [10]. As EFA is the first step in building scales, it operates on the principle that measurable and observable variables can be reduced to fewer latent variables that share a common variance. Factors are rotated to have better interpretation as unrotated factors are ambiguous. The rotation is done to attain an optimal simple structure which attempts to have each variable load on as few factors as possible, but maximizes the number of high loadings on each variable [33]. This is done using Principal Component Analysis technique and Varimax method of rotation. Kaiser’s Eigen value-based criterion aids to development of latent factors, and it indicates 4 factors having Eigen value more than 1; the result of EFA method can be seen in Table 2. On the basis of tabulation, the next part of the study highlights the results and the implications of the study.
5 Results and Discussion The analysis conducted involves three stages (1) to identify the correlation between the factors, (2) extract a factor, and (3) rotate factors. Items with high correlation are
124
S. Anand and P. Singhal
Table 2 Principal components factor analysis—Varimax rotation factors indicating the latent factors Factor1 DealerAwarenessSuport2
0.84
DealerAwarenessSupport1
0.84
GovtLegislative1
0.67
GovtImplication2
0.63
GovtLegislative2
0.63
GovtImplication1
0.61
Factor 2
SecondaryMarket1
0.79
Depreciation Effect
0.78
SecondaryMarket2
0.77
SecondaryMarket3
0.67
Factor 3
ServiceSupport
0.85
LaunchImpact1
0.63
Leasing2
0.62
Leasing1
0.50
LaunchTime1
Factor 4
0.79
LaunchTime2
0.76
LaunchImpact2
0.71
Eigen value
4.96
2.30
1.66
1.20
Proportion of variance explained
29.19
13.56
9.79
7.08
Cumulative variance explained
29.19
42.75
52.55
59.63
Alpha
0.70
0.79
0.82
0.73
placed in similar constructs as they measure the same concept, while items found in the different constructs have low correlation because it measures different concepts. Extraction of factors is done using SPSS program. Data were analyzed descriptively for reliability (Cronbach Alpha value), and factor analysis is used to obtain solutions. Table 1 shows the values of KMOs (Kaiser–Meyer–Olkin). 83 > 0.6 shows that these items are sufficient for inter-correlation while the Barlett test was significant (ChiSquare = 1936.47, p < 0.05). Tests of KMOs help researchers determine whether the item is suitable for the implementation of factor analysis. This indicates that the data do not have a serious problem of multicollinearity, and then the item is suitable to run for factor analysis. Varimax rotation method conducted on the item to produce meaningful orthogonal factors (uncorrelated) and can be interpreted more accurately. This method can determine the items that are categorized according to the same characteristics of the same measure in order of getting rid to items that are not relevant for further analysis. Factor analysis (Table 2) was conducted using Varimax rotation to confirm 4 constructs. Results of factor analysis (total variance explained), a total of four
Factors Influencing Customer Satisfaction and Belief Toward Car …
125
components (factors) that gives the Eigen value greater than 1. Eigen value indicates the proportion of the variance contribution of each factor extracted by factor analysis. All factors converged into four factors explaining 59.63% of the total Variance. The factors having Eigen values greater than one are considered significant, and all the factors with Eigen value less than one are considered insignificant and disregarded [11]. Following 4 factors are classified: Factor 1: This factor was represented by three items and was named Impact of Agility accounted for the amount of variance 29.19%.This factor included the items Variable LaunchTime1, LaunchTime2, and LaunchImpact2. Factor 2: This factor was represented by four items and was labeled Unorganized secondary market accounted for the amount of variance 13.56%. This factor comprised items representing Depreciation Effect, SecondaryMarket1, SecondaryMarket2, and SecondaryMarket3. Factor3: This factor was represented by six items and was named as Environmental Issue accounted for the amount of variance 9.79%. This factor included the items was Dealer Awareness Support1, Dealer Awareness Support2, Government Legislative1, Government legislative2, Government Implication1, and Government Implication2. Factor 4: This factor was represented by four items and was named Ecoleasing accounted for the amount of variance 7.08% Consisted items were Service Support, Leasing1, Launch Impact 1, and Leasing2. The reliability of measurements is examined using the Cronbach’s α test, where the coefficients are calculated using SPSS. The alpha value of 0.70 or greater indicates an acceptable level of Peng et al. [25]. As shown in Table 2, all alpha values are greater than 0.70, suggesting that the internal reliability of the construct is good. The Overall Cronbach’s α for the scale developed is 0.834. The new trends dominating the global markets are the introduction of electric vehicle leasing and environmentally friendly cars that lead to sustainable development in the car manufacturing industries as well in the overall environmental situation [24]. In a developing country like India, these trends take more time to established but the efforts need to be done keeping in mind that consumer is the key to business. When share of traditional business model in India is still in dominating zone, the evaluations of the factors which could concern the consumer must needed to be consider. The results obtained in the study helps in extracting some of those factors. The whole business communities worldwide now are in search of a sustainable solution for business, which must also be evident for the Indian car sector. Leasing is considered as the new progressive trends toward the use of scientific and technological progress, and leasing has a positive effect on economic growth [1, 31]. Keeping these facts in mind, the factors are developed and analyzed, which relates the customer view on the traditional business model as well as to get customer perception about the acceptance on an alternate business model, which works on the principle of Eco-leasing. The results of study strongly indicate toward the consumer perception for a better business model for Indian car market. The companies on global domain are shifting from just manufacturing a product toward developing capabilities that make it possible for them to offer services and solutions as a complement to their basic
126
S. Anand and P. Singhal
product offering, and this study helps understanding the concept as per the Indian automobile sector, which really needs some boost from the recent down fall.
6 Conclusion The traditional business model that relies heavily on product sales to customers’ needs to be revised to make the whole business process and process more sustainable. The rapid state of business and the various laws imposed by the government on the age of a particular type of car compel the customer to deal with the secondary car market in advance by selling his existing car to get a new improved model. The secondary car market in a country like India is not yet organized and is largely controlled by inexperienced mediators. All of these issues affect a car customer who is at a level of satisfaction during and after use. This study attempts to analyze factors that actually determine the level of satisfaction and belief of the customer in a focused situation. For this, a tool is needed to assess the perception of existing customers on a proposed issue. The outrage of the research mainly lies in the fact that the list of questions is being tried and validated to determine the results of the hypothesis that has been made. As a result, various questions were developed with the help of available courses, suggestions from experts, and knowledge in the field of study. The developed tool is tested mainly on a sample of respondents, and evaluating results leads to reconstruction and modification of the questionnaire. With the confirmation of the questionnaire test developed, it is used to collect data from various existing car customers. The method adopted was a simple sample of data collection. Since the research conducted is based on the fact that there is not much past history or tools available, it confirms the use of material analysis (EFA) in the analysis of the collected data. Prior to performing the analysis, a tool is needed to validate the measurement, performed by Kaiser–Meyer–Olkin (KMO) sample sufficiency test and Bartlett’s Sphericity test and anti-image correlation matrix, both of which confirm the results as given in the path section. The four factors are divided into categories namely Impact of agility, Unorganized secondary market, Issue Environment, and eco-leasing. With the help of this study, a measurement tool is developed and validated. The data analysis reveals a variety of factors and reflects customer perceptions about the traditional business model and other business models. This study is limited to scale-making and related analysis. These items may help to understand the consumer’s view of the traditional sales business model currently operating in the Indian car market while at the same time these items may be related to the sustainable business model focusing on the Indian car service and car market as it has become a global trend now. These factors can be used further to form the relationship model to develop an alternative business model focusing on leasing of car for the Indian market.
Factors Influencing Customer Satisfaction and Belief Toward Car …
127
References 1. Aizcorbe A, Starr-McCluer M (1997) Vehicle ownership, purchases, and leasing, consumer survey data. Monthly Labor Rev 34–40 2. Anand S, Choudhary AK, Singhal P (2019) Car eco-leasing encouraging product service system with circular economy to help environment. India J Environ Prot 39(4):352–358 3. Barve A (2011) Impact of supply chains agility on customer satisfaction. In: 2010 international conference on e-business, management and economics IPEDRl 3:325–329 4. Berger IE, Corbin RM (1992) Perceived consumer effectiveness and faith in others as moderators of environmentally responsible behaviors. J Public Policy Mark 11(2):79–89 5. Bhakta P (2018) Unorganized players in used car market pose threat to consumer protection, India times press, 10 November 2015. https://auto.economictimes.indiatimes.com/4793. Accessed 19 June 2018 6. Bhandari J, Phadtare M (2013) Study of factors influencing pre owned car purchase. PARIDNYA MIBM Res J 1(1):1–13 7. Bhattacharya S, Mukhopadhyay D, Giri S (2014) Supply chain management in Indian automotive industry: complexities, challenges, and way ahead. Int J Manag Value Supply Chains (IJMVSC) 5:49–62 8. Bolesnikov M, Stijaci MP, Radisi M, Takaci A, Borocki J, Bolesnikov D, Bajdor P, Dziendziora J (2019) Development of a business model by introducing sustainable and tailor-made value proposition for SME clients. Sustainability 11:1–16 9. Brand C (2016) Beyond Dieselgate: implications of unaccounted and future air pollutant emissions and energy use for cars in the United Kingdom. Energy Policy 97:1–12 10. Child D (2006) The essentials of factor analysis, 3rd edn. Continuum International Publishing Group, New York, NY 11. Costello AB, Osborne JW (2005) Best practices in exploratory factor analysis: four recommendations for getting the most from your analyses. Pract Assess Res Eval 10(7):1–9 12. Freund FL (2015) Sustainable business models for eco-design and innovation—the case of River simple. In: The challenges of eco-innovation EcoSD annual workshop, pp 57–67. 13. George D, Mallery P (2003) SPSS for windows step by step: simple guide and reference 11.0 update, 4th edn. Allyn and Bacon, NY 14. Hair JF Jr, Black WC, Babin BJ, Anderson RE, Tatham RL (2006) Multivariate data analysis, 6th edn. Pearson Prentice Hall, New Jersey 15. Haq G, Weiss M (2016).CO2 labeling of passenger cars in Europe: status, challenges, and future prospects. Energy Policy 95:324–335 16. Henson RK, Roberts JK (2006) Use of exploratory factor analysis in published research: common errors and some comment on improved practice. Educ Psychol Measur 66(3):393–416 17. Jung S, Lee S (2011) Exploratory factor analysis for small samples. Behav Res Methods 43(3):701–709 18. Kanda Y, Nakagami Y (2006) What is Product-Service Systems (PSS)?—a review on PSS researches and relevant policies, IGES Kansai research centre discussion paper. Institute for Global Environmental Strategies 19. Khavul S, Bruton GD (2012) Harnessing innovation for change: sustainability and poverty in developing countries. J Manage Stud 50(2):285–306 20. Kolar E, Lindström L (2018) Future business model for OEMs in the automotive industry, Thesis. Department of Technology Management and Economics, Division of Entrepreneurship and Strategy, Chalmers University of Technology, Gothenburg, Sweden 21. Kumar A, Anand S (2012) Status of vehicular pollution in NCT of Delhi. Int J Adv Res Manage Soc Sci 1(3):85–100 22. MacVaugh J (2010) Limits to the diffusion of innovation: a literature review and integrative model. Eur J Innov Manage 13(2):197–221 23. Mahadevan B (2007) Operations management—theory and practice. Pearson Prentice-Hall, New Delhi
128
S. Anand and P. Singhal
24. Mehta J (2019) Car leasing In India: Breaking the stereotypical definition of luxury, July 12. Accessed 24 Jan 2020. https://vinodkothari.com/2019/07/car-leasing-in-india-breaking-the-ste reotypical-definition-of-luxury/ 25. Peng B, Tu Y, Elahi E, Wei G (2018) Extended producer responsibility and corporate performance: effects of environmental regulation and environmental strategy. J Environ Manage 218:181–189 26. Ram S, Sheth JN (1989) Consumer resistance to innovations: the marketing problem and its solutions. J Consum Mark 6(2):5–14 27. Reio TG, Shuck B (2014) Exploratory factor analysis: implications for theory, research, and practice. Adv Dev Hum Resour 17(1):12–25 28. Schneberger J, Luedeke J, Vielhaber M (2018) Agile transformation and correlation of customer-specific requirements and system-inherent characteristics—an automotive example. In: 28th CIRP design conference-journal procedia CIRP, Nantes, France, vol 70, pp 78–83 29. Sharma K (2012) Pre-owned car market in India: a study of marketing strategies of car makers. Int J Eng Sci Manage 2(2):80–184 30. Shevchenko A, Lévesque M, Pagell M (2016) Why firms delay reaching true sustainability. J Manage Stud 53(5):911–935 31. Vakhitov DR, Aidar Z (2014) Leasing as a factor of economic growth. In: 2nd global conference on business, economics, management and tourism, Prague, Czech Republic, pp 30–31 32. Wells P (2008) Alternative business models for a sustainable automotive industry. In: Tukker A, Charter M, Vezzoli C, Andersen MM (eds) Perspectives on radical changes to sustainable consumption and production. Greenleaf, Sheffield, pp 80–98 33. Yong AG, Pearce S (2013) A beginner’s guide to factor analysis: focusing on exploratory factor analysis. Tutorials Quant Methods Psychol 9(2):79–94
A Review on Musculoskeletal Disorders and Design of Ergonomics Aids with Relevance to Lower Back and Lumbopelvic Pain in Pregnant Women Nikhil Yadav, M. L. Meena, G. S. Dangayach, and Yashvin Gupta Abstract This is an era of “Women Empowerment” spreading overall in various sectors. The factors like drastic change in economic conditions and other social perspective led to the employment of pregnant women rendering full time service. There has been a requirement aroused for the ergonomic design of the workplaces reducing the risk factors associated with the health of pregnant women and ensuring her greatest productivity. This paper reviews various ergonomic interventions that could be employed to maintain the postural stability of the pregnant women. A lot of researches have been done by people from medical backgrounds and physiotherapists in drafting out appropriate bodily postures leading to sustainability of the pregnant women in different work cultures. Centre of mass of the body changes with uneven distribution of mass and shifting of centre of gravity causing postural imbalance in the body. The work has mainly focused on prolong standing and sitting postures required at various tasks. Due to standing for long hours, the legs are prone to varicose veins that get exaggerated in pregnant women due to increase in weight of the body. In the same context sitting for long hours causes musculoskeletal disorder in hip and pelvic girdles. Some of the ergonomic interventions like designing footrest, antifatigue mats, sit-stand stools etc. designed for preoperative settings could be redesigned in relevance to workplace designing for pregnant women. The usability index of maternity support belt used to maintain proper posture is yet to be investigated. Since postural stability equilibrium decreases with the advent of pregnancy during its third trimester, a lot of tests along with the biomechanical solutions have to be manifested for the same. Due to decline of the equilibrium the musculoskeletal disorders are reported to be significantly increased which calls for the ergonomic interventions in workplaces for pregnant women. Keywords Pregnant · Musculoskeletal · Ergonomics · Lower back pain · Lumbopelvic pain N. Yadav (B) · Y. Gupta Department of Mechanical Engineering, Government Women Engineering College, Ajmer, India M. L. Meena · G. S. Dangayach Department of Mechanical Engineering, Malaviya National Institute of Technology, Jaipur, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_12
129
130
N. Yadav et al.
1 Introduction The participation of women towards the development of a nation is remarkably higher. In that case their health issues cannot be taken casually particularly when they are undergoing the stage of pregnancy. Mechanical and industrial engineers have to redesign the workplace along with certain ergonomic interventions that could improve the productivity of the unit. The main area of focus in this paper is at investigating the factors responsible for inducing postural loads causing stress at the lower back and lumbar regions. Herniated disk is one of the forms of lower back pain. This feature accelerates with the old age because the muscles are no longer capable of sustaining the musculoskeletal disorder. Now we can treat the pregnant situation as an analogy to this by considering the fact that the muscles during pregnancy. The postural load in the lower and upper extremity due to wrong practising of the posture is retained throughout the life and thus affecting the quality of living. The literature review has been done on the articles on postures at different stages like sitting, standing, walking, lifting a load, doing a repetitive task etc. Studies have also suggested that the postural changes have been varying individually with remarkable trend variations in lumber lordosis. The nature of jobs also affects the postural behaviour of pregnant women. The main body parts that used to get affected is trunk and arm posture. The constraints imposed to these parts affects the postures and behavioural pattern. The posture of the trunk segments is affected by the restriction imposed on anterior tilt of the pelvis by the increasing size of lower trunk. Researches are yet trying to find the effect of change in trunk dimensions on the upper part of the body especially while sitting. It would not be right to think that all the effects on posture would be reverted after postpartum. The musculoskeletal disorders caused due to change in the behavioural pattern would remain intact and could cause serious problems in the long run of life. The novelty of the work lies in digging up the factors and facts that actually affect the life of pregnant women postpartum. Women workers demand protection during the period of pregnancy as it leads to inability to cope up with many work factors requiring intense manual labour and awkward working conditions. Further the solution of the problems under study lies in ergonomic interventions that could be recommended at the workplaces for the smooth flow of the job without hampering the physical and mental health of the pregnant women.
2 Methodology The preferred reporting items for systematic reviews and Meta-analyses (PRISMA) guidelines were used [1]. Keywords searched on databases were biomechanics, lower back pain and lumbopelvic pain, girdle pain, musculoskeletal disorders in pregnant women and anthropometry of pregnant women. One hundred and fifty-three articles were selected based on the title of the articles and abstracts. Further screening process was based on content of articles like methodology used and conclusion given on the
A Review on Musculoskeletal Disorders and Design of Ergonomics …
131
Fig. 1 Flowchart showing the process of screening and reviewing the articles as per (PRISMA) [1]
musculoskeletal disorders in pregnant women. Duplicate articles were screened off minimizing the number to one hundred and seventeen, out of which fifty-seven were found non-relevant. Sixty articles were checked under the eligibility criteria that the contents must be based on the lumber pelvic pain and lower back pain. Nineteen articles on pregnant women regarding lower back pain and lumber pelvic pain as well as based on medical aspects of pregnancy without any relevance to biomechanics were discarded (Fig. 1).
3 Literature Review A systematic literature review has been done in exploring the findings done by various researchers, doctors, physiotherapist and gynaecologists. Some of the key references along with their findings are tabulated as below (Table 1).
132
N. Yadav et al.
Table 1 Literature review Authors
Year Findings
Bullock-Saxton [2]
1991 Investigation has been done on determining certain postural changes. These have been found in relevance to change in spinal morphology other anthropometrical changes subsequently with the advent of the pregnancy. Postural changes due to change in spinal morphology is accompanied by lower back pain and lumbopelvic pain. Several postures are attained by the pregnant women unintentionally to get themselves accommodate according to the workplace design which needs to be modified so that to lessen the musculoskeletal disorders
Morrissey [3]
1998 Author has identified certain risk factors with the postural point of view that could induce the pain in the load bearing parts of the body. A few ergonomic interventions could be recommended to decrease the postural load and hence the musculoskeletal disorders in lumber region and the lower back pain
Hattori et al. [4]
2000 The researchers have studied the changes in bodily posture as per the demand of the tasks and advent of stress on the different body parts with the nature of posture and plane of symmetry of vertical loading
Gilleard et al. [5]
2002 Authors have investigated the alignment of the upper body in different postures and found that there has been week correlation between the anthropometrical changes in the upper portion of the body and the progression period of pregnancy however the spinal cord has been reported to get flattered postpartum
Wang et al. [6]
2005 An experimental analysis was done based on regression model in predicting the strength of the posterior region, mainly back. It can be inferred that with the advent of the pregnancy the load bearing capacity of the bones, ligaments and tendons decreases and thus causing musculoskeletal disorders in the body. Attaining wrong postures introduces postural mechanical load causing stress
Cheng et al. [7]
2006 The authors have done a case study in determining the adverse effects of repetitive tasks on posture and its impact on the musculoskeletal disorders
Skaggs et al. [8]
2006 Authors have studied the impact of socio-economic status on the conditions during pregnancy. The prevalence of lower back pain and lumbopelvic pain seems to be exaggerated due to lack of awareness, training and workload (continued)
A Review on Musculoskeletal Disorders and Design of Ergonomics …
133
Table 1 (continued) Authors
Year Findings
Motozawa et al. [9]
2008 The authors have studied the anthropometric characteristics of pregnant drivers in Japan and the effect of seating postures. They have done a comparative study with American drivers than with the Japanese ones and found a significant difference that could lead to the anthropometric design of car seats as per the users’ requirement
Dumas et al. [10]
2008 Researchers have made the ergonomic interventions in the form of desk board attachment for workstations. It improved the posture by reducing kyphosis and hence making them to sit in the upright posture but produced discomfort at the lower back and pelvis area. The impact of this intervention was found to be adverse in relevance to the upper part of the body
Ho et al. [11]
2008 Usability index of maternity belt is investigated by the authors. It has been investigated that the efficiency of maternity belts in lowering back pain and pelvic girdle is quite lower and also might be having adverse effect on it
Dumas et al. [12]
2009 Researchers stresses on the increased tendency of the back muscles against fatigue during pregnancy. According to the article, the muscles train themselves in adapting the situations during pregnancy in handling postural load but the rate of adaptation is slow that back pain persists during the duration and postpartum as well
Ponnaupula and Boberg [13]
2010 The researcher has investigated the effects on lower extremities with the advent of pregnancy and found that the changes depend upon the anthropometrical characteristics of the body
Robinson et al. [14]
2010 Authors have investigated the pain location in pregnant women by doing an experimental survey. The subjects were subjected to straight leg rest test and by the help of pain diagram and disability rating index they come to a comparative conclusion between the lower back pain and pelvic girdle pain
Beaucage-Gauvreau et al. [15] 2012 Authors have done case study on pregnant women in Benin, West Africa to find the cause of lower back pain due to different postural movements of trunk as per the workplace demand and found that the repetitive changes in the trunk posture are the main factors of fatigue in women causing lower back pain Anan and Shinkoda [16]
2013 The authors have investigated the adverse effect of start of gait cycle immediately after standing off the chair and recommended for an ergonomic intervention that could avoid the same (continued)
134
N. Yadav et al.
Table 1 (continued) Authors
Year Findings
Brown and Johnston [17]
2013 Researchers have investigated the severity of the lower back pain and other similar musculoskeletal disorders postpartum and during delivery. It was found that due the persistence of this disorder the women at are at larger risk. It invites further the ergonomic interventions and other proper solution to combat with the back pain
Mohseni-Bandpei et al. [18]
2013 Authors have facilitated the future researchers for diagnosing the lower back pain effectively by incorporating the use of surface electromyography
Inanir et al. [19]
2014 Investigation of changes in hormones during pregnancy and its effect on dynamic stability was studied. The risk of falls was quantified during all the three trimesters and probable factors causing the risks were observed. The critical stage of risk was found more in third trimester comparatively
Ozturk et al. [20]
2015 Authors studied the correlation of lower back pain and postural and dynamic stability and found that with the advent of lower back pain the stability gets hampered
Cakmak et al. [21]
2015 The authors have studied the various physiological changes occurring during pregnancy and the factors occurring due to these changes which are responsible for increasing the risk of fall
Opala-Berdzik et al. [22]
2015 The authors have made a comparative study on the postural stability in static conditions during various stages of pregnancy and postpartum and found that there is a significant change in the same
Branco et al. [23]
2016 The researchers have studied the effect of certain bodily characteristics like body composition that change during the advent of pregnancy causing the change in gait pattern and inducing sway movement
Carvalho et al. [24]
2016 Authors have made a survey on finding the severity level, frequency and the most probable period of pregnancy in which the lock back mostly occurs. Frequency is observed to be quite high, mostly during nights. The symptom reported by the subjects is in the form of burning sensation. The lower back pain is most likely to occur in second trimester of the pregnancy
Javadian et al. [25]
2016 Authors have suggested the ways to manage the pelvic pain on medical background. But yet the scope of research is opened for the biotechnology engineers to interfere with the ergonomic interventions that could minimize the pain by decreasing postural load and make then to carry out tasks at an ease (continued)
A Review on Musculoskeletal Disorders and Design of Ergonomics …
135
Table 1 (continued) Authors
Year Findings
Roossien et al. [26]
2017 Authors have collaborated in this interdisciplinary research by introducing a smart chair and hence setting an example of automation based on machine learning. The feedbacks on the postures attained by the subjects could be given by the chair in improving the same and hence catering for the healthy working conditions at the workplaces. Though the study has not been done on the pregnant women but yet the work has given a wide scope of research to be pursued further
Bey et al. [27]
2018 Author has investigated the usability index of maternity belts with the advent of pregnancy for the stabilization of pelvic girdle and its effect on musculoskeletal disorders. The effect of the same has been reported satisfactorily in the first two trimesters, however with the increase in the mass of the body and shifting of centre of gravity in third trimester, it lead to decrease in the effectiveness of maternity support belt
Danna-Dos-santos et al. [28]
2018 Authors have focused on certain strategies that would help the pregnant ladies to maintain an upright posture. The anthropometry of the body plays a major role in balance mechanism where the larger body would have more sway movement as compared to the average one. The risk of falls has been observed to be greater in the last two trimesters of pregnancy
Aoyama et al. [34]
2018 Author have studied the effect of lumbopelvic pain on gait pattern described by pregnant women. Asymmetry in the movement has been noticed in the form of enhancement of rotational and translational movements with the advent of lumbopelvic pain leading to the dynamic instability
Forczek et al. [29]
2019 Authors investigated the effect of gain in weight with the progression of pregnancy period which found to affect the gait stability to a larger extent. The authors have done a follow up study to establish a correlation between base of support in increasing the gait stability and opens a wide scope of research further
Pardo et al. [30]
2018 The researchers have done work on find the correlation of pain experienced and change in foot posture with the advent of pregnancy. There has been no significant relation thus found but postural stability could be accessed on the changes in foot posture during the period (continued)
136
N. Yadav et al.
Table 1 (continued) Authors
Year Findings
Bertuit et al. [31]
2018 Authors have done a comparative study on gait analysis with and without girdle pain and also compared the two types of pelvic belts. They found that pregnant women with pelvic girdle pain and with belt showed the same gait cycle as the pregnant women without pelvic girdle pain and hence emphasizing the usability of pelvic belts
Takeda et al. [32]
2018 The researchers have focused on the factors that would cause the risk of fall of pregnant women and found the inability of joints to maintain the balance and musculoskeletal changes of the body in the lead role for causing such risks
Mackenzie et al. [33]
2018 Authors have tried to find out the experiences, feelings and other emotional attributes shown by the pregnant women undergoing pelvic girdle pain. The condition would affect the productivity of the workplace and also influence the quality of life. This further necessitated the designing of workplaces and introduction of ergonomic interventions
Tanigawa et al. [34]
2018 Authors have established a relation between lumbopelvic pain and gait characteristics. Result shows that translational and rotational motion of the pregnant women having lumbopelvic pain is affected to a great extent
Reliquias and Kuebler [35]
2019 The researchers have studied the effectiveness of ergonomic intervention in the form of sit-stand workstation especially at the workplaces requiring sitting for long hours and hence a rest period in the form of stand is being required on the employee part. The effectiveness of the sit to stand workstation has not been found significant but it has dormant impact which could be further modified to be thus exploited
Christensen et al. [36]
2019 Researchers have focused on dynamic stability of the pregnant women in the progression period of 2nd trimester. The gait is observed to be affected significantly with the pain in pelvic girdle making them to describe swayed movements and slower pace
Catena [37]
2019 Author conducted the study to investigate the changes in anthropometry succeeded by change in dislocation of centre of mass and posture after pregnancy. The distribution of mass in the pregnant ladies was significantly studied in anterior, posterior and lateral directions (continued)
A Review on Musculoskeletal Disorders and Design of Ergonomics …
137
Table 1 (continued) Authors
Year Findings
Starzec et al. [38]
2019 Authors have made a case study on Polish and Norwegian women to investigate the factors contributing to the pelvic girdle pain. This way a scope of wide research has been exposed for the future researchers to determine the correlation between the anthropometrical and bodily characteristics with the musculoskeletal disorders. The morphological characteristics is observed to be changed depending upon the region and community
Kokanali and Caglar [39]
2019 Authors have done a substantial work in correlating striae gravidarum and lock back pain. The evaluation of the low back pain cannot be done quantitatively but qualitatively only by pain questionnaires and other media. Striae gravidarum can assist in prediction of severity of low back pain
Haddox et al. [40]
2020 Author has done a tremendous work by doing the study on a biomechanical model of pregnant women to find the factors that could be responsible for the risk of falls. The generic model is said to be of a little use to predict the same as the anthropometry, musculoskeletal morphology as well as other bodily characteristics may vary from person to person. Author has studied the gait cycle on the model by varying the centre of mass, distribution of mass and inertia. Such a model would help the researchers in biomechanical field to predict other dormant factors leading to dynamic instability
4 Results and Discussions Certain physiological changes that occur in the body of the women with the advent of pregnancy tend them to retain the certain postures to make themselves at ease. But the postures attained thus might not be recommended as per the changes in anthropometrical characteristics and hence could induce the postural load followed by lower back pain and lumbopelvic pain. Pregnant women working as work force in certain industries especially the education sector could have adverse effect in the form of decreased productivity. Literature review traversed across revealed the result of the investigation on different types of tasks and the workplaces that are more prone to postural in stability and hence becomes the major source of musculoskeletal disorders. Repetitive tasks and the asymmetrical nature of the job poses postural imbalance jobs requiring diverse postures for long durations cause pain in certain parts of body. Prepartum conditions can handle the postural load as the strength of the body is high enough whereas in pregnant condition due to release of certain hormone, the laxity of the ligaments increases and hence the strength bearing capacity of the body decreases.
138
N. Yadav et al.
A lot of research has been done on recommending the ergonomic interventions. The researcher’s community comprise of physiotherapists, biotechnology engineers, biomedical engineers etc. who have done substantial work in determining the factors based on physiological, hormonal and anthropometrical changes. The work left now on the part of mechanical and industrial engineers to reduce the factors causing musculoskeletal disorders by incorporating the proper technical reasoning. There has been no proof of work on designing the leg rest properly.
5 Conclusions A lot of work has been done in this area by physiotherapists, gynaecologist and doctors in the form of statistical analysis on the survey data and certain experimental results. A lot of work is yet to be done in finding the correct postures that could minimise the postural loads acting on the body. Further it would not help only by finding the correct postures but also the incorporations of ergonomic interventions is required at a large scale. Certain ergonomic interventions suggested by researchers are maternity belt, pelvic belts, slippers for pregnant women etc. that could minimise the postural load and assist in carrying out the pregnancy. The main research area to be focussed upon is to reduce the musculoskeletal disorders especially in the form of lower back pain and lumbopelvic pain.
References 1. Moher D, Liberati A, Tetzlaff J, Altman DG, The PG (2009) preferred items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6:e1000097. https://doi.org/ 10.1371/journal.pmed.1000097 2. Bullock-Saxton JE (1991) Changes in posture associated with pregnancy and the early postnataal period measured in standing. Physiol Theor Pract 7:103–109 3. Morrissey SJ (1998) Work place design recommendations for the pregnant worker. Int J Ind Ergon 21:383–395 4. Hattori Y, Ono Y, Shimaoka M, Hiruta S, Shibata E, Ando S, Hori F, Takeuchi Y (2000) Effects of box weight, vertical location and symmetry on lifting capacities and ratings on category scales in Japanese female workers. Ergonomics 16(26):2031–2042 5. Gilleard WL, Crosbie J, Smith R (2002) Static trunk posture in sitting and standing during pregnancy and early postpartum. Arch Phys Med Rehabil 83:1739–1744 6. Wang M, Leger AB, Dumas GA (2005) Prediction of back strength using anthropometric and strength measurements in healthy females. Clin Biomech 20:685–692 7. Cheng PL, Dumas GA, Smith JT, Leger AB, Plamondon A, McGrath MJ, Tranmer JE (2006) Analysis of self-reported problematic tasks for pregnant women. Ergonomics 49(3):282–292 8. Skaggs CD, Prather H, Gross G, Jeorge JW, Thompson PA, Nelson DM (2007) Back and pelvic pain in an undeserved United States pregnant population: a preliminary descriptive survey. Nat Univ Heal Sci. https://doi.org/10.1016/j.jmpt.2006.12.008 9. Motozawa Y, Hitosugi M, Tokudome S (2008) Analysis of seating position and anthropometric parameters of pregnant Japanese drivers. Traffic Inj Prev 9(1):77–82
A Review on Musculoskeletal Disorders and Design of Ergonomics …
139
10. Dumas GA, Upjohn TR, Delisle A, Chrpentier K, Leger A, Plamondon A, Salazar E, McGrath MJ (2009) Posture and muscle activity of pregnant women during computer work and effect of an ergonomic desk board attachment. Int J Ind Ergon 39:313–325 11. Ho SSM, Yu WWM, Lao TT, Chow DHK, Chung JWY, Li Y (2008) Effectiveness of maternity support belts in reducing low back pain during pregnancy: a review. J Clin Nurs 18:1523–1532 12. Dumas GA, Leger A, Plamondon A, Charpentier KM, Pinti A, McGrath M (2010) Fatigability of back extensor muscles and low back pain during pregnancy. Clin Biomech 25:1–5 13. Ponnaupula P, Boberg JS (2010) Lower extremity changes experienced during pregnancy. J Foot Ankle Surg 49:452–458 14. Robinson HS, Mengshoel AM, Bjelland EK (2010) Pelvic girdle pain, clinical tests and disability in late pregnancy. Man Ther 15:280–285 15. Beaucage-Gauvreau E, Dumas GA, Lawani M (2012) Trunk postural demands of occupational activities of some merchant pregnant women in Benin, West Africa. Ergonomics 55(10):1218– 1228 16. Anan YSM, Shinkoda K (2013) Biomechanics of rising from a chair and walking in pregnant women. Appl Ergon 44:792–798 17. Brown A, Johnston R (2013) Maternal experience of musculoskeletal pain during pregnancy and birth outcomes: significance of lower back and pelvic pain. Midwifery 29:1346–1351 18. Mohseni-Bandpei MA, Watson JM, Richardson B (2013) Application of surface electromyography in the assessment of low back pain: a literature review. Phys Ther Rev 5(2):93–105 19. Inanir A, Cakmak B, Hisim Y, Demirturk F (2014) Evaluation of postural equilibrium and fall risk during pregnancy. Gait Posture 39:1122–1125 20. Ozturk G, Kulucu DG, Aydog E, Kaspar C, Ugurel B (2015) Effects of lower back pain on postural equilibrium and fall risk during the third trimester of pregnancy. J Matern Fetal Neonatal Med ISSN 1476-7058 21. Cakmak B, Ribeiro AP, Inanir A (2015) Postural balance and the risk of fall during pregnancy. J Matern Fetal Neonatal Med ISSN 1476-7058 22. Opala-Berdzik A, Blaszczyk JW, Bacik B, Cieslinska-Swider J, Swider D, Sobota G, Markiewicz A (2015) Static postural stability in women during and after pregnancy: a prospective longitudnal study. PLoS ONE 10(6):e0124207 23. Branco M, Santos-Rocha R, Vieira F, Silva M, Aguiar L, Veloso AP (2016) Influence of body composition on gait kinetics throughout pregnancy and postpartum period. Hindawi Publishing Corporation Scientifica, pp 1–12 24. Carvalho MECC, Lima LC, Terceiro CADL, Pinto DRL, Silva MN, Cozer GA, Couceiro TCDM (2017) Low back pain during pregnancy. Rev Bras Anestesiol 67(3):266–270 25. Jvadian P, Shobeiri A (2016) Management of Pelvic pain. Semin Colon Rectal Surg 27:59–63 26. Roossien CC, Stegenga J, Hodselmans AP, Spook SM, Koolhas W, Brower S, Verkerke GJ, Reneman MF (2017) Can a smart chair improve the sitting behavior of office workers? Appl Ergon 65:355–361 27. Bey ME, Arampatizs A, Loegeerlotz K (2018) The effect of a maternity support belt on static stability and posture in pregnant and non-pregnant women. J Biomech 75:123–128 28. Danna-Dos-santos A, Magalhaes AT, Silva BA, Duarte BS, Barros GL, Silva MFC, Silva CS, Mohapatra S, Degani MA, Cardoso VS (2018) Upright balance control strategies during pregnancy. Gait Posture 66:7–12 29. Forczek W, Ivanenko Y, Curylo M, Fraczek B, Maslon A, Salamaga M, Suder A (2019) Progressive changes in walking kinemeatics throughout pregnancy—a follow up study. Gait Posture 68:518–524 30. Pardo FJV, Amo ALD, Rios MP, Gijon-Nogueron G, Yuste CC (2018) Changes in foot posture during pregnancy and their relation with musculoskeletal pain: a longitudinal cohort study. Women Birth 31:e84–e88 31. Bertuit J, Leyh C, Feipel V (2018) Pelvic belts and pregnanacy-related pelvic girdle pain: influence on temporal and spatial gait parameters. Int Biomech 5(1):104–112
140
N. Yadav et al.
32. Tkeda K, Yoshikata H, Imura M (2018) Changes in posture control of women that fall during pregnancy. Int J Women’s Health Reprod Sci 6(3):255–262 33. Mackenzie J, Murray E, Lusher J (2018) Women’s experiences of pregnancy related pelvic girdle pain: a systematic review. Midwifery 56:102–111 34. Tanigawa A, Morino S, Aoyama T, Takahashi M (2018) Gait Analysis of pregnant patients with lumbopelvic pain using inertial sensor. Gait Posture 65:176–181 35. Reliquias LLR, Kuebler JC (2019) The behavior of pain in response to sit-stand workstations: a systematic review. Phys Ther Rev 24(5):223–228 36. Christensen L, Veirod MB, Vollestad NK, Jakobsen VE, Stuge B, Cabri J, Robinson HS (2019) Kinematic and spatiotemporal gait characteristics in pregnant women with pelvic girdle pain, asymptomatic pregnant and non-pregnant women. Clin Biomech 68:45-52 37. Catena RC, Campbell N, Wolcott WC, Rothwell SA (2019) Anthropometry, standing posture, and body center of mass changes up to 28 weeks postpartum in Caucasians in the United States. Gait Posture 70:196–202 38. Starzec M, Baszak AT, Tamowski A, Rongies W (2019) Pregnancy-related pelvic girdle pain in polish and Norwegian women. Nat Univ Heal Sci. https://doi.org/10.1016/j.jmpt.2019.03.002 39. Kokanali D, Cagler AT (2019) Hidden association between the presence and severity of striae gravidarum and low back pain in pregnancy. Eur J Obstet Gynecol Reprod Biol 233:49–52 40. Haddox AG, Hausselle J, Azoug A (2020) Changes in segmental mass and inertia during pregnancy: a musculoskeletal model of the pregnant woman. Gait Posture 76:389–395
Comparative Analysis of Supplier Selection Based on ARAS, COPRAS, and MOORA Methods Integrated with Fuzzy AHP in Supply Chain Management Josy George, Pushkal Badoniya, and J. Francis Xavier Abstract Supplier selection has been a fundamental issue for any assembling association. It is constantly been a difficult undertaking for managers to discover the most appropriate supplier for supply among the numerous numbers, which depends on the various sorts of assessment interaction and rule. Albeit the quantity of various models dynamic (MCDM) strategies is accessible for tackling the MCDM issue, and it is seen that in the greater part of these techniques utilized for positioning the elective give variety in outcomes when the progressions are made for measures loads. This paper plans to portray the utilizations of three MCDM strategies for taking care of the provider choice issue. Additive Ratio Assessment (ARAS), Complex Proportional Assessment (COPRAS), and Multi-Objective Optimization dependent on Ratio Analysis (MOORA) Method utilizing Fuzzy Analytical Hierarchy Process (F-AHP) used to do the individual loads of the measures. In this paper with the assistance of an illustrative model, the positioning presentation of the referenced techniques is determined and contrasted and one another. In the illustrative model, an assembling firm is hoping to choose the most reasonable provider for supply among the fourteen-provider dependent on seven unique rules like Part Per Million (PPM), Quality, Price/Cost, Standardization, Service, Flexibility, and Delivery. Keywords ARAS · COPRAS · MOORA · MCDM · Supply chain management · Supplier selection
1 Introduction In the present scenario, supply/service provider assessment and determination are the absolute most testing undertakings of production network the board, which keep up the connection among provider and assembling firm, with the goal that a drawn out responsibility can be set up. MCDM are scientific techniques which empower
J. George (B) · P. Badoniya · J. F. Xavier School of Mechanical Engineering, Vellore Institute of Technology, Bhopal, M.P., India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_13
141
142
J. George et al.
the synchronous appraisal of numerous quantifiable and non-quantifiable strategies and operational components and furthermore remember numerous individuals for the decision-making. Multi-Criteria Decision-Making (MCDM) strategies typically include the leader to survey choices concerning the models for the choice and furthermore to allot the standards loads of importance. At that point the most ideal choice can be picked dependent on the assigned loads to the measures and afterward positioning to the other options. Particularly, with the assistance of ms-excel, the techniques have gotten simpler for the clients and analyst, so they have discovered incredible acknowledgment in numerous zones of decision in economy or the executives, as per scientist quantities of strategies are determined for weightage and rank figuring. Among numerous multi-rules procedures MAXMIN, MINMAX, SAW, AHP, TOPSIS, SMART, VIKOR, and ELECTRE are the most habitually utilized strategies. MCDM strategies will assist with improving the nature of choices by settling on the dynamic cycle more express, reasonable, and effective as indicated by Lourenzutti and Krohling [16], some complicated selection problem explained by Podvezko and Sivileviˇcius [18], some cases discussed by Tavana et al. [22] regarding supplier selection, [10, 20] also based on different MCDM problems. MCDM might be a process pointed toward finding the least complex option among the entirety of the adequate other options. In the greater part of the issues, the plenitude of rules for the examination of options has gotten far reaching. At the end of the day, leaders try to unwind the different issues raised by MCDM.
2 Theoretical Analysis of MCDM Techniques As discussed above, different researchers used and implemented MCDM techniques in a large variety of territories. Like in recent years, numerous MCDM tools have been able to discover their applications [14, 19]. In this paper a mathematical model is utilized to examine the MCDM methods, we attempted to execute and looked at the changed MCDM procedures for the choice of the most appropriate provider for the assembling business. The embraced strategies are Additive Ratio Assessment (ARAS), Complex Proportional Assessment (COPRAS), Fuzzy Analytical Hierarchy Process (F-AHP), and Multi-Objective Optimization dependent on Ratio Analysis (MOORA) Method. This part comprises of the multitude of hypothetical subtleties of F-AHP, ARAS, COPRAS, and MOORA strategy. All the means are clarified in detail underneath under Sects. 2.1, 2.2, 2.3, 2.4.
2.1 Fuzzy Analytic Hierarchy Process (F-AHP) The fuzzy AHP which was proposed by Van Laarhoven and Pedrycz [23] can be depicted as an augmentation of the Analytic Hierarchy Process (AHP), which remains as an incredible MCDM tool for tackling both quantitative and subjective issues. The
Comparative Analysis of Supplier Selection Based on ARAS …
143
fuzzy AHP is remarkable tool, its capacity to manage the fluffiness and dubiousness of phonetic decisions by building up a viable prioritization. The fuzzy AHP technique was a result of the failure of the AHP to manage imprecision and emotional ness in the pair-wise examination measure talked about by Aikhuele and Turan [2] and Badizadeh and Khanmohammadi [4]. AHP was created by Saaty [21], and AHP has been generally used to address various models dynamic (MCDM) issues. It allots needs to different choice standards by playing out a pair-wise examination between options. In a conventional AHP model, the primary level indicates the objective; the rules and sub-rules (assuming any) are in the third and fourth levels individually, and the fourth level contains the other options. In F-AHP, semantic factors addressed by three-sided fuzzy numbers are being used to perform pair-wise correlations among the models and choices themselves. This is accomplished by building a fuzzy judgment lattice. Laarhoven and Pedcrycz [23] were one of the initial analysts to incorporate fuzzy rationale into AHP. They acquainted the triangular fuzzy numbers enrolment work to be utilized in F-AHP for pair-wise examination. Buckley [9] acquainted another strategy with register fuzzy loads and explicitly used three-sided enrolment capacities. Different researcher acquainted new techniques with utilize three-sided participation capacities in pairwise examinations. This examination uses the technique depicted by Buckley [9] and utilizes the three-sided fuzzy participation capacity to ascertain relative loads of rules just as choices. The explanation behind utilizing the three-sided participation work is that while meeting the case organization which is talked about in the following segment, and all the estimated values for every basis as depicted by the buying staff were around a solitary incentive rather than any norm or a scope of qualities. Following are the steps to be performed: Step 1: Comparing criteria and alternatives using linguistic variables shown in Table 1. As we can see from Table 1, the linguistic terms are mapped to triangular fuzzy numbers. Suppose if the expert suggests that “Criterion 1 (Cr1 ) is strongly important than criterion 2 (Cr2 )”, then it will take (4, 5, 6) fuzzy triangular value. On the other Table 1 Linguistic term and the corresponding triangular fuzzy numbers Linguistic variables
Saaty value Fuzzy triangular values
Equally dominant
1
(1, 1, 1)
Slightly dominant
3
(2, 3, 4)
Strongly dominant
5
(4, 5, 6)
Very strongly dominant
7
(6, 7, 8)
Extremely dominant
9
(9, 9, 9)
The intermittent values between two adjacent scales (2, 4, 2 6, 8) 4
(1, 2, 3)
6
(5, 6, 7)
8
(7, 8, 9)
(3, 4, 5)
144
J. George et al.
hand, while constructing a pair-wise matrix, the comparison of Cr2 to Cr1 will have a fuzzy triangular value (1/6, 1/5, 1/4). The sample pair-wise comparison matrix “D” is shown below. Here d ij indicates the comparison of ith criterion with jth criterion using fuzzy triangular values as mentioned in Table 1. For the above example, if Cr1 is strongly important than Cr2 , d 12 value represents this comparison and will have to be equal to d 12 = (4, 5, 6). As we can see from Table 1, the linguistic terms are planned to triangular fuzzy numbers. Assume if the master recommends that “Standard 1 (Cr1 ) is emphatically significant than basis 2 (Cr2 )”, at that point it will take (4, 5, 6) fuzzy triangular value. Then again, while building a couple astute framework, the examination of Cr2 to Cr1 will have a fuzzy triangular value (1/6, 1/5, 1/4). The example pair-wise correlation grid “A” is appeared in condition 1. Here d ij demonstrates the examination of ith model with jth rule utilizing fluffy three-sided values as referenced in Table 1. For the above model, if Cr1 is emphatically significant than Cr2 , d 12 esteem addresses this examination and should be equivalent to d 12 = (4, 5, 6), ⎡
d11 · · · ⎢ .. . . D=⎣ . . dn1 · · ·
⎤ d1n .. ⎥. . ⎦ dnm
Step 2: The mean value of fuzzy comparison values is calculated for each criterion that is shown with in the equation below, ri =
n di j , i = 1, 2, . . . , n
( j=1)
Step 3: Find the vector summation of each r i . Then locate the reciprocal summation vector and change the fuzzy triangular fee to make it in growing order. Then locate the fuzzy weight of every criterion i (wi ) via multiplying each r i with this reverse vector. W eight (wi ) = ri ⊗ (r1 ⊕ r2 ⊕ · · · ⊕ rn )−1 = (lwi , mwi , uwi ). Operations on fuzzy numbers are defined as follows: a1 ⊕ a2 = (l1 + l2 , m 1 + m 2 , u 1 + u 2 ), a1 ⊗ a2 = (l1 . l2 , m 1 .m 2 , u 1. u 2 ), a1−1
=
1 1 1 , , , u 1 m 1 l1
Comparative Analysis of Supplier Selection Based on ARAS …
145
1 1 1 1 a1n = l n , m n , u n . Step 4: In this step, the weights are calculated, triangular fuzzy numbers (lwi , mwi , uwi ) are to convert them into crisp values, and the center of gravity method was applied: Mi =
(lwi + mwi + uwi ) . 3
Step 5: M i is a non-fuzzy member that needs to be normalized using the above equation, Mi N i = n
i=1 Mi
.
After finding the normalized weights of all the factors and therefore the alternatives, the score is calculated by multiplying each various weight with the connected criteria. The choice with the very best score is stratified first and may be selected by the decision-maker.
2.2 Additive Ratio Assessment (ARAS) A standard MCDM problem is the project of ranking a small range of choice alternatives, each and every specifically described by using capability of a number determination requirements which want to be systematically taken into account. In this paper, the ARAS approach is utilized for the universal overall performance contrast of suppliers for a manufacturing firm. According to the ARAS approach, an account characteristic figuring out the problematic relative vicinity of a possible choice is straight away proportional to the relative have an effect on of values and weights of the predominant requirements seen in a problem. The technique of fixing troubles by means of ability of the use of ARAS approaches, in situations when MCDM problem consists of amazing criterion and non-beneficial criterion, in accordance to [24], Additive ratio assessment (ARAS) has the following steps. Step 1: At first, the related decision/evaluation matrix is formulated. In any MCDM problem (discrete optimization problem), the relevant data is represented by the decision matrix showing preferences from feasible alternatives rated on n criteria (attributes).
146
J. George et al.
⎡
x12 ⎢x ⎢ 21 ⎢ ⎢ . X =⎢ ⎢ . ⎢ ⎣ . xm1
x12 x22 . . . xm2
... ... ... ... ... ...
⎤ x1n x2n ⎥ ⎥ ⎥ . ⎥ ⎥, . ⎥ ⎥ . ⎦ xmn
where m is the quantity of choices, n is the quantity of models portraying every other option, and x ij is the worth addressing the presentation of ith elective as for jth basis. Step 2: Determine the optimal value of all criterion considered for evaluation. Let x0 j be the optimal value of jth criterion. If the optimal value of jth criterion is known, then x0 j = max xi j for beneficial criterion, x0 j = min xi j for non-beneficial criterion. Now, taking into account the optimal values of all the considered criteria, the original decision matrix is reformulated as follows: ⎡
x01 ⎢ . ⎢ ⎢ ⎢x X = ⎢ i1 ⎢ . ⎢ ⎣ . xm1
x0 j . xi j . . xm j
... ... ... ... ... ...
⎤ x0n . ⎥ ⎥ ⎥ xin ⎥ ⎥. . ⎥ ⎥ . ⎦ xmn
Step 3: In this step, all the initial criteria values are normalized while employing the following equations: x For beneficial criteria, ri j = m i j xi j . i=0
r∗
For non-beneficial criteria, ri∗j = x1i j , ri j = m i j r ∗ . i=0 i j Step 4: From the normalized decision matrix, the corresponding weighted normalized decision matrix is developed using the following equation: vi j = w j ∗ ri j , i = 1, 2, 3 . . . , m, where w j is the weight of jth criterion and ri j is the normalized performance of ith alternative with respect to jth criterion. Step 5: In this stage, the optimality function value is determined. Si =
n
( j=1)
vi j , i = 0, 1, 2, 3, . . . , m,
Comparative Analysis of Supplier Selection Based on ARAS …
147
where S i is the estimation of optimality work for ith elective. The most noteworthy estimation of S i consistently connotes the best other option, while the most reduced S i esteem recognizes the most unfavored other option. Considering the computational interaction of the ARAS technique, it tends to be uncovered that the optimality work S i has an immediate and corresponding relationship with x ij qualities and loads wj of the thought about measures and their overall effect on the eventual outcome. The needs of the choices would thus be able to be resolved dependent on S i value. Thus, it is advantageous to assess and rank the choice choices utilizing S i value. Step 6: The level of alternative utility degree is dictated by contrasting and a variation, which is frequently taken as the preferably best worth (S 0 ). The utility degree U i of ith alternative can be calculated employing the following equation: Ui =
Si , i = 1, 2, . . . , m. S0
It is very clear that the determined estimations of U i lie in the interval of [0, 1] and can be requested in an expanding arrangement to give a total positioning of the thought about other options. The mind boggling relative proficiency of the plausible choices can likewise be resolved by the utility capacity esteems.
2.3 Complex Proportional Assessment (COPRAS) The Complex Proportional Assessment (COPRAS) was firstly introduced by Zavadskas et al. [25] firstly introduced method, which takes account of the separately discussed minimizations and maximization parameters of [17] that influence the proportional dependency and utility levels of the alternatives [1, 13]. This approach contrasts alternatives and sets their expectations by taking into account parameters weights according to the contradictory criteria. The importance and utility (priority) of the alternatives are directly and proportionally dependent. The following steps are defined by Adali and Isik [1] in the COPRAS procedure. The Complex Proportional Assessment (COPRAS) was first and foremost presented by Zavadskas et al. [25]. In this technique, which considers the limiting and the boosting models independently talked about by Madic et al. [17] which impacts the corresponding reliance and level of utility of the options [1, 13]. This technique looks at the other options and decides their needs under the clashing rules by considering the models’ loads. It accepts immediate and corresponding conditions of the importance and utility degree (need) of the other options. Coming up next are the means of the COPRAS strategy as portrayed by Adali and Isik [1]. Step 1: The initial decision matrix, represented by X.
148
J. George et al.
⎡
X = [xi j ]m∗n
x12 ⎢x ⎢ 21 ⎢ ⎢ . =⎢ ⎢ . ⎢ ⎣ . xm1
x12 x22 . . . xm2
... ... ... ... ... ...
⎤ x1n x2n ⎥ ⎥ ⎥ . ⎥ ⎥ . ⎥ ⎥ . ⎦ xmn
where x ij is the evaluation value of ith alternative for jth criterion, m is the number of alternatives, and n is the number of criteria. Step 2: The Normalization of the decision matrix X done by using the following equation: xi j R = [ri j ]m∗n = m i=0
xi j
.
Step 3: Calculation of the weighted normalized decision matrix, D, with the aid of the use of the following equation: D = [yi j ]m∗n = ri j · wi j , (i = 1, . . . , m and j = 1, . . . , n), where r ij is the normalized overall performance price of ith selection on jth criterion and wj is that the weight of jth criterion. The total of weighted normalized values of each criterion is continually up to the weight for that criterion: m
yi j = wi j .
i=1
Step 4: In this step, the sums of weighted normalized values are calculated for each the useful and non-beneficial criteria by using the subsequent equations: S+i =
n
j=1
y+i j S−i =
n
y−i j
j=1
where y+i j and y−i j are the weighted normalized values for the beneficial and nonbeneficial criteria, respectively. Step 5: Determination of the relative significances of the alternatives, Qi , by using the subsequent equation: Q i = S+i
m S−i mini S−i i=1 + , m S−i i=1 mini S−i /S−i
Comparative Analysis of Supplier Selection Based on ARAS …
149
where S− min is the minimum value of S−i . Step 6: Calculation of the quantitative utility, U i , for ith alternative by using the following equation: Ui =
Qi · 100% Q max
where Qmax is the maximum relative significance value. As a final result of the equation, utility values of the picks vary from 0 to 100%. The higher is the fee of U i , the greater is the precedence of the alternative. Based on the alternative’s utility values, a whole rating of the aggressive preferences can be obtained.
2.4 Multi-Objective Optimization Based on Ratio Analysis (MOORA) The MOORA approach refers to the proportion examination and reference point approach. This methodology is chiefly utilized for multi-property improvement. This methodology was introduced by Brauers and Zavadskas [6]. Numerous scientists like [3, 5, 7, 8, 11, 12, 15] utilized MOORA in various ways as per application. The MOORA strategy is a moderately new multi-models dynamic technique dependent on proportion framework and dimensionless estimation. MOORA technique is made out of five significant advances. Step 1: The initial decision matrix, represented by X. The technique begins with a choice network of reactions of various options in contrast to assessment measures, and the information given is addressed as grid X m*n . x ij is the presentation proportion of ith elective on jth characteristic, m is the quantity of choices, and n is the quantity of properties. ⎡
x12 ⎢x ⎢ 21 ⎢ ⎢ . X =⎢ ⎢ . ⎢ ⎣ . xm1
x12 x22 . . . xm2
... ... . . . ...
⎤ x1n x2n ⎥ ⎥ ⎥ . ⎥ ⎥, . ⎥ ⎥ . ⎦ xmn
where x ij is the performance measure of ith alternative on jth criteria, m is the number of alternatives, and n is the number of criteria. Step 2: Normalization of the decision matrix X. MOORA refers to a proportion framework where every reaction of an option on models is contrasted with a denominator, which is agent for all options worried that unbiased. This proportion can be communicated as follows:
150
J. George et al.
xi j xi∗j =
m i=1
xi2j
,
where xi∗j is a dimensionless number that belongs to the interval [0, 1] representing the normalized performance of ith alternative on jth criteria. Step 3: For multi-objective optimization. These normalized performances square measure delivered within the case of maximization (for helpful attributes) and deducted within the case of decrease (for non-beneficial attributes). Then the improvement problem becomes yi =
g
xi∗j −
j=1
n
xi∗j ,
j=g+1
where g is the quantity of characteristics to be expanded, (n − g) is the quantity of properties to be limited, and yi is the standardized evaluation estimation of ith elective concerning all the ascribes. At times, it is regularly seen that a few ascribes are a higher priority than others. To give more significance to a trait, it very well may be increased with its comparing weight (importance coefficient). Step 4: Determine the weighted assessment value. Generally, it is frequently seen that some choice measures are a higher priority than others. To build the need of rules, it very well may be duplicated with its weight. At the point when these models loads are thought about becomes as follows: yi =
g
j=1
w j · x i∗j −
n
w j · x i∗j
j=g+1
where wj is the priority of jth criteria, which can be assigned using different multicriteria decision-making methods. Step 5: Ranking of alternatives. Decision alternatives have to be compelled to be hierarchal within the selection order in accordance to lowering values of yi *. Assessment value may be positive or negative looking forward to standards situation and priority values.
3 Illustrative Example An Original Equipment Manufacturing (OEM) company wants to choose its best supplier for the supply of raw materials. For the selection of a suitable supplier, various criteria are to take into account, so the application MCDM approaches determine which supplier will meet OEM requirements best according to the criteria values set by management. The various criteria involved in the decision-making
Comparative Analysis of Supplier Selection Based on ARAS …
151
Table 2 Decision matrix for alternatives with criteria type Suppliers
PPM
Quality
Price/Cost
Standardization
Service
Flexibility
Delivery
S1
0.10
7.6
390
5
46
18
11
S2
0.05
5.5
360
4
32
21
11
S3
0.05
5.3
290
6
32
21
11
S4
0.05
5.7
270
5
37
19
9
S5
0.10
4.2
240
7
38
19
8
S6
0.10
4.4
260
4
38
19
8
S7
0.10
3.9
270
2
42
16
5
S8
0.05
7.9
400
3
44
20
6
S9
0.05
8.1
380
5
44
20
6
S10
0.10
4.5
320
9
46
18
7
S11
0.05
5.7
320
8
48
20
11
S12
0.05
5.2
310
7
48
20
11
S13
0.10
7.1
280
5
49
19
12
S14
0.50
6.9
250
6
50
16
10
Criteria type
Min
Max
Min
Max
Max
Max
Max
process are PPM (Part per million) customers, Quality, Price/Cost, Standardization, Service, Flexibility, and on-time delivery. Here criteria PPM and Price/Cost are nonbeneficial and the attributes about other criteria are beneficial, which is shown in Table 2.
3.1 Weightage Calculation of Criteria Using Fuzzy AHP The traditional AHP is inadequate for managing fluffiness and vulnerability in multirules dynamic (MCDM), in view of fragmented data, inaccuracy of human decisions, and fluffy climate. Subsequently, the fluffy AHP strategy can be seen in Table 3 as a high level scientific technique created from the traditional AHP.
3.2 Supplier Ranking Using ARAS Method A definite normalization process is explained in the theories section, where the beneficial attributes and reciprocal of all the criteria are considered for non-beneficial attributes. Based on the mentioned equation in the section, the calculation of the degree of utility for all suppliers is carried out which is shown in Table 4.
152
J. George et al.
Table 3 Fuzzy geometric mean value, fuzzy weight, average weight, and normalized weight data Fuzzy geometric mean value (r i )
Fuzzy weight (wi )
Average weight (M i )
Normalized weight (N i )
PPM
(2.56, 3.38, 4.12)
(0.20, 0.35, 0.57)
0.3741
0.3373
Quality
(1.69, 2.32, 3.06)
(0.13, 0.24, 0.43)
0.2659
0.2397
Price/Cost
(1.26, 1.69, 2.27)
(0.10, 0.17, 0.32)
0.1966
0.1772
Standardization
(0.65, 0.96, 1.35)
(0.05, 0.10, 0.19)
0.1124
0.1013
Service
(0.46, 0.63, 0.85)
(0.04, 0.06, 0.12)
0.0732
0.0660
Flexibility
(0.32, 0.42, 0.57)
(0.03, 0.04, 0.08)
0.0489
0.0441
Delivery
(0.26, 0.32, 0.44)
(0.02, 0.03, 0.06)
0.0381
0.0344
Total
(7.20, 9.71, 12.67)
1.1093
1
Inverse value
(0.14, 0.10, 0.08)
Increasing value
(0.08, 0.10, 0.14)
Table 4 Performance index (S i ) values, degree of utility (U i ) values, and ranking
Suppliers
Si
Ui
Rank
S0
0.1970
1
S1
0.1197
0.6075
12
S2
0.1630
0.8276
7
S3
0.1740
0.8835
3
S4
0.1771
0.8991
1
S5
0.1335
0.6777
8
S6
0.1261
0.6400
10
S7
0.1196
0.6072
13
S8
0.1642
0.8338
6
S9
0.1689
0.8577
5
S10
0.1228
0.6237
11
S11
0.1745
0.8861
2
S12
0.1734
0.8803
4
S13
0.1328
0.6743
9
S14
0.0999
0.5073
14
3.3 Supplier Ranking Using COPRAS Method This technique thinks about the other options and decides their needs under the clashing models by considering the rules loads. It accepts immediate and corresponding conditions of the importance and utility degree (need) of the other options. In this technique, the impact of boosting and limiting measures on the assessment result is considered independently. The determination of the best option depends on considering both the ideal and the counter ideal arrangements appeared in Table 5.
Comparative Analysis of Supplier Selection Based on ARAS …
153
Table 5 Priority values (Qi ), quantitative utility (U i ), and ranking Suppliers
S+
S−
Qi
Ui
Rank
S1
0.0472
0.0518
0.0880
0.7876
11
S2
0.0385
0.0393
0.0923
0.8256
9
S3
0.0437
0.0336
0.1066
0.9540
4
S4
0.0414
0.0320
0.1074
0.9612
3
S5
0.0440
0.0396
0.0974
0.8716
6
S6
0.0360
0.0412
0.0872
0.7808
12
S7
0.0283
0.0420
0.0785
0.7027
13
S8
0.0404
0.0425
0.0901
0.8060
10
S9
0.0464
0.0409
0.0980
0.8772
5
S10
0.0512
0.0461
0.0970
0.8683
7
S11
0.0531
0.0360
0.1117
1.0000
1
S12
0.0494
0.0352
0.1094
0.9788
2
S13
0.0476
0.0428
0.0969
0.8672
8
S14
0.0486
0.1208
0.0661
0.5913
14
3.4 Supplier Ranking Using MOORA Method The end results of the MOORA technique that suggests the appraisal esteems with positioning ascertained via vector standardized preference lattice for share association of MOORA methodology reference issue strategy and full increasing reasonably MOORA methodology-based examinations. The connected weighted standardized want grid is likewise wanted for the proportion association of the MOORA strategy, reference issue approach, and full increasing reasonably MOORA approach-based examinations appeared in Table 6.
4 Result and Conclusion Multi-Criteria Decision-Making is generally utilized for dynamic issues where there are a few factors in acquiring the best arrangement and various techniques are utilized for tackling an unpredictable issue. The issue is to locate the best provider who satisfies all the rules of the producer with the ideal condition. Numerous calculations are accessible in the MCDM approach, where ARAS, COPRAS, and MOORA are the most used techniques for this issue and contrasted with one another. The Ranking of options was performed by applying ARAS, COPRAS, and MOORA strategies, a joint model of weighted collection of the F-AHP techniques. Figure 1 shows the individual aftereffects of ARAS, COPRAS, and MOORA consolidate with F-AHP. All these MCDM techniques distinguished that provider 4,
154
J. George et al.
Table 6 Weighted assessment values (Y i *) and ranking Y = (Max − Min)
Suppliers
Maximum
Minimum
S1
0.1455
0.1177
0.0278
Rank 9
S2
0.1145
0.0837
0.0308
8
S3
0.1219
0.0732
0.0487
5
S4
0.1202
0.0702
0.0501
4
S5
0.1131
0.0951
0.0180
11
S6
0.1011
0.0981
0.0030
12
S7
0.0831
0.0996
–0.0165
13
S8
0.1347
0.0897
0.0450
6
S9
0.1463
0.0867
0.0596
2
S10
0.1275
0.1072
0.0203
10
S11
0.1415
0.0777
0.0638
1
S12
0.1315
0.0762
0.0553
3
S13
0.1431
0.1011
0.0419
7
S14
0.1422
0.3324
–0.1901
14
Fig. 1 Result comparison of ARAS, COPRAS, and MOORA method
provider 11 for ARAS Method and COPRAS and MOORA Method individually for the firm, however when the information examination concentrated farther than no one but, we can comprehend the distinction in the two strategies, it showed that there were huge contrasts in the outcomes and high measure of vulnerability for different providers.
Comparative Analysis of Supplier Selection Based on ARAS …
155
References 1. Adali EA, I¸sık AT (2016) Air conditioner selection problem with COPRAS and ARAS methods. Manas Sosyal Ara¸stırmalar Dergisi 5(2):124–138 2. Aikhuele DO, Turan FM (2017) A subjective and objective fuzzy-based analytical hierarchy process model for prioritization of lean product development practices. Manage Sci Lett 7:297– 310 3. Attri R, Grover S (2014) Decision making over the production system life cycle: MOORA method. Int J Syst Assur Eng Manage 5(3):320–328 4. Badizadeh A, Khanmohammadi S (2011) Developing a Fuzzy model for assessment and selection of the best idea of new product development. Indian J Sci Technol 4(12):1749–1762 5. Brauers WKM, Zavadskas EK (2010) Project management by MULTIMOORA as an instrument for transition economies. Technol Econ Dev Econ 16(1):5–24 6. Brauers WK (2004) Optimization methods for a stakeholder society. Kluwer Academic, Boston 7. Brauers WKM, Zavadskas EK (2009) Robustness of the multi-objective MOORA method with a test for the facilities sector. Technol Econ Dev Econ 15(2):352–375 8. Brauers WKM, Zavadskas EK (2011) MULTIMOORA optimization used to decide on a bank loan to buy property. Technol Econ Dev Econ 17(1):174–188 9. Buckley JJ (1985) Fuzzy hierarchy analysis. Fuzzy Sets Syst 17:233–247 10. Chai J, Liu JNK, Ngai EWT (2013) Application of decision-making techniques in supplier selection: a systematic review of literature. Expert Syst Appl 40(10):3847–4272 11. Chakraborty S (2011) Applications of the MOORA method for decision making in manufacturing environment. Int J Adv Manuf Technol 54(9–12):1155–1166 12. Chand M, Raj T, Shankar R (2014) A comparative study of multi-criteria decision-making approaches for risks assessment in supply chain. Int J Bus Inf Syst 18(1):67–84 13. Chatterjee P, Chakraborty S (2014) Flexible manufacturing system selection using preference ranking methods: a comparative study. Int J Ind Eng Comput 5:315–338 14. Durán O, Aguilo J (2008) Computer-aided machine tool selection based on a fuzzy-AHP approach. Expert Syst Appl 34:1787–1794 15. Gadakh V, Shinde V, Khemnar N (2013) Optimization of welding process parameters using MOORA method. Int J Adv Manuf Technol 69(9–12):2031–2039 16. Lourenzutti R, Krohling RA (2013) A study of TODIM in a intuitionistic fuzzy and random environment. Expert Syst Appl 40(16):6459–6488 17. Madic M, Markovic D, Petrovic G, Radovanovic M (2014) Application of copras method for supplier selection. In: The 5th international conference “Transport and Logistics” 18. Podvezko V, Sivileviˇcius H (2013) The use of AHP and rank correlation methods for determining the significance of the interaction between the elements of a transport system having a strong influence on traffic safety. Transport 28(4):389–403 19. Rostamzadeh R, Sofian S (2011) Prioritizing effective 7Ms to improve production systems performance using fuzzy AHP and fuzzy TOPSIS (Case Study). Exp Syst Appl 38:5166–5177 ˇ Turskis Z (2014) Integrated evaluation of external 20. Ruzgys A, Volvaˇciovas R, Ignataviˇcius C, wall insulation in residential buildings using SWARA-TODIM MCDM method. J Civil Eng Manage 20(1):103–110 21. Saaty TL (1980) The analytic hierarchy process. McGraw-Hill, New York 22. Tavana M, Momeni E, Rezaeiniya N, Mirhedayatian SM, Rezaeiniya H (2013) A novel hybrid social media platform selection model using fuzzy ANP and COPRAS-G. Exp Syst Appl 40(14):5425–5786 23. Van Laarhoven PJM, Pedrycz W (1983) a fuzzy extension of Saaty’s priority theory. Fuzzy Sets Syst 11:229–241
156
J. George et al.
24. Zavadskas EK, Turskis Z, Vilutiene T (2010) Multiple criteria analysis of foundation instalment alternatives by applying Additive Ratio Assessment (ARAS) method. Arch Civil Mech Eng 10(3):123–141 25. Zavadskas EK, Kaklauskas A, Turskis Z, Tamošaitien J (2008) Selection of the effective dwelling house walls by applying attributes values determined at intervals. J Civ Eng Manag 14(2):85–93
Comparative Efficiency Measurement of Indian Hospitals Using Data Envelopment Analysis: A Proposed Model Dilip Kushwaha and Faisal Talib
Abstract In engineering and management, efficiency measurement of the various decision-making units (DMUs) of an organization is a key requirement for economic performance and effective utilization of resources. The comparative efficiency of different DMUs indicates the level of improvement required for various inefficient units with other efficient units; hence, necessary action can be taken especially for the welfare of inefficient units. It is not easy to measure comparative efficiencies of different units of an organization when the units of the organization are multiinput and multi-output types. The hospital industry is one of these types of industries where various professional specialists and machinery with different technologies are involved to cure different types of patients. To measure the efficiency and performance of such industry, a non-parametric and linear programming model has been proposed by (Charnes et al. in European Journal of Operations Research 2:429–441, 1978 (Charnes et al. in Eur J Oper Res 2:429–441, 1978 [1]), known as the Data Envelopment Analysis (DEA). In terms of measurement of efficiency and comparative performance of Indian hospitals, only a few papers are available that can be counted on the finger, hence, there is a need to bridge up this gap and therefore, the present study was designed to fill this gap. In this study, a comparative performance assessment framework is proposed for Indian hospitals and according to the framework comparative performance is measured. To carry out this objective, the present study utilizes the DEA methodology. For this, 15 public sector hospitals which provides basic healthcare are selected from Delhi, in which a number of doctors, nurses, other paramedical staffs and beds are considered as inputs and number of annually outpatients, inpatients and major surgeries are deemed to be outputs for hospitals efficiency and their relative performance assessment. The conclusion along with the limitations and future research directions are also presented at the end of this paper. D. Kushwaha (B) Department of Mechanical Engineering, Zakir Husain College of Engineering and Technology, Faculty of Engineering & Technology, Aligarh Muslim University, Aligarh, India F. Talib (B) Department of Mechanical Engineering, Zakir Husain College of Engineering & Technology, Faculty of Engineering & Technology, Aligarh Muslim University, Aligarh, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_14
157
158
D. Kushwaha and F. Talib
Keywords Data envelopment analysis (DEA) · Public sector hospitals · Decision-making units (DMUs) · Efficiency · Comparative performance · Input/Output
1 Introduction In the era of advancement in new digital healthcare technologies and rapid growth in technological innovations, hospitals play an important role by providing primary, secondary, and tertiary levels of patient care [2, 3]. Hospitals are the important subpart of health care systems, and evaluation of hospitals is one of the most versatile requirements for health policy-makers [2, 4]. Today health care sector is increasing worldwide, and it receives a major portion of gross domestic product (GDP). India is a fastest growing nation, which has made rapid progress in the healthcare and wellbeing of its people over the last several decades. However, much progress has to be achieved to increase research, people awareness and facilitating the quality health care services to the mass public [5]. Today the government of Indian is huge under pressure to increase the performance of the healthcare systems [6]. Indian government wants to improve public health facilities spending to 2.5% of the Indian GDP by 2025 [6]. Therefore, the world’s largest government-funded healthcare schemes like Ayushman Bharat was launched on 23 September 2018, Pradhan Mantri Jan Arogya Yojana (PMJAY) was launched to provide basic health insurance of at least 100 mn families and Mission Indra Dhanush for immunization [6, 7]. It is expected that healthcare sector become one of India’s largest market—both in terms of revenue and employment. It is expected to increase US$ 372 billion by 2022 [6, 7]. There are two major category of Indian healthcare delivery system—public (government) and private. Government healthcare system focuses on providing basic healthcare facilities in the form of primary healthcare centres (PHCs) and community healthcare (CHCs). As healthcare costs are augmented rapidly, it is become more important to evaluate the performance of hospitals across countries. India has a large population, and the health condition of the country population affects the progress of the country. Therefore, assessing the performance of hospitals has a significant impact on improving the quality of health services and better utilization of available resources. Hence, the performance measurement of hospitals is the main aim of this study and therefore, first of all this study proposes a model to estimate the overall technical efficiency of public sector hospitals in Delhi and then measure the comparative performance of hospitals. Typically, public sector hospitals operate with the same funding model and different benefits. Therefore, the main focus of this study is to analyse the comparative efficiency of various public sector hospitals in Delhi without considering various geographical locations, sizes, status and other salient features. Central Pollution Control Board (CPCB) shows that Delhi had only five ‘good’ AQI days in the last 4 years (2015–2018). Diseases like Cancer, infections and other respiratory diseases increasing rapidly and the number of respiratory deaths over the years, with such a large disease burden in the capital city, it becomes even more necessary to have
Comparative Efficiency Measurement of Indian Hospitals …
159
a strong public healthcare system that can provide affordable and easily available preventive, primary and curative healthcare. And therefore, it is becoming mandatory to measure the performance/efficiencies of government hospitals because about 50% of people provide medicine in government hospitals. Comparative performance measurement of various public sector hospitals is most important to understand the utilization of resources. It also indicates the significant information for the advancement of hospitals reform to promote efficiency, as well as improve accountability of health polices makers, as well as improve funding systems. In this study the performance analysis is based on the theory of production in economics in which hospital services as a production process where inputs (e.g., number of doctors, number of nurses, number of other paramedical staff and number of beds) are transformed into different outputs (e.g., annually number of outpatient, number of inpatient and number of major surgeries, etc.). To evaluate the performance of decision-making units (DMUs) (here hospitals) which converts multiple inputs into multiple outputs, one of the mostly used methods is the Data Development Analysis (DEA) method. Since the hospital sector provided a range of services by handling with multiple inputs, estimating hospital efficiency through the DEA is attractive and one of the most popular approaches in the available literature. DEA was initiated by Charnes et al. [1] based on the measurement of productive efficiency originated by Farrell [8] in their research paper by Charnes et al. [1] and introduced DEA as a linear programming mathematical tool to measure the comparative efficiency and productivity of various homogeneous DMUs and generally applied for comparison of hospitals and further developed for piece-wise linear technologies by Banker et al. [9]. DEA is a frontier estimator which is based on a linear programming model. DEA calculates a comparative ratio of outputs to inputs for each DMU as a linear programming procedure, which is reported as the relative efficiency score. DEA relates to ‘best’ or ‘efficient’ rather than central tendency behaviour. The relative efficiency score is generally represented, either as a number between 0–1 or 0–100%. A DMU with a score of less than 1 or 100% is said to be an inefficient DMU relative to other DMUs. In recent years, there has been a rapid increase in publications related to the theory and the applications of DEA. DEA is now recognized as a modern tool for instant performance measurement and has been successfully applied in many contexts around the world. Agriculture, banking, supply chain, transportation, as well as public policy are the top-five application fields of DEA with the greatest numbers of journal articles in between 2015 and 2016 [33]. While another research shows that banking, health care, agriculture and farm, transportation, and education are the top-five industries [10]. Objectives of the Study The main objectives of the current study are to provide a comprehensive review of literature related to the measurement of hospital’s efficiency by DEA and propose a framework for efficiency measurement and according to the framework measure the comparative performance of hospitals so that it may assist physicians and healthcare
160
D. Kushwaha and F. Talib
managers with a deeper understanding of various input/output variables of efficiency measurement in their hospitals. Thus, the study focuses on the following objectives: • To provide a comprehensive review of sources and characteristics of efficiency measurement typically found in the literature; • To identify performance indicators for assessing the performance of selected Indian hospitals; • To propose a comparative performance assessment framework for Indian hospitals; • To measure the comparative efficiency and effectiveness of selected Indian hospitals; • To compare the performance of the selected Indian hospitals and identify the most efficient hospital; and • To improve the overall efficiency of Indian hospitals by understanding valuable insights on performance measurement and effectiveness. The remainder of this paper is arranged in the following manner. In Sect. 2, an extensive literature review was conducted that describes some important research work related to the assessment of hospital efficiency. Section 3 discusses the research methodology in detail, selection of criteria for choosing input/output variables. Section 4 presents the results and discussions of this study. Section 5 highlights the brief conclusions of this study and discussed the limitations and possible future scope of the study and give an important idea about possible future research in this area.
2 Literature Review This study utilizes online search databases namely Google Scholar, Research Gate, etc. for the works published on the topic covering ‘measurement of hospital efficiency’ with the aim to identify the research gaps present in the area of study by identifying input/output variables of efficiency measurement in healthcare sector. Table 1 illustrates some important research work related to the assessment of hospital efficiency using DEA. In health care, [29] has used DEA in 1983 and published the first paper in this category, in which they focusing on measuring the nursing service efficiency by DEA and comparison of this with cost per patient day. Sherman [28] used DEA for identifying the inefficient hospitals in a category of seven educational hospitals and verifies the result by the advice of a panel of hospitals specialist. Banker et al. (1986), use DEA and Translog method to compare the hospital cast and production correspondences on the same empirical data and found that DEA has more wide and diverse behaviour. Grosskopf and Valdmanis [27] measures the relative efficiency of Californian public hospitals. Fizel and Nunnikhoven [26] employs DEA for individual nursing homes and purges with Regression analysis and find that private hospitals are
Comparative Efficiency Measurement of Indian Hospitals …
161
Table 1 Selected research literature related to the measurement of hospital efficiency Researchers Methodology
Input variables
Output variables
[11]
Various DEA model 37 Indian Private with MPI and Tobit sector hospitals Regression from 2010 to 2014
Sample size
Cost of labour, net fixed assets, current assets, and other operating expenses
Total income and profit after tax
[12]
DEA and VRS methodology
The number of physicians, nurses, active beds and equipment
Bed occupancy rate, number of discharged patients, bed price and physicians’ fees
[13]
An output oriented The regional 4-year window DEA efficiency of model healthcare facilities in Slovakia is measured for the period 2008-15in 8 regions
Two stable inputs (number of beds and medical staff), three variable inputs (number of all medical equipment, MR and CT
Two stable outputs (use of beds, average nursing time)
[14]
Bayesian-Combined 117 Greek public application of a hospitals in 2009 chance-constrained DEA and meta frontier analysis
Doctors, other personnel (Nurses, administrative and support staff), beds and operating cost
Annual numbers of IPD and OPD
[15]
DEA with Truncated Regression Approach with double bootstrap
113 acute care hospitals in Ontario for the years 2003–2006
Administrative staff hours, nursing hours, staffed beds, medical-surgical supplies costs, non-medical supplies costs, equipment expenses
Ambulatory visits, case-mix weighted inpatient days
[16]
Slack based DEA model and Jackknifing analysis
36 public sector hospitals in Uttarakhand (India) for the year 2011
Number of beds, doctors, and PMS (paramedical staff)
Total number of OPD, IPD, major and minor surgeries
Intensive care units (ICUs) of 7 training hospitals in 2015
(continued)
162
D. Kushwaha and F. Talib
Table 1 (continued) Researchers Methodology
Sample size
[17]
DEA with the input-oriented VRS and MPI
101 hospitals for Total number of the years 1998 to beds, total 2006 from Vietnam number of full-time physicians and non-physicians employed
Input variables
Output variables Total Number of OPD visits, IPD stayed days and Surgical operations
[18]
Data envelopment analysis (DEA) and the Malmquist productivity index (MPI)
29 hospitals of Catalonian, 9 are small (with beds < 200), 11 are medium-sized (with 200 < beds < 400), and 9 are big (with beds > 400)
Physicians, other staff, beds and materials (money)
4 desirable outputs (acute, long stay, intensive, visits) and 1 undesirable outputs (infections)
[19]
Malmquist indexes 53 acute hospitals of TFP, in turn based in Scotland in the on DEA years 1992 -97
Medical staff, nursing staff, other staff, total number of beds
Inpatients surgery, outpatients day cases and day patients
[20]
DEA and SFA
Acute public hospitals in Ireland between 92 and 2000,
Average number IPD and OPD of beds, people cases employed in each hospital
[21]
Five regression methods and DEA are used to compare efficiency with 3 different models
90 Family health services authorities providing primary health care in England during two financial years between 1993 and 1995
Remunerate cost of GPs and practice nurses, no. of GPs, Number of practice nurses,
[22]
DEA and SFR
Acute care hospitals in Florida over the period 1982–1993, 186 hospitals for each year
6 cost categories 6 output variables in the DEA model and the sum of the 6 items in the SFR model
[23]
DEA and SFA
UK, NHS hospitals Three cost (Trusts) indices (CCI, 2CCI and 3CCI) for NHS Trusts
No. of patients registered, No. of deaths, Area of the FHSA in hectares, etc.
Patients treated within each Healthcare Resource Group, teaching, research (continued)
Comparative Efficiency Measurement of Indian Hospitals …
163
Table 1 (continued) Researchers Methodology
Sample size
Input variables
Output variables
[24]
SFA and panel data and Non-parametric models and panel data (DEA and MPI)
43 (Five university teaching and 38 other public hospitals) acute care hospitals for the years 1988–1994 in Finland
Cost variable, fixed factor variable, price variables, exogenous variables
Outpatient and Inpatient treatment, Teaching and Research variables
[25]
Output oriented data 140 nursing homes envelopment from Connecticut, analysis (DEA) with for period 1982–83 additional constraint
Dietary, housekeeping and laundry staff, director of nursing, registered nurse (RN), licensed practical nurse (LPN) and nurses’ aides hours total expenditure on non-labour
Medicare patient days, Medicaid patient days, Private patient days, and other patient day
[26]
DEA with regression analysis
163 Michigan nursing homes, of which 104 are for-profit and 59 are non-profit homes for 1987
Registered nurse and licensed practical nurse hours, and aides’ and orderlies’ hours
Patient days for skilled and intermediate-care patients
[27]
Relative efficiency of hospitals using DEA
22 public and 60 private, hospitals operating in California in 1982
No. of physicians, non-physician labour, admission, and net assets
Acute, intensive care, surgeries, ambulatory and emergency care
[28]
DEA and assessment of DEA results by a panel of hospital expert
A set of seven teaching hospitals in Massachusetts (USA) during financial year 1976
Full-time non-physician, Total dollar Supply, Number of bed days
Patient days with age > 65 and < 65 years, Nurses trained, Interns/residents trained
[9]
DEA and Translog Methods
114 hospitals in North Carolina (USA) in 1978
Services (Nursing ancillary, administrative, general) and capital
Utilization of inputs by patients days for IPD age < 14, aged (14–65), and aged > 65 (continued)
164
D. Kushwaha and F. Talib
Table 1 (continued) Researchers Methodology
Sample size
Input variables
Output variables
[29]
Comparison of 17 hospitals in Wisconsin (USA) for the years 1978–79 by grouping with 3 different criteria
Total inpatient routine costs
Total routine aged and paediatric days, Total routine maternity days, All other routine days
Two-phase analysis, DEA with CRS is used in phase one
inherently more efficient than government hospitals. Chattopadhyay and Ray [25] uses DEA with additional constraints. DEA with Malmquist productivity index and stochastic frontier models with panel data was used by Linna [24] to measure the hospital cost efficiency in Finland. Jacobs [23] in their paper “Alternative Methods to Examine Hospital Efficiency: DEA and SFA” concludes that each method has particular strengths and weaknesses. DEA with different regression analysis is used for comparing the efficiency and performance of primary health care services in England by Giuffrida and Gravelle [21]. Gannon [20] uses DEA and SFA to measure efficiency of acute public hospitals in Ireland and comparison between DEA and SFA methods. Ferrari [19] measures the changes in productivity and efficiency of hospitals during reform DEA and calculating MPI. Prior [18] examined Spanish hospitals to incorporate the postulates of Total Quality Management into DEA models. Chowdhury and Zelenyuk [15] use DEA with truncated regression to estimate the efficiency score of hospitals and distributions of efficiency across various environmental and other factors and the effect of different factors on efficiency. In the context of India, Mogha et al. (2015) use a slack based DEA model on 36 government hospitals in Uttarakhand. Gandhi and Sharma [11] use various DEA models to measure the technical efficiency of 37 private hospitals in India for the years 2010– 2014. Gearhart and [30] examine cross-county healthcare efficiency rankings using two stages DEA. Stefko et al. [13] introduced the window approach as an extension to the basic DEA models. In a latest study [14, 34] used DEA-VRS model, and Tobit regression compared the efficiency levels of training and research hospitals in Turkey and identified the reasons which affect the efficiency. Gaps Identified Although a large number of research articles are available on measuring the hospital efficiency but there are only a handful of research articles that are available in the context of India. There is a need to bridge up this gap, and therefore, the present study was designed to fill this gap. Hence, there is a lack of studies performed on the Indian health sector on comparison of efficiencies and this study will help physicians and healthcare managers to utilize hospitals facilities efficiently for formulating policies for government hospitals in India.
Comparative Efficiency Measurement of Indian Hospitals …
165
3 Research Methods 3.1 Introduction to DEA DEA is a mathematical framework that facilitates the measurement of comparative efficiencies of various multi-input and multi-output DMUs of an organization. All the DMUs are performing a similar task, and they use similar types of resources hence public sector hospitals of a particular region are suitable for their relative efficiency measurement through DEA. Relative efficiency measurement is significantly very useful in practical contexts where sufficient information is not available for absolute efficiency measurement. In basic DEA model one constructs a feasible input–output set related to the DMUs which is being to be assessed here, and there is no need to hypotheses a functional form between input–output. The graphical representation of DEA for various DMUs that have one input and one output is shown in Fig. 1, a linear piece-wise boundary is constituted by efficient DMUs, and the boundary occupies a space in which other inefficient DMUs coordinate (input, output) lies, in other words, this linear piece-wise boundary envelops the whole other inefficient DMUs data and therefore this method is known as DEA. For multi-input and multioutput DMUs the methods work in the same way by linear programming method. Therefore, DEA constitutes a peace-wise frontier according to data which is known as efficient frontier or boundary and by applying this efficient frontier DEA computes a maximum efficiency measure for each DMU with respect to that of all other DMUs and each efficient DMUs lies on the piece-wise frontier with efficiency score 1 and all other inefficient DMUs enveloped by piece-wise frontier with efficiency score between 0 and 1. From basic CRS-DEA model proposed by Charnes et al. [1] a lot of literature extensions have been developed in which variable returns to
Fig. 1 Graphical representation of CRS and VRS-DEA models for single input and the single output of various DMUs (A, B, C, D, E, E, F)
166
D. Kushwaha and F. Talib
Table 2 The CRS and VRS values of an inefficient DMU ‘C’.
DEA-methods
CRS VRS
Approaches Input-saving efficiency
Output-increasing efficiency
hi hC hj hC
mC mk mC ml
scale (VRS) is one of them. VRS-DEA model is developed by Banker et al. [9], by relaxations in the assumption of the CRS-DEA model. The VRS-DEA model is more appropriate in the context of real life. These models are known as basic DEA models. For graphic presentation, assume that hospitals produce a single output with the help of a single input, a constant return to scale (CRS) and a variable return to scale (VRS) technology can be illustrated according to Fig. 1. The straight line shows the efficiency frontier under the assumption of CRS, while the piece-wise linked line shows the efficiency frontier under the assumption of VRS. There are two approaches to measuring technical efficiency in both basic DEA models. The first one is input-saving efficiency which means the possible reduction in the use of inputs for a given level of outputs (e.g., a given number of treated patients), and the second one is the output-increasing efficiency which means the possible increase in outputs for the given level of inputs. Table 2 represents the CRS and VRS values of an inefficient DMU ‘C’ for both approaches. Therefore, this study progress step by step according to procedure shown in Fig. 2 in step 1 choice of frontier for efficiency measurement, i.e., input oriented or output oriented approach. Step 2 in this step, selection of input and output was done which is more responsible for hospital efficiency and in step 3, measuring and comparing the efficiency of different hospitals on various input/output parameters is performed. All these steps are shown in Fig. 2. Farrell [8] measure the technical efficiency of the single input, single output variables case, and DEA generalizes the concept of Farrell and measure the multipleinput and multiple-output case by measuring a comparative score as a ratio of a virtual output to a virtual input. There are a lot of field to DEA application while bootstrapping estimates for sophisticated testing is the upgraded DEA methodology which is used recently but it is still the format is same, therefore in this study we does not going to make a methodological advancements and therefore used the basic DEA model. In DEA we measure the relative efficiency by computing the ratio of weighted
Step-1
Step-2
Step-3
CHOICE OF EFFICIENCY MEASUREMENT APPROACH i.e. Input oriented or output oriented approach
SPECIFICATION OF INPUTS AND OUTPUTS In this step we select the inputs and outputs which is to be used in the selected approach.
MEASUREMENT OF EFFICIENCY AND THEIR COMPARISON In this step we also analyse the impact of different inputs and outputs on efficiency
Fig. 2 Steps in measuring and analysing hospitals efficiency
Comparative Efficiency Measurement of Indian Hospitals …
167
outputs to weighted inputs for each of the DMUs. Farrell [8] explained that the DMU which score is less than one reflects how long the radial distance of that DMU to a calculated production frontier. Let in DEA ‘N’ no. of DMUs to be analysed, where each DMU uses ‘m’ different inputs and produce ‘s’ different outputs. To determine the highest score for ‘N’ DMUs, comparative efficiency is estimated, for every test DMU ‘J 0 ’ with ‘i’ inputs and ‘r’ outputs, by computing the Model 1, as suggested by Charnes et al. [1]. Model 1. DEA ratio model: Maximise : h jo =
s r =1
u r Yr jo /
m
Vi X i jo .
i=1
Subject to: s r =1
u r Yr j /
m
Vi X i j ≤ 1 j = 1 . . . ... jo . . . . . . N ,
i=1
ui ≥ 0
r = 1 . . . . . . s,
Vi ≥ 0
i = 1 . . . . . . m,
Or
u r , Vi ≥ ε > 0,
where, hj0 is the efficiency value of the evaluated DMU (j0 ), No. of DMUs here hospitals (j) = 1, 2,…, j0 , …, N. weight assigned to output r = ur , weight assigned to input i = V i , output obtained by the jth DMU = Y rj , input used by the jth DMU = X ij , and ε is an infinitesimal (+ve) value. DEA pick up the most favourable set of weights for each DMU and bisect the DMUs into 2 types as efficient and inefficient. The DEA has intractable nonlinear and non-convex properties [1], therefore, model 1 is not used for actual computation of the efficiency scores. Hence model 1 is stated as an optimal linear programming model by constraining either the numerator or the denominator of the efficiency ratio to be equal to one. Therefore model 1 becomes either maximising weighted output with weighted input equal to one or minimising weighted input with weighted output equal to one [31]. In this study for analysis, the DEA model for assessing the comparative input efficiency of DMU j0 is as given below. Let us suppose that we have ‘N’ DMUs here hospitals (j = 1,…, j0 , …, N) consuming ‘m’ inputs to secures ‘s’ outputs. Let us represents X i j and Y rj the level of the ith input and rth output respectively observed at DMU j and X i j0 and Yr j0 for DMU j0 . For solving the fractional Model 1, it is required to transform it into a linear programming model as in Model 2 which is given below.
168
D. Kushwaha and F. Talib
Model 2. DEA linear programming model: Max h j0
s
u r Yr j0 .
r =1
Subject to: m
Vi X i j0 = 1,
i=1 s r =1
u r Yr j −
m
Vi X i j < 0 j = 1 . . . . . . j0 . . . . . . N ,
i=1
ur ≥ 0
r = 1 . . . . . . s,
Vi ≥ 0
i = 1 . . . . . . m.
In this paper we uses the input-orientation CRS-DEA models which is described above, which is keeping a question in the mind “By how much can input quantities be proportionally reduced without changing the output quantities produced?” for this study the input-oriented model is selected, as inputs are no. of doctors, nurses, paramedical staffs, and beds are the primary decision variable over which the govt. has most control. These inputs are also reflects as the resources of disparity in efficiency across hospitals. Therefore, the main focus of this study is to measure the input-oriented pure technical efficiency of hospitals (DMUs). Table 3 describes the variables employed throughout the analysis. Table 3 Explanation of the input/output variables employed in the analysis Variable
Explanation
Inputs Doctors (V1 )
Total number of doctors (Specialists + medical officers + Sr. residents + Jr. residents)
Nurses (V2 )
Total number of nurses (Nursing sisters + Nursing staff)
Other paramedical staff (V3 ) Total number of other paramedical staff (pharmacist + technician + Technical Attendant + Lab Attendant + radiographic technician + dark room assistant + physiotherapist etc.) Beds(V4 )
The total number of hospital beds
Outputs OPD Visits (U1 )
The total number of annually visit patients to outpatient departments and Emergency rooms
IPD Visits (U2 )
The total number of annually IPD admissions patients
Major surgeries (U3 )
The total number of major surgeries done in the hospital in one year
Comparative Efficiency Measurement of Indian Hospitals …
169
Let we want to measure the relative efficiency J 0 , which is obtained from the solution to the following value-based DEA linear programming (LP) model 3 of DEA in which Model 1 was extended by Banker et al. [9] is written as in Model 3. Model 3. DEA linear model with CRS and VRS Max h jo
s
u r Yr j0 + u 0 .
r =1
Subject to: m
Vi X i j0 = 1,
i=1 s r =1
u r Yr j −
m
Vi X i j + u 0 ≤ 0 j = 1 . . . . . . j0 . . . . . . N ,
i=1
ur ≥ 0
r = 1 . . . . . . s,
Vi ≥ 0
i = 1 . . . . . . m,
u r ≤ 0 or ≥ 0, where u 0 represents VRS values. u 0 may take a zero, positive, or negative value. Model 3 is run N times to obtaining the comparative efficiency score of all DMUs. Therefore, u0 is dual in linear programming term. If u 0 > 0 or u 0 < 0 then variable returns to scale (VRS), i.e., increasing returns to scale and decreasing returns to scale respectively and if u 0 = 0 then it means constant returns to scale (CRS). Different methods can be used to solve the linear programming models. In which Simplex Method is most applied method. Probably the most readily available software for this purpose is in Excel, under Solver in the Tools menu, i.e., DEA-SOLVER-LV8 (2014-12-05).xlsm [shared] micro-soft excel which is used for this analysis. Some other examples of specialist software are ‘XPRESSMP’ and ‘LINDO’ and Warwick DEA Software, etc. The complete conceptual framework for comparative efficiency measurement which is explained above can be outlined in the form of as depicted in Fig. 3.
3.2 Input/Output Variables and Data Selection In DEA, comparative efficiency can be performed of DMUs which operate on the same input and produced the same output implying that they perform similar tasks in similar geographical environmental and economic location. Therefore, one cannot
170
D. Kushwaha and F. Talib
Inputs • Number of doctors • Number of nurses, and • Other paramedical staff • Number of beds
Transformation process: Hospitals
Outputs • Outpatient visits • Inpatients visit, and • The number of major surgeries
Efficiency measurement of hospitals w.r.t. the inputs and outputs variables
DEA approach selection CRS – DEA Approach 1. Input- oriented 2. Output- oriented
VRS – DEA Approach 1. Input- oriented 2. Output- oriented Significant input/output variable and minimum number of DMUs selection and data collection
Analysis on Warwick DEA Software 'XPRESSMP' and 'LINDO' software DEA-SOLVER-LV8 (2014-12-05).xlsm [shared] micro-soft excel etc.
Determining • Efficient hospitals • Inefficient hospitals • Comparative efficiency • Impact of various inputs/outputs
Fig. 3 A comparative performance assessment framework for Indian hospitals
compare a multi-specialty hospital with a small hospital due to this here one cannot select any teaching hospital in all selected hospitals. In this study, sample hospitals were selected from Delhi so the environmental effects on the sample are almost negligible, and the selected hospitals do not perform any education-related activities; this reduces the problem associated with teaching outputs. Identifying inputs and outputs is as difficult as evaluating the DMUs. The inputs must occupy all resources that have a significant impact on outputs. In output categories it should be ensured that it contain all useful outputs on which this work evaluates the DMUs. In addition, any environmental factors that affect resource changes in outcomes should also be reflected in inputs or outputs, depending on the effect direction. There the main idea about input variable selection is those parameters which are common among all the hospitals which hospitals are used in their daily functioning and according to the requirement of efficiency measurement the inputs are the number of doctors,
Comparative Efficiency Measurement of Indian Hospitals …
171
nurses, other paramedical staffs and beds and outputs are outpatient visits, inpatients visit and the number of major surgeries. These variables are chosen on the basis of hospital literature, ideal and possible of data in hand. Therefore data accuracy is fundamental of such analysis as wrong data in the DEA methodology will affect not only that hospital’s efficiency score but also potentially the efficiency rankings of other hospitals as well. Overall, this study has selected four inputs and three outputs. These inputs and outputs have the most impact on hospital performance therefore these are considered as variables for DEA analysis. The thumb rule “the number of DMUs is expected to be larger than twice the sum of inputs and outputs” [32] is applied for the selection of the number of hospitals according to the inputs and outputs. The selected 3 outputs for each hospital were (i) the number of OPD patients Visits, (ii) the number IPD patients Visits, and (iii) the number of major surgeries. The four inputs included are (i) the number of doctors, (ii) the number of nurses, (iii) the number of other paramedical staff, and (iv) the number of beds. There are 38 State hospitals in Delhi. The data will be obtained for one year only from 15 public sector hospitals in Delhi. All the selected hospitals have at least 50 beds and not more than 500 beds. The other 23 hospitals which are left in this analysis are the special type like cancer hospitals, Mother and Child Hospitals, Eye Centre, Super Speciality Hospital, Institute of Dental Sciences, Ayurvedic & Unani Tibbia, Homeopathic, etc. Data like numbers of beds, numbers of doctors, number of other paramedical staff and other statistics related to the hospitals obtained online from the Ministry of Health and Family Welfare website, govt. of NCT of Delhi link (web.delhi.gov.in and health.delhigovt.nic.in).
3.3 Assumptions in Variables and Data Selection One of the most important assumptions in DEA is that DMUs must be homogeneous, so only those public sector hospitals are selected for this study that performs approximately the same function. The age, gender, and status of doctors/nurses/patients have a negligible impact on hospital performance, so these factors are not considered during the efficiency analysis. Organizational factors such as team resources and administrative support, leadership, management, and accountability issues, etc. also influence the efficiency of hospitals which is not considered here.
172
D. Kushwaha and F. Talib
Table 4 Selected public sector hospitals of Delhi for DEA analysis with input/output data of one year Hospitals (DMUs)
(O) No. of OPD (U1 )
(O) No. of IPD (U2 )
(O) No. of major surgeries (U3 )
(I) No. of doctors (V1 )
(I) No. of nurses (V2 )
(I) No. of other paramedical staff (V3 )
(I) No. of beds (V4 )
H1
520,642
5282
1200
32
41
27
50
H2
622,791
15,748
1569
113
89
27
103
H3
219,459
5366
394
18
25
13
100
H4
796,903
49,581
4366
118
96
134
100
H5
1,172,681
53,426
8141
199
403
96
500
H6
701,759
33,263
5021
124
125
46
325
H7
691,785
55,279
6963
107
93
112
100
H8
266,305
9012
785
42
7
66
200
H9
800,681
20,254
3467
131
149
60
200
H10
473,032
16,902
2708
112
109
51
100
H11
793,297
28,379
7990
156
103
55
100
H12
310,359
8832
1092
36
15
17
150
H13
397,531
10,456
1483
63
87
48
100
H14
250,943
20,336
685
30
121
23
200
H15
776,949
47,729
5489
185
236
110
300
4 Result Discussion 4.1 Introduction of Data and Analysis Table 4 contains all the data that we have provided for DEA analysis. This means that these are the data that we need to add to the original file for DEA analysis, and Table 5 represents the very basic summary of DEA CCR-I model analysis.
4.2 Relative Efficiency Score and Rank of Hospitals Table 6 represents the score of each hospital, and the hospitals are ranked according to the score. The hospitals which score one get rank-1 each that means these are the efficient hospitals and construct efficient frontier. Other hospitals scores less than one, these are inefficient hospitals, and these are within efficient frontier. Figure 4 represents the relative efficiency of all the 15 hospitals in which we can easily see that six hospitals having relative efficiency less than one and other nine hospitals having relative efficiency one.
Comparative Efficiency Measurement of Indian Hospitals …
173
Table 5 Summary of the DEA—CCR-I MODEL analysis DEA model = DEA-solver LV8.0/ CCR(CCR-I) Problem = Hospitals No. of DMUs = 15 Returns to scale = Constant (0 = < Sum of Lambda < Infinity) No. of input items = 4 Input (1) = No. of Doctors (V1 ) Input (2) = No. of nurses (V2 ) Input (3) = No. of other paramedical staff (V3 ) Input (4) = No. of Beds (V4 ) No. of output items = 3 Output (1) = No. of OPD (U1 ) Output (2) = No. ofIPD (U2 ) Output (3) = No. of major surgeries (U3 ) Statistics on Input/Output Data V1
V2
V3
V4
U1
U2
U3
Max
199
403
134
500
1,172,681
55,279
8141
Min
18
7
13
50
219,459
5282
394
Average
97.7333
113.267
60.4667
175.2
586,341
25,323
3423.53
SD
56.2049
95.4138
35.6761
115.703
260,447
17,553.9
2646.76
Correlation V1
V2
V3
V4
U1
U2
U3
V1
1
0.79743
0.67143
0.56482
0.91147
0.7576
0.85705
V2
0.79743
1
0.48136
0.81782
0.79038
0.67391
0.66556
V3
0.67143
0.48136
1
0.27602
0.67232
0.86365
0.65896
V4
0.56482
0.81782
0.27602
1
0.54005
0.49305
0.46047
U1
0.91147
0.79038
0.67232
0.54005
1
0.77244
0.86208
U2
0.7576
0.67391
0.86365
0.49305
0.77244
1
0.84223
U3
0.85705
0.66556
0.65896
0.46047
0.86208
0.84223
1
No. of efficient DMUs = 9 No. of inefficient DMUs = 6
4.3 Average Projection of Inputs and Outputs Variables Projection is to push or pull the output and input variables respectively of inefficient DMUs according to efficient frontier so that they can improve themselves as an efficient DMUs. Projections of efficient DMUs are zero because they are already on efficiency frontier. Other 6 DMUs (H2, H5, H9, H10, H13 and H15 ) are not on efficiency frontier so for them we have some suggestions, for example inefficient hospitals on an average for input V1 we are working on 98 but to achieve efficiency frontier we
1
H1
1
1
No
DMU
Score
Rank
12
0.7831
H2
2
1
1
H3
3
1
1
H4
4
10
0.9153
H5
5
1
1
H6
6
1
1
H7
7
1
1
H8
8
11
0.7896
H9
9
Table 6 Relative efficiency score and Rank of hospitals which is assigned in the DEA analysis
14
0.6873
H10
10
1
1
H11
11
1
1
H12
12
15
0.5838
H13
13
1
1
H14
14
13
0.7497
H15
15
174 D. Kushwaha and F. Talib
Comparative Efficiency Measurement of Indian Hospitals …
175
DMU
Hospitals H15 H14 H13 H12 H11 H10 H9 H8 H7 H6 H5 H4 H3 H2 H1 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Efficiency
Fig. 4 Graph between DMUs (hospitals) and their relative efficiencies
must be working on ~85 of input V1 and from this, we can achieve 10.87% reduction in input V1 which shows the conservation of resources. We can achieve on an average 14.42% reduction for input V2 , 9.942% reduction in V3 , and 14.24% reduction in V4 (Table 7). In output projection, we can see in Table 8 that all the efficient and inefficient DMUs do very well for the output U1 and U2. This is the good sign that all the hospitals do very well on these output parameters. Now for output U3 inefficient hospitals are working on an average 3424 surgeries but projection shows that we need to push to ~3799 surgeries that mean there is the scope of improvement is 17.2%.
4.4 Average Slack Value of Inputs and Outputs We know that slacks are the resources that will go unutilized and what is the lack of in output. We know that efficient units are utilized 100% of their resources therefore for the efficient unit, the slack values of each input and output should be zero. So there is no any slack for any input and output variables of efficient units. Table 9 shows average slacks of input V1 and V2 and output U3 data for inefficient DMUs. These slacks are occurring because we are concentrated on other parameters.
0.901
1
0.584
0.142
Average
Max
Min
S. D
Score
5.94
1
15
5.6
Rank
58.18
18
51.046
18 98.8
7
−41.62 15.255
403
0
113
182.14
84.567
−10.87
199
97.73
V2 Data
Diff. (%)
Data
Projection
V1
Table 7 Projection of inputs variables
64.75
7
258.06
87.98
Projection
19.016
−47.23
0
−14.42
Diff. (%)
V3
36.93
13
134
60.47
Data
35.8491
13
134
54.1436
Projection
14.241
−41.62
0
−9.942
Diff. (%)
V4
120
50
500
175
Data
94.664
50
360.71
151.75
Projection
14.96
−41.62
0
−11.24
Diff. (%)
176 D. Kushwaha and F. Talib
Comparative Efficiency Measurement of Indian Hospitals …
177
Table 8 Projection of outputs variables Data
U1
U2
U3
Projection Diff. Data (%) Average Max
586,341
Projection Diff. Data (%)
Projection Diff. (%)
586,341
0
25,323 25,323
0
3424 3798.84
17.2
1,172,681 1,172,681
0
55,279 55,279
0
8141 8141
161
Min
219,459
219,459
0
St Dev
269,588
269,588
0
5282
0
18,170 18,170
5282
0
394 394
0
2740 2734.23
41.8
Table 9 Slack values of inputs and outputs Score
Rank
Slack
Slack
Slack
Slack
Slack
Slack
Slack
V1
V2
V3
V4
U1
U2
U3
Average
0.9006
5.6
1.401
11.0049
0
6.4628
0
0
375.3097
Max
1
15
11.06
110.804
0
96.942
0
0
2529.068
Min
0.5838
1
0
0
0
0
0
0
0
St Dev
0.1424
5.94
3.283
28.5228
0
25.03
0
0
718.8379
5 Conclusions In this study, a comparative performance assessment model/framework was proposed and then use input-oriented DEA model to measure the comparative efficiency of 15 public sector hospitals in Delhi, India. The results of input-oriented DEA-CRS model show that out of 15 hospitals, and 9 (60%) are pure technical efficient means they are converts their inputs into outputs efficiently. However, six hospitals (40 per cent) are technical inefficient due to the unserviceable resources. The hospital H13 has the least relative efficiency with least score of 0.5838 and gets the last ranking, and means H13 has the maximum effect of unserviceable resources on its efficiency score. It shows that this hospital inputs can get better excessively more than average. The projection results of all the inefficient DMUs illustrate that all the inputs have the considerable possibility of reduction and projection of outputs shows that for two outputs U1 and U2 all the DMUs (efficient and inefficient) are doing well; therefore, there are no to improvement but for output U3 all the inefficient hospitals (DMUS) have the significant scope of augmentation except DMU H5 . Input-oriented DEA model indicates that on an average, inefficient hospitals may be able to reduce 10.871% of doctors, 14.417% of nurses, 9.942% of other paramedical staff, 11.235 per cent of beds, and there is no need to expand of out-door patients and in-door patients but 17.147 per cent Major surgeries to be extended. These can help the government to set a policy agenda regarding which hospital need immediate attention and at the same time fixing gaps in the public health delivery mechanism.
178
D. Kushwaha and F. Talib
The slack values of different inefficient hospitals show that some input resources will go unutilized and from the results input, slack values of inefficient hospitals (DMUS) shows that on an average 1.4 doctors, 11 nurses, and 6.4 beds get unutilized in these inefficient hospitals and output slack values shows that these inefficient hospitals are lag behind on an average 375 major surgeries. This shows that despite several facilities, affordability in healthcare continues to be a serious concern. Further, despite available infrastructure and a huge government spending, there is a poor evidence for its impact due to lack of a centralised monitoring mechanism and no proper structure for implementing healthcare policies. The hospital industry is one of these types of industries where comparative efficiency analysis is utmost important for the administration to policymaking and curing patients. That is why measuring a hospital’s efficiency is one of the most essential tool for health policy maker authorities to validate their impact of policies because now a days huge resources are spend on medical services.. At present healthcare is becoming the most versatile requirement for every human being. Therefore, the government is making a large investment in it. On the basis of these results the government can make health improvements and also promote quality and responsibility. The efficiency of hospitals of this paper gives an idea to the government about fund allocation to the hospitals according to their services, i.e., efficiency-based funding system and also the introduction of performance-related pay for hospital staffs termed the “bonus” system. Profit-organizations, i.e., private sector hospitals are usually more efficient than government sector hospitals. Therefore, the government is encouraging the public–private partnership (PPP) model to increase the efficiency and performance of medical services. Various input/output drivers, DEA approaches and determination of response variables which have a significant impact on hospital efficiency are observed in the analysis.
6 Limitations It is difficult to measure the impact of any scheme or initiative on hospital performance immediately, as any plan gives a significant result over a period of time but this study considers the collection of only one year of data which may be not advisable for getting better results. The re-admission rate of patients in the hospital is another important variable for regulation of hospital performance, but in this study, it is not considering because of complexity. Other variables such as environmental impact, different geographical location, size of hospitals, and teaching status are also not considered in this analysis. In this study, the categorization of feasible input/output data and omitting undesirable data like infections, death etc. and the then the collection of data is one of the necessary and extremely challenging tasks. A proper categorization of variables and validation of the model can improve the result of this study. The efficiency of hospitals deviates from the actual one due to the absence of relevant data related to the input and output variables. To identify inefficient hospitals and make them efficient by input reduction or output growth or a combination
Comparative Efficiency Measurement of Indian Hospitals …
179
of both, authentic data is required. Measurement of hospitals’ efficiency depends on the inputs and outputs variables therefore the selection of variables requires great care as the entire analysis is based on it.
7 Future Work As per the above discussion, DEA can be carried out with any one of these two models either CRS or VRS. In the above we calculate the efficiency frontier with the CRS but we can also calculate the efficiency frontier with VRS model. By dropping one by one input or output variables, several alternative DEA models can be created [15], so this study can measure and compare the performance of hospitals on various combinations of input/output variables. We can compute the efficiency score with another method i.e. stochastic frontier analysis, panel data analysis and compare the result with Data envelopment analysis. Other DEA models like a variable return to scale (VRS), increasing return to scale, and decreasing return to scale analysis can be performed by different assumption and by little modification. A survey can be conducted with questionnaires in which the data will be analysed by statistical tools, and one can compare the result with the DEA result. For a period of time, DEA Malmquist productivity indices (MPI) can be used. This method gives a great understanding related to hospital behaviour with respect to productivity and efficiency for a period of time. A sensitivity analysis can give an idea of how the result changes with respect to the input and output variables i.e. when there is little change in the input or output data. This sensitivity analysis can give an excellent understanding of resource utilization and output augmentation. Considering other important input/output variables such as re-admission of patients, hospitalized deaths, environmental variables, etc., the study can be conducted which can give a good result related to the quality of performance of hospitals and help physicians and healthcare managers. Researchers can also examine hospital efficiency across geographic locations (rural, urban), size (small, medium, big) and teaching status (teaching, non-teaching hospitals).
References 1. Charnes A, Cooper WW, Rhodes E (1978) Measuring the efficiency of decision-making units. Eur J Oper Res 2(6):429–441 2. Azam M, Rahman Z, Talib F, Singh KJ (2012) A critical study of quality parameters in health care establishment: developing an integrated quality model. Int J Health Care Qual Assur 25(5):387–402 3. Talib F, Rahman Z, Azam M (2010) Total quality management implementation in the healthcare industry: a proposed framework. In: Proceedings of second international conference on production and industrial engineering (CPIE-2010) organized by Department of Industrial and Production Engineering, Dr. B.R. Ambedkar National Institute of Technology, Jalandhar (NITJ), Punjab, India during December 03–05, pp 1361–1368
180
D. Kushwaha and F. Talib
4. Azam M, Rahman Z, Talib F (2012) Core quality and associated supportive quality parameters/dimensions: a conceptual quality frame work in healthcare establishment. Int J Bus Excell 5(3):238–277 5. Talib F, Azam M, Rahman Z (2015) Service quality in healthcare establishments: a literature review. Int J Behav Healthc Res 5(1/2):1–24 6. IBEF, 2020. India Brand Equity Foundation. https://www.ibef.org/industry/healthcare-india. aspx. Accessed 15 Apr 2020 7. Talib F, Asjad M, Attri R, Siddiquee AN, Khan ZA (2019) Ranking model of total quality management enablers in healthcare establishments using the Best-Worst Method. TQM J 31(5):790–814 8. Farrell MJ (1957) The measurement of productive efficiency. J R Stat Soc 120(3):253–281 9. Banker RD, Conrad RF, Strauss RP (1984) A comparative application of data envelopment analysis and translog methods: an illustrative study of hospital production. Manage Sci 32(1):30–44 10. Liu JS, Lu LYY, Lu WM, Lin BJY (2013) A survey of DEA applications. Omega 41:893–902 11. Gandhi AV, Sharma D (2018) Technical efficiency of private sector hospitals in India using data envelopment analysis. Benchmarking Int J. https://doi.org/10.1108/BIJ-06-2017-0135 12. Bahrami MA, Rafiei S, Abedi M, Askari R (2018) Data envelopment analysis for estimating efficiency of intensive care units: a case study in Iran. Int J Health Care Qual Assur 31(4):276– 282 13. Stefko R, Gavurova B, Kocisova K (2018) Healthcare efficiency assessment using DEA analysis in the Slovak Republic. Health Econ Rev 8(6). https://doi.org/10.1186/s13561-018-0191-9 14. Mitropoulosa P, Talias MA, Mitropoulos I (2015) Combining stochastic DEA with Bayesian analysis to obtain statistical properties of the efficiency scores: an application to Greek public hospitals. Eur J Oper Res 243:302–311 15. Chowdhury H, Zelenyuk V (2015) Performance of hospital services in Ontario: DEA with truncated regression approach. Omega. https://doi.org/10.1016/j.omega.2015.10.007 16. Mogha SK, Yadav SP, Singh SP (2015) Slack based measure of efficiencies of public sector hospitals in Uttarakhand (India). Benchmarking Int J 22:1229–1246 17. Pham TL (2011) Efficiency and productivity of hospitals in Vietnam. J Health Organ Manag 25(2):195–213 18. Prior D (2006) Efficiency and total quality management in health care organizations: a dynamic frontier approach. Ann Oper Res 145:281–299 19. Ferrari A (2006) Market oriented reforms of health services: a non-parametric analysis. Serv Ind J 26(1):1–13 20. Gannon B (2005) Testing for variation in technical efficiency of hospitals in Ireland. Econ Soc Rev 36(3):273–294 21. Giuffrida A, Gravelle H (2001) Measuring performance in primary care: econometric analysis and DEA. Appl Econ 33(2):163–175 22. Chirikos TN, Sear AM (2000) Measuring hospital efficiency: a comparison of two approaches. HSR: Health Serv Res 34(6) 23. Jacobs R (2000) Alternative methods to examine hospital efficiency: data envelopment analysis and stochastic frontier analysis, CHE Discussion Paper 177 24. Linna M (1998) Measuring hospital cost efficiency with panel data models. Health Econ 7:415– 427 25. Chattopadhyay S, Ray SC (1996) Technical, scale, and size efficiency in nursing home care: a nonparametric analysis of Connecticut homes. Health Econ 5:363–373 26. Fizel JL, Nunnikhoven TS (1992) Technical Efficiency of For-profit and Non-profit Nursing Homes. Manag Decis Econ 13:429–439 27. Grosskopf S, Valdmanis V (1987) Measuring hospital performance: a non-parametric approach. J Health Econ 6(1):89–107 28. Sherman HD (1984) Hospital efficiency measurement and evaluation-empirical-test of a new technique. Med Care 22:922–938
Comparative Efficiency Measurement of Indian Hospitals …
181
29. Nunamaker TR (1983) Measuring routine nursing service efficiency: a comparison of cost per patient day and data envelopment analysis models. Health Serv Res 18:183–208 30. Gearhart NM, Michieka NM (2018) A comparison of the robust conditional order-m estimation and two stage DEA in measuring healthcare efficiency among California counties. Econ Model 395–406 31. Parkin D, Hollingsworth B (1997) Measuring production efficiency of acute hospitals in Scotland, 1991–94: Validity issues in data envelopment analysis. Appl Econ 29:1425–1433 32. Tyagi P, Yadav SP, Singh SP (2009) Relative performance of academic departments using DEA with sensitivity analysis. Eval Program Plann 32(2):168–177 33. Emrouznejad A, Yang G (2018) A survey and analysis of the first 40 years of scholarly literature in DEA: 1978–2016. Socioecon Plann Sci 61:4–8 34. ˙Ilgün G, Konca M (2019) Assessment of efficiency levels of training and research hospitals in Turkey and the factors affecting their efficiencies. Health Policy Technol 8(4):343–348
An Adoption of Blockchain Technology in Agri Food Supply Chain: An Overview Ashutosh Kumar, D. J. Ghode, and Rakesh Jain
Abstract Blockchain Technology (BT) is anticipated to bring a drastic change in the manner transactions are carried in the agri food supply chain (AFSC). BT provides transparent operations by avoiding trust-related issues in the whole supply chain (SC). In this paper, a systematic literature assessment is taken as the main methodology to analyse and survey the recent literature. The main aim is to be identified as the pros and cons of BT related to its adoption in AFSC. This research paper also indicates the reason behind the slow adoption of blockchain (BC) as well as hesitation regarding the implementation in the industries. There are many ongoing projects which adopt BT in their SC, but still, there are some obstacles which are needed to be sorted out before implementation. Due to the barriers and some difficulties, its popularity is not spreading as much between the farmers and the implementing authority. BT deserves much popularity in any SC because it has a better capacity to handle transparency, traceability, privacy, security, and many more characteristics. The key finding of this paper is the potential challenges in the adoption of BT in AFSC. This study has proposed the solution to the various issues in the implementation of BT in AFSC. Keywords Blockchain · Transparency · Traceability · Supply chain management · Agri food supply chain
1 Introduction Supply Chain (SC) is a series of (planning and implementation) processes and (material flow, information flow, and money flows) that goal to meet end customer requirements, which take place within and between different stages along a sequence, from production level to final consumption level [24]. The SC includes not only producers and their suppliers but also depends upon the logistic flows, distributers, storehouses, wholesaler’s retailers, and end users themselves. In the wide-ranging terms
A. Kumar · D. J. Ghode (B) · R. Jain Department of Mechanical Engineering, Malviya National Institute of Technology, Jaipur, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_15
183
184
A. Kumar et al.
SC also involve new product design, processing, marketing, distribution, finance, and customer service. In recent time, attentiveness toward supply chain management (SCM) has been enlarging in the agri food industries both in the developing and developed countries. Agri products are used as raw materials for producing end products with higher added value in agri food chain. In general, preservation and governing processes expand the shelf life of the product. Due to shelf life constraints of agricultural and food products and increase customer attention to safe and eco-friendly manufacturing methods, there is a need of much efficient and effective SC in food industry [13]. SC is more a need than an option for most industries or firms. Modern AFSCs are very complex and carry several stakeholders, with every doing specific roles concerning food productions. Information or record should be available from ‘farm to fork’ and its beyond including detailed information about agriculture procedures on farms, as well as information referring to packaging shipping and storing conditions until costumer buy to trace food within the SC. Current structures of AFSCs are still causing inflexibility and inefficiencies. By enforcing the BT in the SC, the bullwhip effect may be minimized [24]. By adopting BT in the SC, product source and processes can be tracked. BT along with advanced IT, communications, and IOT have adopted for the development of agri food value chain management in four major aspects: Traceability, Information security, inoperability, manufacturing, and transparency. There is massive challenge to keep the food safe and at good quality because food is traveling larger distances. For better control and monitoring over SC and quality of product, a collaborative traceability is required and it also enable source detection of product. BC has been proven that for faster handling times be must have to improve tracking processes [20]. The main objective of this study is to provide challenges in adoption of BT in AFSC. After rigorous literature review, this research paper provides better understanding about how BT implementation in AFSC. Further, in Sect. 2, methodology to achieve goal of this paper is provided. Section 3 covers the literature review. Section 4 discusses the various challenges in adoption of BT in AFSC. In Sect. 5, conclusion and future research directions have been given.
2 Methodology In this paper, a systematic literature assessment is taken as the main methodology to analyse and survey the presently literature. A systematic assessment is a systematic and comprehensive way to take and select the most relevant theory and practices in the premier literature in the field. This review article has been larger landscape of research on BT in AFSC and then to analyse the findings and their vital application in AFSC [27]. Six databases were selected for this literature search that include Science Direct, Web of Science, Taylor & Francis Online, Emerald; Google Scholar, and IEEE
An Adoption of Blockchain Technology in Agri Food Supply Chain: An Overview
185
Table 1 Search string for the database findings Keyword used
‘Blockchain’, ‘information transparency’, ‘agriculture’, ‘agricultural’, ‘agribusiness’, ‘agriculture safety’, ‘agriculture security’, ‘agriculture traceability’, ‘food safety’, ‘food security’, ‘food traceability’, ‘food supply chain’, ‘food trade’, ‘food value chain management’, ‘agriculture value chain management’, ‘agri food value chain management’,
Database used
Science Direct, Web of Science, Taylor & Francis Online, Emerald, Google Scholar, IEEE Xplore
Related search string ‘Agriculture’ or ‘Agricultural’ or ‘Agribusiness’ or ‘agriculture safety’ or ‘agriculture security’ or ‘food safety’ or ‘food security’ or ‘food traceability’ or ‘food supply’ or ‘food trade’ or ‘agriculture traceability’ Or ‘food value chain management’ or ‘agriculture value chain management’ or ‘agri food value chain management’ or ‘agriculture trade’
Xplore. The main motive of the research is to detect the important issues and challenges related to the applications of BT in agri food chain management. The combination of “blockchain” and “agri food value chain management”, with all these related terms was used to search keywords, title, and abstract in the above databases (Table 1).
3 Literature Review 3.1 Current Scenario of AFSC Agri food industries or organization are related to the production and distribution of animal-based products and vegetable product. There are mainly two types to distinguish agri food [5]. I.
II.
First one is for “fresh agriculture products” such as flowers, fruits, and fresh vegetable. The main processes in this AFSC are Handling, Conditioning storage, Packaging, transporting, and trading of these products. And the second one is for “processed food products” such as snacks, juice, meat, and desserts. In such chains agri products are used as raw material for producing customer products with greater added values. In many cases condition and conservation processes enlarge shelf life of products.
Actors of both type of chains can know the original good quality of food product and quality decay may happened due to the inappropriate action taken by the other actors in the SC. For example, milk cane may lose their good quality due to the sun light. It may unfit for processing. So, there is a need of strong communication chain between its actors so that such unnecessary thing does not happen in SC. There is need of such SC in which communication or information sharing between its own participants can happen.
186
A. Kumar et al.
The previous couple of years have seen a bang of research and development processes around the BT. It in an exceedingly cloud storage situation, which is used to access 3 level of information, (i) provenance, (ii) assortment, (iii) validation, and (iv) storage [3]. The result of this paper concludes that BC is best in term of privacy of its user plus tamper proof datasets which occupy lesser storage. Implementing the ‘internet of things’ such as sensors and other technology in the SCM sector have been inspired many research interests in the past few years. Both IOT and BC are still under explored for the research in the AFSC domain. Although the record has to be arranged by IOT devices and BC [4], the BC would not require any other devices such as hardware which are used to store data in very low cost. BC is used to store data in unit level that makes it possible to track every single unit even in complex foods [9]. Further BT needs IT infra in every SC to access internet that is difficult to implement at this very moment because some of the raw material provider is very remote to reach. Agri Food industries are facing some bigger challenges to attain reliability and transparency in SCM. Actors involved in such challenges are as follows: • Donar/supplier: suppliers of raw materials, like seed and nutrient, however additionally pesticides and chemicals. • Maker/Manufacturer: sometimes farmers are accountable of process from planting the seed to the harvesting of seed. • Processors: participants might do varied action from easy packing to additional complicated tasks (ex-pressing of the olive) • Dealer/Salesperson/Distributor: participants is responsible for transporting o/p of the processors. Ex-product • Shopkeeper: Actors are accountable for merchandising the merchandise. They have very little shop or huge super markets. • Customer: Ultimate part of the series.
4 Blockchain in AFSC Food supply and agriculture are interlinked to each other and the products obtained by agriculture are treated as raw material for various SC in which customer are generally the first client [14]. The AFSC are mostly multi actors and distributed network chain. In this network different participants are involved to do specific work such as farming, shipping, distributing, and retailing. But these systems are not working efficiently and also unreliable. Example: when a person purchase goods locally they do not have sufficient knowledge about goods origin and the environmental foot print of production. BT can be used to solve real practical problem which arises in agriculture and food SC chain. These initiatives can be parted in four main types such as written below:
An Adoption of Blockchain Technology in Agri Food Supply Chain: An Overview
A.
B.
C.
D.
187
Food security Food security defines by FAO (the food and agriculture organization) situation when “all people, at all times, have physical, social and economic access to sufficient, safe and nutritious food that meets their dietary needs and food preferences for an active and healthy life”. Food Safety Food safety involves the different processes related to the hygienic of food in order to safe from diseases occurring in human bodies. BC could prevent from this by detecting the root cause of the problem and hence it can be easily removed. Food Integrities Food integrity relates the simple exchange of food material in the SC. Every participant should have given details about their origin of goods. Food integrity is mainly concerned in China [22]. Food integrity and food safety can be improved through transparency and traceability [14]. Through BC, food industries could mitigate food fraud and could easily detect and return back to their original source. Customer has to be done only to scan the QR code to see every detail about the products. The QR code can be scanned by simple android phone, and we can see the small detail about the products. Smaller Farmer Supports Smaller co-operatives of farmer constitute strong method in order to increase competitiveness in developed country. Helping individual farmers could give them high value of crop what they are cultivating. There are some new start-ups which support smaller farmers by giving them equipment, this will help them in tracking of seeds and other nutrients which is good for their crops [14]. A BC is a virtual/digital transaction databook that is maintained by network of various computing devices that are not dependent on any trusted third-party intermediaries. Every transaction databooks known as blocks are processed through certain software platform that gives permission to the data to transmit, process, store, and presenting in human readable form. BT shares all the document for keeping record [23]. All the necessary data is collected at every transaction step in AFSC. BC is easy to understand, and document can be shared in any given time frame by any stakeholders in whole SC. BT is still in premature state due to this there are many challenges that needs to be sorted out before the implementing in AFSC. Movement of agri food can be track along the whole SC and this process of tracking is known as traceability. There are two different concepts regarding track and trace. Tracking means backward processes; in this, origin of product can be found by maintaining of record along whole SC. Tracing means forward processes in which last users or customers are found by position in the whole SC [2]. Tracing follows upstream path to find the origin of product, whereas tracking refers to downstream in the AFSC. Tracking of food at the farm level as well as farmer processing level and also at the transportation and marketing level along whole SC can be profit to trace fraud in food product and to minimize food wastages, minimize risk of food safety and level food transparency. Bhat
188
A. Kumar et al.
and Joudu [2] have given different drivers as well as motivators for traceability in AFSC. Motivational factors can be detected by food quality, food safety, and food visibility.
4.1 Application of BT in AFSC One of the primary applications of BT was bitcoin. A peer to peer electronic cash transfer [19]. After that smart contracts, the transparencies come in the picture to gain the trust of the people, which is only focus on the gaining trust of user [6]. Out of cent percent, there are more or less eighty percent of the academic paper depends on bitcoin network and other twenty percent focuses on another BC application [25]. Tian [22] has proposed a model stated that agri food based on RFID and BC serves farmer of china to trace their product. It also ensures food safety as well as trustrelated to the product. Clone of RFID tag can be created in the BC system. Walmart’s company has invested lot of money and human resources to achieve transparencies and traceability of the product within the SC [27]. Tracking of their product origin by the use of BT is the main achievement of Walmart. Zhao et al. [26] have drawn a structure of production based on information services by used of BT principle and also by using cloud computing.
5 Challenges in Adoption of BT in AFSC BT is in its infancy stages. So, there are many issues related to the implementation of them. For better implementation of BT in AFSC, there need to be solved some serious problem before thinking of its implementation. From the plethora of literature review, the most commonly occurring challenges while adoption of BT in AFSC are as follows: • Scalabilities and store capacities Scalabilities and store capacities are commonly occurring issues while implementing BT. [12] have said that scalabilities are the main reason behind the low adoption of BT. He also mentioned that storage is the main challenges and sent negative thoughts in mind while implementing BT. • Privacies fraud BT always gives us high level of transparency and to help in gaining trust among stakeholders [21]. But it also has some disadvantages related to the saving from privacy fraud or in other words user privacy is not secure however. Kosba et al. [15] also proposed that BT could not promise about the transactional privacies even in the permissioned BC or private BC. • Rule and regulation problem BT and agri food expand all over the world. So there needs to be made rule and regulation to handle all the processes and activities in proper way without any disturbance. There is necessity to introduce strong law in the SC [7].
An Adoption of Blockchain Technology in Agri Food Supply Chain: An Overview
189
• Problem of high cost Lin and Liao [17] have proposed that BT needs lot of expanses and human factors to implement in the AFSC. Yli-Huumo et al. [25] also proposed that when the SC become more complex BT needs more power and more human effort to sort out problems. • Latencies and throughput problem Tian [22] mentioned the challenges of latency and throughput as (i) capacity of transactions per sec is limited to seven and with the growth of BC size dealing with it is very important, (ii) while in case of latencies issues, most of the BC system requires some time to operate transaction [8]. • Low level of skill Due to the infancy stage of BT, most of the people are not having deep knowledge about how to operate the BC in AFSC [1]. For gaining depth knowledge about BT, [10] has done many interviews and attain many meetings with the skill full person to know about the characteristics of the BT (Table 2).
6 Conclusion and Direction for Future Research In this study, we conducted systematic literature review (SLR) to view the pre-work on BT in AFSC. This review was conducted by some question formulation; locate study, relevant studies selections and their evaluation. We identified 27 publications that include various journals as well as books and conference papers. The main targeted publications for this review covered theme related to traceability, BC based information securities, food security, food safety, food integrities, etc. This paper covers the timeframe of 11 years (2008–2019) related to BT in AFSC. This research article enhances knowledge mainly in three points in terms of BC: • This research paper gives an idea about the evolution or origin of BT and its applications in AFSCM. • This paper highlighted main challenge and gives primary solution for implementing BT to AFSC. • Few research gaps have been identified and there after research direction was suggested for implementing BT in AFSCM. The research finding of this study gives us strong framework for further study in the current areas of BT in AFSCM. There are mainly two area in the literatures to use BT to increase efficiency and to enhance performance of AFSC. First one is by using BT alone, and the second one is by using BT merge with other technologies such as RFID and IoT. But it has some own limitations while implementing with the BT. It generates larger volume of data and required connectivity with the computational power. Computational power required high demand of electricity which is not worth full. Hybrid approach may be considered to implement with the BT. The area is open for future researcher to examine the advantages and disadvantages of hybrid concepts and choose the optimal solutions to improve AFSC.
190
A. Kumar et al.
Table 2 Solutions for the above challenges of the BT in the AFSC Main challenge Main solutions possible Scalability and store capacity
Main advantages
Removal of older Security can be transactions [11] improved in the network
Main disadvantages
Direction for future research
Transaction speed is disturbed by more data store in the BC
Adding AI and Deep Learning concepts with the present algorithm
Privacy leakage Double Chain architecture [16]
1. Users privacy can be increased 2. Increase speed of transaction 3. Through put rate is increased
Public BC consensus are very less
Need to conduct high empirical research to enhance the performance of AFSCM
Problem of high cost
Need to perform proof of stake algorithms [18]
1. Computing time is very less required 2. Short of latency is required
1. Consensus speed is low; 2. Accumulation of node is happened even though node is not attached to the structure
Conduct comparative analysis of the given algorithm by used of empirical studies in AFSCM
Rule and regulations problem
Clear cut regular framework is required to design and implement proper BT [10]
Increase the processes of rule and regulations policy co-related with BT implementation such that each actor adopts BC easily and simply
Instead of getting suitable results government had invested lot of money and resources
Examine the effect of BCT on AFSC More importantly in the western world of agri food trades and then have to take some decision on making rule and regulations towards this field
Throughput problem and Latency issues
Block size can be 1. Can be increased [25] improved the throughput without disturbing other characteristics unaltered for ex-security, privacy 2. Speed of transaction is increased
Bigger/larger block can require more expanses for management
Use of proper consensus algorithm that helps to reduce the transaction time and improve throughput
An Adoption of Blockchain Technology in Agri Food Supply Chain: An Overview
191
References 1. Atlam HF, Alenezi A, Alassafi MO, Wills GB (2018) Blockchain with Internet of Things: benefits, challenges and future directions. Int J Intell Syst Appl 10(6):40–48 2. Bhat R, Jõudu I (2019) Emerging issues and challenges in agri-food supply chain. Sustain Food Supply Chains 23–37 3. Caro MP, Ali MS, Vecchio M, Giaffreda R (2018) Blockchain-based traceability in Agri-Food supply chain management: a practical implementation. In: 2018 IoT vertical and topical summit on agriculture—Tuscany, IOT Tuscany 2018, pp 1–4 4. Casino F, Dasaklis TK, Patsakis C (2019). A systematic literature review of blockchain-based applications: current status, classification and open issues. Telematics Inform 36(November 2018):55–81 5. Castro JAO, Jaimes WA (2017) Dynamic impact of the structure of the supply chain of perishable foods on logistics performance and food security. J Ind Eng Manage 10(4 Special Issue) 6. Christidis K, Devetsikiotis M (2016) Blockchain and smart contracts for the internet of things. IEEE Access 4(2016):2292–2303 7. Crosby M, Nachiappan, Pattanayak P, Verma S, Kalyanaraman V (2016) Blockchain technology: beyond bitcoin. Appl Innov Rev (2):6–19 8. Fernandez-Carames TM, Fraga-Lamas P (2018) A review on the use of Blockchain for the internet of things. IEEE Access 6(2018):2169–3536 9. Francisco K, Swanson D (2018) The supply chain has no clothes: technology adoption of blockchain for supply chain transparency. Logistics 2(1):2. https://doi.org/10.3390/logistics 2010002 10. Ge L, Brewster C, Spek J, Smeenk A, Top J (2017) Blockchain for agriculture and food. Wageningen Economic Research, Netherland 11. Hamida EB, Brousmiche KL, Levard H, Thea E (2017) Blockchain for enterprise: overview, opportunities and challenges. In: The thirteenth international conference on wireless and mobile communications, Nice, France, pp 23–27 12. He Y, Li D, Wu SJ, Shi C (2018) Quality and operations management in food supply chain. J Food Qual (Editorial) 13. Hua J, Wang X, Kang M, Wang H, Wang FY (2018) Blockchain based provenance for agricultural products: a distributed platform with duplicated and shared bookkeeping. In: IEEE intelligent vehicles symposium, proceedings, 2018-June(iv), pp 97–101 14. Kamilaris A, Fonts A, Prenafeta-Bold´ FX (2019) The rise of blockchain technology in agriculture and food supply chains. Trends Food Sci Technol (September) 15. Kosba A, Miller E, Shi Z, Wen C (2016) The blockchain model of cryptography and privacypreserving smart contracts. In: IEEE symposium on security and privacy. San Jose, USA, 22–26 May 2016. USA, Washington 16. Leng K, Bi Y, Jing L, Fu HC, Nieuwenhuyse IV (2018) Research on agricultural supply chain system with double chain architecture based on blockchain technology. Future Gener Comput Syst 86(2018):641–649 17. Lin IC, Liao TZ (2017) A survey of blockchain security issues and challenges. Int J Netw Secur 19(5):653–659 18. Lu Q, Xu X (2017) Adaptable blockchain-based systems: a case study for product traceability. IEEE Softw 34(6):21–27. https://doi.org/10.1109/MS.2017.4121227 19. Nakamoto S (2008) Bitcoin: a peer-to-peer electronic cash system. Accessed July 16, 2019, https://bitcoin.org/bitcoin.pdf 20. Pearson S, May D, Leontidis G, Swainson M, Brewer S, Bidaut L, Zisman A (2019) Are distributed ledger technologies the panacea for food traceability? Glob Food Sec 20(February):145–149 21. Reyna A, Martin C, Chen J, Soler E, Diaz M (2018) On blockchain and its integration with IoT challenges and opportunities. Future Gener Comput Syst 88:173–190
192
A. Kumar et al.
22. Tian F (2016) An agri-food supply chain traceability system for China based on RFID & blockchain technology. In: 13th international conference on service systems and service management, ICSSSM 2016 23. Tse D, Zhang B, Yang Y, Cheng C, Mu H (2018) Blockchain application in food supply information security. In: IEEE international conference on industrial engineering and engineering management, December 2017, pp 1357–1361 24. Van der vorst J, Silva da C, Trienekens J (2007) Agro-industry supply chain: concepts and applications. Food and agriculture organization of the United Nations. 25. Yli-Huumo J, Ko D, Choi S, Park S, Smolander K (2016) Where is current research on blockchain technology? A systematic review. PLoS ONE 11(10):1–27 26. Zhao G, Zhang G, Liu M, Yu S, Liu Y, Zhang X (2018) Research on manufacturing service behaviour modelling based on blockchain theory. AIP Conf Proc 2018:30–35 27. Zhao G, Liu S, Lopez C, Lu H, Elgueta S, Chen H, Boshkoska BM (2019) Blockchain technology in agri-food value chain management: a synthesis of applications, challenges and future research directions. Comput Ind 109:83–99
Prediction Model for Talent Management Analysis Through e-HRM Nasreen Nasar, Sumati Ray, Abdul Wadud, and Saiyed Umer
Abstract A novel method for talent management analysis has been proposed in this paper. Talent management is one of the important processing in the human resource management systems, which supports the retention, development, and motivation of employees in the organization through the electronics human resource management (e-HRM) system. The implementation of the proposed system has been divided into two parts. In the first part, various activities and their respective challenges for talent management have been demonstrated and discussed. In the second part of the implementation, a prediction model has been derived by analyzing the data for talent management. This data analytics has been performed in three components: (i) data preprocessing, (ii) feature analysis, and (iii) data classification. Extensive experimentation has been performed using two standard talent activity datasets from the ‘kaggle website’, and the performance has been justified and elaborated for each activity of talent management through e-HRM. Keywords Talent management · Human resource management · Data analytics · Prediction model · e-HRM
1 Introduction Talent is an important driver for business performance between human resources and business leaders. Talent management is the process of the identification of the talent shortages and talent surpluses, locating and relocating talents, and compensating the talent present in the organization. It ensures (i) an optimum level of talent in the organizations, i.e., it ensures that there is neither shortage nor surplus of the N. Nasar Department of Management and Business Administration, Aliah University, Kolkata, India S. Ray Indian Institute of Social Welfare and Business Management, Kolkata, India A. Wadud · S. Umer (B) Department of Computer Science & Engineering, Aliah University, Kolkata, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_16
193
194 Talent strategy
Talent analysis
N. Nasar et al. Workforce planning Potential development
Talent acquisition Leadership measure
Capability development Capability performance
Fig. 1 Working-flow diagram of a talent management system
talent, (ii) timely location and acquisition of talent in the organizations for achieving organizational goals, (iii) formulating and implementation of contemporary compensation strategy formulation for attracting and retaining the talent, (iv) long-term goal achievement by having talent competitive enough to keep the organizational performance high and growing, and (v) looking after the interests of both the employees who have the talent and the organization for which the talent is being used. Talent Management (TM) practices are always at the strategic priority [1] and it works in a sequential manner as shown in Fig. 1. Thunnissen [2] had explained and identified the happenings of Talent Management (TM) practices with various influence factors at the organizational, institutional, and individual levels. Dimitrov and Kiril [3] had clarified various perspectives of TM essence and had proposed the fish-bone diagram for TM essence. There is Ambidexterity of HRM (Human Resource Management) that measures the ability of HRM to both explore and exploit talented employees needed to adapt to intense competitiveness in the changing business world. It has been argued that the electronics human resource management system (e-HRM) significantly influences the relationships between HRMA and talent management [4]. This e-HRM moderately supports the HRMA to perform talent management (TM) tasks. Organizations with the use of modern technologies, like the internet and webservices are now able to manage talent by use of Talent Management Systems (TMS). By use of such software, the information on the current talent inventory in the organization is maintained and retrieved when required. Based on the current inventory level and environmental (both internal and external) scanning, the forecast for the gap in the demand and supply of the talent can be assessed and based on that action plans like locating and adjusting the excess talent or find the demanded talent for the forecasted period are prepared. With the use of software and web-services, talent management has become more and more rewarding and is considered a strategic tool for any organization. Talent Management as a part of Human Resource Management finds ways to engage, train, and motivate employees to get the best performance from them and is directed toward the organizational goals and objectives. The e-HRM integrates all the HRM activities through transforming HR managers to make strategic planning in the organizations using web-based technology [5]. The e-HRM uses various activities that are influenced by several factors which are (a) information flow, (b) positive/negative interaction between subordinates, (c) perseverance controlling by individuals, and (d) acceptance of negative attitude by the employees. So, the acceptance and effectiveness of e-HRM software have been
Prediction Model for Talent Management Analysis Through e-HRM
195
performed by introducing media and message characteristics in the e-HRM activities [6] which are e-Compensation, e-Recruitment, e-Learning, e-Selection, and ePerformance Management while these activities affect the maximization the strength of talented employees for increasing the effectiveness and productivity in the organization [7]. These e-HRM activities directly or indirectly support the working principle of the talent management system (TMS). The right employees are identified as talent subjects, and these measurements are generated by the combined effect of e-HRM and talent identification technology. The e-HRM software is eligible to (i) recognize and reward a diverse array of skills and capabilities within the business units and across the organization and (ii) facilitate, promote, and support a one-size-fits-all approach for TMS [8]. Nowadays both TMS and HRM have been seen to emphasize the integration of business strategies, the importance of appropriate role allocation, and both covered the same functional areas of recruitment, selection, training and development, appraisal, and reward management [9] activities for TMS. However, while there are some similarities, there are also some differences [10]: (i) HRM has been seen to emphasize egalitarianism in contrast to the segmentation policy of TM, (ii) HRM has also been seen to focus on management functions, and (iii) TM on the people involved with a particular focus on the retention, attraction, and development of talent. Hence the organization increasingly recognize the value of effective talent management (TM) practices that are informed by data and analytics techniques rather than intuition, the need to understand the role of information technology becomes more pertinent, and so the objective of this paper is to derive the prediction model for discussion and demonstration of the talent management analysis (TMA) scheme in an organization. The contributions of this paper are as follows: • Here the implementation of the proposed system has been divided into two parts. In the first part, the various activities and their respective challenges for talent management systems have been demonstrated and discussed. In the second part, the prediction models have been derived by analyzing the data for the talent management analysis, and for this, the steps of data analytic tasks have been performed. • During data analytics, the experimentation of the talent management system is divided into three segments which are (i) processing the employed data, (ii) analyzing the parameters as features from that data, and (iii) finally, classifying the data based on the selected features. • Here the processing of talent management analysis in the organization has been shown experimentally through available datasets, and the use of these datasets is highly correlated with measuring the talented employee in the organization. • During data preprocessing, the considered datasets are preprocessed in such a way that the preprocessed data can be used for analytics purpose, and its every row shows samples while each column shows feature. • During feature analysis, the dataset has two types of features (i) learning features and (ii) target features. The learning features are used to train the classifier while target features are assumed to the class labels of that target feature. At
196
N. Nasar et al.
the data classification stage, with respect to the target feature (class label) and the learning features are trained to the classifiers to derive some prediction model. In this work, some different classifier techniques such as SVM (Support Vector Machine), CART (Classification and Regression Tree as Decision Tree), kNN (kNearest), LR (Logistic Regression), and Random Forest have been used to obtain the performance of the proposed TMS. • For experimental purposes, two datasets available on the Kaggle website have been employed to obtain and elaborate the performance for Talent Management Analysis in an excellent manner. This paper is organized as follows: Sect. 2 describes the proposed system for the Talent Management Analysis. Section 3 discusses the experimental results with discussions about the methodology of the proposed system. Finally, the paper is concluded in Sect. 4.
2 Proposed analysis 2.1 Analysis for Talent Management System In this section, we have analyzed different functionalities and activities of a talent management system (TMS) in the organization and also validate these activities through data analytic tasks. For theoretical justifications of functionalities and activities of the talent management system, we have demonstrated some TMS models used in the organization. During the implementation of TMS, the researchers have derived these models which are based on the activities and functionalities of TMS. Among these models, Lewis and Heckman [10] have explained the talent management review system through the systematic and strategic ways of implementation in the talent management system which is shown in Fig. 2. From this figure, it has been observed that the TMS has several components such as (i) strategy for sustainable competitiveness, (ii) talent implementation, (iii) talent polling, (iv) talent management, and (v) talent prediction. Here each component has its outcomes: (i) based on Opportunity Resources, the TMS finds the strategies for finding competitive candidates, (ii) talent implementation is based on Talent analysis and fungibility activities, (iii) Performance, Compensation policies, and Carrier ladder are derived by Talent polling system, (iv) Competency architectures and Enterprise-wide data systems are handled by TMS, and (v) Selection, Recruitment, Performance management, and Compensation administration activities are handled by a talent prediction model. These activities are also handled by the e-HRM model. The objectives of TM are to manage the talent of all employees or to manage the talents of high potential and performing employees, and it is the most controversial thing that comes for TM. The topics of TM are discussed including identification of talent required for business operations, managing top management talents, linking the strategic business operations to TM practices, etc. Previously, the main objective
Prediction Model for Talent Management Analysis Through e-HRM
197
Talent Management Components Talent
Talent pool
strategy
Resources
Talent analysis Talent fungibility
Talent
system
Competency Performance architectures Compensation Enterprise-wide policy data systems Carrer ladder
Selection Recruitment Performance management Compensation administration
Talent Management Outcomes Fig. 2 A talent management review system proposed by Lewis and Heckman [10]
of TM was to focus on the recruitment of the most intelligent and capable talented employees with better managerial skills. Over time, the objectives of TM have been extended. Collings and Mellahi [11] had been defined TM as processes and activities which help to identify the advantages of sustainable competition in the organization, a talent pool development for the high performing incumbents. The recruitment activity is also being part of TM activities, which have been managed based on the recruitments of the role in question and it is implemented by the internal development and external recruitment combination. The organization needs to generate work motivations, commitment for the organization, and extra-role performance among the employees to get their best talent approaches. These approaches are (i) differentiated approach which is limited to higher potential employees and (ii) inclusive approach that is available to all employees. Hence, the theories of TM have defined that maximizing the talents of employees is the key benefit of sustained competitive advantages and this links the TM to HRM practices in the organization. Nowadays, many multinational companies have adopted TM strategies with small and medium-sized companies. So, based on the TM strategies, Collings and Mellahi [11] defined strategic talent management as defining a theoretical model for strategic talent management. Mathematically, this model is defined as AMO architecture, where ‘A’ represents ability, ‘M’ represents motivations, and ‘O’ represents opportunity. The combined effect of ‘A’, ‘M’, and ‘O’ will measure the performance of an employee i.e., P = f (A, M, O). This model architecture has been demonstrated in Fig. 3. The above discussed theoretical aspects of the talent management system activities have been used for measuring the talents of employees for the success and effectiveness of the organization and the data due to these activities have been experimented with through the data analytic tasks which are discussed in the following section.
Commitment
Organization performance
N. Nasar et al.
Pivotal positions
198
Behaviour
Fig. 3 The strategic talent management model proposed by Collings and Mellahi [11]
2.2 Data Analysis for Talent Management System The analysis of data for the talent management system has been performed for measuring the talent of the employees and for this the available dataset has been analyzed for the purpose to get the best effective discriminant features or parameters which play an important role in talent measurement. The key objectives of this analysis are (i) to derive some important and useful relationships between various TM activities and (ii) to better represent and understand these various TM activities in the real world scenario [12]. In the data analysis for the talent management system, computer science plays a very important role through mathematics and statistics to use and modify the various techniques of machine learning and data science algorithms. Data science [13] techniques support for better predictability about the sense of data coming for analysis. These techniques include multidisciplinary research fields from computer science, statistics, mathematics, physiology, and philosophy [14]. Here the mathematics and statistics give mathematical support for predicting the ways for analyzing the captured data whereas the computer science subjects with artificial intelligence and machine learning techniques provide support for training the captured and preprocessed data to predict some models. These models are used as prediction models for solving the problems by the domain experts from various fields such as business data, medical healthcare data, economic data, and financial expert data [15] for their particular branch of studies. In data science, for deriving the prediction models, there are some basic steps: (i) preprocessing the captured data which needs cleaning, integrating, and normalizing processes of data preprocessing task, (ii) feature analysis which involves feature understanding, feature distribution, and feature selection, and lastly (iii) data classification which involves the model prediction and performance-driven for the data analytic tasks. The data cleaning is the process for handling duplicated, incorrectness, missing and garbage values, and incompleteness problems. Data integration is the process of merging data from different sources. Finally, data normalization is the process of transforming the integrated and cleaned data into a proper scale such that values of each feature in the
Prediction Model for Talent Management Analysis Through e-HRM
Data preprocessing
Dataset for TM
199
Data classification
Preprocessed TM data
Analyzed TM features
Cassification models for TMS
Fig. 4 Data-flow diagram of data analytic task for the talent management system
data should lie in the common range. These data preprocessing techniques derive a better prediction model for the talent management system [16]. The Machine learning techniques [17] have been widely used in practice for solving problems in various research areas such as medical data analysis, cognitive computing, business data analysis, financial data analysis, natural language processing, computer vision, and pattern recognition problems. These techniques are employed for various learning tasks to learn the system by using reasoning and inferencing strategies. The machine learning techniques have been divided into four categories: (i) learning for supervised, (ii) learning for unsupervised, (iii) learning for semi-supervised, and (iv) learning for reinforcement learning strategies [18]. The supervised leanings are associated with the labeled data problems e.g., regression 7 classification. The unsupervised learning techniques are with unlabeled data problems e.g., probability density estimation, clustering, and dimensionality reduction. So, the data flow diagram for data analytics tasks in the proposed system has been shown in Fig. 4. • Data preprocessing: The data are collected from various sources that may contain some artifacts such as noises, human errors, mall functions, and missing values [19]. The data preprocessing task is responsible to make the collected data used for further processing. In this work, during data preprocessing, each row is considered as samples while each column is a parameter of a feature. Here a dataset is represented as which may contain integer or real numerical values, categorical, missing, or garbage values. Any dataset containing categorical values is transformed into numerical values and then some missing and garbage values are identified and replaced by the mean value of that feature column and if any feature contains more missing values then that feature (column) is removed from. • Feature analysis: Here, during feature analysis, the unsupervised machine learning technique [20] has been employed where the irrelevant information has been removed and makes it useful for further processing. This task is more responsible to contribute to the solution of (i) the over-fitting problem, (ii) feature selection problem, (iii) improving accuracy, and (iv) reducing training time for classification model building. There exist several methods for feature analysis techniques and among them, dimensionality reduction to extract more discriminant features from the given dataset is most important. In this work, we have obtained the distribution of every feature and have performed the analysis of those features concerning the classification task.
200
N. Nasar et al.
• Data classification: Here, during data classification tasks, we have employed some supervised learning techniques [21] such as classification and regression. The working principle of classification or regression techniques is to classify a sample to its particular class. In the classification or regression, the classifiers have been trained using data samples based on that the prediction models are obtained. These prediction models are used to test unknown samples to obtain the performance of the proposed system. The processing of classification is based on three components: (i) Input, (ii) learning algorithm, and (iii) heuristic function f (Model). The obtained prediction model performs either binary or multi-class methods. In this work, we have employed a multi-class classification technique. Here during classification, the classifiers are trained using learning features with respect to the label feature. During classification we have employed (i) Logistic Regression, (ii) Decision Tree (CART), (iii) k- Nearest Neighbour (KNN), (iv) Support Vector Machine (SVM), and (v) Random Forest (RM). • Performance evaluation: During data classification, the learning features and the target feature y from the respective dataset have been divided randomly with 50% data for training set and the remaining 50% data for testing set. The training sets are used to train the classifier while the testing-sets are used to obtain the performance in terms of accuracy.
3 Experimental Results and Discussions 3.1 Talent Management Analysis Using Dataset 1 The demonstration of the proposed Talent Management Analysis has been performed using two benchmark datasets: (i) Job Classification Dataset (DS 1 ) [22] and (ii) Human Resources Dataset (DS 2 ) [23]. Both these datasets have been downloaded from ‘Kaggle’ website. The description of DS 1 dataset is shown in Table 1. From Table 1 it has been observed that the dataset DS 1 contains 845 samples while each sample has 14 features. The distribution of features in this dataset DS 1 has been Table 1 Description of job classification dataset (DS1) Samples
Features
Actual features
Purpose
845
14
u1 = ID, u2 = JobFamily, u3 = JobFamilyDescription, u4 = JobClass, u5 = JobClassDescription, u6 = PayGrade, u7 = EducationLevel, u8 = Experience, u9 = OrgImpact, u10 = ProblemSolving, u11 = Supervision, u12 = ContactLevel, u13 = FinancialBudget, u14 = PG
TMA1
Preprocessed features u6 = PayGrade, u7 = EducationLevel, u8 = Experience, u9 = OrgImpact, u10 = ProblemSolving, u11 = Supervision, u12 = ContactLevel, u13 = FinancialBudget
Prediction Model for Talent Management Analysis Through e-HRM
201
(b)
(a)
(c) Fig. 5 Feature distribution of selected features in the DS1 dataset
shown in Fig. 5. From this figure it has been observed that u6 = PayGrade, u7 = EducationLevel, u8 = Experience, u9 = OrgImpact, u10 = ProblemSolving, u11 = Supervision, u12 = ContactLevel, and u13 = FinancialBudget are more discriminant features while the distributions of u1 = ID, u2 = JobFamily, u3 = JobFamilyDescription, u4 = JobClass, u5 = JobClassDescription and u14 = PG are consistent with some level and less discriminant. So, after preprocessing these features are dropout from DS 1 and hence the reduced dataset DS1 ∈ R845×8 has been obtained. The dataset DS1 ∈ R845×8 undergoes to data classification tasks to derive the prediction models due to different selection of target (y) and learning features (X) which have been shown in Table 2. In Table 2 each dataset has been divided into 50% training set and 50% testing set respectively and then the machine learning classifiers such as Logistic Regression, CART, KNN, SVM, and Random Forest have been applied. The data classification performance due to DS1 ∈ R845×8 dataset has been shown in Table 3. From Table it has been observed that Exp1 has obtained better performance, and it is due to y = u9 = OrgImpact target feature and X = (u6 , u7 , u9 , u10 , u11 , u12 , u13 ) learning features. Hence, the impact of (y = u9 = OrgImpact) ≥ (y = u10 = ProblemSolving) ≥ (y = u6 = PayGrade) ≥ (y = u11 = Supervision) ≥ (y = u13 = FinancialBudget) ≥ (y = u7 = EducationLevel) (y = u12 = ContactLevel) ≥ (y
202
N. Nasar et al.
Table 2 Different selections of features with training–testing sets using DS1 ∈ R845×10 dataset for TMA1 experimentation Target feature
Learning feature
Dataset
y = u6 = PayGrade
X = (u7, u8, u9, u10, u11, u12, u13)
E x p11
y = u7 = EducationLevel
X = (u6, u8, u9, u10, u11, u12, u13)
E x p21
y = u8 = Experience
X = (u6, u7, u9, u10, u11, u12, u13)
E x p31
y = u9 = OrgImpact
X = (u6, u7, u8, u10, u11, u12, u13)
E x p41
y = u10 = ProblemSolving
X = (u6, u7, u8, u9, u11, u12, u13)
E x p51
y = u11 = Supervision
X = (u6, u7, u8, u9, u10, u12, u13)
E x p61
y = u12 = ContactLevel
X = (u6, u7, u8, u9, u10, u11, u13)
E x p71
y = u13 = FinancialBudget
X = (u6, u7, u8, u9, u10, u11, u12)
E x p81
Table 3 Performance for TMA1 experimentation Classifier
Dataset for experiment E x p11
E x p21
E x p31
E x p41
E x p51
E x p61
E x p71
E x p81
Logistic regression
71.42
50.23
35.89
64.67
57.38
36.19
50.23
57.28
KNN
42.85
42.85
57.29
78.25
78.91
72.17
42.29
50.67
CART
71.43
71.43
71.39
71.83
78.98
50.12
43.28
42.67
SVM
50.34
35.71
35.89
85.45
71.45
57.29
58.98
64.35
Random forest
42.19
71.42
42.76
71.34
71.46
71.29
51.27
57.89
Avg
69.93
54.33
48.65
74.31
71.64
57.41
49.21
54.57
= u8 = Experience). So, these features are important for talent management system (TMS) and based on these parameters the TMS will retain or recruit the talented employees.
3.2 Talent Management Analysis Using Dataset 2 For the second experiment, the Human Resources Dataset (DS 2 ) has been considered to perform another talent management analysis TMA2 . This dataset contains 310 samples (rows) while each sample contains 35 features. The description of DS 2 has been shown in Table 4. During preprocessing, (v1 = EmployeeName), (v2 = EmpID), (v13 = Position), (v15 = Zip, v16 = DOB), (v22 = DateofHire), (v23 = DateofTermination), (v27 = ManagerName), (v29 = RecruitmentSource), (v34 = LastPerformanceReviewDate), (v35 = DaysLateLast30) features have been dropout as these features are less effective for TMA2 experimentation. Finally, the dataset DS 2 has been reduced to DS2 ∈ R310×24 which is demonstrated in Table 4. The distribution of these preprocessed features of DS2 ∈ R310×24 dataset has been shown in Fig. 6. From these figures it has been observed that each feature has its own distinctiveness
Prediction Model for Talent Management Analysis Through e-HRM
203
Table 4 Description of human resources dataset (DS2) Samples Features Actual features
Purpose
310
35
v1 = EmployeeName, v2 = EmpID, v3 = MarriedID, v4 = TMA2 MaritalStatusID, v5 = GenderID, v6 = EmpStatusID, v7 = DeptID, v8 = PerfScoreID, v9 = FromDiversityJobFairID, v10 = PayRate, v11 = Termd, v12 = PositionID, v13 = Position, v14 = State, v15 = Zip, v16 = DOB, v17 = Sex, v18 = MaritalDesc, v19 = CitizenDesc, v20 = HispanicLatino, v21 = RaceDesc, v22 = DateofHire, v23 = DateofTermination, v24 = TermReason, v25 = EmploymentStatus, v26 = Department, v27 = ManagerName, v28 = ManagerID, v29 = RecruitmentSource, v30 = PerformanceScore v31 = EngagementSurvey, v32 = EmpSatisfaction, v33 = SpecialProjectsCount, v34 = LastPerformanceReviewDate, v35 = DaysLateLast30
310
24
v3 = MarriedID, v4 = MaritalStatusID, v5 = GenderID, v6 = EmpStatusID, v7 = DeptID, v8 = PerfScoreID, v9 = FromDiversityJobFairID, v10 = PayRate, v11 = Termd, v12 = PositionID, v14 = State, v17 = Sex, v18 = MaritalDesc, v19 = CitizenDesc, v20 = HispanicLatino, v21 = RaceDesc, v24 = TermReason, v25 = EmploymentStatus, v26 = Department, v28 = ManagerID, v30 = PerformanceScore, v31 = EngagementSurvey, v32 = EmpSatisfaction, v33 = SpecialProjectsCount
Preprocessed features
v1
v2
v3
v4
v5
v6
(a) v13
v14
-
-
(d)
180
v16
80
v10
(c)
(b) v15
-
v9
v8 v7
0
-
(f)
(e)
Fig. 6 The feature distribution of selected features in the DS 2 dataset
v11
v12
204 Table 5 Different selection of features with training–testing sets using DS2 ∈ R310×24 dataset for TMA2 experimentation
N. Nasar et al. Target feature
Learning feature
Dataset
v3 = MarriedID
[V − v3]
E x p12
v4 = MaritalStatusID
[V − v4]
E x p22
v5 = GenderID
[V − v5]
E x p32
v6 = EmpStatusID
[V − v6]
E x p42
v7 = DeptID,
[V − v7]
E x p52
v8 = PerfScoreID
[V − v8]
E x p62
v9 = FromDiversityJobFairID
[V − v9]
E x p72
v10 = PayRate
[V − v10]
E x p82
v11 = Termd
[V − v11]
E x p92
v12 = PositionID
[V − v12]
2 E x p10
v14 = State
[V − v10]
2 E x p11
v17 = Sex
[V − v10]
2 E x p12
v18 = MaritalDesc
[V − v18]
2 E x p13
v19 = CitizenDesc
[V − v19]
2 E x p14
v20 = HispanicLatino
[V − v20]
2 E x p15
v21 = RaceDesc
[V − v21]
2 E x p16
v24 = TermReason
[V − v24]
2 E x p18
v25 = EmploymentStatus
[V − v25]
2 E x p19
v26 = Department
[V − v26]
2 E x p20
v28 = ManagerID
[V − v28]
E x p11
v30 = PerformanceScore
[V − v30]
2 E x p21
v31 = EngagementSurvey
[V − v31]
2 E x p22
v32 = EmpSatisfaction
[V − v32]
2 E x p23
v33 = SpecialProjectsCount
[V − v33]
2 E x p24
while finding these effectiveness over TMA2 , we have performed various experi2 due to these features. The different selection of ments such as E x p12 , . . . , E x p24 2 features with training and testing sets for E x p12 , . . . , E x p24 experiments has been shown in Table 5. Here the data classification tasks have been performed to derive the prediction models due to different selection of target (y) and learning features 2 experiments. (X) for E x p12 , . . . , E x p24 In Table 5 each dataset has been divided into 50% training set and 50% testing set respectively and then the machine learning classifiers such as Logistic Regression, CART, KNN, SVM, and Random Forest have been applied. The data classification performance due to DS2 ∈ R310×24 dataset has been shown in Table 6. From this table it has been observed that due to (v11 = Termd) target feature the TMA2 has achieved better performance. Hence, the impact of (v11 = Termd) ≥ (v7 = DeptID) ≥ (v5 = GenderID) ≥ (v21 = RaceDesc) ≥ (v30 = PerformanceScore)
24.19
32.25
30.82
31.96
CART
SVM
Random forest
Avg
83.87
37.09
Logistic regression
35.48
2 E x p14
2 E x p13
KNN
92.85
89.42
Avg
99.89
82.90
83.87
83.87
82.25
80.64
98.38
99.67
99.34
99.56
99.45
CART
72.58
93.54
Random forest
64.51
E x p22
SVM
83.87
KNN
E x p12
Dataset
Logistic regression
Classifier
85.48
85.48
83.87
83.87
87.09
87.09
2 E x p15
97.16
99.56
99.89
99.72
87.09
99.56
E x p32
Table 6 Performance for TMA2 experimentation
97.09
99.77
99.82
99.45
87.09
99.34
2 E x p16
95.77
96.77
96.84
99.77
91.93
93.54
E x p42
92.74
98.38
99.78
99.45
74.19
91.93
2 E x p18
98.75
99.73
99.87
99.56
95.16
99.45
E x p52
88.70
91.93
91.93
75.80
91.93
91.93
2 E x p19
94.02
95.16
99.86
99.32
82.25
93.54
E x p62
88.38
91.93
90.32
75.80
91.93
91.93
2 E x p20
91.93
90.32
91.93
91.93
91.93
93.54
E x p72
52.89
51.61
59.67
43.54
51.61
58.06
E x p11
17.56
20.89
23.89
21.45
11.25
10.34
E x p82
96.96
99.89
98.38
99.45
91.93
95.16
2 E x p21
99.41
99.72
99.89
99.67
98.38
99.43
E x p92
96.04
95.16
98.38
99.56
90.32
96.77
2 E x p22
62.57
70.96
66.12
74.19
45.16
56.45
2 E x p10
93.99
95.16
99.67
99.34
82.25
93.54
2 E x p23
22.55
24.78
25.78
24.67
21.78
15.78
2 E x p11
17.71
23.67
20.78
21.67
10.56
11.89
2 E x p24
23.91
29.90
23.66
21.98
22.67
21.35
2 E x p12
Prediction Model for Talent Management Analysis Through e-HRM 205
206
N. Nasar et al.
≥ (v31 = EngagementSurvey) ≥ (v6 = EmpStatusID) ≥ (v8 = PerfScoreID) ≥ (v32 = EmpSatisfaction) ≥ (v4 = MaritalStatusID) and considering these features for target features will be beneficial for the talent management analysis. It has also been observed that (v10 = PayRate), (v14 = State), (v17 = Sex), (v18 = MaritalDesc), (v28 = ManagerID), and (v33 = SpecialProjectsCount) may not been considered as target features as their performance are worst for talent management system. Even these features may not improve the performance if they are considered to be learning features. The remaining other features may be used as learning features for talent management system TMA2 experimentation.
4 Conclusions A novel talent management analysis has been performed in this paper. This talent management analysis is very important for the organization to preserve the employees by their retention, motivation, and performance measurement in the organization. The selection of the right candidate as a talent subject is the key role of the talent management system, and this is supported by the combined effect of e-HRM and talent identification technology. Here the implementation of the proposed system has been divided into two parts: (i) the activities and their challenges for talent management systems have been discussed in the first part and (ii) the data analysis for talent management system has been shown experimentally through available datasets in the second part. During data analytics, talent management experimentation has been performed into three sub-components such as (i) data preprocessing, (ii) feature analysis, and (iii) data classification. During data preprocessing, the employed datasets have been preprocessed while during feature analysis, the distribution of feature sets in the dataset has been analyzed with proper justification and their contributions for talent management systems have been shown through data classification tasks. During classification, various combinations of the target feature and the learning features have been selected and various classifiers such as Logistic Regression, kNearest Neighbor (KNN), CART, Support Vector Machine (SVM), and Random Forest have been employed to obtain the discrimination and distinctive nature of the selected features. For experimental purposes, two datasets available on the Kaggle website are employed and their experimentation and the performance have been elaborated and discussed excellently.
References 1. Freidberg M, Kao T (2008) The state of talent management: today’s challenges, tomorrow’s opportunities’. Hewitt’s Hum Capital, 1–37 2. Thunnissen M (2016) Talent management: for what, how and how well? An empirical exploration of talent management in practice. Empl Relat 38(1):57–72 3. Dimitrov K (2018) Talent management-an etymological study. arXiv:1810.02615
Prediction Model for Talent Management Analysis Through e-HRM
207
4. AlKerdawy MMA et al (2016) The relationship between human resource management ambidexterity and talent management: the moderating role of electronic human resource management. Int Bus Res 9(6):80–94 5. Voermans M, Van Veldhoven M (2007) Attitude towards e-HRM: an empirical study at philips. Pers Rev 36(6):887–902 6. Stone DL, Lukaszewski KM (2009) An expanded model of the factors affecting the acceptance and effectiveness of electronic human resource management systems. Hum Resour Manage Rev 19(2):134–143 7. Stone DL, Stone-Romero EF, Lukaszewski K (2006) Factors affecting the acceptance and effectiveness of electronic human resource systems. Hum Resour Manage Rev 16(2):229–244 8. Wiblen S (2016) Framing the usefulness of e-HRM in talent management: a case study of talent identification in a professional services firm. Can J Adm Sci/Revue Canadienne des Sciences de l’Administration 33(2):95–107 9. Horváthová P (2011) The application of talent management at human resource management in organization. In: 3rd international conference on information and financial engineering, IPEDR, vol 12, pp 50–54 10. Lewis RE, Heckman RJ (2006) Talent management: a critical review. Hum Resour Manage Rev 16(2):139–154 11. Collings DG, Mellahi K (2009) Strategic talent management: a review and research agenda. Hum Resour Manage Rev 19(4):304–313 12. Tarique I, Schuler RS (2010) Global talent management: literature review, integrative framework, and suggestions for further research. J World Bus 45(2):122–133 13. Dhar V (2013) Data science and prediction. Commun ACM 56(12):64–73 14. Ramsay JO (2004) Functional data analysis. Encycl Stat Sci 4 15. Han J, Pei J, Kamber M (2011) Data mining: concepts and techniques. Elsevier 16. Kotsiantis SB, Kanellopoulos D, Pintelas PE (2006) Data preprocessing for supervised leaning. Int J Comput Sci 1(2):111–117 17. Shalev-Shwartz S, Ben-David S (2014) Understanding machine learning: from theory to algorithms. Cambridge university press 18. Mohammed M, Khan MB, Bashier EBM (2016) Machine learning: algorithms and applications. CRC Press 19. Famili A, Shen W-M, Weber R, Simoudis E (1997) Data preprocessing and intelligent data analysis. Intell Data Anal 1(1):3–23 20. Cai D, Zhang C, He X (2010) Unsupervised feature selection for multi-cluster data. In: Proceedings of the 16th ACM SIGKDD international conference on knowledge discovery and data mining, pp 333–342 21. Hal Daumé III. A course in machine learning. Publisher, ciml. info, 5:69, 2012. 22. Job classification dataset. https://www.kaggle.com/HRAnalyticRepository/job-classificationdataset, 2017 23. Job classification dataset. https://www.kaggle.com/rhuebner/human-resources-data-set, 2019
Study of Biomedical Waste Management Performance Indicators in Indian States Anurag Deepak, Astha Sharma, Dinesh Kumar, and Varun Sharma
Abstract Due to exponential population growth, biomedical waste in developing countries like India is increasing at an alarming rate affecting human health and the environment. Biomedical waste management (BMWM) consisting of segregation, collection, transportation, treatment, and disposal processes, which acts as an essential exercise to tackle adverse effects on humans and the environment. Harmful dioxins emitted during these processes play a considerable role in polluting the environment and creating health hazards. This paper discusses the cause and effect of biomedical waste using BMWM indicators. The computation of indicators includes different geographical regions in India. The studied BMWM indicators have two dimensions: (i) Environment (Collection efficiency, waste diversion rate, uncollected waste, and GHGs emissions) and (ii) Social (population density, number of workshops). The critical indicators determined using Principal Component Analysis (PCA) are mentioned as population density (PD), collection efficiency (CE), workshops held in a year (WS), and GHGs emissions (GHE). The analysis represents that the states can develop a cleaner environment and reduce health hazards by increasing waste collection, public awareness, and reducing emissions. In this regard, the crucial states of concern include Rajasthan, Uttarakhand (minimum CE and WS), and Maharashtra (maximum GHE). Performance indicators may help frame the guidelines for the state pollution boards to work in an integrated manner. Keywords Waste management · Biomedical waste · PCA · Indicators · Healthcare waste
Abbreviations BMWM GHE
Biomedical medical waste management Greenhouse emissions
A. Deepak · A. Sharma · D. Kumar (B) · V. Sharma Department of Mechanical and Industrial Engineering, IIT Roorkee, Roorkee, Uttarakhand 247667, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_17
209
210
A. Deepak et al.
HCFs CBWTFs BMW WHO IPCC CPCB TQM HCWM HCWGR ISM MICMAC AHP ANP MCDM
Healthcare facilities Common biomedical waste treatment facilities Biomedical waste World health organization Intergovernmental panel on climate change Central pollution control board Total quantity management Healthcare waste management Healthcare waste generation rate Interpretive structural modeling Matrice d’Impacts Croisés Multiplication Appliquée á un Classement Analytic hierarchy process Analytic network process Multi-criteria decision-making
Nomenclature WGR CO2 PD dm CF FCF OF MT
Waste generation rate in (kg/day/bed) Carbon dioxide in (MT/year) Population density in (people per square kilometres) Dry matter content in the waste incinerated (fraction) Total carbon content (fraction) Fraction of fossil carbon in the total carbon (fraction) Oxidation factor (fraction) Metric tonne
1 Introduction The growing population has made healthcare India’s most important sector in terms of employment and revenue. This sector consists of hospitals, medical devices, clinical trials, health insurance, etc. [14]. It provides health services through clinics, nursing homes, hospital facilities, which generate waste comprising hazardous and non-hazardous nature. The term healthcare waste or biomedical waste was defined by Chartier et al. [9] as “all the waste generated within healthcare facilities, research centers and laboratories related to medical procedures.“ Waste produced has two components: hazardous waste in the range of 10–25%, which pose environmental and health hazards, while remaining are non-hazardous waste comparable to municipal waste. The generated biomedical waste in various Healthcare facilities (HCFs)
Study of Biomedical Waste Management Performance Indicators …
211
imposes risks to human health and the environment if not appropriately managed by emitting different harmful dioxins during multiple stages of management processes. The healthcare waste generation rate across developing countries ranges from 0.181 to 6.03 kg/bed/day [3]. With the increase in healthcare waste, decisions regarding sustainable management are crucial for society’s well-being. To manage biomedical waste with rapid urbanization, India prepared biomedical waste management and handling rules first in 1998 and recently amended in 2019 for effective waste management practices [23]. The noted key Performance Indicators (KPIs) for biomedical waste management include segregation, collection and storage, transportation, treatment, and disposal processes for all kinds of waste produced in healthcare facilities [21]. Numerous tools practice the evaluation of sustainable management in industries. Singh et al. [27] have categorized the sustainable assessment tools into three categories, namely, indicators or indices, product-related assessment, and integrated assessment. Indicators are useful for decision-making and in communicating information, which helps compare the performance and further benchmarking the indicator value. The management system’s dynamic complexity can be summarized with the help of indicators [13]. These are further classified into integrated and nonintegrated. The non-integrated indicators help in comparison purposes, ignoring the environmental parameters, while integrated indicators aggregate different dimensions of the environment. Indicators should be continuously monitored for tracking sustainability trends in the changing environment. Thus, in the present study, healthcare waste management is assessed with integrated indicators signifying various dimensions. Various studies were dependent on the KPIs for improving biomedical waste management practices. Chaerul et al. [8] developed a system dynamics model for interactions among critical factors in hospital waste management. Population and number of beds signify a direct relationship with HCW. Askarian et al. [4] implemented a TQM approach to reduce the volume of infectious waste in Iran. Improved segregation methods, drawing up new guidelines, and increasing awareness reduce the waste by 26%. Ciplak and Barton [10] prepared a system dynamics model by considering population, bed inventory, type of HCFs, incidence rates, and rigorous segregation method as independent variables to determine the future quantities of healthcare waste and treatment plant capacities. Thakur and Anbanandam [28] identified the barriers that hinder HCWM and classified them into clusters depending on their driving and dependence power using ISM & fuzzy MICMAC analysis. Minoglou and Komilis [22] identified the variables affecting HCWGR and prepared a model to predict the waste generation rate. CO2 and life expectancy at birth are positively correlated to WGR, while other variables are critically examined using PCA. Aung et al. [5] prepared an HCWM evaluation framework to evaluate management practices in various hospitals using the MCDM technique. Segregation was weighted highest, followed by awareness and training as per AHP, while segregation system and treatment disposal have the highest impact on alternatives based on ANP. Pires and Martinho [24] proposed a waste hierarchy index in the context of circular economy to measure waste hierarchy for MSW management. The literature suggests that healthcare waste management KPIs are essential for preventing and managing
212
A. Deepak et al.
waste efficiently for a safe environment and decreasing health hazards. In order to measure the performance number of variables impact the management of biomedical waste. To reduce the complexity of the performance assessment, indicators serve as the critical decision tools. The indicators measure the performance of the management system over time and provide best–worst practices. Thus, this research aims to study indicators for BMWM in Indian scenario for effective waste management practices at the meso level and to examine indicators using the multivariate analysis method. The multivariate approach used in the study is PCA, a dimension reduction technique, which reduces the number of variables for the concerned study.
2 Research Background In this section, literature concerning the indicators for different waste management streams is shown for evaluating the performance of management strategies across the countries. The indicators presented involve the management streams of BMW, Municipal solid waste (MSW), Construction and demolition waste (C&D), Electronic waste (E-waste), respectively. For instance, [17] provided critical environmental and socio-economic indicators for the sustainable development of developing countries. Wen et al. [30] reviewed the performance indicators for recycling in Taiwan and observed that collection rate, recycling rate, and recovery rate are crucial for measuring recycling conditions. To estimate the collection rate, survival analysis, and ownership approaches were implemented for electrical and electronic waste. Ramachandran [25] prepared a list of indicators for urban development in India by measuring performance monitoring of water supply, sewerage, solid waste management, and stormwater drainage. Bertanza et al. [7] proposed the indicators for municipal solid waste that affect the collection procedure of the service. The different collection policy is quantified and compared in the various localities of the region. The focus of these indicators is related to operational efficiency and economic accomplishment. Loukil and Rouached [18] shed light on the meso level indicators for solid waste collection in the cities. The study develops the composite index, which helps estimate the various towns’ performance systems’ performance regarding some benchmarked indicators values. Concerning to our waste stream, Maamari et al. [19] presented a list of sustainable development indicators for the HCW treatment sector. The indicators are divided into five aspects as strategic commitment, environmental, social, economic, and safety performance. Barbosa and Mol [6] presented the healthcare waste indicators applicable to a Brazilian healthcare center for continuous monitoring of the management process. The different waste groups take the form of indicators for improving the segregation practices in the healthcare institution. Also, Ferronato et al. [12] examined the HCWM using indicators at the hospital and city levels. Yuan [32] identified the Construction and Demolition waste (C&DW) management indicators from four perspectives, i.e., waste generation, environmental, economic, and social, for the overall effectiveness of waste in a holistic manner. Manfredi and
Study of Biomedical Waste Management Performance Indicators …
213
Goralczyk [20] developed the indicators based on the life cycle impact and assessment approach for 12 different waste streams of waste management. MSW imposes the highest environmental impact on their treatment, while environmental benefits are associated with waste streams of high recycling rates. Zaman and Lehmann [34], on the other hand, developed the zero waste index to measure the waste management system’s performance in a city. The index estimates the virgin material and energy efficiency, GHGs emissions reduction, and water-saving by resources recovered from the different waste compositions. Wilson et al. [31] also presented a set of indicators for sustainable performance of integrated solid waste management at city level. The indicators comprises of physical component, i.e., collection, recycling, disposal and governance component: inclusivity, financial sustainability and proactive policies. Zaman [33] identified zero waste management (ZWM) indicators with the use of online survey for assessing performance of ZWM and 56 potential indicators was selected. Ikhlayel [15] proposed a number of indicators for sustainable development in MSW and WEEE by classifying the level of waste management into traditional, integrated solid waste management and sustainable manner. Bertanza et al. [7] evaluated the descriptive, economic performance indicators for solid waste collection system for improving the collection strategies of towns in Italy. Labour, vehicles and container components are separately considered to justify their roles on whole system. da Silva et al. [26] analyzed and selected the indicators for solid waste management based on sustainability and applicability criteria. Due to a lack of secondary sources, only 11 out of 49 indicators were observed for Brazilian municipal areas. The indicators mentioned above related to biomedical waste management have influence at the micro-level, i.e., hospital level, and limited literature at the meso level, i.e., regional level prohibit in measuring the service provided by regional pollution boards for disposal of biomedical waste. The extensive literature focused on indicators for the complex nature of waste management regarding economic, environmental, social, and business aspects in several streams of waste. The finite number of biomedical waste management indicators at the meso level has motivated this study to develop the indicators for measuring the waste management performance in the Indian scenario. Therefore, the paper aims to study the waste management indicators for biomedical waste management concerning the social and ecological dimensions. For validating the indicators, quantification of indicators is performed for the different geographical states of India. Moreover, to observe the critical indicators, multivariate models, i.e., principal component analysis, are used for the study.
3 Definition of Indicators for BMWM The indicators act as the tools for examining the performance of a particular activity or process over time. They can lead to better decisions by simplifying and making aggregated information available to decision-makers [11]. The studied indicators follow the national guidelines, literature above, and regional documentation concerning waste
214
A. Deepak et al.
management. The indicators undertaken has two components- environmental and social indicators. Environmental indicators include Collection Efficiency (CE), Waste diversion rate (WDR), Uncollected Waste (UW), and GHGs emissions. In contrast, social indicators comprise Population Density (PD), and the number of workshops (WS) held at the HCFs and CBWTFs. Biomedical waste generation is dependent on several factors. The study [29] explained the dependency of waste generation on the type of HCFs and bed occupancy. On the other hand, life expectancy, HDI (Human Development Index), mean year of schooling, and CO2 emissions are suggested [22] as factors responsible for waste generation. Also, the population, as an exogenous variable, is highly dependent on the waste generation rate. The description of the indicators is explained in the following section.
3.1 Collection Efficiency The collection of waste is an essential step in reducing health hazards and environmental risks. The waste segregated at HCFs should be collected within 24 h by central treatment and disposal facilities [9]. Higher the collection of BMWs will lower be the inventory of waste at nearby places. The collection efficiency is examined as the quantity of BMW reaching CBWTFs to the total BMW generated at the HCFs. Collection E f f iciency =
Quantit y o f B M W r eaching C BW T Fs × 100 T otal B M W generated
3.2 Waste Diversion Rate 3R (Reduce, Reuse, Recycling) are the top priority of the waste hierarchy for waste management. Recycling activity helps in diverting the amount of waste to be disposed of in a landfill that imposes a negative impact on the environment. Measurement is done as the total treated BMW disposed of through authorized recyclers in Kg/day to the total quantity of BMW generated. W aste diver sion rate =
T otal amount o f waste diver ted away f r om land f ill T otal quantit y o f B M W generated
Study of Biomedical Waste Management Performance Indicators …
215
3.3 Uncollected Waste As per BMWM guidelines in India, all the HCFs are required to send their BMW to CBWTF but can treat their waste itself only if the central facility is not available within a radius of 75 km. The leading cause of uncollected waste is the authorization of HCF with competent authority and knowledge or awareness about the severity of BMW around the premises. The waste leftover by the non-collection of CBWTFs is an uncollected waste. If waste remained uncollected, it will either be mixed with MSW, making the total waste infectious or disposed of with other streams of waste.
3.4 GHGs Emissions The conventional treatment method of BMW in India is through incineration, which emits harmful gases, including CO2 , NOx, and other dioxins affecting the environment. The leading cause of these harmful emissions is treatment technology used and untreated BMW mixed with MSW disposed of in landfills. The calculation of CO2 emissions is as follows: MT 44 G H G emissions = (B M W i) × dm × C F × FC F × O F × year 12 i
3.5 Population Density The number of HCFs is dependent on the population of the states, which also justifies the number of beds in the particular HCFs. The population density of the state is the population over the land area of that state. The more the PD higher will be waste generation and more GHGs emissions. The population density data follows the census of India [2].
3.6 Number of Workshops The training/workshops are necessary for proper management of BMW by doctors, staff members in the HCFs, and at CBWTFs for treating the waste according
216
A. Deepak et al.
to applied technologies. The total number of workshops in a year includes the workshops by state pollution boards, HCFs, and CBWTFs for health workers. Data Sources For the computation of indicators, the data follows the state annual reports for 2017, published by CPCB in 2019. According to BMWM regulations in India, every HCFs publish its information by the end of the year. The factors data is publicly available and accessed on January 20, 2020. The survey in the state of Uttarakhand allows us to understand the nature of the flow of BMW and problems faced by HCFs and CBWTF. The specialist’s member includes doctors and medical staff. The IPCC national greenhouse gas inventories guidelines 2019 are used to calculate the CO2 emissions. To minimize the uncertainty of the PD indicator, the actual census of 2011 is undertaken. Research Method Pearson and Hotelling first proposed Principal Component Analysis (PCA) as a statistical technique for dimension reduction by retaining as much variation in the data [16]. This technique reduces the dimensionality of data by transforming the inter-related variables into uncorrelated variables. PCA comes under the category of interdependence models of multivariate analysis, which considers all variables under one category. There are two categories of variables in dependence models, i.e., response variables and explanatory variables. This method reduces the data from two perspectives: lowering the dimensions of variables and maintaining the orthogonality of new principal components. The stepwise approach for the principal component analysis is presented in Fig. 1. The first step of the method is the data adequacy test which tests whether there is a correlation among variables or not; if the correlation is not present in the variables, the analysis is inappropriate. In our study, this is tested with the help of Bartlett’s sphericity adequacy test. The second step normalizes the data as the variables are estimated in different units. The variables are comparable with others with the help of this procedure. Now to understand the correlations between variables, in Step 3, computation of covariance matrix is made. Higher the correlations among the variables, more redundant information are embedded in the model. The positive correlation suggests an increase or decrease in variables together, and the same goes with a negative correlation. The eigenvectors and eigenvalues of the covariance matrix in Step 4 are computed, and the number of these represents the number of principal components. The eigenvectors are the directions of the axes in which there is maximum variance, and eigenvalues are the corresponding coefficients representing the variance carried by each principal component. Once the maximum variance axis is established, the principal components are extracted by ranking the highest to lowest eigenvalues. To extract the principal components, various procedures are followed. In our study, principal components are extracted with a scree plot and a loading plot of the variables included in the analysis. Bartlett’s sphericity test checks the significant difference between the identity matrix and the correlation matrix. The null and alternative hypotheses are provided below. The Scree plot signifies the cumulative variability of individual principal components. The principal
Study of Biomedical Waste Management Performance Indicators …
217
Fig. 1 The step-by-step procedure for the implementation of principal component analysis
component with Eigenvalue more exceptional than one or with the highest value is selected for further analysis. The analysis is done using the loading plot of variables to extract the variables in the principal component. For our study, PCA was performed by XLSTAT 2020.1.3 [1]. H0: There is no correlation significantly different from 0 between the variables. H1: At least one of the correlations between the variables is substantially different from 0.
4 Results and Discussion 4.1 Indicators Values for Different States Table 1 shows the computed indicator values for the Indian states. The normalization method transforms the indicator values into comparable quantities. The method used for normalizing values is the proportionate method. The states’ waste generation rate varies significantly, as the number of beds ranges from 0.010 to 0.329 in Uttarakhand and Maharashtra. Delhi has the highest waste generation rate of 0.207 kg/day/bed, and the lowest was in AP as 0.042 kg/day/bed. The population density was lowest in Himachal Pradesh with 0.0086 and 0.798 in Delhi, respectively. For effective management of BMWM, awareness among practitioners is important,
218
A. Deepak et al.
Table 1 Computed indicators values for the states S. no
States
WGR
PD
WS
CE
WDR
UW
GHGs
Beds
1
Delhi
0.207
0.798
0.0026
0.118
0.096
0.0096
0.120
0.056
2
Haryana
0.103
0.040
0.062
0.0812
0.075
0.193
0.051
0.054
3
HP
0.094
0.0086
0.069
0.082
0.082
0.187
0.011
0.015
4
Punjab
0.099
0.038
0.123
0.12
0.121
0
0.062
0.074
5
Rajasthan
0.086
0.014
0.120
0.0715
0.042
0.241
0.104
0.125
6
UK
0.138
0.013
0.0031
0.0727
0.303
0.235
0.007
0.010
7
AP
0.042
0.021
0.209
0.12
0.067
0
0.086
0.121
8
Maharashtra
0.090
0.025
0.069
0.115
0.064
0.024
0.384
0.329
9
MP
0.073
0.016
0.027
0.106
0.0464
0.066
0.101
0.097
10
Telangana
0.064
0.022
0.313
0.111
0.10
0.042
0.069
0.114
and it is maximum in Telangana (0.313) and minimum in Uttarakhand (0.0013) in a year. The collection efficiency is dependent on the waste collected by CBWTFs from all the HCFs, including secondary sources of BMW. It is 0.12% in Punjab and AP and 0.0715% in Rajasthan. Uttarakhand diverts their maximum treated waste away from landfills as recyclables with 0.303%, while Rajasthan is still struggling in recycling rate with only 0.042%. The treatment of BMW through incineration process, toxic emissions pollute the environment. The GHGs emissions in Maharashtra are 0.384 MT/year and 0.0076 MT/year in Uttarakhand, respectively. Wi j N or mali zing values = n i Wi j where Wi j is the respective indicator value for ith indicator and jth state.
4.2 Principal Component Analysis Bartlett’s sphericity test interpretation indicates that the computed p-value (twotailed) is lower than the significance level alpha = 0.05; therefore, one should reject the null hypothesis and accept the alternative hypothesis. If the computed p-value is less than the significance level of 95%, then there is no use in analyzing the critical indicators. The alternative hypothesis states that there is a correlation between the indicators, and therefore principal component analysis is applicable for the study. Scree Plot-Eigen values: Fig. 2 shows the principal components with their cumulative variability and eigenvalues. The eigenvalues’ components are more significant than one are: F1 (2.673) and F2 (1.399). F1 explains 44.55% of the total variability, and F2 shows 23.316%, respectively. Thus, the highest variability component, F1, is being further analyzed using the factor loading values and loading plot.
Study of Biomedical Waste Management Performance Indicators …
219
Scree plot
100
3
2.5
Eigenvalue
2 60 1.5 40 1 20
0.5
0
Cumulative variability (%)
80
0 F1
F2
F3
F4
F5
Principal Components
Fig. 2 Scree plot of principal components
Factor loading of the variables: From above, F1 is a principal essential component. The loading values provide the correlation among the principal component axis and indicators. During this stage, a more considerable length of line and smaller angle w.r.t. F1 axis can find the indicators in the F1 component. Figure 3 displays positive correlated F1 components as collection efficiency, workshops held, population density, and GHGs emissions. The negatively correlated indicators concerning the F1 axis are uncollected waste and recycling rate.
5 Conclusions The paper studied the BMWM performance indicators for measuring the condition in various states of India. The principal component analysis is an advanced multivariate analysis tool for dimension reduction and transforming variables into unrelated principal components. F1 principal component (CE, PD, GHGs, and WS) shows the highest eigenvalue and considered for analysis. The concerned states are Rajasthan, Uttarakhand, and Maharashtra. By increasing the efficiency of CBWTFs and their waste collection path, the lower value of CE in Rajasthan would show improvement. There is a need for separate allocation funds for improving the number of workshops held, significant segregation, and diversion of waste away from landfills which helps in the reduction of GHGs emissions. The studied performance indicators are valuable tools for building effective waste management guidelines in an integrated manner.
220
A. Deepak et al.
Variables (axes F1 and F2: 67.87 %) 1 WS 0.75
F2 (23.32 %)
0.5 0.25 UW 0 GHGs
-0.25
CE
WDR -0.5 -0.75 PD -1
-1
-0.75
-0.5
-0.25
0
0.25
0.5
0.75
1
F1 (44.56 %)
Acve variables Fig. 3 Loading plot for indicators correspond to the F1 component
Future studies can include more sustainability indicators for further improving the waste management system.
References 1. Addinsoft (2020) XLSTAT statistical and data analysis solution. Boston, USA 2. Anon (2011) Census. Retrieved January 12, 2020, https://www.census2011.co.in/density.php. 3. Ansari M et al (2019) Dynamic assessment of economic and environmental performance index and generation, composition, environmental and human health risks of hospital solid waste in developing countries; a state of the art of review. Environ Int 132.https://doi.org/10.1016/j.env int.2019.105073 4. Askarian M, Heidarpoor P, Assadian O (2010) A total quality management approach to healthcare waste management in Namazi Hospital, Iran. Waste Manage 30(11):2321–2326. https:// doi.org/10.1016/j.wasman.2010.06.020 5. Aung TS, Luan S, Xu Q (2019) Application of multi-criteria-decision approach for the analysis of medical waste management systems in Myanmar. J Clean Prod 222:733–745. https://doi. org/10.1016/j.jclepro.2019.03.049
Study of Biomedical Waste Management Performance Indicators …
221
6. Barbosa FCL, Mol MPG (2018) Proposal of indicators for healthcare waste management: case of a Brazilian public institution. Waste Manage Res 36(10):934–41. https://doi.org/10.1177/ 0734242X18777797 7. Bertanza G, Ziliani E, Menoni L (2018) Techno-economic performance indicators of municipal solid waste collection strategies. Waste Manage 74:86–97. https://doi.org/10.1016/j.wasman. 2018.01.009 8. Chaerul M, Tanaka M, Shekdar AV (2008) A system dynamics approach for hospital waste management. Waste Manage 28(2):442–449. https://doi.org/10.1016/j.wasman.2007.01.007 9. Chartier Y et al (2014) Safe management of wastes from health-care activities. Geneva, Switzerland 10. Ciplak N, Barton JR (2012) A system dynamics approach for healthcare waste management: a case study in Istanbul Metropolitan City, Turkey. Waste Manage Res 30(6):576–586. https:// doi.org/10.1177/0734242X12443405 11. DiSano J (2007) Indicators of sustainable development: guidelines and methodologies, 3rd edn. United Nations Publication, New York, Unted Nations 12. Ferronato N, Alarcón GPP, Lizarazu EGG, Torretta V (2020) Assessment of municipal solid waste collection in Bolivia: perspectives for avoiding uncontrolled disposal and boosting waste recycling options. Resour Conserv Recyc (June). https://doi.org/10.1016/j.resconrec. 2020.105234 13. Godfrey L, Todd C (2001) Defining thresholds for freshwater sustainability indicators within the context of South African water resource management. In: 2nd WARFA/waternet symposium: integrated water resources management: theory, practice, cases. Cape Town, South Africa 14. IBEF (2019) India healthcare report, December 2019, vol 2019. New Delhi 15. Ikhlayel M (2018) Indicators for establishing and assessing waste management systems in developing countries: a holistic approach to sustainability and business opportunities. Bus Strat Dev 1(1):31–42. https://doi.org/10.1002/bsd2.7 16. Jolliffe IT (2002). Principal component analysis, 2nd edn. Springer, New York 17. Linster (2001) OECD environmental indicators: towards sustainable development, Paris 18. Loukil F, Rouached L (2020) Waste collection criticality index in African Cities. Waste Manage 103:187–197. https://doi.org/10.1016/j.wasman.2019.12.027 19. Maamari O, Maurice J, Brandam C, Lteif R, Salameh D (2017) Indicators of sustainable development for health care waste treatment industry. Environ Res Eng Manag 73(2):7–20. https://doi.org/10.5755/j01.erem.73.2.13399 20. Manfredi S, Goralczyk M (2013) Life cycle indicators for monitoring the environmental performance of European waste management. Resour Conserv Recycl 81:8–16. https://doi.org/10. 1016/j.resconrec.2013.09.004 21. Manyele SV, Lyasenga TJ (2010) Factors affecting medical waste management in low- level health facilities in Tanzania. Afr J Environ Sci Technol 4(5):304–318 22. Minoglou M, Komilis D (2018) Describing health care waste generation rates using regression modeling and principal component analysis. Waste Manage 78:811–818. https://doi.org/10. 1016/j.wasman.2018.06.053 23. Patil AD, Shekdar AV (2001) Health-care waste management in India. J Environ Manage 63(2):211–220. https://doi.org/10.1006/jema.2001.0453 24. Pires A, Martinho G (2019) Waste hierarchy index for circular economy in waste management. Waste Manage 95:298–305. https://doi.org/10.1016/j.wasman.2019.06.014 25. Ramachandran M (2010) Handbook of service level benchmarking, New Delhi 26. da Silva L, Prietto PDM, Korf EP (2019) Sustainability indicators for urban solid waste management in large and medium-sized worldwide cities. J Clean Prod 237:117802.https://doi.org/10. 1016/j.jclepro.2019.117802 27. Singh RK, Murty HR, Gupta SK, Dikshit AK (2012) An overview of sustainability assessment methodologies. Ecol Ind 15(1):281–299. https://doi.org/10.1016/j.ecolind.2011.01.007 28. Thakur V, Anbanandam R (2016) Healthcare waste management: an interpretive structural modeling approach. Int J Health Care Qual Assur 29(5):559–581. https://doi.org/10.1108/IJH CQA-02-2016-0010
222
A. Deepak et al.
29. Thakur V, Ramesh A (2018) Analyzing composition and generation rates of biomedical waste in selected hospitals of Uttarakhand, India. J Mater Cycles Waste Manage 20(2):877–890. https://doi.org/10.1007/s10163-017-0648-7 30. Wen L, Lin C, Lee S (2009) Review of recycling performance indicators: a study on collection rate in Taiwan. Waste Manage 29(8):2248–2256. https://doi.org/10.1016/j.wasman.2009. 03.022 31. Wilson DC, Rodic L, Cowing MJ, Velis CA, Whiteman AD, Scheinberg A, Vilches R, Masterson D, Stretz J, Oelz B (2014) ‘Wasteaware’ benchmark indicators for integrated sustainable waste management in cities. Waste Manage. https://doi.org/10.1016/j.wasman. 2014.10.006 32. Yuan H (2013) Key indicators for assessing the effectiveness of waste management in construction projects. Ecol Ind 24:476–484. https://doi.org/10.1016/j.ecolind.2012.07.022 33. Zaman AU (2014) Identification of key assessment indicators of the zero waste management systems. Ecol Ind 36:682–693. https://doi.org/10.1016/j.ecolind.2013.09.024 34. Zaman AU, Lehmann S (2013) The zero waste index: a performance measurement tool for waste management systems in a ‘Zero Waste City.’ J Clean Prod 50:123–132. https://doi.org/ 10.1016/j.jclepro.2012.11.041
Determining Manpower Requirement for Material Handling in a Factory Sreenija Kompally and V. Madhusudanan Pillai
Abstract The number of material handlers required within a factory plays a crucial role in the operational efficiency. Many times, the management is not sure of the number of material handlers required for the movement of the material within the factory. Thus, they end up deploying more or less number of material handlers than necessary. For one of the factories operated by a company XYZ, a case study is conducted. Pick and place (attaching and detaching) elemental tasks of various material handling equipment are listed. Standard data available for different elemental activities are used for estimating standard time. Data is collected pertaining to the time required for a trip and number of trips along with the layout of the plant. Curve fitting is used to determine the time required for a trip, and finally, the number of workers required is calculated. An Excel VBA-based tool having different modules is developed and can be used for determining the number of material handlers in a similar type of plant. Keywords Material flow · Material handling equipment · Curve fitting · Excel VBA
1 Introduction 1.1 Background In an automobile manufacturing environment, the sequence of products to be produced changes every day, and as a result, the flow of the parts also changes. The parts are stored at a centralized location or a decentralized location in the factory based on the replenishment strategy to minimize line side inventory. Due to lack of clear understanding of the flow of parts, and the number of trips required for S. Kompally · V. M. Pillai (B) Department of Mechanical Engineering, National Institute of Technology Calicut, kozhikode, Kerala, India e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_18
223
224
S. Kompally and V. M. Pillai
the material handling activities, the factories deploy a greater number of material handlers than required. The forecast is used to determine the product mix, and the bill of materials describes the parts needed for it. If the time taken for the movement of the parts among the workstations is known, the number of employees required can be determined from the available data.
1.2 Introduction to Material Flow Products can be assembled on-time without any bottlenecks, if the required material is available on-time. The products are usually assembled at point-of-use (POU). Once the material reaches the dock, it will be unloaded near the dock area. Based on the criticality, the part is either stored at any of the intermediary locations or sent to the line side directly. The intermediary location could be a supermarket, where the parts are stored and later utilized when necessary. Sometimes the parts are kept at different locations other than the line side due to space constraint called staging area. The part could also be sent to a kitting location, where many parts which are required for the final assembly are present. As per the Bill of Materials (BOM), these parts are kitted and sent to the assembly area. Figure 1 gives an overview of the movement of material handling devices.
1.3 Objectives The initial part of the study deals with the calculation of material handlers within a factory of a company XYZ and then making the calculation generic so that it is applicable to any factory of the particular company. There were four forklift operators, six hand pallet operators, two tugger operators, and three manual operators in the factory for unloading, material handling, and kitting activities in the undertaken case. Fig. 1 Movement of material handling devices
Determining Manpower Requirement for Material Handling …
225
The factory requires a tool to determine the number of operators for material handling activities as per the operation level of the factory by using plan for every part (PFEP) such that, whenever the forecast gets updated, the parts required for each product mix gets updated and the number of operators required for each material handling equipment changes. The following are the objectives considered for the study: • To analyze the material flow in the factory and to study the elemental activities associated with loading and unloading tasks for different material handling equipment used in the factory. • To develop a procedure to determine the number of material handling operators required in the factory. • To develop a generic tool so that it can be applicable for any factory of the company with minimum inputs from the user, using Excel VBA. Rest of this paper is organized as follows: Section 2 presents a literature review. Section III describes the solution methodology. In Sect. 4, the different equations for estimating various parameters of transportation related to material handling are given. In Sect. 5, the modular approach is discussed for the case considered. In Sect. 6, the generic equations for the necessary calculations are illustrated. The results and conclusions of the case study are discussed in Sect. 7.
2 Literature Review 2.1 Review on Material Flow Related Literature The material handling resource utilization by [7] for a stamping plant using a simulation study and found that creating teams that share duties across press lines tends to smooth out the utilization of drivers and requires fewer total material handling resources to accommodate a typical press schedule. [5] developed a simulated model which gives the benefits with costs associated with material flow with Kitting and the negative aspects of Kitting which include storage at several stock areas. [4] developed the guidelines and design options by categorizing the parts based on consumption volume and grouping the low-volume parts as separate segments. [2] explained how the kitting supply system resulted in the reduction of line side storage space, inventory value, operator walk time, line side replenishment, and overall lead time. [1] have given the application of PFEP for creating a flow of purchased parts by developing a supermarket location for every part which leads to minimize inventory cost.
226
S. Kompally and V. M. Pillai
2.2 Review on Material Handling Equipment Related Literature By developing an arena simulation model, Scholz et al. [6] identified the advantage of using Tugger train, how it reduces the number of trips by transporting large volume of parts at once, and the spatial nearness of the production units with high-frequented and small-scaled supply using autonomous transport entities. The metabolic energy expenditure of Tugger drivers in manufacturing plants is observed by [3] and developed a model using Excel for providing the relaxation times for proper recovery which reduces the physical fatigue.
3 Solution Methodology For estimating the number of material handlers required within a factory, the time required for the loading and unloading tasks while using a material handling equipment and the time required for the movement of the material handling equipment from one place to another are required. The company has a standard tool for estimating time for each elemental activity. The standards are developed by using the MTM approach. The standard tool cannot be integrated with other tools to read the values directly. Hence, the elemental activities associated with loading and unloading tasks of each material handling equipment are listed and the time required for these activities to carry out are taken from the standard tool.
3.1 Pick and Place Time for Forklift The picking and placing activities and their estimated time while loading the container(s) on the forks of the forklift and unloading the container(s) from the forks of the forklift are listed and given in Table 1. Table 1 shows that the time for pick and place the load is 0.744 min.
3.2 Attach and Detach Time for Tugger A tugger is a material handling equipment which can carry many carts simultaneously. Required number of containers or parts are placed on the carts and the carts are attached one with another. The number of trips required can be reduced by using a tugger. The number of carts that can be carried concurrently depends on the aisle width in the factory.
Determining Manpower Requirement for Material Handling … Table 1 The elemental activities and time values for a forklift
Activity
227 Time (min)
While picking the container on the forks The forklift driver will slow down the forklift
0.031
Move the forklift towards the load
0.088
Align and insert the forks into the pallet
0.124
Lift the load to a small height
0.045
Take reverse and accelerate the forklift
0.122
While placing the container off the forks The driver will slow down the forklift
0.031
Drive towards the place where the load should be 0.05 kept Lower the forks
0.045
Place the load
0.05
Remove the forks out of the pallet
0.088
Take reverse and accelerate
0.07
Total
0.744
The elemental activities and their estimated time for the tugger operation while attaching the cart(s) to the hook of the tugger and while detaching the cart(s) from the hook of the tugger are listed and given in Table 2. We are assuming that each cart is six feet length and the carts are disengaged in an order from behind and an average of ten feet distance is considered from the place where the tugger is present to the point-of-use or the supply location. Table 2 shows that the time for a tugger operator to attach and detach one cart is 0.792 min. Similarly, we can estimate that the time taken for a tugger operator to attach and detach two carts are 1.447 min, three carts are 2.131 min, and four carts are 2.867 min.
3.3 Pick and Place Time for Hand Pallet The elemental activities and their estimated time for the hand pallet while picking the container(s) onto the forks of the hand pallet and placing the container(s) off the forks of the hand pallet are listed and given in Table 3. From the above-listed activity time values, the total pick and place task time for hand pallet is 0.462 min.
228
S. Kompally and V. M. Pillai
Table 2 The elemental activities and time values for a tugger Activity
Time (minutes)
While attaching the cart to the tugger The operator should decelerate the tugger
0.031
Come out of the tugger
0.035
Move towards the cart (average distance of 10 feet)
0.084
Move the cart towards the tugger (average distance of 10 feet)
0.096
Align and attach the cart
0.05
Push the button so that wheels of the cart are raised to some height
0.025
Again, move towards the tugger door entrance (for one cart)
0.071
Enter the tugger
0.035
Accelerate the tugger
0.031
While detaching the cart to the tugger The operator should come out
0.035
Move towards the cart which is to be unloaded (for one cart)
0.071
Remove the cart from the tugger
0.03
Keep the cart aside
0.096
Again, move towards the tugger (for one cart)
0.071
Accelerate the tugger
0.031
Total
0.792
Table 3 The elemental activities and time values for a hand pallet
Activity
Time (minutes)
While picking the container on the forks Move the hand pallet towards the load and align
0.05
Insert the forks into the pallet/container
0.042
Push the handle so that the wheels are aligned properly
0.2631
While placing the container on the forks Take reverse and start moving with the load
0.057
Push the button and remove the forks simultaneously and turn
0.05
Total
0.462
4 Transporting Parameter Estimation For determining the time taken for the movement of material handling equipment to move from one place to another place, the various values of speed considered are 4 miles per hour (MPH), 5 MPH, …, 10 MPH. Under these speeds, the time taken for
Determining Manpower Requirement for Material Handling …
229
transportation distances such 1-m, 5-m, 10-m, …, 100-m is determined. From these values, the equations for the transportation time of material handling equipment at a particular speed are developed using curve fitting so that it could be incorporated into the tool to be developed. The equation developed for time at a particular speed is a function of distance. For any distance of transportation under a speed, the time can be calculated from these equations. The equations for the transportation time of the material handling equipment from 4 to 10 MPH are developed and are given below in Eqs. (1)–(7). In Eqs. (1)–(8), Y represents the time taken for the transportation and x represents the distance in meters. The equation of the polynomial curve for a speed of 4 MPH is as follows: Y = −(5E − 05x 2 ) + (0.0236x) + 0.0092
(1)
The equation of the polynomial curve for a speed of 5 MPH is as follows: Y = (7E − 07x2) + (0.0131x) + 0.0072
(2)
The equation of the polynomial curve for a speed of 6 MPH is as follows: Y = −(8E − 09x 4 ) + (1E − 06x 3 ) − (7E − 05x 2 ) + (0.0121x) + 0.0099 (3) The equation of the polynomial curve for a speed of 7 MPH is as follows: Y = −1E − 06x 2 + 0.0091x + 0.0051
(4)
The equation of the polynomial curve for a speed of 8 MPH is as follows: Y = −1E − 06x 2 + 0.0074x + 0.0019
(5)
The equation of the polynomial curve for a speed of 9 MPH is as follows: Y = 1E − 08x 4 − 2E − 06x 3 + 1E − 04x 2 + 0.0056x + 0.0114
(6)
The equation of the polynomial curve for a speed of 10 MPH is as follows: Y = 1E − 06x 2 + 0.0058x − 0.0004
(7)
For manual pushing of trolleys, the company had data (time for manual transportation) for different distances. These data are used for the polynomial curve fitting and the relationship is given in Eq. (8). Y = −(1E − 05x 2 ) + (0.0128x) + 0.051
(8)
230
S. Kompally and V. M. Pillai
5 Modular Approach Modular approach, i.e., separate modules, is prepared for determining the number of operators required for each task. Three different modules are created like Unloading module, Material Handling module, and Kitting module.
5.1 Unloading Module Definition Unloading module determines the number of operators required for getting down the containers from the truck/ trailer to the staging area. Hand pallets for unloading containers and pallets from the truck to the staging area and workers manually moving trolleys from the truck to the staging area are used for unloading at any dock area which is shown in Fig. 2. The time taken and the number of operators required are calculated for an average distance from the trailer center to the unloading area. For one shift having 450 working minutes, the data is collected. It is observed that the medium sized containers are taken three at once which are stacked one above another whereas all other container types are taken one at once. An average distance from the trailer to the unloading area and the number of carts for tugger if used are taken as initial inputs. Assuming an average speed of 6MPH for Forklift and Tugger and 4MPH for Hand Pallet. A matrix is being developed for the collection of input of the data in the tool which includes the quantity of each container type unloaded in the shift and on an average how many such container types are carried at once, which are the inputs which the user should enter for the given container type and material handling equipment.
Fig. 2 Overview of the layout of the factory
Determining Manpower Requirement for Material Handling …
231
6 Material Handling Module Definition In the material handling module, the number of material handlers required for the movement of the material from one location to another location is determined. For material handling, the parts moved from four supply locations such as dock-1, dock-2, supermarket-1, and staging area-1 are considered (see Fig. 2). The plant has three assembly areas (lines) and one of the areas has seven workstations marked as 1, 2, …, 7. The parts from the above said supply locations are moved to five different point-of-use locations such as workstations 1, 2, 3, 4 of the assembly area and sub-assembly area S-A(1) (see Fig. 2). The parts are moved from the supply area to the point-of-use in containers with a certain quantity of parts in it. There are seven different kinds of containers in which the parts are kept and moved. In the seven different types of containers, two types are small containers which are moved by carrying twenty containers at once, one type is medium sized container which is carried twelve at once, one large sized container which is carried one at once are moved per trip and one pallet on which the part is kept and is moved one per trip. Two other container types are carts, one for handling small parts and the other for handling large parts. The carts are moved one at once if manually pushed and three carts are hinged one behind the other if a tugger is used for movement. The distances from one location to another location are taken from AutoCAD layout considering the shortest path and this layout is given in Fig. 2. The areas where there is more congestion, the speed is being reduced and is considered for calculation. The data is collected from a plan for every part (PFEP) and the containers are moved with different material handling equipment like forklift, tugger, hand pallet, and manual pushing.
7 Kitting Module Definition Kitting module determines the number of operators required for delivering the kit carts from the kitting area to the assembly line and the number of operators required for preparing the kit carts. For preparing the kit cart the assumptions considered are as follows: • The movement for getting the empty kit cart is neglected. • One bend and one arise is assumed while picking and placing the part in the kit cart. • The carts required for a trip are picked from one place near the kitting area and are transported/distributed to a different point-of-use locations. Generally, there could be no bend and arise for picking and placing the part or one bend and one arise either while picking the part or placing the part on the kit cart
232
S. Kompally and V. M. Pillai
or two bend and two arise while picking the part and also while placing the part on the kit cart. As it would be difficult to know which parts are at waist level and which are not at the waist level, an assumption that the kitter should bend and arise once is considered. Two scenarios are being considered for kitting. They are as follows: • Picking parts by reading the list • Picking parts by the indication of light. The material handling equipment considered for transporting the kit carts to the point-of-use is Tugger and the manual pushing of the cart. If Tugger, the number of carts carried at once by the tugger and the speed should be mentioned in the tool. If manually, at once only one cart can be carried and the time required for pushing the kit cart to the point-of-use is calculated accordingly. The amount of time required for preparing the kit cart is calculated based on the weight of the part and an average distance of three feet is assumed to be moved for moving to the next part after picking one part within the kitting area for preparing the kit cart. The kit carts prepared at kitting zone-1 (see Fig. 2) are considered for the kitting pilot. The kit carts are assumed to be prepared and kept either towards the north or towards the south of the layout based on the nearness to the station. The distance is also taken accordingly where the point-of-use is nearby either from the north part of the kitting zone in the layout or from the south of the kitting zone on the layout. Fifteen different type kit carts are being prepared per shift in kitting zone-1. Each kit cart is having a difference in number being prepared, which is decided according to the forecast of the product mix manufactured and their sequence, and then moved to different point-of-use locations. All the empty kit carts are weighing 130 kg. Each kit cart carries different kits with varying quantities. It is assumed that each kit has one kg weight. The kit carts are moved using a tugger by carrying three carts at once by hinging one behind another.
8 Calculations 8.1 Unloading Module Calculation The generic calculations for the unloading module are demonstrated below: The number of hand pallet operators required is calculated using Eq. (9). (N )u = (T · T )u ÷ n s where N u = number of unloading operators required (T · T )u = total time taken for unloading the container type n s = number of working minutes per shift
(9)
Determining Manpower Requirement for Material Handling …
233
Total time taken for unloading the container type is calculated using Eq. (10). (T · T )u = (t1)u + (t)2
(10)
where (t1)u = total time taken for traveling from the center of the trailer to the center of the unloading area. (t)2 = total time for loading and unloading activities. (t1)u = t × 2 × n
(11)
where t = time taken for traveling from the center of the trailer to the unloading area. n = number of trips. By using Eqs. (1)–(8), and based on the specified speed for a particular distance, the ‘t’ value is calculated. n = qu ÷ a
(12)
where qu = quantity of container type per shift. a = average number of container types carried at once. (t)2 = (t)3 × n
(13)
where t 3 = loading and unloading time for the material handling equipment which are discussed in Sect. 3. By using Eqs. (9)–(13), the number of operators required for the unloading activity is calculated.
9 Material Handling Module Calculation The calculation for the number of forklift operators required for material handling activity in the factory is demonstrated below. Nm = (T · T )m ÷ n s
(14)
where Nm = number of material handling operators required. (T.T )m = total time taken for the material handling equipment operator. Total time taken for the material handling is calculated using Eq. (15). (T · T )m =
((t1)m + (t)2 )
(15)
234
S. Kompally and V. M. Pillai
where (t1)m = total time taken for traveling for the movement. (t1)m = tm × 2 × n
(16)
where tm = time taken for the material handling equipment to move from the supply area to the point-of-use area at a specified speed. n =q ÷a
(17)
where q = quantity of containers required. q = pr ÷ q p
(18)
where pr = quantity of parts required. q p = quantity of parts in each container. By substituting the values in Eqs. (14)–(18), the number of material handling operators for a particular material handling equipment is calculated.
10 Kitting Module Calculation The generic calculations for the kitting module are demonstrated below. In the kitting module, the number of operators for the preparation of the kit carts and for the movement of the kit carts from the kitting zone to the point-of-use calculations is given below [see Eqs. (19)–(24)]. (N k) p = (T · T )k ÷ n s
(19)
(N k) p = number of operators required for the preparation of the kit carts. Time taken for preparation of the kit carts is calculated using Eq. (20). (T · T )k =
((tk × f ) + (tm ) + (tr ))
(20)
where (T · T )k = time taken for the preparation of the kit carts. tk = total time to pick and place the parts in the kit cart. f = frequency of kit cart required in the shift. tm = time taken for the movement within the kitting area. tr = time taken to read the part names and quantities from the list. tk = t p × q p
(21)
Determining Manpower Requirement for Material Handling …
235
where t p = time taken to pick and place the part (in the undertaken case study, all the parts have weight up to 1 kg and the standard time to pick and place a part having weight up to 1 kg is 0.038 min). q p = quantity of parts to pick and place. tm = td × (n p − 1)
(22)
td is taken from the standard data available based on the weight of the kit cart. The weight gets accumulated as the parts are getting added to the cart (in the case study undertaken, the weight of the kit cart with parts was up to 180 kg and the average distance moved within the kitting area to pick the next part is taken as 3 feet and is kept constant, therefore the time taken from standard data is 0.075 min). n p = number of different parts in the kit cart. tr is taken from the standard data available by considering a constant of ten letters to read (the time taken to read each letter is 0.01 min from the standard data. Therefore, tr = 0.1 min). By substituting the values in Eqs. (19)–(22) the time required for the preparation of kits is known. Nkm = (T · T )km ÷ n s
(23)
where Nkm = number of material handling operators required for the movement of the kit carts. (T · T )km = the total time taken for the movement of the kit carts. For the movement from the kitting area to the point-of-use, the average distance is being calculated by taking the frequency of the carts prepared during the shift. See the equation for average distance calculation: da = (
(d × f )) ÷ f
(24)
where d a = average distance. d = distance from the kitting zone to the point-of-use. For the average distance in Eq. (24) and the speed, the time taken for the movement of the kit carts is calculated using any one of the Eqs. (1)–(8). If the material handling equipment used is a tugger, then the number of carts carried at once should be considered.
236
S. Kompally and V. M. Pillai
Table 4 Results from the tool cx
Number of operators required Material handling task
Unloading task
Total Kitting task
Forklift
3
–
–
3
Hand pallet
1
6
–
7
Tugger
1
–
1
2
Manually
1.18
0.48
1
2.66
11 Results and Future Scope 11.1 Results and Conclusions Using the approach discussed in this paper, an Excel tool is developed. Using this tool, the estimates obtained for the number of operators for various tasks of a shift for the undertaken case study are given in Table 4. An overview of the Excel tool developed is provided in Appendix. The number of manual operators is taken in fraction so that the management could think of using the manual operator who are not completely required for the tasks can be used for other purposes also if necessary. Hence, whenever the PFEP gets updated, the number of material handlers required may increase or decrease based on the number of products manufactured. Using the calculations made for the case study, a generic tool is prepared for the factory such that once the PFEP is updated the number of material handlers required is calculated and it can be applied to similar plants of the company.
11.2 Future scope The results could be verified by developing a simulation model. Another scope could be developing one more module on repacking activity.
Appendix The overview of the generic tool prepared using Excel VBA for the calculation of material handling module and kitting module is illustrated below. Figures 3 and 4 are the screenshots of the dialogue boxes created using Excel VBA userform for the material handling module. The first tab consists of the information which we could get from any transaction which includes, part name /number, supply area, point-of-use area, container
Determining Manpower Requirement for Material Handling …
237
Fig. 3 Basic data
Fig. 4 Material handling
type, part requirement, quantity of parts in each container, and the material handling equipment used for transporting to the point-of-use. Figure 5 gives an overview of
Fig. 5 Part requirement
238
S. Kompally and V. M. Pillai
Fig. 6 Distances
Fig. 7 Unit quantity of containers
the screenshot of the part requirement sheet which need to be updated whenever the data gets updated to get the results. In second tab, the different containers will be populated and the quantity of each container carried at once should be entered. The third tab consists of the distances in which the different supply locations and point-of-use locations will populate and the user should enter the distances along the path, a maximum of five different path distances can be entered and the minimum of all will be retrieved. The distances from one location to another location are taken from AutoCAD layout (See Fig. 2) considering the shortest path for the pilot data. Figure 6 and 7 give an overview of the screenshots of the distances sheet and unit quantity of containers sheet which are one time input sheets by the user. The screenshot of the sample code of the material handling module is as shown below in Fig. 8. For calculating the time required for preparing each kit cart and the time required for transporting the kit carts to the point-of-use four tabs are used in which the first tab consists of the information as follows: Kit cart name, point-of-use where it should be delivered, number of such kit carts moved per shift, distance from the kitting area to the point-of-use and the weight of the empty kit cart. Figure 9 shows the screenshot of the kitting module related sheet in which the data should be entered. The second tab consists of the information which is solely for calculating the time required for preparing the kit cart. The data like the parts which are placed in the kit cart, the weight of the kit cart, and quantity of parts which are required in the kit cart are present. Figure 10 shows the screenshot f the kitting data sheet. The third tab consists of the standard data values necessary for the calculations, of the standard time taken for the movement of the kit cart (up to 180 kg) within the kitting area from a distance of one feet to thirty feet and a table in which the time required for picking and placing the part with varying weights with one bend and one arise are also present. The following Fig. 11 shows the screenshot of the buttons and the details to be entered for getting the kitting results.
Determining Manpower Requirement for Material Handling …
239
Fig. 8 Screenshot of the code for the material handling module
Fig. 9 Kitting basic data
Fig. 10 Kitting data
The screenshot of the sample code of the material handling module is as shown below in Fig. 12.
240
S. Kompally and V. M. Pillai
Fig. 11 The buttons and details to enter
Fig. 12 Screenshot of the code for the kitting module
References 1. Abdelhadi A, Seifoddini H, Almomani M (2012). Part and inventory control analysis using plan of every part concept: A case study at elba, inc., USA. In: International conference on industrial engineering and operations management, Istanbul 2. Gajjar JM, Thakkar HR (2014) Improvement in material feeding system through introducing kitting concept in lean environment of MSME: a review study. Int J Eng Res Technol 2(1):891– 900 3. Ganesan AP (2018) Automated ergonomics assessment of material handling activities 4. Karlsson A, Markus S (2016). Parts feeding of low-volume parts to assembly lines in the automotive industry, Master’s thesis 5. Karlsson E, Thoresson T (2011) A comparative study of the material feeding principles kitting and sequencing at Saab Automobile, Trollhättan: creation of guiding principles of which articles to be supplied with kitting, Master’s thesis
Determining Manpower Requirement for Material Handling …
241
6. Scholz M, Serno M, Franke J, Schuderer P (2017). A hybrid transport concept for the material supply of a modular manufacturing environment. In: 27th international conference on flexible automation and intelligent manufacturing, Italy, pp 1448–1453 7. Williams EJ, Ulgen OM, Bailiff S, Lote R (2006) Material handling resource utilization simulation study for stamping plant. In: Simulation conference IEEE, pp 1940–1945
Data Analytics for NIRF Ranking of Indian Institutions to Check the Consistency and Validity of Ranking Framework Mamta Yadav, Arvind Bhardwaj, and Kapil Kumar Goyal
Abstract The National Institutional Ranking Framework (NIRF) used the weight and sum approach to the combined scores of indicators in the NIRF scores, which were ranked by the institutes. This approach assumes that all indicators contribute independently to the NIRF score in the specified proportions, but this assumption is doubtful as the indicators tend to correlate with each other and some highly so. This indicates a multi-collinearity problem that makes some indicators redundant, and some indicators contribute much less to the NIRF score, which makes them non-contributing. These overlapping and noncontributing indicators make the results invalid. Using data for top 100 engineering institutions of NIRF 2019, these problems are demonstrated, and a solution is proposed. This resulted in a new NIRF Score calculated by only two out of five indicators namely RPC (Research, Professional Practice, and Collaborative Performance) and Perception. Keywords Correlation · Multiple regression · Multicollinearity · Validity
1 Introduction The institutional ranking is a very important academic exercise in the highly competitive world of today, especially for higher educational institutions. Because the rankings are being published based on the performance of the institutions for certain parameters, which measures performance. In case any higher educational institution lacks in the respective parameter, it shall be reflected in their institution ranking, thus the performance of the institution in each parameter is of utmost crucial for achieving a good ranking. If colleges and universities focus on best performance it is a healthy process, and its impact can be seen in the overall development of the country because everything is interlinked. If institutes participate in institutions ranking, it also affects the student’s admission in that particular institution. Whether it is THE rankings or M. Yadav · A. Bhardwaj · K. K. Goyal (B) Department of Industrial and Production Engineering, Dr. B. R. Ambedkar National Institute of Technology, Jalandhar, India e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_19
243
244
M. Yadav et al.
QS rankings, all the surveys are being published in international mass media, which attracts students from different countries in a particular institution according to its ranking [1–3]. National Institutional Ranking Framework (NIRF) There exist a few higher education ranking systems worldwide as earlier discussed, in which, it is hard to find a good place for Indian institutions. Therefore, the Ministry of Human Resource Development has taken initiative to propose indigenous ranking systems with custom framework. The objective of this ranking framework is to rank the higher educational institutions functional in India. The method adopted for the analysis of such institutions has been thoroughly discussed and deliberated before the implementation. The MHRD has proposed a core committee for the identification of different parameters and related mandatory requirements. The performance parameters considered include: Teaching, Learning, and Resources (TLR), Research, Professional Practice and Collaborative Performance (RPC), Graduation Outcomes (GP), Outreach and Inclusivity (OI), and Perception.
2 Research Background In this section, the authors have provided a brief review of available literature related to NIRF ranking of colleges. To name a few, [4] constructed validity maps by using data of top 100 colleges in India and determined how the top colleges are placed with each other and obtained a quantitative estimate by using pierce’s measure to predict if the use of one construct measure to predict another is acceptable or not and in another investigation Prathap [5] found Scientometric and econometric sense out of NIRF 2017 data by performing a comparative end to end research evaluation of leading engineering institutions of India in this he conclude Madras as best engineering college of India. Prathap [6] states that the NIRF exercise is a perfect example of what the Nigerian author Chimamanda Ngozi Adichie called ‘the risk of a single narrative.’ All life is a dynamic accumulation of many stories, but the human propensity is to cram all of this into a simple, one-dimensional narrative. NIRF reduces the immense complexity of higher education into one score. Sheeja et al. [7] investigated whether there is a relationship between academic performance and institutional ranking, based on the National Institutional Ranking Framework (NIRF) of India. Kumar and Tiwari [8] gave an overview and score for the 2016 Indian Rankings (overall and Teaching, Learning, and Resources), which included numerous risks and problems for advancement over the next execution of the NIRF interaction. The objective of this review was to inform people among educational institutions regarding these campaigns. Sivakumaren [9] proposed a new parameter called h-index to determine the h-index of the institutions, departments. Brahma and Verma [10] evaluated the selected university library websites by the 2017 National Institutional Ranking Framework (NIRF). The paper investigated the domain authority, multiple web pages, links, and measures the Web Impact Factor of the library websites of 23 universities. Sivakumaren and
Data Analytics for NIRF Ranking of Indian Institutions …
245
Rajkumar [11] suggested adding additional criteria such as colleges, departments, and contributors’ h-index in the evaluation of rank awarding universities. Aithal and Revathi [12] showed the need for increased institutions in the Indian higher education system, which provide higher education to most of the country’s qualified population, and the performance of the US private university model persuaded the Indian government to allow access to build the country’s private universities.
3 Research Methods Data: NIRF uses five broad indicators with different weightings, which are used to calculate the overall NIRF score, which can take a maximum value of 100. NIRF uses data submitted by universities in a prescribed format, which are desirous of participating in ranking exercise. • • • • •
Teaching, Learning and Resources (Ranking weightage = 0.3) Research Productivity, Impact and IPR (Ranking weightage = 0.3) Graduation outcome (Ranking weightage = 0.2) Outreach and Inclusivity (Ranking weightage = 0.1) Perception (Ranking weightage = 0.10).
The first two indicators are directly relevant to the university, and these are given a total weight of 70%. Data used for this study were taken from NIRF 2019 ranking website (NIRF.org). Only NIRF score and five parameters’ data for 100 top engineering institutions are used for our study. Analysis: Statistical techniques of different types are used to determine the validity and consistency of ranking results. To begin with, a correlation analysis was used to determine the relationship between the indicators and the NIRF score. Multicollinearity testing was followed by multiple regression, and the resulting collinearity diagnostics values were used to check the severity of the problem of redundant indicators. Indicators that are not redundant selected for the model were used in stepwise regression analysis to determine the predictive power of the indicators. Using selected indicators, a new NIRF score is calculated, and the institutes have been ranked. Comparison has been made between the new and the old rankings.
4 Application of Research Methods For ranking purposes, indicator scores are used to calculate the overall score, which is further used for ranking institutes. First, the results of correlation among the indicators are discussed as shown in Table 1. It is observed that the correlations between the overall NIRF score and indicators are highest for the case of Research and Professional Practice (RPC) and least with
246
M. Yadav et al.
Table 1 Correlation among indicators NIRF Score
TLR
RPC
GO
OI
Perception
NIRF Score
1
0.719
0.930
0.695
0.231
0.871
TLR
0.719
1
0.525
0.290
0.318
0.562
RPC
0.930
0.525
1
0.552
0.033
0.766
GO
0.695
0.290
0.552
1
0.240
0.598
OI
0.231
0.318
0.033
0.240
1
0.153
Perception
0.871
0.562
0.766
0.598
0.153
1
Outreach and Inclusivity. Teaching, learning, and Resources (TLR) has a very high correlation with RPC and Perception while less with GO (Graduation Outcome) and Outreach and Inclusivity (OI). RPC has a very high correlation with TLR, GO, and Perception while the lowest correlation with OI. OI has a non-significant correlation with all other indicators as compared to other indicators, which indicate that the diversity of students (regional/foreign), women students, socially, economically, and physically challenged students has nothing to do with the performance and productivity of an institution. TLR and RPC have a high correlation between each other, which shows that research and professional practices depend on teaching, learning, and resources, and a high TLR score can be an outcome of this. High RPC and GO also show a high correlation between each other, which shows that the institutes with good research capabilities can give good placements and career opportunities to students. The lack of higher correlation between TLR and GO shows that placements of students do not only depend upon the TLR but also depend upon so many other factors that include the relationship of placement of institutions with corporate world and also reputation of institutions in terms of perception.. The high correlation among some indicators shows that there can be multicollinearity due to which some indicators become irrelevant. Now to verify whether the problem of multicollinearity exists among the various indicators and if it exists, how severe, the multiple regression is applied. The results of multiple regression analysis of various indicators in different combinations are shown from Tables 2, 3, 4, 5, 6 and 7. It can be seen from Table 2 that the variance of overall NIRF score is explained by all the five indicators with 100% adjusted R square value. From the coefficient, Table 3, it can be seen that unstandardized b weights are same as assigned weights (TLR = 30%, RPC = 30%, GO = 20%, OI = 10%, Perception = 10%) but standardized beta weights are different from the weights assigned by NIRF (TLR = 25.5%, RPC = 54.3%, GO = 19.7%, OI = 5.6%, Perception = 18.6%). Standardized beta weights show that all indicators do not contribute to the overall NIRF score in the proportion as assigned by the NIRF, so the overall calculated NIRF score is misleading. Although all the five indicator values contribute to the overall NIRF score in statistical significance, the results show the problem of multicollinearity, which means that there is overlapping of indicators, and some indicators are not useful for calculating
Data Analytics for NIRF Ranking of Indian Institutions …
247
Table 2 Multiple regression analysis of all five indicators B weight Intercept
Beta weight
t
Sig
0.001
0.157
Tolerance
0.875
Condition index 1.000
TLR
0.300
0.255
4011.289
0.000
0.576
RPC
0.300
0.543
6768.073
0.000
0.361
3.395 7.698
GO
0.200
0.197
3091.505
0.000
0.571
16.021
OI
0.100
0.056
1047.500
0.000
0.803
24.248
Perception
0.100
0.186
2226.078
0.000
0.334
29.540
Tolerance
Condition index
R square = 1.00, Adjusted R square = 1.00
Table 3 Multiple regression of TLR, RPC, GO, and perception B weight
Beta weight
t −4.175
0.000
TLR
0.347
0.295
21.224
0.000
0.627
4.325
RPC
0.353
0.639
41.226
0.000
0.505
14.891
Intercept
−5.307
Sig
1.000
GO
0.247
0.243
17.676
0.000
0.639
22.380
Perception
0.102
0.058
4.690
0.000
0.803
25.289
Tolerance
Condition index
R square = 0.988, Adjusted R square = 0.988 Table 4 Multiple regression analysis of TLR, RPC, and GO B weight
Beta weight
Sig
−2.038
0.044
TLR
0.375
0.319
22.342
0.000
0.724
4.085
RPC
0.341
0.618
37.703
0.000
0.550
13.351
GO
0.266
0.262
17.960
0.000
0.696
21.885
Intercept
−2.531
t
1.000
R Square = 0.986, Adjusted R Square = 0.985 Table 5 Multiple regression analysis of RPC, GO, and perception B weight Intercept
Beta weight
22.633
t
Sig
13.571
0.000
Tolerance
Condition index 1.000
RPC
0.328
0.593
15.921
0.000
0.399
GO
0.187
0.185
6.186
0.000
0.622
6.304
Perception
0.165
0.307
7.913
0.000
0.369
17.094
R square = 0.947, Adjusted R square = 0.945
3.133
248
M. Yadav et al.
Table 6 Multiple regression analysis of RPC and perception B weight Intercept
Beta weight
t
13.817
Sig 4.653
Tolerance
0.000
Condition index 1.000
RPC
0.510
0.923
29.424
0.000
0.999
3.385
Perception
0.356
0.200
6.389
0.000
0.999
18.714
Tolerance
Condition index
R Square = 0.905, Adjusted R Square = 0.903
Table 7 Multiple regression analysis of RPC and GO B weight Intercept
Beta weight
t
18.393
Sig 9.107
0.000
RPC
0.434
0.785
21.762
0.000
0.696
3.644
1.000
GO
0.266
0.262
7.251
0.000
0.696
14.338
R square = 0.912, Adjusted R square = 0.910
the overall NIRF score. If this problem is more severe then it can make the overall NIRF score invalid. The problem of multicollinearity is shown in the Condition Index Column of Tables 4, 5, and 6, which shows that there are values greater than 15, which infers that there exists a problem of multicollinearity. Therefore, all indicators are not needed for calculating the overall score. Therefore, some indicators need to be retained, and some need to be removed, which creates another problem as to which indicators need to be retained and which need to be removed. Table 1 shows that the problem of multicollinearity is found mainly among the four indicators, namely, TLR, RPC, GO, and Perception. It can also be analyzed from Table 1, which indicators are retained and which are removed, which are discussed below. 1.
2. 3.
TLR and RPC have high correlation between themselves as well as with overall NIRF score. RPC has high correlation with overall NIRF score compared to TLR so RPC must be retained. GO and Perception have a high correlation with each other while with overall NIRF score, the latter one has high correlation, therefore, it must be retained. OI has low correlation with other indicators and with overall NIRF score, so it can be easily removed.
After removing the indicators in order to form a model free from multicollinearity, it is observed that only two indicators are retained. But this model cannot be used for further analysis due to the following reasons: 1. 2.
Two indicator model cannot have a very high predictive power. The predictive power value of the perception indicator is low as it is calculated by using surveys.
Data Analytics for NIRF Ranking of Indian Institutions …
249
Therefore, in order to counter these problems, a three-indicator model is selected, which also includes GO, as it does not create the problem of multicollinearity as can be seen from the collinearity diagnostics, Table 7. After selecting these three indicators, a stepwise Multi Regression is run and analyzed, which can be seen in Tables 8, 9, 10, and 11. This method reveals that RPC is the powerful predictor while Perception and GO are weak predictors for calculating the overall NIRF Score. It removes the problem of multicollinearity by selecting RPC and GO as the predictors for calculating the overall NIRF Score, which can be seen in the collinearity diagnostics Table 8 Stepwise multiple regression analysis of RPC, GO, and perception Step 1
Step 2
Step 3
Adjusted R square
0.863
0.910
0.932
R square change
0.864
0.048
0.022
RPC
0.930
0.785
0.807
GO
–
0.262
0.213
Perception
–
–
0.153
Beta weight
Table 9 Collinearity diagnostics of RPC, GO, and perception Model 1 2
3
Dimension
Eigen value
Condition index
Variance proportions (Constant)
RPC
Perception
GO
1
1.818
1.000
0.09
0.09
2
0.182
3.160
0.91
0.91
1
2.628
1.000
0.04
0.02
0.02
2
0.285
3.035
0.73
0.01
0.23
3
0.087
5.510
0.23
0.97
0.74
1
3.538
1.000
0.00
0.01
0.01
0.00
2
0.360
3.133
0.02
0.04
0.21
0.01
3
0.089
6.304
0.00
0.94
0.67
0.00
4
0.012
17.094
0.97
0.01
0.12
0.99
Table 10 Stepwise multiple regression analysis of RPC and GO Step 1
Step 2
Tolerance
Adjusted R square
0.863
0.910
–
R square change
0.864
0.048
–
RPC
0.930
0.785
0.696
GO
–
0.262
0.696
Beta weight
250
M. Yadav et al.
Table 11 Collinearity diagnostics of RPC and GO Model 1 2
Dimension
Eigen value
Condition index
Variance proportions (Constant)
RPC 0.09
GO
1
1.818
1.000
0.09
2
0.182
3.160
0.91
0.91
1
2.777
1.000
0.00
0.03
2
0.209
3.644
0.03
0.75
0.01
3
0.014
14.338
0.96
0.23
0.99
0.00
Column Table 11. RPC and GO as good indicators for calculating the overall NIRF score without the problem of multicollinearity and also explain that 91% variation of overall NIRF score is a good prediction.
5 Result Discussion Table 12 shows that institutes with higher RPC score tend to gain in position whereas institutes with higher NIRF Score on the basis of other four indicators tend to lose thereby getting lower ranking. This shows that RPC has become critical in the new NIRF Score (New Ranking). The top 10 institutes have small positive or negative changes in positions and those beyond have a much wider range of gain or loss. Table 12 Institute-wise new rankings Institute name IIT Madras
New rank 1
NIRF rank 1
Gain 0
NIRF score
Predicted NIRF score
89.05
83.96811729
IIT Delhi
2
2
0
85.36
83.03636287
IIT Kharagpur
3
4
−1
84.4
82.25223702
IIT Roorkee
4
6
−2
79.41
80.96360503
IIT Bombay
5
3
2
77.57
80.45908761
IIT Kanpur
6
5
1
74.57
78.9312561
IIT Guwahati
7
7
0
70.87
69.69202568
Jadavpur University
8
14
−6
63.92
68.35014024
Institute of Chemical Technology
9
12
−3
63.12
66.07984511
NIT Rourkela
10
16
−6
61.62
65.25381391
IIT(ISM) Dhanbad
11
15
−4
59.8
63.09439964
IIT (BHU) Varanasi
12
11
1
59.73
62.29757662
Anna University
13
9
4
59.35
61.01697834 (continued)
Data Analytics for NIRF Ranking of Indian Institutions …
251
Table 12 (continued) Institute name
New rank
NIRF rank
Gain 6
NIRF score
Predicted NIRF score
59.09
60.3361931
IIT Hyderabad
14
8
NIT Tiruchirappalli
15
10
5
58.18
59.99353391
Jamia Millia Islamia
16
27
−11
57.75
59.68726053
NIT Karnataka
17
21
−4
56.8
59.56426518
IIT Indore
18
13
5
56.46
59.43974308
Vellore Institute of Technology
19
18
1
56.35
58.6558137
Thapar Institute of Engineering and Technology
20
23
−3
56
56.16434552
International Institute of Information Technology Hyderabad
21
39
−18
55.25
55.68452623
NIT Calicut
22
28
−6
54.86
55.54550055
Birla Institute of technology and science
23
25
−2
54.73
54.87866653
NIT Warangal
24
26
−2
54.32
54.24925787
VNIT Nagpur
25
31
−6
53.82
53.20541688
IIT Mandi
26
20
6
53.21
52.71178369
IIT Patna
27
22
5
53.13
52.63427672
DTU
28
34
−6
52.69
52.60680881
IIT Bhubaneswar
29
17
12
52.37
52.15118389
IIEST
30
19
11
52.12
52.09889281
NIT Durgapur
31
46
−15
51.27
50.53252471
Indraprastha Institute of Information Technology
32
55
−23
50.39
49.52500963
Sanmugha Arts Science Technology & Research Academy
33
38
−5
50.1
49.40040433
SRM Institute of science and technology
34
36
−2
49.97
49.34279544
Panjab University
35
54
−19
49.03
49.27074311
MNNIT Allahabad
36
42
−6
48.95
48.91926028
IIT Ropar
37
29
8
48.94
48.66339485
Siksha O Anusandhan
38
32
6
48.83
48.65417948
MNIT (Jaipur)
39
53
−14
48.73
48.40581846
MANIT (Bhopal)
40
62
−22
47.74
47.89755111
NIT Kurukshetra
41
41
0
47.58
47.89495498
Sri Sivasubramaniya Nadar College of Engineering
42
37
5
47.49
47.84791579
(continued)
252
M. Yadav et al.
Table 12 (continued) Institute name
NIRF score
Predicted NIRF score
College of Engineering Pune
New rank 43
NIRF rank 49
Gain −6
47.31
47.28291319
NIT Silchar
44
51
−7
47.28
47.23541087
Birla Institute of technology and science
45
33
12
47.13
47.17644741
Sathyabama Institute of Science & Technology
46
47
−1
46.47
46.9658383
IIIT Allahabad
47
82
−35
46.07
46.33732659
IIITDM Jabalpur
48
75
−27
45.94
46.09996331
Aligarh Muslim University
49
40
9
45.89
45.92741009
SVNIT (Surat)
50
58
−8
45.75
45.91866388
IISST (Indian Institute of Space Science and Technology)
51
30
21
45.61
44.70138374
IIT Gandhinagar
52
24
28
45.38
44.48808352
NIT Hamirpur
53
60
−7
45.2
43.93772662
Amity University
54
35
19
43.59
43.69669759
Kalinga Institute of Industrial Technology
55
48
7
43.49
43.150743
Jawaharlal Nehru Technological University
56
45
11
43.38
42.87117113
College of Engineering Vishakhapatnam
57
59
−2
43.29
42.26357927
Koneru Lakshamaiah Education Foundry University (K L College of Engineering)
58
52
6
41.88
42.19696929
IIT Jodhpur
59
50
9
41.72
42.0844489
Atal Bihari Vajpayee Indian Institute of Information Technology and Management Gwalior
60
81
−21
41.48
41.70515431
Dhirubhai Ambani Institute of Information and Communication Technology
61
92
−31
41.26
41.64989796
International Institute of Information Technology Bangalore
62
65
−3
40.98
41.5141463
(continued)
Data Analytics for NIRF Ranking of Indian Institutions …
253
Table 12 (continued) Institute name
NIRF score
Predicted NIRF score
National Institute of Industrial Engineering Mumbai
New rank 63
NIRF rank 66
Gain −3
40.88
41.08407535
PEC University of Technology
64
78
−14
40.54
40.90901472
Guru Govind Singh Indraprastha University New Delhi
65
73
−8
40.51
40.79813459
Thiagarajar College of Engineering
66
56
10
40.48
40.2436839
R.V. College of Engineering
67
63
4
40.32
39.65065502
University College of Engineering Hyderabad
68
83
−15
40.1
39.59723467
Jaypee Institute of Information Technology Noida
69
80
−11
39.67
39.5110474
NIT Raipur
70
74
−4
39.53
39.35006207
Manipal Institute of Technology
71
43
28
39.42
39.2318675
Coll. of Engineering Thiruvananthapuram
72
71
1
39.37
38.91480413
Army Institute of Technology Pune
73
91
−18
39.21
38.76821489
M.S. Ramaiah Institute of Technology
74
64
10
39.09
38.53151098
Sri Krishna College of Engineering and Technology Coimbatore
75
97
−22
38.94
37.91661143
PSG College of Technology
76
44
32
38.56
37.68319934
B.M.S. College of Engineering
77
69
8
38.55
37.43066684
C.V. Raman College of Engineering Bhubaneswar
78
94
−16
38.42
37.06883436
Siddaganga Institute of Technology
79
79
0
37.98
36.99066201
Kalasalingam Academy of Research and Higher Education
80
61
19
37.87
36.77879355
Kumaraguru College of Technology
81
77
4
37.82
36.72110723 (continued)
254
M. Yadav et al.
Table 12 (continued) Institute name
NIRF score
Predicted NIRF score
Sri Venkateshwara University Tirupati
82
89
−7
37.74
36.45066777
NIT Agartala
83
70
13
37.68
36.34115351
Defence Institute of Advanced Technology
84
57
27
37.59
36.1477634
Mepco Schlenk Engineering College
85
88
−3
37.32
36.06177739
Bharati Vidyapeeth Deemed University College of Engineering Pune
86
93
−7
37.22
36.00338383
Karunya Inst. of Tech. and Sciences Coimbatore
87
72
15
37.06
35.63390507
NIT Goa
88
87
1
36.92
34.08232552
Chaitanya Bharathi Institute of technology Hyderabad
89
100
−11
36.66
33.95272296
Bannari Amman Institute of Technology Sathyamangalam
90
98
−8
36.42
33.9501504
Pondicherry Engineering College
91
76
15
36.39
33.75960627
Dayalbagh Educational Institute
92
84
8
36.39
33.58922112
Coimbatore Institute of Technology
93
68
25
36.14
33.39408398
Government College of Technology Coimbatore
94
90
4
36.12
33.32717653
Shri Mata Vaishno Devi University
95
85
10
36.11
32.73613252
NIT Meghalaya
96
67
29
35.99
32.70962747
Vel Tech Rangarajan Dr. Sagunthala R & D Institute of Science and Technology
97
86
11
35.97
31.45224241
Kongu Engineering College Perundurai
98
99
−1
35.9
31.30957429
Hindustan Institute of Technology and Science Chennai
99
95
4
35.89
31.09229305
100
96
4
35.86
12.56555871
Punjab Technical University Kapurthala
New rank
NIRF rank
Gain
Data Analytics for NIRF Ranking of Indian Institutions …
255
6 Conclusion With the five NIRF indicators, the validity of the NIRF score is suspicious and lacks conceptual clarity, as there are highly overlapping indicators as well as noncontributing indicators that are simply weighted and summed up. These indicators, therefore, make the overall meaning vague and uncertain. In this study, RPC is computed to be the most important indicator, which reflects that the performance of primary functions of the institute like generation of knowledge is invaluable. Multicollinearity and non-contributing indicators for NIRF are earmarked in the present study. The investigation carried out for NIRF may also be implemented for other WUR (world university ranking) systems and other educational and social ranking frameworks provided WAS (weight and sum) approach is being used for computing the weights of indicators. With the case study discussed for NIRF data and the pros and cons discussed in the present investigation for using lesser parameters for clarity, the ranking framework may be revisited.
References 1. Soh KC (2015) Multicolinearity and indicator redundancy problem in world university rankings: an example using THEWUR 2013–2014 data 2. Prathap G (2018) Eighteen National Institutes of Technology in the top 100 NIRF Engineering Ranking. Curr Sci 115(3):369–371 3. Mathew S, Cherukodan S, Sheeja NK (2017) Scholarly communication and institutional ranking: a study based on NIRF 4. Prathap G (2019) Construct validity maps and the NIRF 2019 ranking of colleges. Curr Sci 117(6):1079 5. Prathap G (2017) Making scientometric sense out of NIRF scores. Curr Sci 112(6):1240–1242 6. Prathap G (2017) Danger of a single score: NIRF rankings of colleges. Curr Sci 113(4):550–553 7. Sheeja NK, Mathew S, Cherukodan S (2018) Impact of scholarly output on university ranking. In: Global knowledge, memory and communication 8. Kumar A, Tiwari SK (2016) India rankings 2016: ranking model for Indian higher educational institutions. In: 2016 international conference on ICT in business industry & government (ICTBIG). IEEE, pp 1–6 9. Sivakumaren KS (2017) Contributions of publications of Indian Institute of Management in Ranking Institutions in National Institutional Ranking Framework: a study. J Libr Inf Sci 7(2) 10. Brahma K, Verma MK (2018) Evaluation of selected universities library websites listed by National Institutional Ranking Framework (NIRF) during the Year 2017: a webometric analysis. J Sci Res 7(3):173–180 11. Sivakumaren KS, Rajkumar T (2019) Publications of Indian Universities in National Institutional Ranking FrameWork (NIRF) system: a study. Libr Philos Prac 1–10 12. Aithal PS, Revathi R (2017) Comparison of private universities in India based on NIRF ranking and fee charging strategies. Int J Case Stud Bus IT Educ (IJCSBE) 1(2):72–85
Design Feature Assessment for Fused Deposition Modeling Using Supervised Machine Learning Algorithms Rahul Bansal, Sukhdeep Singh Dhami, and Jatinder Madan
Abstract This paper presents the development of a supervised machine learning model to predict dimensional accuracy for different parts fabricated using fused deposition modeling (FDM) technique. Supervised learning models, namely, polynomial regression, decision trees, and random forest regression were used to predict the overall dimensional accuracy as well as that of individual part features. All three selected algorithms performed satisfactorily when random datasets from existing data were provided for predicting the expected accuracy of the parts produced using FDM, which validates the proposed model. The results also show that the polynomial regression model provided the best results by predicting the dimensions most accurately. The proposed machine learning approach could be further implemented to develop models for recommending appropriate additive manufacturing process parameters for attaining best accuracy. Keywords Supervised machine learning · Regression analysis · Fused deposition modeling (FDM) · Dimensional accuracy
Nomenclature AM ANN BP-NN CNN DEM
Additive Manufacturing Artificial Neural Network Back Propagation Neural Network Convolutional Neural Network Discrete Element Method
R. Bansal · S. S. Dhami (B) Department of Mechanical Engineering, National Institute of Technical Teachers Training & Research, Chandigarh, India e-mail: [email protected] J. Madan Department of Mechanical Engineering, Chandigarh College of Engineering and Technology, Chandigarh, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_20
257
258
FDM FF-NN GA LPBF ML PLA RP S/N SVM
R. Bansal et al.
Fused Deposition Modeling Feed Forward Neural Network Genetic Algorithm Laser Powder Bed Fusion Machine Learning Polylactic Acid Rapid Prototyping Signal-to-Noise Support Vector Machine
1 Introduction In today’s fast-changing world of technological advancements, the demand for shorter development time and reduced product life cycle, along with higher quality and reliability of products, has led to the rise of an advanced and rapid fabrication technique, called Additive Manufacturing [1]. It is defined as “a process of joining materials to make objects from 3D model data, usually layer upon layer, as opposed to subtractive manufacturing methodologies” [2]. With improvements in the part quality, accuracy, and material properties, direct fabrication of functional end-use products has become the main trend of additive manufacturing [3]. Fused Deposition Modeling (FDM), an affordable additive manufacturing technique, is used to print functional prototypes in small or medium quantities in an office-friendly environment. It provides adaptable and reliable finished parts for modeling, prototyping, and production applications [4]. In this process, an entity is digitally defined by a computer-aided-design (CAD) model, which is then created using layer-by-layer deposition of a feedstock plastic filament material heated to melting temperature and extruded through a nozzle [5]. FDM offers unparalleled design freedom and an excellent strength-to-weight ratio for the fabricated parts, compared to conventional manufacturing techniques [6]. Despite all the growth and advancements, dimensional accuracy and surface quality remains a challenge for AM fabricated parts [7]. The fundamental reason for this inconsistency is the dependency of the part characteristics on numerous process parameters such as layer thickness, raster width, print speed, infill density, and infill pattern, etc. Therefore, to achieve functionally reliable and optimum quality customized parts, several studies are focused on learning the relationship between the part and process characteristics [8]. Advancements in sensor technologies, process measurements, and simulation techniques, created a new wave of AM-related data analysis [9]. Extracting knowledge and making sense from vast amounts of AM data was a tedious task, so there was an immediate need for advanced analytical tools that helped with data mining and analysis. So to transform this available data into some actionable and insightful data, machine learning was used [10].
Design Feature Assessment for Fused Deposition Modeling …
259
The primary aim of an ML algorithm is to allow the computers to learn automatically without human intervention or assistance and adjust actions accordingly [11]. With large and reliable training datasets, ML models learn and make inferences based upon that knowledge. Machine learning techniques can be broadly classified into four main categories: supervised learning, unsupervised learning, semi-supervised, and reinforcement learning. In supervised learning, machines are trained using well “labeled” training data, and on basis of that data, machines predict the output. In unsupervised learning, there is no labeled training data set available. Instead, the model itself tries to find the hidden patterns and insights from the given data, and groups them into different clusters [12]. The most common ML applications in the AM field are defect detection, quality assessment, anomaly predictions, and process parameter optimization, etc. [13]. This paper contributes to the application of ML algorithms for predictions in the AM domain. The organization of this paper is as follows: A summary of the literature related to the present study is presented in Sect. 2. The methodology adopted to carry out the present work is given in Sect. 3. Lastly, Sect. 4 presents concluding remarks and the scope of future work.
2 Summary of the Research A review of literature on additive manufacturing process parameters, response characteristics, and machine learning techniques was carried out to have insight into the problem. A tabular summary of the literature is presented in Table 1. Some of the inferences drawn after reviewing multiple research papers were (1) extensive research has been performed on various process parameters such as layer thickness, print speed, build orientation, raster width, air gap, and infill density, while less analyzed parameters were infill pattern, shell width, and extrusion temperature. (2) process parameter optimization using various tools and methods such as ANOVA, Central composite design (CCD), Finite Element Analysis, Fuzzy inference system, and Grey relational analysis. (3) ML techniques such as backpropagation, neural networks, genetic algorithm, SVM and decision trees have been used for porosity detection, fault detection, and build precision prediction. (4) ML algorithms have been widely used for anomaly detection, but there is a wide scope for prediction and monitoring in the AM domain.
260
R. Bansal et al.
Table 1 Overview of AM and ML techniques used to optimize process parameters References Methods/Tools
Process parameters
[14]
Genetic algorithm (GA)
Part deposition orientation Surface finish, Build time
Response
[15]
Taguchi, S/N ratio, ANOVA
Air gap, Layer thickness, Raster angle
[16]
Finite element model, Road width, Layer Central composite design thickness, Scan speed (CCD), ANOVA
Stress distribution, Dimensional accuracy
[17]
Artificial neural networks Z-height, Part volume, (ANNs) Bounding-box volume
Build time prediction
[18]
Grey Taguchi method, ANN, Fuzzy logic
Part orientation, Road width, Layer thickness, Air gap, Raster angle
Dimensional accuracy (length, width, thickness)
[19]
Taguchi, S/N ratio, ANOVA
Road width, Air gap, Layer thickness, Raster angle
Surface finish, Dimensional accuracy
[20]
Taguchi method, Fuzzy logic
Wire-width compensation, Dimensional error, Print Speed, Filling Warpage deformation velocity, Layer thickness
[21]
Genetic programming (GP), ANN
Layer height, Build orientation, Raster angle, Raster width, Air gap
Wear strength prediction
[22]
Taguchi, ANOVA, Regression
Layer thickness, Infill density, Post-processing heat-treatment time
Ultimate Shear and Yield strength, Shear modulus, Fracture strain
[23]
Full Factorial design, Desirability function approach (DFA)
Layer thickness, Print speed, Infill density
Dimensional accuracy
[24]
ASTM D638, ASTM D760 standards
Layer thickness, Print speed
Tensile and Flexural strength
[25]
CNN, Random forest classification
SEM microstructural Classification of AM images of powder material powders
[26]
DEM, BP-NN
Spreader speed
Powder spreading prediction
[27]
SVM with RBF kernels
Feed rate
PID process control for FDM
[28]
SVM ensemble classifier
Multiple images collected at each build layer
Defect (porosity, cracks, or inclusions) detection in PBF using in situ sensors coupled with CT scans
[29]
SVM, K-Nearest Neighbors, Bagged Trees, FF-NN
Laser power, Hatch spacing, Velocity
Porosity detection in LPBF
Elastic performance
Design Feature Assessment for Fused Deposition Modeling …
261
3 Experimental Methodology The major aspect of this research was to compare the prediction of expected accuracy using three machine learning algorithms, namely—polynomial regression, decision tree, and random forests. The part, shown in Fig. 1, was selected for the experimentation having common features such as a cavity, protrusion, and circularity. Figure 2 depicts the methodology adopted for the experimental work.
Fig. 1 CAD drawing of the part
Fig. 2 Flowchart for experimental study
Selection of ML algorithms Data Collection Data Pre-Processing and Visualization Evaluation of ML models Predictive Analysis
262
R. Bansal et al.
Table 2 Required dimensions of the fabricated parts Shape
Diameter
Side 1
Side 2
Height
Square cavity
–
8 mm
8 mm
–
Cylindrical cavity
8 mm
–
–
–
Solid square protrusion
–
5 mm
5 mm
10 mm
Solid cylindrical protrusion
5 mm
–
–
10 mm
Base cuboid
–
25 mm
25 mm
10 mm
The dimensions of the various features in the CAD model are given in Table 2. The dimensional measurements of the fabricated parts were the output parameters.
3.1 Selection of Machine Learning Algorithms Dimensional measurement predictions for different shapes were done using three supervised machine learning algorithms, namely, (i) polynomial regression, (ii) decision tree, and (iii) random forests to predict the measurements for different shapes. Polynomial regression is a form of linear regression, which establishes a relationship between the input variables and the output variable in the form of a polynomial equation. It helps to fit a nonlinear model to the data, by adding all combinations of features up to the given degree [30]. Decision Tree is a supervised learning algorithm, used for solving both regression and classification problems. It is constructed from the root node (data set), which represents the entire population or sample. It further splits into left and right internal nodes (subsets of a data set). Each internal node represents a test on an attribute, each branch represents an outcome of the test, and each leaf node (terminal node) represents the final prediction value [31]. Random Forest Regression, is an ensemble learning technique that performs regression tasks by growing multiple decision trees, which run in parallel and do not interact with each other. Each tree in a random forest learns from a random sample of the data points. This technique predicts the final output based on the average of each tree output [32]. Python, which is an open-source language, was used for implementing the machine learning algorithms.
3.2 Part Fabrication and Data Collection Eighty-one test specimens were fabricated using a full factorial design (34 ), with four input variables namely layer thickness (LT), print speed (PS), infill density (ID), and infill pattern (IP) and three levels of each parameter, Table 3.
Design Feature Assessment for Fused Deposition Modeling …
263
Table 3 Factors and levels for input parameters Levels
Input parameters (x) Layer thickness (mm)
Infill density (%)
Print speed (mm/s)
Infill pattern
1
0.1
20
50
Lines
2
0.2
35
60
Rectilinear
3
0.3
50
70
Concentric
Output parameter (y) Dimensional Measurements of square and cylindrical shapes
The experiments are conducted using Polylactic Acid (PLA) material, 1.75 mm filament on a Raise 3D printing machine. Some of the fabricated parts are shown in Fig. 3.
Fig. 3 Parts fabricated using 3D printing
Table 4 Measured dimensions for the square cavity S. No Input parameters
1
Output parameter: dimensional measurements
LT (mm) ID (%) PS (mm/s) IP
Square cavity side 1 Square cavity side 2 – 8 mm (X) – 8 mm (Y)
0.1
8.133
8.168 8.127
20
50
Lines
2
0.1
20
50
Rectilinear 8.257
….
….
….
….
….
….
….
43
0.2
35
70
Lines
8.290
8.192
44
0.2
35
70
Rectilinear 8.283
8.148
45
0.2
35
70
Concentric 8.161
8.003
….
….
….
….
….
….
80
0.3
50
70
Rectilinear 8.292
8.130
81
0.3
50
70
Concentric 8.022
8.050
….
264
R. Bansal et al.
The dimensions of the fabricated parts were measured using a machine vision system that uses MSU3D measuring software. Table 4 shows a sample of the measurement data for the square cavity for 81 combinations of process parameters. The information about the process parameters and accuracy details was used for designing an ML-based model, which is subsequently used to do predictive analysis.
3.3 Data Pre-processing and Visualization A well-structured data is the most important parameter for any machine learning algorithm because the quality and quantity of data directly determine the accuracy of the predictive model. Data pre-processing techniques, including formatting and cleaning the data, are performed to remove incomplete variables and to understand more about data [33]. The available dataset consists of a categorical feature, i.e., Infill Pattern consisting of non-numeric values viz. lines, rectilinear, concentric, which must be converted into an array format so that it could be understood by the ML algorithm. This was achieved using the One-Hot Encoding approach. Further, a data visualization technique was employed to identify different patterns and trends in the given data for splitting the data in 80:20 ratio for training and testing, respectively.
3.4 Evaluation of ML Models Root Mean Squared Error (RMSE) was used as the evaluation criteria for the comparison between the measured dimensions and the predicted dimensions of the fabricated part. Root mean square error is defined as the square root of the mean of the square of all of the errors [34]. Mathematical notation for RMSE is given as [35]. RMSE =
n i=1
yi − yˆ n
2
where yi = measured/original part dimension, y = predicted dimension by ML algorithm, and n = total no of test samples. Table 5 shows this value for different shapes, using three ML algorithms. Provided RMSE values are from the first run of the ML model because with each run the model keeps on updating and improving itself, thus altering the RMSE value. The comparison of prediction performance of the three ML algorithms in terms of RMSE is depicted in Fig. 4. It is observed that the RMSE values for polynomial regression are least for dimensional prediction of the square cavity, cylindrical cavity, square protrusion, and base cuboid. For the cylindrical protrusion, the RMSE value for polynomial regression
Design Feature Assessment for Fused Deposition Modeling …
265
Table 5 RMSE for different shapes for the three ML algorithms Shape
Polynomial regression
Decision tree regression
Random forest regression
Square cavity
0.00745
0.00905
0.00814
Cylindrical cavity
0.01405
0.02089
0.01638
Solid square protrusion
0.00442
0.00913
0.00538
Solid cylindrical protrusion
0.01402
0.01345
0.01384
Base cuboid
0.00497
0.00778
0.00508
0.022 Polynomial Regression
0.02
Decision Tree Regression
0.018
Random Forest Regression
RMSE
0.016 0.014 0.012 0.01 0.008 0.006 0.004 Sq Cavity
Cyl Cavity
Solid Sq Protrusion Solid Cyl Protrusion
Base Cuboid
Fig. 4 Performance comparison of ML algorithms based on RMSE
almost coincides with those for decision tree regression and random forest regression algorithms. Further, the range of RMSE for polynomial regression is from 0.0044 to 0.0140, it is 0.0077 to 0.0209 for decision tree regression and 0.0051 to 0.0164 for random forest regression. A lower value of RMSE indicates a better fit of the model. Hence, polynomial regression is the best fit model for this data set.
3.5 Predictive Analysis In the final step, the dimensions of desired sides were predicted using polynomial regression, decision tree, and random forest model using the testing dataset. Table 6, shows the percentage error in predicting the dimensions of side 1 and side 2 for the square cavity by the three ML algorithms.
266
R. Bansal et al.
Table 6 Percentage error for square cavity using three ML algorithms S. no
% Error for square cavity = ((Actual − predicted)/Actual) * 100
Input parameters
Polynomial regression
Decision tree
Random forest
Side 1
Side 2
Side 1
Side 2
Side 1
Side 2
0.91
−1.80
2.76
−1.94
1.69
−1.56
LT (mm)
ID (%)
PS (mm/s)
IP
1
0.3
20
50
Rectilinear
2
0.1
20
60
Concentric
0.41
−0.27
−0.06
−0.26
0.67
−0.23
3
0.2
20
70
Linear
−0.64
−0.44
−0.60
−1.34
−1.07
−1.18
4
0.1
50
70
Linear
−0.23
−0.91
−0.20
−0.44
−0.33
−0.89
5
0.1
20
60
Linear
0.92
−0.52
−0.45
−1.03
0.61
−0.91
6
0.3
35
70
Concentric
−0.99
−0.60
−1.77
−0.61
−0.95
−0.19
7
0.1
50
60
Concentric
0.26
−0.40
1.49
0.00
1.26
0.16
8
0.1
35
70
Rectilinear
−0.23
0.38
0.28
0.20
−0.35
0.52
9
0.3
20
50
Concentric
0.72
0.44
1.51
0.86
0.77
0.91
10
0.1
20
50
Concentric
−0.52
0.83
−1.60
0.48
−0.58
1.02
11
0.2
20
70
Concentric
−1.41
−0.84
−0.57
0.20
−1.35
−0.44
12
0.3
50
70
Concentric
−1.70
−0.79
−2.59
−1.06
−2.24
−0.72
13
0.1
35
60
Linear
0.09
−0.59
0.36
−1.17
0.70
−0.95
14
0.2
20
50
Concentric
−2.32
−0.06
−1.65
−0.57
−1.92
−0.04
15
0.1
35
50
Rectilinear
0.38
−0.26
0.91
−0.24
0.48
−0.16
16
0.2
35
60
Concentric
−0.77
−0.70
1.13
0.15
−0.55
−0.96
17
0.3
35
60
Linear
1.42
0.89
0.14
0.87
0.67
0.64
Thus in predicting the dimensions for the square cavity, the mean absolute errors for side 1 were observed to be 0.82, 1.06, and 0.95 for Polynomial Regression, Decision Tree Regression, and Random Forest Regression, respectively, whereas the same for side 2 were 0.63, 0.67, and 0.68, respectively. Similar results were obtained for dimensional prediction of the cylindrical cavity, square protrusion, cylindrical protrusion, and the base cuboid. Thus, the prediction results for the complete set of dimensions also showed that Polynomial Regression was the best out of the three tested ML algorithms in predicting dimensions. Figure 5, shows the prediction accuracy of polynomial regression for various features based on the testing data set. The normalized error deviation for the square and cylinder cavity dimension ranges from −0.02 to +0.02. The same trend is observed for the horizontal solid dimensions. For vertical solid dimensions, only two values of normalized error out of a total of 51 values fall outside this range, whereas the remaining values are observed to remain within the range ±0.02. Therefore, it may be concluded that polynomial regression exhibits good performance for predicting dimensions of parts fabricated with PLA material using 3D printing with different values of layer thickness, print speed, infill density, and infill pattern.
Design Feature Assessment for Fused Deposition Modeling …
267
Fig. 5 Normalized error for a horizontal cavity dimensions b horizontal solid dimensions c vertical solid dimensions, using Polynomial regression
4 Conclusions An approach to predict dimensional accuracy for different features and shapes using supervised machine learning algorithms has been presented. A total of 81 parts were fabricated with Polylactic Acid on a Raise 3D printing machine with varying four process parameters. The dimensions of the various features viz. square and cylindrical cavities and projections of the fabricated part were measured. The data set of input parameters and measured dimensions were used for training and testing of three ML algorithms—polynomial regression, decision tree regression, and random forest regression. Python, which is an open-source language, was used for implementing the machine learning algorithms. The results revealed that the polynomial regression model provides the best prediction values with minimum error for this particular
268
R. Bansal et al.
data set. The prediction accuracy achieved was within the limit ±0.02 for all the dimensions of the 3D fabricated parts with polynomial regression.
References 1. Mangat AS, Singh P, Singh S (2017) Investigations on dimensional accuracy of FDM based natural fibre embedded poly-lactic-acid structures. Int J Adv Multidiscip Res (IJAMR), 39–42 2. Standard Terminology for Additive Manufacturing Technologies. Accessed Oct 04, 2020, https://web.mit.edu/2.810/www/files/readings/AdditiveManufacturingTerminology.pdf 3. Guo N, Leu MC (2013) Additive manufacturing: technology, applications and research needs. Front Mech Eng 8(3):215–243 4. 3D Printing with FDM Technology. Accessed Oct 04, 2020, https://www.sculpteo.com/en/mat erials/fdm-material/ 5. Mohamed OA, Masood SH, Bhowmik JL (2015) Optimization of fused deposition modeling process parameters: a review of current research and future prospects. Adv Manuf 3:42–53 6. Peng A, Xiao X, Yue R (2014) Process parameter optimization for fused deposition modeling using response surface methodology combined with fuzzy inference system. Int J Adv Manuf Technol 73:87–100 7. Gao W, Zhang Y, Ramanujan D, Ramani K, Chen Y, Williams CB, Wang CCL, Shin YC, Zhang S, Zavattieri PD (2015) The status, challenges, and future of additive manufacturing in engineering. Comput Aided Des 69:65–89 8. Vyavahare S, Teraiya S, Panghal D, Kumar S (2019) Fused deposition modelling: a review. Rapid Prototyping J. https://doi.org/10.1108/RPJ-04-2019-0106 9. Mani M, Lane B, Donmez A, Feng S, Moylan S, Fesperman R (2015) Measurement science needs for real-time control of additive manufacturing powder bed fusion processes. In: National Institute of Standards and Technology, NIST IR 8036 10. Wuest T, Weimer D, Irgens C, Thoben KD (2016) Machine learning in manufacturing: advantages, challenges, and applications. Prod Manuf Res 4(1):23–45 11. Machine Learning. Accessed Oct 06, 2020, http://www.contrib.andrew.cmu.edu/~mndarwis/ ML.html 12. Hastie T, Tibshirani R, Friedman J (2009) The elements of statistical learning: data mining, inference, and prediction. Springer 13. Meng L, McWilliams B, Jarosinski W, Park HY, Jung YG, Lee J, Zhang J (2020) Machine learning in additive manufacturing: a review. JOM 14. Thrimurthulu K, Pandey PM, Reddy NV (2004) Optimum part deposition orientation in fused deposition modeling. Int J Mach Tools Manuf 44:585–594 15. Lee BH, Abdullah J, Khan ZA (2005) Optimization of rapid prototyping parameters for production of flexible ABS object. J Mater Process Technol 169:54–61 16. Zhang Y, Chou K (2008) A parametric study of part distortions in fused deposition modelling using three-dimensional finite element analysis. Proc Inst Mech Eng, J Eng Manuf 222:959–968 17. Munguía J, Ciurana J, Riba C (2008) Neural-network-based model for build-time estimation in selective laser sintering. Proc Inst Mech Eng, J Eng Manuf 223:995–1003 18. Sood AK, Ohdar RK, Mahapatra SS (2009) Improving dimensional accuracy of fused deposition modelling processed part using grey Taguchi method. Mater Des 30:4243–4252 19. Nancharaiah T, Raju DR, Raju VR (2010) An experimental investigation on surface quality and dimensional accuracy of FDM components. Int J Emerg Technol 1(2):106–111 20. Jinwen Z, Anhua P (2012) Process-parameter optimization for fused deposition modeling based on Taguchi method. Adv Mater Res 538–541:444–447. https://doi.org/10.4028/www.scientific. net/AMR.538-541.444 21. Garg A, Tai K (2014) An ensemble approach of machine learning in evaluation of mechanical property of the rapid prototyping fabricated prototype. Appl Mech Mater 575:493–496
Design Feature Assessment for Fused Deposition Modeling …
269
22. Torres J, Cotelo J, Karl J, Gordon AP (2015) Mechanical property optimization of FDM PLA in shear with multiple objectives. JOM 67 23. Akande SO (2015) Dimensional accuracy and surface finish optimization of fused deposition modelling parts using desirability function analysis. Int J Eng Res Technol (IJERT) 4:196–202 24. Jaya Christiyan KG, Chandrasekhar U, Venkateswarlu K (2016) A study on the influence of process parameters on the mechanical properties of 3D printed ABS composite. In: IOP conference series: materials science and engineering, vol 114. https://doi.org/10.1088/1757899X/114/1/012109 25. Ling J, Hutchinson M, Antono E, DeCost B, Holm EA, Meredig B (2017) Building data-driven models with microstructural images: generalization and interpretability. Mater Discovery 10:19–28 26. Zhang W, Mehta A, Desai PS, Higgs III CF (2017) Machine learning enabled powder spreading process map for metal additive manufacturing (AM). In: Proceedings of the 28th annual international conference 27. Liu C, Roberson D, Kong Z (2017) Textural analysis-based online closed-loop quality control for additive manufacturing processes. In: Industrial and systems engineering conference 28. Gobert C, Reutzel EW, Petrich J, Nassar AR, Phoha S (2018) Application of supervised machine learning for defect detection during metallic powder bed fusion additive manufacturing using high resolution imaging. Addit Manuf 21:517–528 29. Imani F, Montazeri M, Gaikwad A, Rao P, Yang H, Reutzel EW (2018) Layerwise in-process quality monitoring in laser powder bed fusion. In: ASME, 13th international manufacturing science and engineering conference 30. Géron A (2017) Hands-on machine learning with Scikit-learn and TensorFlow, 1st edn 31. James G, Witten D, Hastie T, Tibshirani R. An introduction to statistical learning with applications in R. https://doi.org/10.1007/978-1-4614-7138-7 32. Bratsas C, Koupidis K, Salanova JM, Giannakopoulos K, Kaloudis A, Aifadopoulou G (2020) A comparison of machine learning methods for the prediction of traffic speed in urban places. MDPI J Sustain 12:142. https://doi.org/10.3390/su12010142 33. The 7 steps of machine learning. Accessed Oct 08, 2020, https://towardsdatascience.com/the7-steps-of-machine-learning-2877d7e5548e 34. Chai T, Draxler RR (2014) Root mean square error (RMSE) or mean absolute error (MAE)? Geoscientific model development 35. Root-mean-square error in R programming. Accessed Oct 08, 2020, https://www.geeksforg eeks.org/root-mean-square-error-in-r-programming/
Development of a 3D Printed Orthopaedic Cast for Wrist Fracture Mohd Ahad Islam, Mukul Shukla, and Yogesh Tripathi
Abstract Wrist fracture is reported as the most common orthopaedic injury. Traditional fracture management for such cases includes immobilization of the wrist with a traditional calcia plaster or fibreglass cast. These methods are often associated with various disadvantages like heavyweight, not water resistant and medical complications such as compartment syndrome and other skin related discomforting issues. The present work aims at developing a custom made, lightweight FDM 3D printed cast which is more effective in overall healing by overcoming the above complications. For developing this patient specific cast, the injured limb is first 3D scanned using a handheld Infrared scanner to precisely capture the wrist geometry. Thereafter the CAD model is constructed using this scan data in Fusion 360 CAD software. The final design is 3D printed with Polyethylene Terephthalate Glycol (PET-G) resulting in an anatomically accurate and waterproof cast which is 31.06% lighter than the traditional cast. The developed cast is then reviewed by the orthopaedics for its clinical efficacy. Keywords 3D printed · Wrist fracture · Orthopaedic cast · FDM · Custom cast · PETG
1 Introduction Wrist fractures are one of the most common orthopaedic injuries, the traditional clinical procedure for rehabilitation in such injuries is done using calcia plaster cast. However, it does immobilization job satisfactorily but its usage has major setbacks. For example, it’s heavy and not waterproof therefore proper hygiene cannot be maintained. This may result into unwanted situation like itching of patient’s skin, bacterial and fungal borne infections [1]. Moreover, its usage may lead to lowered patient satisfaction during the rehabilitation. This situation demands a better cast design to support the fractured wrist. Here “better cast design” reflects the need of anatomically M. A. Islam (B) · M. Shukla · Y. Tripathi Department of Mechanical Engineering, Motilal Nehru National Institute of Technology-Allahabad, Prayagraj, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_21
271
272
M. A. Islam et al.
fit and lightweight casts. Such designs will support the fractured wrist and as well as proper hygiene can be maintained to overcome any itching and infection problem thereby enhancing the level of patient satisfaction. Thanks to advancement in medical scanning/imaging, CAD technology and 3D printing. The combined usage of such digitized technologies has resulted into realization of functional medical devices. The central idea of this paper is to put forward the detailed workflow for development of anatomically accurate, and lightweight wrist cast for fracture immobilization. The developmental procedure consists of three major steps, enlisted below: a. b. c.
Acquiring patient specific wrist surface using a 3D scanner. CAD modelling of the cast. Physical realization of the cast using 3D printing.
2 Methodology 2.1 3D Scanning of the Arm In order to begin the development, process the patient is first 3D scanned since the cast is patient specific, i.e., it is custom made for each individual. In our case we have used an IR (Infrared) based Microsoft Kinect device [2] to capture the 3D point cloud. While scanning it was ensured that the positioning of the wrist is in the clinically recommended 20–30° extension position. To accurately capture the complete arm from all angles, the scanner is made to do a 360° rotation around the subject at varying elevations. The scanner captures data within a defined bounding box at 30 fps and gives a live feedback to the connected laptop. With the initials setup Fig. 1a, on mesh reconstruction in the Skancet software a face count of 0.3 million was achieved along with some irregularities Fig. 1b.
SCANNER
MONITOR
(b) Fig. 1 a 3D scanning of subject b Mesh reconstruction with irregularities
(a)
Development of a 3D Printed Orthopaedic Cast for Wrist Fracture
273
(a)
(c) (b) Fig. 2 a 3D Printed laptop mount b New scanning setup with more accessible live feedback c Higher resolution scan result with new laptop mount setup
On performing multiple trials, it was found that the main reason for these overlapping irregularities and lower mesh resolution was because of the limitation of our scanning setup which required continuous monitoring of the feedback on the screen while being held in one arm and moving the scanner with the other arm, as can be seen in Fig. 1a, thus resulting in some non-continuous, jerky movements. In order to obtain a higher resolution scanned mesh and to avoid these irregularities, it was important to maintain a stable position of the scanner throughout the scanning process. To achieve this target, mounts were designed so that the scanner module can be attached directly onto the monitoring laptop. This way data acquisition and its feedback both can be done with the same movement. Scanner mounts were designed in Autodesk Fusion 360 software and were 3D printed in PLA polymer. With the new setup, scanned data can be reconstructed at a much higher resolution because of improved stability, as can be seen in Fig. 2c. Face count of over 5 million was achieved in a bounding box of 1 m3 , thus accurately capturing anatomical model of the arm without any irregularities.
2.2 CAD Modelling The reconstructed mesh is imported into Autodesk Meshmixer software and the boundaries of the cast are defined such that the movement of knuckle joints, thumb
274
M. A. Islam et al.
along with that of elbow is not restricted. After defining the boundaries, the cast is extruded along its surface normal to a uniform thickness of 3 mm (Fig. 3). This surface mesh model of the cast is then converted into a BREP model using Autodesk Netfabb software, so that further geometry modifications can be made in a CAD software. The converted BREP model is verified by performing a surface deviation study in GOM Inspect software. The result of the study validates that the converted BREP model is true to the scan data. The cast’s final design (Fig. 5) was decided based on an iterative design approach. Prototypes of different designs were made and tested for their accessibility to the patient and also their clinical efficacy, these were evaluated by a team of Orthopaedics. Based on their suggestions, design iterations were made and were retested until the optimal solution was reached. Some of the major design iterations have been shown in Fig. 4. To start off, we had cut the cast in two sections with a helicoidal plane in front of the palm section but since this might be a weak but since the flexion strength of the wrist is more than the extension therefore this frontal opening could serve as a weak point in the design. As a result, the opening of the cast was decided to be set at the back of the hand. Finally, in order to accommodate various stages of inflammation during the application period of the cast it was recommended by the team to go for a single piece cast instead of a two-piece (full cast) which is cut along the backside away from the Radius bone, so that it is protected, and there’s enough room for the patient to easily slide its arm. The cast is held on to the patient’s arm with the help of medical Velcro straps which can be adjusted depending upon the level of inflammation. Three-point fixation technique [3] was recommended by the orthopaedics therefore six strap mounts are modelled on to the cast, three on each side. During the initial simulations it was found that the junction of strap mounts with the cast, serves as stress concentration points, therefore, appropriate fillets were introduced in the design.
(a)
(b)
Fig. 3 a Cast boundary selection b Conversion of initial cast mesh into BREP model
Development of a 3D Printed Orthopaedic Cast for Wrist Fracture
275
Fig. 4 Deviation of BREP model from the original scanned mesh
Fig. 5 Back view of the prototypes
2.3 Static FEA To validate the design of cast, static FEA analysis of the design is performed in Ansys software. Forces acting on a cast are experimentally recorded in the research work done by Yan et al. [4]; Liu et al. [5]. They evaluated these forces during the movement of the wrist with the help of membrane pressure sensors fixed at key positions (Fig. 6) in the cast, using their data Table 1, a static structural study is created (Fig. 7).
276
M. A. Islam et al.
Fig. 6 Rendered image of the final cast design
Table 1 Forces acting on the cast
Forces (N) A
34.77
I
86.01
B
13.42
J
73.81
C
18.91
K
95.16
D
90.28
L
31.72
E
48.19
M
73.81
F
34.16
N
92.11
G
54.29
0
3.66
H
68.32
P
63.44
The six fixation points highlighted with blue in Fig. 8 will serve as mounting points for the medical Velcro strap. Therefore, these are modelled as fixed support, i.e., zero displacement in the simulation model. Stress and displacement distribution obtained from the simulation model is shown in Figs. 9 and 10, respectively. The induced stresses in the cast are well within the safe zone and are maximum, 20.082 MPa, at the junction of the Velcro strap mount and the cast. Whereas the maximum deformation is of 0.71 mm, which is much less than 1.5–2.5 mm in the traditional cast [5], thus validating the light weight design to be medically safe.
Development of a 3D Printed Orthopaedic Cast for Wrist Fracture Fig. 7 Load case for the cast
Fig. 8 Fixed boundary condition
277
278
M. A. Islam et al.
Fig. 9 Deformation distribution
2.4 3D printing Layer-by-layer manufacturing, i.e., 3D printing gives the freedom to the designer to create complex organic shapes which were earlier difficult to produce with traditional manufacturing, along with that it allows for mass customization. In our work we have leveraged this additive manufacturing to produce our lightweight, patient specific cast on a Creality Cr10 FDM 3D printer with PETG polymer (Fig. 12). In order to get the model printed it is first sliced into multiple layers, with the help of a Slicing software. Thickness of each layer is defined along with other parameters in the software. For this the STL file is imported in CURA software and various printing parameters are defined, as mentioned in Table 2. Print orientation is also defined in the software, in our case the model is printed standing up. On slicing the model, the software automatically generates the G-code file, containing all the defined print settings and the travel moves for the 3D printer. Layer height is set as 0.3 mm which is half of the nozzle diameter (0.6 mm) for optimal balance between speed and inter layer strength [6]. Prototypes were printed with different infill%, 100%, 30%, and 0% (Fig. 11) based on their flexural performance 0% infill was selected with the outer wall thickness of 1.2 mm. Since the area near the junction of the Velcro mount is a weak point in the design (based
Development of a 3D Printed Orthopaedic Cast for Wrist Fracture
279
Fig. 10 Stress distribution Table 2 Slicer settings for the 3D printing of cast Layer height
Wall thickness
Nozzle temp
Build plate temp
Print speed
Infill
0.3 mm
1.2 mm
250 °C
85 °C
40 mm/s
0%
(c) 0% Fig. 11 Infill % prototypes
(b) 100%
(a) 30%
280
M. A. Islam et al.
(a)
(b)
Fig. 12 a 3D Printed prototype b Velcro straps attached to the cast
on the stress analysis), infill modifiers were added so that full solid layer, i.e., 100% infill, instead of a hollow shell is printed in that particular region (Fig. 12). The complete manufacturing process took around 6.5 h with a 0.6 mm extruder nozzle. After fabrication, three Velcro straps were added through the designed mounts according to the three-point fixation technique.
3 Results and Conclusion The developed 3D printed patient specific cast overall weight is 162 g while the traditional cast weighs around 235 g [4]. As a result, there is 31.06% reduction in the total weight. The developed cast was also reviewed by the orthopaedics’ some of their positive observations were that since the cast is patient specific, therefore it offers better localized support and it can be altered with each patient, based on the clinical requirements. As it is water resistant the patient can maintain proper hygiene thereby avoiding various skin complications. And with the reduction in weight, the overall rehabilitation period will become more effective and convenient for the patient. In order to further reduce the weight of the cast as well as provide more comfort to the patient in future study we would design the ventilating pattern using the FEAbased topology optimization. Along with that the cast’s clinical effectiveness and the
Development of a 3D Printed Orthopaedic Cast for Wrist Fracture
281
overall patient satisfaction needs to be clinically tested and, based on these results necessary modifications have to be made in the design. Acknowledgements This work is supported by the Design Innovation Centre “DIC” (Letter No. 658/R&C/19-20), under MHRD (Ministry of Human Resource Development), India. The authors wish to acknowledge the important contributions of Dr. D.Dey, Dr. R.P.Singh, and Dr. Avinash Jaiswal in analysing the feasibility of the cast’s design and for their valuable suggestions.
References 1. Halanski M, Noonan KJ (2008) Cast and splint immobilization: complications. JAAOS-J Am Acad Orthop Surg 16(1):30–40 2. Cui Y et al (2011) 3D body scanning with one kinect. In: 2nd international conference on 3D body scanning technologies, vol 10 3. Alemdaro˘glu et al (2010) Three-point index in predicting redisplacement of extra-articular distal radial fractures in adults. Injury 41(2):197–203 4. Yan W et al (2019) Lightweight splint design for individualized treatment of distal radius fracture. J Med Syst 43(8):284 5. Lin H, Shi L, Wang D (2016) A rapid and intelligent designing technique for patient-specific and 3D-printed orthopedic cast. 3D Printing Med 2(1):1–10 6. Kuznetsov VE et al (2018) Strength of PLA components fabricated with fused deposition technology using a desktop 3D printer as a function of geometrical parameters of the process. Polymers 10(3):313
Effect of IoT in Supply Chain Management—A Review Gagandeep, Nikhilesh Singh, Abhishek Charak, Mohit Tyagi, and R. S. Walia
Abstract In the twenty-first century, the supply chain management (SCM) has been widely implemented in the manufacturing as well as service industries. Traditionally, SCM practitioners and researchers faced various difficulties like uncertainty of demand, fluctuations of cost and complexity during the execution of SCM in several manufacturing/service industries. This causes further difficulties in control and management of the widespread development of supply chains scenario. In previous years, the handling of the uncontrolled big data is the biggest problem faced by the various industries during the implementation of SCM. Internet of Things (IoT) plays a vital role and key tool in supply chain management. IoT provides realtime shipment, inventory visibility and makes easier to analyse the big data, it also contributes several technologies such as Radio Frequency Identification (RFID), Big Data Analytics, cloud computing and furthermore, it also ensures transparency, traceability and security in order to enhance the execution and the performance of SCM methodology. This paper emphasises on use of effective techniques and a better guideline for effective implementation of IoT in SCM. Keywords Supply chain management (SCM) · Cloud computing · Internet of things (IoT) · Big data analytics · Radio frequency identification (RFID)
1 Introduction Nowadays, an individual company needs to fulfil the demand of customer, to sustain in this competitive market. An appropriate balance of supply and demand plays a key role to satisfy the customers. The company is needed to adopt the SCM for maintaining the adequate balance between the supply and demand of the services as Gagandeep (B) · N. Singh · A. Charak · R. S. Walia Department of Production and Industrial Engineering, Punjab Engineering College, Chandigarh, India M. Tyagi Department of Industrial and Production Engineering, National Institute of Technology, Jalandhar, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_22
283
284
Gagandeep et al.
well as goods. SCM involves the flow of information and goods from suppliers to consumers through several entities such as suppliers, factories, distributors, retailers and consumers. A model named as Supply Chain Operations Reference (SCOR) explains that strategies, such as planning, source, making, delivery, return and enable, are accompanied with the satisfaction of customer’s demand [1]. SCM is a set of approaches which can be effectively used to integrate suppliers, factories, and distributors, retailers to distribute the appropriate products, in appropriate quantity, at right time, to right place and to right customer [2]. Traditional SCM approaches usually face a number of difficulties like uncertainty of demand, fluctuation of cost, handling of big data, transparency, tracking and security of goods, during execution of SCM in any organisation. The handling of big data is associated with the handling with a lot of paper work. Accordingly, there is a need to adopt the application of Internet of Things (IoT) in SCM to resolve such difficulties. The concept of Internet of Things (IoT) was developed by Kevin Ashton in 1999, at Britain [3]. Basically, IoT is a network that comprises one or more software, algorithms and a set of physical components including controller, storage device and sensors, etc., the physical elements are connected with each other for sensing and communicating the data within the system and/or among the systems. IoT is a progressive move in the area of computer technology and communication which is focussed to share the information at any time, at any place by virtue of any media. IoT technologies, such as Cloud computing, Big Data Analytics and RFID, make it easier to analyse the huge and structured/unstructured data. These technologies also provide inventory visibility, real-time transportation monitoring, agility and adaptability to resolve several problems of SCM that ensures transparency, traceability and security [4]. Further, the use of IoT reduces the time required for data capturing and decisionmaking which enhances the level of quickness in SCM. This paper is focused to study the importance of Internet of Things in SCM, by reviewing the most relevant available literature. Section 2 describes the research background regarding Internet of Things in SCM and, IoT technologies. Moreover, a methodology as followed to obtain the relevant literature is described in Sect. 3. The detailed analysis of the selected papers is provided in the Sect. 4. Furthermore, Sect. 5 contains the discussion on reviewed papers. Finally, the concluding points are summarised in Sect. 6.
2 Research Background In past few years, numerous research and review articles have been published in the area of Supply Chain Management (SCM). Countless research is going on particularly on the implications of the Internet of Things in SCM. However, the range of influence of IoT on all SCM processes is not clearly known. Therefore, the impact of IoT is required to be individually examined on each process of SCM. According to SCOR model, the processes of supply chain are plan, source, make, deliver, return and enable. IoT has capabilities to identify, trace, track, secure and share the real-time
Effect of IoT in Supply Chain Management—A Review
285
information and securing the data as well as objects [5]. A study was conducted to combine the RFID tags with Global Positioning System (GPS) for instant tracking of the indoor and outdoor products at any place [6]. IoT was used to establish a Time–Temperature Indicator (TTI) for controlling the temperature in cold supply chain processes [7]. Yan et al. [8] had introduced an approach including cloud of objects which enables to share and collaborate the resources among supply chain partners. IoT technologies such as big data analytics and cloud computing are used with Cyber-Physical Production Systems (CPPS) in the automated smart manufacturing [9]. The Cyber-Physical Production Systems (CPPS) was introduced to link the physical and cyber world by connecting the hardware components with the CloudBased Manufacturing (CM). RFID technologies and sensors are main IT enablers which are used for security and privacy concern in supply chain. Zelbst et al. [10] presented a research regarding the use of RFID technology and its influences on the efficiency of manufacturing and supply chain [10]. A study was conducted to improve the sustainability of hydrogen supply chain by using DEMATEL wherein different barriers exist in the green supply chain management in the manufacturing industries and were examined using DEMATEL-based technique [11, 12]. Further, the use of the cloud computing is inevitable in the SCM. Cloud computing provides real-time inventory monitoring and sales information to the central system which can be easily circulated across the whole supply chain network without any delay [13]. Cloud computing also provides several payment methods that permit the suppliers to measure their services in accordance with the competitive scenario and business planning [14]. The disturbances in the supply chain can be also captured with the applications of the IoT and smart technologies that permit to remotely observe and control the location and condition of order from manufacturing to the end consumer [15]. The supply chain transparency and cooperation may depend on knowledge and information shared in between the members of supply chain [16, 17]. Another study was conducted to examine the influence of cloud computing on the distribution of real-time information among various supply chain associates [18]. Finally, the above literature shows that RFID and Cloud computing are widely implemented as effective IoT technologies in the several supply chain processes.
2.1 IoT Technologies IoT technologies are considered as wonderful tools to deal with problems faced in the traditional SCM. IoT technologies involve Radio Frequency Identification (RFID), Cloud computing Big data analytics etc., which provides real-time shipment, inventory visibility and makes easier to analyse the big data and, also ensures transparency, traceability and security in order to enhance the execution and the performance of SCM. IoT comprises main four layers and each layer consisting of different IoT technologies used for different purposes [19]. First layer is a sensing layer configured for the collection of data using various technologies RFID tags, sensors, actuators and bar codes. These technologies are used for real time tracking and identification of
286
Gagandeep et al.
goods. Second layer is a networking layer used for exchange of information among the entities of Supply Chain, with the help of wireless networks or wired networks. Third layer is a service layer which established as implementation and control of the IoT services by concerning with the quality and performance of IoT services and applications. Fourth layer is an interface layer which is used for providing the information to customer and also permits interaction with system. Main IoT technologies are implemented in SCM to make it effective and more agile, are Radio Frequency Identification (RFID), Cloud computing and Big data analytics. RFID is imperative technology for the identification and tracking of products. It also transmits the information about goods to the supplier. RFID tags are comprising of microchips which are similar to adhesive stickers. In this technology, the tags and labels began the communication in two ways on the activation of the microchips. First way is Tag Talk First communication and second way is Interrogator Talk First communication. In Tag Talk First (TTF) communication, the tags first transfer information to interrogator without any request. In Interrogator Talk First (ITF), tags transfer information on the request of interrogator. In this technology, three main types of tags are used such as passive, active and semi-active. The passive tags are activated by the reader according to the requirement. The active tags are always activated and continuously transfer information because active tags have battery. The semi-active tags are hybrid type of passive and active tags wherein semi-active tags consist of battery but activated at the request of operator [20]. Cloud Computing (CC) is Internet-based platform which connects the number of server resources on single cloud platform and allows to share information with each other which can be accessed at anytime and anywhere [21]. This platform includes various resources such as computing servers, storage servers and application servers. Cloud computing is effectively used to instantly analyse, monitor and store the large amount of data emitted by IoT devices. Cloud computing services are considered as effective alternative for manage and establish the various data centres [22]. A cloud computing model has different servers connected with each other through Internet at cloud platform, as shown in Fig. 1. In cloud computing, the serves are only interconnected but also capable of storing and analysing the data effectively.
Fig. 1 Cloud computing model
Effect of IoT in Supply Chain Management—A Review
287
The scalability of RFID technology system can be done by combining the RFID technology architecture with Cloud Computing [23]. A software named as Middleware is used to permit software developers to interact with variety of devices such as sensors, actuators, RFID tags through wired or wireless sensor networks. The wireless sensor networks include a set of sensors configured to monitor and trace the status of different goods/products such as their location and flow in supply chain. Further, the sensors are used for various estimations like as temperature measurement, pressure measurement, flow detection, level measurement, noise detection, air pollution measurement, closeness, humidity measurement and speed testing, wherein the data is collected using sensors, and further processed through cloud computing [24]. In view of above, it is revealed that the IoT technologies play key role to efficiently improve the SCM by developing transparent and end-to-end communication among all entities of SCM.
3 Methodology In this review paper, the most relevant papers are recapitulated from the available literature at various data bases such as Google Scholar and ScienceDirect. The keywords, such as Internet of Things (IoT), Supply Chain Management (SCM), Modern SCM, Data analytics and cloud computing, were used. The search strings, such as Internet of Things (IoT) in Supply Chain Management (SCM), Cloud computing in SCM, Modern SCM and Data analytics in SCM, are formed by using above-mentioned keywords. Initially, the papers are selected by using search strings such as Internet of Things (IoT) in Supply Chain Management (SCM), modern supply chain management and data analytics in SCM. Total 66 papers were selected as relevant from 2014 to 2020 using above-mentioned search strings. These papers include both research and review papers which are mainly focussed on the impact of the applications of the IoT technologies in SCM implementation in various sectors. IoT has the ability to reduce several problems facing in SCM and also helps to enhance as cost-saving, inventory accuracy, traceability, transparency and easy analysis of the big data. The selected papers are analysed in accordance with their publication years, as shown in Fig. 2. The maximum number of papers as published is found in 2018. Further, the papers as selected are categorised in accordance with the countries where the research was conducted as shown in Fig. 3. The selected papers are further examined by reading their title, abstract and conclusions of papers and 30 papers are found as more relevant to the present study. Moreover, a final scrutiny is carried out to select the most relevant papers for the systematic and detailed analysis. In this scrutiny, total 14 papers are selected for detailed and systematic analysis. These papers are strictly focussed towards the use of IoT technologies such as RFID and Cloud computing in SCM, IoT applications in SCM and also address the problems which are faced in traditional SCM. The methodology is followed to select most relevant papers which is described in Fig. 4.
288
Gagandeep et al.
Fig. 2 The selected papers are distributed in accordance with the years 2014–2020
Fig. 3 The selected papers are distributed in accordance with the countries
4 Detailed Analysis of Selected Papers Ben-Daya et al. [24] was conducted a detailed study in year 2019 in the field of IoT in Supply chain management. This paper emphasises on the aspects of IoT and their effects in SCM. The authors had collected the papers which are mainly focussed on RFID technology and objects, sensors of the RFID. Further, this paper has identified that the majority of researchers are concentrating on the delivery supply
Effect of IoT in Supply Chain Management—A Review
289
Fig. 4 Methodology followed to select the most relevant papers
chain processes in the field of food and manufacturing sector. The paper has defined IoT in SCM in terms that the IoT is a grid of visible entities linked digitally for sensing, detecting and communicating within the organisation as well as among the organisations, that makes the supply chain processes smart for executing real time tracking and sharing of information. This helps in planning, managing and executing the supply chain processes in better way. The basic supply chain strategies are planning, source, making, delivery, return and enabling. However, most of the research is carrying out in delivery and make process in accordance with this research. Abdel-basset et al. [1] introduced a framework for building the smart, secure and efficient systems, using IoT in supply chain, and further analyse the influences of IoT in Supply chain management. In this particular study, the authors had specified various problems faced by traditional supply chain such as fluctuation of demand, sale fluctuation, transparency between the supplier and consumer, instant tracking and security. To deal with these problems, IoT can be implemented in the SCM for creating a smart and secure system. A website was established by authors through which the information can be shared between the supplier and manager. The supplier can also use the RFID technology to identify and trace the products and exchange the same information with manger using website. The whole lifecycle of goods is monitored which helps to improve the transparency between the members of supply chain. From security point of view, the proposed framework includes combination of two
290
Gagandeep et al.
techniques such neutrosophic Decision-Making Trail and Evaluation Laboratory (nDEMATEL) technique and Analytics Hierarchy Process (AHP). The n-DEMATEL technique gives causes and effects of standards and sub-standards and, AHP calculates the contribution of criteria and sub-criteria. This combination effectively deals with the problem of security. Moreover, this combination saves time and provides good results. Tiwari et al. [25] were performed a literature study in the year 2017, which is related to the Big data analytics (BDA) in supply chain management in industrial sector between 2010 and 2016. This paper reveals that the dealing with Big data is concerning issue for the supply chain members. Big data includes huge number of datasets which is difficult to handle as it expands the technical abilities of the storage space system which become difficult to process and interpret. Supply chain practitioners and experts were searching for a technique which help them to how data is produced, captured, organised and examined and find important information. This paper focussed on the technique named as Big Data analytics, which is one of the best methods to overcome the aforementioned problems. Big Data analytics includes predictive analysis, data mining and statistical investigation. Big Data analytics allows the supply chain practitioners and experts to collect, store, share and analyse the huge number of datasets within the firm as well as among the firms. Big Data analytics examines the large amount of data to get useful information about the patterns and correlations, trends. Further, this paper emphasises on the supply chain analytics which includes the use of data analytics in SCM. Moreover, this paper discussed the descriptive, predictive and prescriptive analytics. The authors define that the descriptive analytics helps to analyse the past data for understanding the changes which has occurred in business. This analysis is useful to show average investments per customer, variations in sale and total stock in inventory per year. Predictive analytics is used to make predictions about the future by using the current and past data. Prescriptive analytics is the third face business analytics which include descriptive and predictive analytics. In prescriptive analytics decision should be taken based on the predictive and descriptive analytics. This review shows the applications of BDA in different supply chains such as finance, healthcare and manufacturing. Dhumale et al. [23] conducted a project-oriented study in which RFID is used as an important technique in cloud-based supply chain management system, for collecting and distributing the updated information about the delivery of products, cost of products and identification of products. The authors proposed a supply chain management system which helps to share all the information about the product among the supply chain actors. The system uses the RFID technique with the help of cloud platform. This system is basically divided into three units such as supplier unit, container unit and distributor unit. The supplier unit contains all information like as product ID, delivery time and date, this information is stored in catalogue which further share among suppliers and distributors using clouding computing platform. The clouding computing platform is handled by using a programming language Python, for uploading all information regarding product in google spreadsheet. The container unit consists of several elements such as a controller, Global System for Mobile
Effect of IoT in Supply Chain Management—A Review
291
(GSM) communication, Global Positioning System (GPS), display, Keypad, Integrated Circuits (ICs) and motor. The position of container is monitored by using GPS. This location tracking message is sent to the supplier as well as distributor through the GSM. The distributing unit uses supplier’s password to open the container. If the password is correct then the container is open, otherwise wrong password is shown on the display. The container is opened or closed with the help ICs and further ID of products in container are confirmed with the help of processor. After that information is uploaded on the google spreadsheet and also sent back to suppliers. Accorsi et al. [26] were conducted a research in the field of food supply chain using IoT, in year 2017. This paper is aimed to elaborate the approaches and goals to design and construct an IoT architecture for management, scheduling and control the food supply chain operations. For this, authors implement IoT in food supply chain (FSC) and proposed a simulate tool. This tool emphasises on different stages and processes of FSC and envision the different items, entities, resources and infrastructures. The tool permits the gaming and planning of the food supply chain processes in a userconvenient manner. This system permits the real-time analysis of the food supply chain for transporting the food in economical way without wastage of food and causes much environmental pollution. Cai et al. [27] were conducted an examination in the year 2019, on the construction of agricultural products in China, using IoT-based SCM system. This paper is aimed to analyse and visualise various problems such as collecting, storing and analysis of the information agriculture products being transported from source to end customer. The authors had proposed an IoT-based agriculture product SCM system in which the agriculture products wear RFID tags connected with reader through a wireless antenna. The antenna transfers the RF signal of the tags to the reader for tracing the agriculture product. The proposed system conveniently recognises distance more than 30 metrs with proper security during the transportation of the product. Further, the sensors and cameras are installed in the agriculture land to get real time information about the growth of crops, humidity, temperature, CO2 contents in the land, using cloud computing. Further, the mangers can access all the information about the product by using a simple button on the cloud computing platform. Kothari et al. [28] studied the several barriers in the supply chain management and how IoT acts as a solution. Further, this research particularly highlights the applications of IoT on supply chain management to facilitate the end-to-end visibility of inventory. The authors have proposed a conceptual model of IoT that enables the supply chain experts to achieve real-time tracking of inventory without any kind of duplication. Leng et al. [29] were conducted a research in the year 2018, on the IoT-based supply chain inspection system for inspecting agricultural products. This paper discussed the identification of agricultural products using RFID and also inspection of agricultural products in supply chain. Further, this paper more emphasises on the use of RFID in the manufacturing, processing, storing, transportation and inspection of the agricultural product supply chain. The authors have used the RFID technology for monitoring humidity of soil, composition, temperature and humidity of air, content of CO2 and more parameters. The RFID tags are connected with a
292
Gagandeep et al.
wireless sensor system, to collect and share the information to further instruments. The information can be controlled with automatic control of various parameters that help to get proper control on the production of crops and greenhouse. The findings of this research reveal the importance of the RFID technology in the performance of agricultural supply chain management and supply chain testing. Evtodieva et al. [30] were conducted a study in the year 2019, on the possibilities of IoT applications in intelligent SCM. This paper discussed various challenges faced by the different entities of supply chain due to the increase in material and information flow. Particularly, these challenges are sale forecasting, cost of products, inventory level and safety. To overcome of these challenges, IoT technologies are used in SCM. The authors also explain that IoT-based SCM is interface between the customer IoT and industrial IoT. The customer IoT includes mobiles and tablets, smart house technology, wearable fitness sensors, etc. and industrial IoT relates to Industry 4.0. The industrial IoT is the combination of automated production technology, Big Data Analytics and M2M technology. The industrial IoT is expressly adopted as compare to customer IoT. Awwad et al. [31] did a literature review in the year 2018, to discuss several applications and advantages of the Big Data analytics in supply chain management. The paper describes that the big data associated with the supply chain is huge amount and variety of high-velocity data that may be in structured or unstructured form. The big data analytics includes prescriptive analytics, predictive analytics and descriptive analytics which helps to effectively analyse the big data. This improves the way to predict customer demand, traceability, risk assessment and performance of supply chain. Further, the authors highlight that the Big Data analytics improves the reaction time of the organisations by 4.25 time. The implementation of Big Data Analytics in SCM still have challenges such as data quality, data scalability, security issues and time consumption so it was observed that the speed of creating infrastructure to endure the increasing data, requires to be improved. De Vass et al. [32] conducted an exploratory study in Australian retail sector. The authors explore the impact of IoT on supply chain integration and performance. In this study semi-structured interviews are conducted with managers to collect data and analysed with NVivo software. Further, it was revealed that implementation of Internet of things in SCM improves the logistics processes internally among the retailers and externally with supplier and customers. Outcomes of research that IoT helps in automatic data collection, inventory visibility in warehouses and information easily sharing between retailers of supply chain. Further, it heightens the performance and sustainability of firms. Use of IoT in SCM has financial impacts, environmental impacts and social impacts which help in sustainability of firm. This study also investigate that there were very limited study has been done on the integration of IoT and supply chain management. Shafique et al. [33] conducted a research to evaluate the impact of IoT abilities on Green Supply Chain Management (GSCM). This research introduced a relationship among IoT abilities, green training, green supply chain practices and energy consumption behaviour. Total 250 participants are involved in this survey. Further,
Effect of IoT in Supply Chain Management—A Review
293
the results of this research are empirically tested by using the introduced relationship. It is found that the implementation of IoT is advantageous for a consumer, supplier integration, and manufacturing organisations in supply chain management for improving its performance and execution. Musa and Dabo [34] conducted a literature review which focus on use radio frequency identification (RFID) in supply chain management in the period 2000– 2015. The finding of this study was showed that number of enterprises adopts RFID to improve the operational processes which helps to improve economic performance and competition in the market. By adopting RFID, costs can be reduced and value addition along the supply chain through process optimization, inventory management and logistics management [34]. Gerami and Sarihi [3] investigated the use of Internet of things in supply chain management to upsurge the productivity and efficiency by increasing the speed and precision in decision-making. IoT may change the key elements of SCM and this transformation will enhance the efficiency and performance of the supply chain. There are benefits to the implementation of SCM but at the same time, some challenges may require to be faced while implementing IoT. Some of the major challenges are information security and high investment. The important information may be stolen from the server by using computer skills so it becomes a security challenge. With the gradual growth and digitalization of the industrial sector, there is an immense need to provide modern education such data science, to the employees, it requires a major investment for execution. Thus, this also be a crucial barrier to the implementation of IoT. The literature summary of the detailed reviewed papers including used techniques, methodology and, their outcomes, are described in Table 1.
5 Discussions It is well known that the IoT plays vital role for tracking, tracing products in supply chain management. IoT technologies are widely adopted in SCM to overcome many problems which were faced by traditional supply chain and logistics. In SCM, the IoT technologies such as RFID and cloud computing are majorly used. Cloud-based RFID is becoming a popular area of research in these days. This is an effective way to develop hand-to-hand transparency in supply chain by using a combination of cloud computing and radio frequency identification technology. This further helps in the traceability of products and checking their status at anytime, anywhere. Also, this helps to execute real time security of the products by eliminating existing issues. Another IoT technique named as Big data analytics is found to be beneficial to analyse the large amount of data which is emitted from different sensors and other devices in supply chain processes. Big data analytics may consist many techniques which are used to clean and analyse the data collected from supply chain entities and processes. Big data analytics is also used for inventory management in the warehouses. Many of researches are mainly focussed on the supply chain processes of the
294
Gagandeep et al.
Table 1 Literature summary of detailed reviewed papers S. no
Author
Year
Technique/methodology used
Major outcomes
1
Ben-Daya et al.
2019
RFID tags, Wireless networks, cloud computing
Mainly focus on challenges faced by SCM and role of IoT to overcome those challenge
2
Abdel-basset et al.
2018
RFID, N-DEMATEL, AHP
Proposed framework is used to improve security criteria and identification process in SCM
3
Tiwari et al.
2017
Supply Chain Analytics
Highlight the major advantages of Big data Analytics in Supply chain
4
Dhumale et al.
2017
RFID, GPS and GSM
Enhances the transparency among the supplier and customers with sending the message to individual customer
5
Accorsi et al.
2017
IoT-based framework for food industries and distribution ecosystem
Facilitates the user-friendly planning of the food supply chain processes
6
Cai et al.
2019
RFID tags, sensors, Tracking of agricultural cameras, cloud computing products along with measurement of temperature, humidity, acceleration, etc
7
Kothari et al.
2018
IoT-based conceptual model
Highlights the importance of IoT in supply chain management to provide the end-to-end visibility of inventory
8
Leng et al.
2019
Different RFID tags and readers are used in the various stages of agricultural supply chain
The used technique helps for tracking the performance of agricultural products which further enhances the transparency in the detection of agricultural supply chain
9
Evtodieva et al.
2019
Analysing the actual informational technologies in supply chain
Discover SCM relationship between customer IoT and Industrial IoT
10
Awwad et al.
2018
Big Data Analytics
Highlights the various applications and barriers of Big Data Analytics in SCM (continued)
Effect of IoT in Supply Chain Management—A Review
295
Table 1 (continued) S. no
Author
Year
Technique/methodology used
Major outcomes
11
De Vass et al.
2020
Exploratory research
Highlights the financial impacts, environmental and social impacts in supply chain management which help in sustainability of firm
12
Shafique et al.
2018
IoT-based conceptual model
Explore benefits of IoT in SCM for a consumer, supplier integration, and manufacturing organisations
13
Musa and Dabo
2016
Literature review
This paper shows the growth of use of RFID from 2000–2015. Also discuss about the benefits, complexities while using RFID
14
Gerami and Saihi
2020
Literature review
Highlights the challenges for the implementation of IoT in SCM, information security and high investment are found as major challenges
manufacturing, transportation, and food supply sector. Few of the researches have been also conducted to smooth the supply chain processes of agriculture sector but to some extent. However, there is a need to perform more research in the field of agriculture supply chain to build the economy of country. Based on this review, it is also observed that the study on the IoT-based SCM is still confined to isolated field of healthcare and its supply chain processes.
6 Conclusion In this paper, the influence of IoT-based SCM is reviewed. Particularly, a detailed review of fourteen papers has been conducted. It is found that the IoT has been widely adopted in the supply chain processes of manufacturing, food supply and agriculture sector. Further, it is observed that the applications of IoT provide the real-time data collection, monitoring, security and transparency in the supply chain processes. Moreover, the combination of cloud computing and RFID provides an interconnection of several supply chain entities that helps in easy real time tracking and quantitative monitoring of inventory and products without any duplication in the supply chain. This leads the organisation towards a profitable end. Big Data Analytics is also found to be beneficial technique to analyse the big data associated with SCM that improves the customer satisfaction and risk assessment in the supply
296
Gagandeep et al.
chain management. Accordingly, this can be concluded that the contribution of IoT is inevitable to tackle with the difficulties associated with the traditional SCM. However, this study can be further extended to examine the influence of IoT including artificial intelligence applications, in the supply chain process of the healthcare sector.
References 1. Abdel-Basset M, Manogaran G, Mohamed M (2018) Internet of Things (IoT) and its impact on supply chain: a framework for building smart, secure and efficient systems. Futur Gener Comput Syst 86:614–628 2. Wu L, Yue X, Jin A, Yen D (2016) Smart supply chain management: a review and implications for future research. Int J Logist Manag 27(2):395–417 3. Gerami M, Sarihi S (2020) The impacts of Internet of Things (IOT) in supply chain management. J Manage Acc Stud 8(3):31–37 4. Ellis S, Morris HD, Santagate J (2015) IoT-enabled analytic applications revolutionize supply chain planning and execution. In: International Data Corporation (IDC) White Paper 5. Ferreira P, Martinho R, Domingos D (2010) IoT-aware business processes for logistics: limitations of current approaches. In: INForum 2010—II Simposio de Informatica, Braga, pp 611–622 6. Yuvaraj S, Sangeetha M (2016) Smart supply chain management using internet of things (IoT) and low power wireless communication systems. In: International conference on wireless communications, signal processing and networking, WiSPNET, pp 555–558 7. Shih CW, Wang CH (2016) Integrating wireless sensor networks with statistical quality control to develop a cold chain system in food industries. Comput Stan Interfaces 45:62–78 8. Yan J, Xin S, Liu Q, Xu W, Yang L, Fan L, Chen B, Wang Q (2014) Intelligent supply chain integration and management based on cloud of things. Int J Distrib Sens Netw 10(3):1–15 9. Thoben KD, Wiesner S, Wuest T (2017) Industrie 4.0 and smart manufacturing—a review of research issues and application examples. Int J Autom Technol 11(1):4–16 10. Zelbst PJ, Green KW, Sower VE, Reyes PM (2012) Impact of RFID on manufacturing effectiveness and efficiency. Int J Oper Prod Manag 32(3):329–350 11. Ren J, Manzardo A, Toniolo S, Scipioni A (2013) Sustainability of hydrogen supply chain. Part I: identification of critical criteria and cause-effect analysis for enhancing the sustainability using DEMATEL. Int J Hydrogen Energy 38:14159–14171 12. Kaur J, Sidhu R, Awasthi A, Chauhan S, Goyal S (2018) A DEMATEL based approach for investigating barriers in green supply chain management in Canadian manufacturing firms. Int J Prod Res 56:312–332 13. Sodero AC, Rabinovich E, Sinha RK (2013) Drivers and outcomes of open-standard interorganizational information systems assimilation in high-technology supply chains. J Oper Manag 31:330–344 14. Benlian A, Hess T (2011) Opportunities and risks of software-as-a-service: findings from a survey of IT executives. Decis Support Syst 52(1):232–246 15. Verdouw CN, Beulens AJM, van der Vorst JGAJ (2013) Virtualisation of floricultural supply chains: a review from an internet of things perspective. Comput Electron Agric 99:160–175 16. Christopher M, Peck H (2004) Building the resilient supply chain. Int J Logist Manag 15(2):1– 13 17. Faisal MN, Banwet DK, Shankar R (2006) Supply chain risk mitigation: modelling the enablers. Bus Process Manage J 12(4):535–552 18. Cao Q, Schniederjans DG, Schniederjans M (2017) Establishing the use of cloud computing in supply chain management. Oper Manag Res 10:47–63 19. Xu LD, He W, Li S (2014) Internet of things in industries: a survey. IEEE Trans Ind Inf 4:2233–2243
Effect of IoT in Supply Chain Management—A Review
297
20. Gnimpieba ZDR, Nait-Sidi-Moh A, Durand D, Fortin J (2015) Using internet of things technologies for a collaborative supply chain: application to tracking of pallets and containers. Proc Comput Sci 56:550–557 21. Jiang W (2019) An intelligent supply chain information collaboration model based on internet of things and big data. IEEE Access 7:58324–58335 22. Bonomi F, Milito R, Natarajan P, Zhu J (2014) Fog computing: a platform for internet of things and analytics. Big Data Internet Things Roadmap Smart Environ 546:169–186 23. Dhumale RB, Thombare ND, Bangare PM (2017) Supply chain management using internet of things. Int Res J Eng Technol 4(6):787–791 24. Ben-Daya M, Hassini E, Bahroun Z (2019) Internet of things and supply chain management: a literature review. Int J Prod Res 57:15–16 25. Tiwari S, Wee HM, Daryanto Y (2017) Big data analytics in supply chain management between 2010 and 2016: insights to industries. Comput Ind Eng 115:319–330 26. Accorsi R, Bortolini M, Baruffaldi G, Pilati F, Ferrari E (2017) Internet-of-things paradigm in food supply chains control and management. Proc Manuf 11:889–895 27. Cai L, Yan Y, Shen J (2019) Examination on the construction of agricultural products supply chain logistics system based on internet of things. Revista de la Facultad de Agronomia de la Universidad del Zulia 36(6):1937–1945 28. Kothari S, Jain S, Venkteshwar A (2018) The impact of IOT in supply chain management. Int Res J Eng Technol 5(8):257–259 29. Leng K, Jin L, Shi W (2019) Research on agricultural products supply chain inspection system based on internet of things. Clust Comput 22:8919–8927 30. Evtodieva T, Chernova D, Ivanova N, Wirth J (2019) The internet of things: possibilities of application in intelligent supply chain management. Adv Intell Syst Comput 908:395–403 31. wwad M, Kulkarni P, Bapna R, Marathe A (2018) Big data analytics in supply chain: a literature review. In: Proceedings of the international conference on industrial engineering and operations management, pp 418–425 32. De Vass T, Shee H, Miah SJ (2020) IoT in supply chain management: a narrative on retail sector sustainability. Int J Logistics Res Appl 1–20 33. Shafique MN, Rashid A, Bajwa IS, Kazmi R, Khurshid MM, Tahir WA (2018) Effect of IoT capabilities and energy consumption behavior on green supply chain integration. Appl Sci 8(12):2481 34. Musa A, Dabo AAA (2016) A review of RFID in supply chain management: 2000–2015. Glob J Flex Syst Manag 17:189–228. https://doi.org/10.1007/s40171-016-0136-2
Toolpath Generation of a Human Anatomical Shape for Double-Sided Incremental Forming Akshay Sahu, Prashant K. Jain, and Puneet Tandon
Abstract In this modern age, incremental sheet forming is a new revolutionary forming technique. This technique does not require any costly dies to form a sheet of metal into a predefined shape. Thus it reduces the time and cost of manufacturing of a die. This technique also makes the manufacturing of customized formable shapes possible with ease. In this research work, toolpath for human anatomical shapes with one feature (i.e., concave or convex) is generated for Double-Sided Incremental Forming (DSIF) setup. Double-Sided Incremental Forming process is the process where instead of one forming tool, two forming tools are in action; one tool is for forming the sheet, and the second tool is to support the sheet, and this second tool is known as supporting tool. The development of methodologies for toolpath, which will be used in the available DSIF setup, is generated by using MATLAB. There are various applications of DSIF fabricated components in the Medical field. Some examples are forming sheet metal cover for fracture in hand instead of bulky conventional plasters, forming of ankle support, skull prosthesis, etc. Keywords ISF · DSIF · Human anatomy · Toolpath
1 Introduction Incremental Sheet Forming (ISF) is one of the most flexible metal forming processes because of its high customization. ISF is the process in which a metal sheet is formed to a required shape using a predefined toolpath. The tool moves over a metal sheet and deforms the sheet to a required shape. In the past, many researchers have investigated the ISF in various aspects to obtain vigorous results. ISF depends on multiple factors and parameters, some of them, which have been identified by the research work, are sheet thickness, clamping force, forming force, working temperature, the thickness of the final product, and A. Sahu · P. K. Jain (B) · P. Tandon Department of Mechanical Engineering, Design and Manufacturing, PDPM Indian Institute of Information Technology, Jabalpur, India e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_23
299
A. Sahu et al.
Fig. 1 Application domains of ISF process
Incremental Sheet Forming
300
Automobile Industry Medical Implant Architectural Features Aerospace Industry
toolpath. One of the critical factors of ISF is toolpath. Surface finish, geometrical accuracy, and final shape of the component are affected by the toolpath in ISF. DISF is the modified variant of ISF. Since the last decade, Incremental Sheet Forming (ISF) has emerged as the most flexible forming technology. Incremental Sheet Forming (ISF) is based on the concept of IT-enabled design and manufacturing technology, which has been accepted by technologists across the globe. ISF is the manufacturing process that manufactures parts that use resources efficiently with a quick time. ISF offering fabricates complex shape parts, which increases its popularity in application fields such as automotive, aerospace, biomedical, and manufacturing industry. Figure 1 shows the applications of ISF in various areas.
1.1 Basic Process of ISF All the different ISF processes have the seven common steps of the process.
1.1.1
Design of CAD Model
A CAD model is one of the main prerequisites step of the ISF process. CAD model of required geometry, which is to be fabricated, is modeled using CAD packages such as SolidWorks, Fusion360, and AutoCAD. This step is indistinguishable for all the ISF processes.
Toolpath Generation of a Human Anatomical Shape for Double-Sided …
1.1.2
301
Pre-processing of CAD Model
In this step, the model generated by the CAD package is processed for the wall angles; all the wall angles should be kept below their yield limits. Otherwise, the ISF process will create cracks on the part.
1.1.3
Generation of STL File
In this step, the model designed by using the CAD package is converted into Standard Tessellation Language (STL) file format. The STL file format contains triangulated surface data of the model. In the ISF process, choosing an STL file format makes model handling easier.
1.1.4
Slicing
The generatedSTL file is exported to MATLAB, which is sliced into layers to get contour information. Following the basic principle of the ISF process, any model is fabricated in a layer-by-layer manner. Hence it is necessary to slice the model (Fig. 2).
1.1.5
Toolpath Generation
With the help of slice data information, the next step is to generate a toolpath for each slice. The generated toolpath is required for each slice. There is no contour filling in the ISF process, which exists in the additive manufacturing process.
1.1.6
Part Fabrication
Sliced data is converted to G and M code format which provides an input to the machine controller for the fabrication of parts. ISF machine fabricates parts layer by layer manner one layer at a time.
Fig. 2 Steps of ISF process
302
1.1.7
A. Sahu et al.
Post Processing
In the post-processing stage, the fabricated part required some finishing process that makes part useful for end-use applications. After fabrication of the part, cleaning and surface treatment is needed. If a single part is divided into two or more sections for fabrication, then after fabrication, they will be welded together for one individual part.
1.2 Classification of Incremental Sheet Forming (ISF) Sided The ISF process has been classified into three main categories. Single Point Incremental Forming (SPIF), Double-Incremental Forming (DSIF), and Two Point Incremental Forming (TPIF). TPIF is further classified into two categories one is Partial Die Two Point Incremental Forming, and another is Full Die Two Point Incremental Forming. This classification is based on the number of tooltip contact. SPIF is the basic Incremental Sheet Forming process with one forming tool and sheet arrangement while in DSIF, two tools and sheet arrangement are there. Two point incremental forming consists of sheet arrangement and a die. Die may have a partial or full shape of the geometry that will be fabricated by using TPIF. Figure 3 shows these processes. Blank Sheet Holder
Forming Tool
Sheet Supporting Tool
(a)
(b)
Blank Holder
(c)
Partial
Full Die
(d)
Fig. 3 Classification of ISF process a SPIF, b DSIF, c Partial die TPIF, d Full die TPIF
Toolpath Generation of a Human Anatomical Shape for Double-Sided …
303
Fig. 4 Double-sided incremental forming process
1.3 Double Sided Incremental Sheet Forming (DSIF) Double-Sided Incremental Forming (DSIF) is one of the processes in which the forming on the metal sheet is done by using two tools, primary and secondary. While the primary tool act as a forming tool secondary tool performs its role of being a supporting tool. DSIF is not a common process for the fabrication of parts among the other ISF process because of its complications in toolpath generation and setup. DSIF provides better surface finish and dimensional accuracy than SPIF. In the DSIF process, the forming tool moves on a predefined toolpath and it makes incremental deformation on the metal sheet while the supporting tool moves on a toolpath which is offset to the toolpath of the forming tool. The offset is such that tooltip canters are in a straight line and this straight line should be perpendicular to the sphered shape tooltip. The forming of the sheet takes place by both the tools simultaneously in a layer-by-layer manner as shown in Fig. 4. Tools move in X–Y planes for generating the first contour and then shifts to the second contour by step size depth in the Zdirection and so on, while the sheet is held on its position by using the blank sheet holders.
1.4 Advantages and Limitations of DSIF Double-Sided Incremental Forming (DSIF) is a typical forming process that fabricates more complex formable shapes. DSIF process has many advantages, mainly it build complex geometries easily with no need of a special tool or die, it has minimum
304
A. Sahu et al.
material wastage than machining, and conventional forming process, and it’s easy to operate and with less maintenance. DSIF process does not require tooling like other CNC operations for the fabrication of parts. DSIF is cheaper than conventional forming processes, and two point incremental forming processes which require costly dies. DSIF gives us freedom of customization which is not possible in the conventional forming process because it always requires a die to fabricate a part from the raw material. Although the DSIF process has various advantages, there are certain limitations associated with this process that restrict its application domain. Layerby-layer deformation of metal sheet leads to the staircase effect which affects surface quality and dimensional accuracy of the fabricated part. Post-processing is required to improve the surface finish of parts. DSIF parts have poor mechanical properties because of layer-by-layer fabrication. Although there are limitations associated with this process, the application of DSIF increased because of its easiness in the field of aerospace industries, automotive industries, architectural features, research field, and biomedical field (implants). Application of the DSIF process has been increased because of its ability to form the complex geometry easily.
1.5 Issues with DSIF Even though there are some advantages of the DSIF process still there is some issue that is accompanying this process which makes it futile for the fabrication of parts in manufacturing industries. Currently, small and very large size part fabrication is one of the issue that limits its use only for small end-use functional applications. The DSIF machine is built of a confined workspace that restricts the part size to a particular dimension. As the dimension of the workspace increases the capital cost of machine increases for large-size parts and due to slow tool movement, this effects on fabrication rate which is very slow for high volume manufacturing. Also as part size increases, have some constraints in the DSIF process, so the fabrication of largesize parts has to study parameters that mainly affect the process compared to the fabrication of small-size parts.
2 Research Background In this section, a brief on available literature related to work in the bio-medical field using Incremental Sheet Forming (ISF) is presented. Ambrogio et al. [1] have fabricated ankle support, which is worn around the ankle to protect it or for immobilization while allowing to heal minor injuries. Potran et al. [2] have fabricated a denture base that is used for replacing missing teeth and surrounding tissues in patients, and Karbowski [3] has fabricated a skull prosthesis, this fabricated prosthesis was the base for prosthesis thermoforming process. In the above research for producing
Toolpath Generation of a Human Anatomical Shape for Double-Sided …
305
a 3D model of required human anatomical shape, researchers have used Reverse Engineering (Laser Scanning), Coordinate Measuring Machine, and Tomographic Images, respectively (Figs. 5 and 6). Later Agbor et al. [4] have investigated “automated flexible forming strategy for geometries with multiple features in DSIF,” and Lingam et al. [5] have investigated “automatic feature recognition and tool path strategies for enhancing accuracy in DSIF.” Both the research is related to features recognition and flexible toolpath generation for features available in geometries. Where feature refers to a geometrical shape having some significance like the concave, convex, frustum of a pyramid, etc. as shown in Fig. 7, both researchers have used a CAD package to produce model having features.
Fig. 5 a Original denture, b First replica (low carbon steel), c Second replica (stainless steel) (Source Potran et al. [2])
Fig. 6 Formed prosthesis (Source Karbowski [3])
Fig. 7 Model having features (Source Lingam et al. [5])
Feature 2 Feature 1
306
A. Sahu et al.
Fig. 8 Formed component of the car skin (Source Scheffler et al. [6])
Scheffler et al. [6] have fabricated the exterior of a car body using Incremental Sheet Forming process. They have gone for two-point incremental forming (TPIF). In their research, it was proven that the sufficient surface qualities for prototypical applications in automobile industries can be produced. The formed metal sheet component of car skin had a deviation of up to 5 mm and was primarily caused by spring-back effect after fine and coarse trimming. They have used a CAD package to develop the CAD model of the exterior skin of the automobile. Figure 8 shows the formed component of the car skin using TPIF process.
2.1 CAD Model (STL Format) Generation Techniques Based on research literature there are four techniques available for the generation of the CAD model in STL format. These techniques are discussed below.
2.1.1
CAD Packages
Computer-Aided Design (CAD) package is software. This CAD packages have ranged from simple 2D vector drafting to surface modeling and 3D solid models. CAD package has been used for a very long time. If we know the dimension of any physical object then it’s possible to create that physical object by using these CAD packages. Major application areas of CAD packages are mechanical design, architectural design, interior design, and aesthetic design, etc. Currently, there are so many CAD packages available in the market, some of them are SolidWorks, Catia, Fusion 360, AutoCAD, etc. If dimensions of a human anatomical shape are available then, these CAD packages are capable of creating a human anatomical shape with ease.
Toolpath Generation of a Human Anatomical Shape for Double-Sided …
2.1.2
307
Coordinate Measuring Machine (CMM)
CMM is a device that measures the geometry of the physical objects by sensing discrete points on the surface of the object with a sensing element a probe. This sensing element is available in various types, depending on the machine this sensing element will be used. Various types of sensing elements are mechanical, laser, optical, and white light. The position of the sensing element may be controlled manually by an operator or controlled by a computer. The sensing element’s position is specified in terms of displacement from a reference point in a three-dimensional Cartesian coordinate system (i.e., XYZ axes).
2.1.3
Computed Tomography (CT) Scan
A computerized tomography (CT) scan is a medical imaging procedure that uses a series of X-ray images taken from different angles around the human anatomical shape and uses computer processing to create cross-sectional images of that specific bone, blood vessel, and tissues of the human anatomical shape. This process provides more information than a single plain X-ray does and on the other hand, it allows users to view inside the human anatomical shape without any external cutting. This technique is very useful when there are internal injuries inside the patient’s body and treatment of the patient can be started in the right direction before cutting the anatomical shape for any operations or implants.
2.1.4
Reverse Engineering (Laser Scanning)
As the name suggests, it’s a technique of regeneration of the CAD model of an already existing physical object with the help of laser. It’s the directed deflection of laser beams, visible or invisible. Reconstruction of a model of the existing physical body using laser scanning is very helpful in medical, aerospace, automobile, and other research fields. This technology has an advantage over other technologies because of its capability of customization. Customization leads towards comfort when it comes to the wearable bio-medical implants.
2.2 Available Toolpath Strategies for Incremental Sheet Forming Based on the literature, there are broadly four toolpath strategies available, which are shown in Fig. 9. Adaptive toolpath is used when we required a high surface quality formed product. Spiral toolpath has been generated by Malhotra et al. [7] in their research work.
308
A. Sahu et al.
(a)
(c)
(b)
(d)
Fig. 9 Toolpath strategies, a Constant height stepped contour toolpath, b Spiral toolpath, c Adaptive stepped contour toolpath, d Adaptive spiral toolpath
2.2.1
Constant Height Stepped Contour Toolpath
This toolpath is used in additive manufacturing, incremental sheet forming and in other CNC systems, i.e., milling, etc. Constant height stepped contour toolpath is one of the simplest toolpath strategies available for ISF part fabrication. In this toolpath strategy, a constant step height (h) is provided to the software. Using the slice height (h) CAD model is sliced and contour is generated. These generated contours will be used directly as a toolpath. Toolpath is shown in Fig. 9a.
2.2.2
Spiral Toolpath
In this toolpath strategy, the tool moves in a spiral motion. The movement of the tool is from inside-out or outside-in. In this strategy, constant step height is provided between the neighbouring revolutions of the spiral. This spiral toolpath is derived from a constant height stepped contour toolpath using interpolations of points. The spiral toolpath minimizes the fabrication time of the components. There is a better surface finish on the fabricated component because of the continuity of the toolpath. Toolpath is shown in Fig. 9b.
2.2.3
Adaptive Stepped Contour Toolpath
This toolpath strategy is a modified variant of constant height stepped contour toolpath. This toolpath strategy is used where the wall angle of the part has a drastic change. Due to this sudden change in the wall angle, the feature on the fabricated
Toolpath Generation of a Human Anatomical Shape for Double-Sided …
309
part will not be there if the part is fabricated by using constant height stepped contour toolpath. To overcome this problem adaptive stepped contour toolpath is used. This toolpath is about the insertion of contours between two already existing contours by calculating the optimum change in the wall angle of the part. These inserted contours will help in producing the shape change feature on the formed components. Toolpath is shown in Fig. 9c.
2.2.4
Adaptive Spiral Toolpath
This toolpath strategy is a modified variant of the spiral toolpath. This toolpath is used where the wall angle of the component has a sudden change. Due to this sudden change in the wall angle, change in the shape of the component will be lost during the forming process. Toolpath is shown in Fig. 9d.
3 Research Methodology In our research, the 3D model in STL (stereo lithography) file format has been used for the simplicity of the algorithm. STL file has several properties that can be used for feature reorganization and toolpath generation. Some properties are the uneven shape of facets, uneven distribution of facets, facets at the boundary, has an unpaired edge, and there is a sudden change in the facet normal when the geometry changes its shape suddenly. The flow chart in Fig. 10 shows the complete process of toolpath generation for human anatomical shape (i.e., forearm) having a single feature.
3.1 Generation of STL File for Required Human Anatomy STL model of human anatomical shape can be produced by using CAD packages and several other methods discussed in the research background section. Here reverse engineering has been used (laser scanning) for the generation of 3D model of a human anatomical shape (forearm), Laser scanner that is used for this purpose is MS KINECT ONE. CAD package that has been used to process the scanned STL file is SolidWorks. The scanned STL model is shown in Fig. 11a. After trimming the extra parts, i.e., the palm portion of the forearm and obtaining the correct geometry of the forearm. The STL model is subdivided into three parts (i.e., blue, red, and green) such that the wall angle of the geometry should not be greater than 60°, as shown in Fig. 11b. After the subdivision of the STL file, some operations on both ends of the forearm have been applied. So that the wall angle should be greater than 60°, then some transformation operations have been implemented on the subdivided parts for proper orientation, one example is shown in Fig. 12 (for the blue part).
310
A. Sahu et al.
Generate STL file of required human
SolidWorks
Input STL model in Extract facet coordinates and facet normal for all facet Provide slice height (h) Contour generaon by MATLAB Divide surface model into features (i.e. concave or convex)
If concave
If convex
First tool act as a forming tool
Second tool act as a forming tool
Toolpath (G-codes) generaon for first tool
Toolpath (G-codes) generaon for second tool
Fig. 10 Flow chart of the complete process
Fig. 11 a Scanned STL file of forearm, b Forearm subdivided into 3 parts
3.2 Slicing and Toolpath Generation The toolpath strategy that has been used in this research is the Constant Height Stepped Contour toolpath strategy. Simplicity is the main reason to choose this toolpath strategy. It’s easier to apply this algorithm of slicing the STL CAD model
Toolpath Generation of a Human Anatomical Shape for Double-Sided …
311
Fig. 12 Proper orientation of blue part
Importing ASCII STL CAD model
Providing constant step height (h)
Generation and Saving of Toolpath in .gcode Format
Providing DSIF process parameters
Slicing of STL CAD model
Fig. 13 Flow chart of the toolpath generation process inside MATLAB
using MATLAB tool. After, slicing and contour generation from the STL CAD model, a new algorithm has been implemented based on the format of the G-code accepted by the DSIF setup. This new algorithm generates the toolpath directly from the STL file of the model. The generated toolpath may be accepted by the other modern CNC machines,.gcode,.gcd, and.txt files can be easily generated using this algorithm..gcode file may have information related to the home position of the tool, spindle speed, coolant on/off, tool radius compensation, incremental or absolute coordinate system, feed rate, constant step height (h), etc. All this information can be added at the time of toolpath generation using MATLAB. Figure 13 shows the flow chart of the toolpath generation process inside MATLAB from the STL CAD model.
3.2.1
Importing STL CAD Model
The STL CAD model having ASCII representation has been imported into the MATLAB. This ASCII representation of the STL file has information for describing the surface of the object as a triangular mesh. Each triangular mesh is also known as
312
A. Sahu et al.
Fig. 14 ASCII representation of the same STL model in text format
a facet. This information has been stored in a simple plain-text format. Fig. 14 shows the ASCII representation of the same STL file in text form.
3.2.2
Providing Constant Step Height (h)
After importing the STL file in MATLAB, and before slicing the CAD model a step height (h) is provided as an input to the MATLAB. Slicing of the CAD model will be done according to the step height provided. This step height affects the fabricated component and the complete forming process. Large step height will decrease the fabrication time, the file size of the generated toolpath and it increases the staircase effect on the fabricated component, while small step height will increase the fabrication time, the file size of the generated toolpath, and it shows the decrease in stair-case effect on the fabricated component. The stair-case effect is the indication of the surface quality of the formed component, less stair-case effect indicates the high surface quality and vice versa.
3.2.3
Providing DSIF Process Parameters
The DSIF process parameters referred to the parameters used by the DSIF experimental setup to form a component from the raw metal sheet. These parameters include the home position of the tool, spindle speed, coolant on/off, tool radius compensation, incremental or absolute coordinate system, feed rate, etc. All the above parameters can be easily controlled using G and M codes. Table 1 shows some G and M codes used for the toolpath generation in this research.
Toolpath Generation of a Human Anatomical Shape for Double-Sided …
313
Table 1 G and M codes used in this research G-codes
Work-done
M-codes
Work-done
G01
Linear interpolation
M03
Spindle on (CW rotation)
G04
Dwell
M07
Coolant on (Mist)
G28
Home position of the tool
M08
Coolant on (Flood)
G90
Absolute positioning
M09
Coolant off
G91
Incremental positioning
M30
End of program, rewind and reset modes
G96
Constant cutting speed or surface speed
M94
Feed per minute
Fig. 15 Slicing of STL model in MATLAB with slice height 0.4 mm
3.2.4
Slicing of the STL Model
The slicing algorithm of an STL file is very simple, which is also used in various rapid prototyping systems using the STL file. A plane is placed at provided slice height. The intersection of facets (triangles) line segments present in the STL file with this plane is found by the program [8, 9], after finding intersection points, these points are joined such that they form a contour using first depth search graph algorithm. Similarly, after getting the first contour at first slicing plane, the programming algorithm moves to upcoming planes and generates contours until the upper dimension of the STL model is reached. The slicing of the STL model is shown in Fig. 15.
3.2.5
Generation and Saving of Toolpath in .gcode Format
After contour generation at each possible slicing plane, a new algorithm has been applied inside the MATLAB for generating the toolpath in.gcode format. The path
314
A. Sahu et al.
Fig. 16 Generated toolpath saved in .gcode format
of the generated contour has been used as a reference for the linear interpolation of the tooltip using G01 G-code. This linear interpolation has been done by using for loop nesting algorithm and generation of the toolpath in.gcode format has been done by using “FileID” and “fprintf” inbuilt commands of the MATLAB. At last, the generated toolpath will be saved by the name provided by the user in any of the three available text file formats. These three text format are .gcode,.gcd, and.txt. Figure 16 shows the saved toolpath in.gcode format and with the file name “example”.
4 Result and Discussion After the generation of toolpath in.gcd file format, we have performed some simulation on an open-source CAM website ncviewer.com, and it’s found that generated toolpath is correct for the available 10 Axes Double-Sided Incremental Sheet Forming Experimental Setup. The simulation of the toolpath is shown in Fig. 17. The generated toolpath is a constant height type of contour toolpath.
Toolpath Generation of a Human Anatomical Shape for Double-Sided …
315
Fig. 17 Simulation of toolpath in ncviewer.com
5 Conclusion This algorithm can rapidly generate a toolpath for a single feature (i.e., concave or convex). We can produce high surface quality on the fabricated part by decreasing the slice height (h), but this will result in increasing the fabrication time. This algorithm gives us freedom of toolpath generation for the customized single featured geometries, having all wall angle less than 60 degrees. The generated toolpath can also be simulated on a CNC setup equipped with a forming tool because the CNC setup also uses the same .gcd or .gcode file, which has been generated by our program for part fabrication. Toolpath generation for multi-featured (i.e., concave and convex) geometry can be implemented soon So that the level of customization increases. Toolpath for supporting tool concerning the toolpath of forming tool (offset toolpath) can be generated so that we can achieve more geometrical accuracy and fewer chances of cracks on the formed component. As the DISF setup is capable enough to form more complex shapes having more than one feature, so it is also possible to generate a unique single toolpath for human anatomical shapes like shoulder and face. Acknowledgements This research was financially supported by Impacting Research Innovation and Technology (IMPRINT) INDIA through the Project of the Government of India (Project Number: 5506). The authors wish to acknowledge the funding agencies.
References 1. Ambrogio G, De Napoli L, Filice L, Gagliardi F, Muzzupappa M (2005) Application of incremental forming process for high customised medical product manufacturing. J Mater Process Technol 162–163(Spec. Iss.):156–162. https://doi.org/10.1016/j.jmatprotec.2005.02.148
316
A. Sahu et al.
2. Potran M, Skakun P, Milutinovic M (2014) Application of single point incremental forming for manufacturing of denture base. J Technol Plast 39(2):15–24 3. Karbowski K (2015) Application of incremental sheet forming. Manage Prod Eng Rev 6(4):55– 59. https://doi.org/10.1515/mper-2015-0036 4. Ndip-Agbor E, Ehmann K, Cao J (2018) Automated flexible forming strategy for geometries with multiple features in double-sided incremental forming. J Manuf Sci Eng Trans ASME 140(3). https://doi.org/10.1115/1.4038511 5. Lingam R, Prakash O, Belk JH, Reddy NV (2017) Automatic feature recognition and tool path strategies for enhancing accuracy in double sided incremental forming. Int J Adv Manuf Technol 88(5–8):1639–1655. https://doi.org/10.1007/s00170-016-8880-1 6. Scheffler S, Pierer A, Scholz P, Melzer S, Weise D, Rambousek Z (2019) Incremental sheet metal forming on the example of car exterior skin parts. Proc Manuf 29:105–111. https://doi. org/10.1016/j.promfg.2019.02.112 7. Malhotra R, Reddy NV, Cao J (2010) Automatic 3D spiral toolpath generation for single point incremental forming. J Manuf Sci Eng Trans ASME 132(6):1–10. https://doi.org/10.1115/1.400 2544 8. Adnan FA, Romlay FRM, Shafiq M (2018) Real-time slicing algorithm for Stereolithography (STL) CAD model applied in additive manufacturing industry. IOP Conf Ser Mater Sci Eng 342(1). https://doi.org/10.1088/1757-899X/342/1/012016 9. Hu J (2017) Study on STL-based slicing process for 3D printing. Solid Free Fabr 885–895
Optimization of Cutting Forces in Dry Turning Process Using Taguchi and Grey Relational Analysis Sumit Verma, Vipin Kakkar, and Hari Singh
Abstract This paper deals with the optimisation of numerous criteria problems in the turning process of EN-8 steel by using Taguchi and Grey Relational Analysis. In order to investigate the machining parameters’ performance characteristics like nose radius, cutting speed, depth of cut and feed, considering multiple responses, i.e., tangential force and feed force, an orthogonal array L18, grey relational coefficients, grey relational grade and analysis of variance (ANOVA) are estimated. The study gives the lower magnitude of the feed force and tangential force and also optimal cutting parameters. The experiment achieved an average Grey Relational Grade of 0.70449, and the prediction of GRG for the optimal setting is 0.86235, which proves the improvement in GRG by 0.15787. The largest Grey Relational Grade is for experiment number 4, which gives the optimum parameter setting for the experiment as A1B2C1D1. The main effect plot (Fig. 3) shows the low values for tool nose radius, feed rate and depth of cut resulting in the minimum tangential force, which is 98.066° N and optimised value of the feed force, which is 107.873° N. The response table and response graphs are shown and ANOVA table reveals that significant factors are depth of cut, feed rate and nose radius in affecting the experimentation at 95 percent confidence level. Keywords Optimization · Turning process · EN 8 Steel · Grey relational analysis · Taguchi · ANOVA
S. Verma (B) Department of Supply Chain & Strategic Sourcing, Hero Motocorp Ltd., Vadodara, India V. Kakkar Department of Mechanical Engineering, Indian Institute of Technology, Delhi, India H. Singh Department of Mechanical Engineering, National Institute of Technology, Kurukshetra, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_24
317
318
S. Verma et al.
1 Introduction From the last few years’ multi-response optimization in manufacturing industries has become a challenging issue. In the present situation, cutting forces in machining operations need to be decreased as the lower cutting forces have many benefits, like surface finish improvement, lower power consumption, burr-free surfaces, better part quality and low-stress components. Turning operation machining requires the operation on a lathe machine and is generally used for producing conical or cylindrical components. Grinding, boring, curved surfaces and flat faces on a lathe machine can be achieved with standard attachments. It is, therefore, necessary to increase tool life, to decrease the cutting forces, to improve surface accuracy and chip thickness through an optimization study in turning operations. Cutting forces are the most important of these four characteristics in terms of power consumption in the performance of a turned component. In order to examine the performance of cutting tools, the relative forces during cutting involved in a turning process are significant in machine tool design and selection. During the process, these forces must withstand by the machine tool and its components without experiencing major vibrations, chatter or deflection. The determination of cutting forces required for distortion of the work material in the zone of shear is essential for a few significant necessities: • To assess the power necessities of a machine tool; • To assess the straining actions that the components, bearings, jigs and fixtures of the machine tool must resist; • To determine the role of different parameters in the cutting forces; • To determine the output in terms of machinability (cutting forces) of any new work material, tool material, techniques, environment etc. Faxial = axial or feed force acting in the tool travel direction Fradial = radial force acting normal to the machined surface Ftangential = tangential or cutting force acting in the cutting velocity direction Cutting speed, doc (depth of cut), nose radius of tool and rate of feed are all turning parameters that have a major impact on efficiency. It is important to choose the most suitable machining conditions for the purpose of refining the quality of machined components, maximise the performance of machining operation, and minimise the costs of machining. It is incredibly difficult to achieve the best possible machine efficiency since there are so many customizable machining parameters. Cutting condition selection necessitates the use of analytical techniques in order to alleviate these machining issues. This article explains how to increase the efficiency of turning parameters with a variety of performance characteristics, like tangential as well as feed force, using factorial architecture and Taguchi-based grey relational analysis. Turning process parameters like nose radius of tool, rate of feed, cutting speed and doc (depth of cut) will impact tangential as well as feed force. The parameter design by Taguchi is an essential tech-tool for sturdy design, providing a methodical strategy to parameter optimization in terms of quality, cost
Optimization of Cutting Forces in Dry Turning Process …
319
and efficiency. This versatile parameter design scheme offers a fast and easy way to construct products and processes that perform reliably and optimally under a variety of conditions. Robust design can (a) make product output insensitive to variations in raw materials, allowing for the use of lower-grade alloys and components in most cases; (b) make designs robust against variations in production, lowering the cost of labour and material for rework and scrap; (c) make the design least vulnerable to variations in operating conditions, resulting in improved product reliability and (d) using a newly structured development process so that engineering time is used more productively. However, the Taguchi technique could be used to optimise process parameters for a single output criterion. The optimization of multiple output characteristics, on the other hand, is more common in industrial processes. The relationships between different factors in a complex and multivariate system like machining are uncertain. Such systems are generally referred to as grey because they provide unreliable, incomplete and uncertain data. In order to solve such problems, grey relational analysis is required. This paper uses Taguchi-based grey relational analysis to optimise the process parameters during the turning operation of EN-8 steel for the purpose of reducing the value of the forces involved in cutting operation.
2 Research Background Turning is a single-point cutting tool method whereby the cutting tool is held parallel to the surface of rotating workpieces in which a certain rate of feed and doc (depth of cut) are provided to the cutting tool to penetrate the workpiece to cause the material removal. Based on the Taguchi method with several output characteristics, Nian et al. [1] suggested using a multi-response signal-to-noise ratio, orthogonal array and ANOVA (analysis of variance) to optimise turning operations. Multiple performance characteristics like cutting power, surface finish and tool life were taken into account when optimising three cutting parameters: doc (depth of cut), cutting speed and rate of feed. Lin [2] proposed a framework for optimising turning process operations with a number of performance characteristics based on the Taguchi method and grey incidence analysis model to achieve a grey relational grade. While machining EN24 steel workpiece with a tool that has TiC-coated tungsten carbide inserts, Singh and Kumar [3] discovered a consistent setting of the parameters of the turning mechanism (rate of feed, doc and cutting speed) that resulted in an optimal cutting force. Aggarwal and Singh [4] used the most up-to-date optimization techniques, such as scatter search, fuzzy logic, RSM, Taguchi approach, and genetic algorithms, to investigate the optimisation of parameters of machining involved in the operation of turning. While machining EN24 steel workpiece with a tool that has TiC-coated tungsten carbide inserts, Singh and Kumar [5] investigated how to set the turning process parameters (feed, depth of cut and cutting speed) to achieve the best feed force. Paiva et al. [6] presented a mixed approach that combined PCA (principal component analysis) and
320
S. Verma et al.
RSM (response surface methodology) to improve responses that are all connected in an AISI 52,100 hardened steel turning phase. In grinding AISI 316 stainless steel, Manimaran and Kumar [7] used Grey relational analysis to optimise multi-response optimization. For hard machining, Gopalsamy et al. [8] have employed grey theory method and ANOVA to optimise process parameter. Abhang and Hameedullah [9] expounded the grey incidence analysis model application for optimization of parameters for the chip thickness and surface roughness of the workpiece by including different turning process parameters, like cutting speed, tool nose radius, feed and solid–liquid lubricant concentration. In dry conditions, the effects of CNC turning process parameters on radial forces, surface roughness and feed of EN19/AISI4140 were investigated by Kabra et al. [10]. To investigate the output characteristics of CNC turning process, the S/N ratio (signal-to-noise ratio), ANOVA (analysis of variance) and OA (orthogonal array) were included. Selvaraj et al. [11] have employed Taguchi method to optimise the dry turning parameters of two separate grades of stainless steel of nitrogen alloyed duplex. The turning operations were performed using TiCN, and TiC-coated carbide tool inserts. The results showed that the rate of feed has a remarkable impact on cutting force and surface roughness. The cutting speed was found to be the most important factor influencing the tool wear. Using scanning electron microscope imaging, tool wear was studied. Confirmation tests were carried out under perfect cutting conditions.
3 Research Methodology 3.1 Design of Experiments The experimental design or design of experiments (DOE) is the method of collecting knowledge in which there is variety, whether or not under the experimenter’s full control. The experimenter is also interested in the impact of some process or action on some objects, which may be humans, plants, animals, materials or something else. As a consequence, experimental design is a discipline with a broad range of applications in the natural and social sciences, as well as engineering. In production, industrial research and development DOE is used in the following areas: • • • • • •
Manufacturing processes optimization Screening as well as recognition of critical factors Method robustness testing Product robustness testing Experiments with formulation Analytical instruments optimization.
Only one variable at a time is varied in the conventional one variable at a time system, with the remaining variables remaining constant in the experiment. The effects of the selected variable on the response under a certain set of conditions are
Optimization of Cutting Forces in Dry Turning Process …
321
revealed by this method of experimentation. The biggest drawback of this strategy is that it does not explain what might occur if the other variable were to shift at the same time. The effect of the interaction among the variables on the response characteristics is not possible to analyse using this approach. An interaction occurs when one factor fails to have an identical impact on the response at various levels of another variable. Complete factorial architecture, on the contrary, necessitates the collection of experimental observations for all conceivable combinations of the factors in the analysis, necessitating the conduct of a large number of trials. As a result, for experiments involving a comparatively larger number of variables, just a small fraction of the combinations of factors that provide the majority of the information are chosen. This is referred to as the fractional factorial design of experiments.
3.2 Taguchi’s Method This method is a well-known method for optimising design parameters. This approach was developed to improve product quality by integrating statistical and engineering principles. The procedure, which is based on the orthogonal array (OA), decreases the experiment’s variance considerably, resulting in the optimal setting of process parameters. With fewer experiment cycles, OA offers a well-balanced range of experiments. This approach is used for data processing as well as predicting the best outcomes. The mean (signal) to the standard deviation (noise) ratio is known as the S/N ratio. This ratio is determined by the product’s quality characteristics to be optimised. Nominal the best (NB), Higher-the-better (HB) and Lower-the-better (LB) are the three most common S/N ratios. The combination of parameters that has the highest S/N ratio is the best configuration. The smaller-the-better theory is used in this study to reduce tangential and feed forces. The corresponding loss function is expressed as follows: S/N = −10 log10
m 1 2 y m i=1 i
(1)
where y represents the observed data, and m represents the number of observations.
3.3 Grey Relational Analysis The Grey Relational theory set up by Dr. Deng incorporates grey modelling, prediction, decision-making and grey relational analysis of a system wherein the model is uncertain or the data is inadequate. This approach offers an effective key result to the multi-input, uncertainty and discrete data type problem. The GRA can be employed to evaluate the relationship
322
S. Verma et al.
between machining parameters and the performance of machining operation. The grey relational grade can give information about the factors affecting response variables. Many researchers improved their experimental results by implementing the Taguchi–Grey relational analysis in a combined way. The grey relational analysis starts with the grey relational generation. The tangential force and feed force are normalised in the range of zero to one during this process. The grey relational coefficient is determined from the normalised data to represent the link between the desired and real tangential and feed forces. The grey relational grade is then evaluated by averaging each performance characteristic’s grey relational coefficient. Complete assessment of a variety of performance characteristics depends on the grey relational grade (GRG). Therefore, optimising complex multiple performance characteristics would be reduced to optimising a single grey relational grade (GRG). The optimum process parameter level is the one that has the largest grey relational grade. In view of the above discussion, the following steps are involved in optimising turning operations with multiple output characteristics utilising the Taguchi approach and grey relational analysis: Step 1: S/N ratio normalisation Initially, it needs to be normalised the S/N ratio so that raw data can be prepared for review or analysis, which includes moving the initial sequence to a comparable sequence. The following formula is used to normalise the S/N ratio. Normalised S/N ratio for the lower the better: max(yi j , i = 1, 2, ..., n) − yi j (2) zi j = max(yi j , i = 1, 2, ..., n) − min(yi j , i = 1, 2, ..., n) Step 2: Deviation sequences estimation, 00j The absolute difference between the comparability sequence y j (l) after normalisation and the reference sequence y0 (l) is the deviation sequence 0j . It is estimated by using Eq. 3 as follows: 0 j (l) = y0 (l) − y j (l)
(3)
Step 3: Grey relational coefficient (GRC) calculation GRC expresses the relationship between the perfect (best) and real normalised Signalto-Noise ratio for all sequences. (min + ξ max ) γ (y0 (l), yi (l)) = 0 j (l) + ξ max
(4)
Optimization of Cutting Forces in Dry Turning Process …
323
where j = 1, 2, 3, 4, …, n here n denotes the number of experimental data items; and l = 1, 2, 3, 4, …, k, here k denotes the number of responses. sequence; y0 (l) is the reference sequence. y j (l) is the specific comparison 0 j (l) = y0 (l) − y j (l) is the absolute difference between y0 (l) and y j (l). min = y j (l) minimum value max = y j (l) maximum value The distinguishing coefficient is represented by ξ, and the value for ξ lies in the range of [0, 1]. Step 4: Weighted grey relational grade (GRG) estimation Complete assessment of the multiple performance characteristics depends on the grey relational grade, and it is defined by Eq. 5 as the average sum of the grey relational coefficients: 1 γi j k i=1 k
γj =
(5)
For jth experiment grey relational grade is represented by γ j and the performance characteristic numbers by k. The higher is the grey relational grade; nearer is the parameter to the ideal sequence. Step 5: Representation of each level of machining parameters in response table and response graph The response table and response graph for every level of machining parameters represent the average of the grey relational grade for every level of cutting parameters as well as the absolute average of the grey relational grade. Step 6: Determine optimum parameters For further study, the grey relational grade determined for each sequence is used as a response. The GRG was analysed using the larger-the-better quality characteristic. Analysis of variance (ANOVA) is used to analyse the grey relational grade obtained using Eq. 5. ANOVA is a statistical method for categorising the influence of individual variables. Henceforth, ANOVA is used to calculate the impact of each factor as well as the percentage contribution of every factor to the responses, using the grey relational grade values. Step 7: Grey relational grade prediction for optimum machining parameters Equation 6 is used to determine the projected grey relational grade by utilising the optimal turning process parameters as γ = γm +
q i=1
(γi − γm )
(6)
324
S. Verma et al.
The number of turning process parameters that have a significant impact on the tangential force and feed force is denoted by q, mean grey relational grade is denoted by γm and grey relational grade at the optimal level by γ i . As a result, optimising multiple factors is reduced to optimising a single grade value.
4 Experimental Work 4.1 Material In this study, a medium strength steel (EN8) with a diameter of 27 mm and a length of 500 mm was used. Table 1 shows the combination of chemical elements for EN8 alloy steel. This steel is medium carbon steel that is unalloyed and has high strength in tensile nature. This type of steel is generally provided in the rolled condition or in cold drawn form. EN8 is typically used in applications that need properties that are superior to mild steel. EN8 has a strong surface hardness and moderate wear resistance when flame or induction hardened. EN-8 steel is medium strength steel, which is suitable for the manufacturing of stressed pins, shaft studs and keys.
4.2 Present Problem Since the analysis used two levels of tool nose radius, therefore Taguchi mixed level design was chosen. The three parameters were selected at three levels. The two-level parameters each had one degree of freedom, and the other three parameters each had two degrees of freedom, for a total of six degrees of freedom. As a result, a total of 7 [=1 × 1 + 3 × 2] degrees of freedom were required. L18 (21 × 37 ) was the most appropriate OA in this case. L18 OA was selected for the experiments implementing Taguchi’s design approach. Table 2 displays the parameters chosen, as well as the designated symbols and its ranges. The machining experiments were performed on a conventional lathe with the following specifications: range of spindle speed—42 to 2040 rpm; range of DOC (depth of cut)—0.2 to 4.8 mm; range of feed rate—0.04 mm/rev. The control range for the machine is 220 V, and the rated current is 23A. The carbide cutting inserts having the geometry of the cutting tool CCMT 060,204 EN/060,208 EN with nose radius of 0.4 and 0.8 mm were used for machining of EN8 steel. The Table 1 EN8 steel chemical elements composition Element
C
Mn
Si
S
P
% Weight
0.36 − 0.45
0.60 − 1.0
0.05 – 0.35
0.06
0.060 Max
Optimization of Cutting Forces in Dry Turning Process …
325
Table 2 Various factors and their levels Symbol
Factor
Levels
A
Nose radius (mm)
2
Level 1 0.4
Level 2 0.8
Level 3
B
Cutting speed (m/min.)
3
41.56
60.65
78.88
C
Feed rate (mm/rev.)
3
0.04
0.08
0.16
D
Depth of cut (mm)
3
0.6
0.8
1
cutting environment throughout the study is dry. For measuring the cutting forces, the dynamometer is used, which is manufactured by the DEE Ltd. To achieve the study’s target, a simplistic multi-criterion methodology using Taguchi’s design approach and GRA (grey relational analysis) is employed. Table 3 displays the observed response parameter values. Table 3 Summary of the test data for the tangential force as well as feed force Experimental data Exp. no
Nose radius
Cutting speed
Feed rate
Depth of cut
Tangential force (N)
Feed force (N)
1
1
1
1
1
102.067
112.776
2
1
1
2
2
147.099
137.293
3
1
1
3
3
362.846
392.266
4
1
2
1
1
98.066
107.873
5
1
2
2
2
191.229
176.519
6
1
2
3
3
318.716
338.329
7
1
3
1
2
132.389
137.293
8
1
3
2
3
201.036
196.133
9
1
3
3
1
225.553
220.649
10
2
1
1
3
137.293
191.229
11
2
1
2
1
132.389
88.259
12
2
1
3
2
230.456
132.389
13
2
2
1
2
107.873
161.809
14
2
2
2
3
171.616
122.583
15
2
2
3
1
137.293
122.583
16
2
3
1
3
122.583
112.776
17
2
3
2
1
137.293
161.809
18
2
3
3
2
176.519
102.969
326
S. Verma et al.
5 Experimental Data and Analysis In this study, the value of both the forces, tangential as well as feed forces, are reported in Table 3. Lower forces indicate better performance in the turning process. Both the forces, tangential as well as feed forces, are selected as ‘lower the better’ for the pre-processing of data in the process of grey relational analysis. Finding the S/N ratio of the responses to be optimised is the first step in the analysation using Taguchi’s approach. For tangential and feed forces, the smaller the better is applied in this turning process. Equation 1 is used for the ‘smaller the better’. After that, the grey relational generation is determined in the grey relational analysis. The tangential force and feed force are normalised in the range from zero to one during this step. The S/N ratio is normalised by the formula used in Eq. 2 for the smaller the better. After the normalisation or grey relational generation, the deviation sequence can be determined utilising Eq. 3. The grey relational generation values or normalised values and deviation sequence values are illustrated in Tables 4 and 5, respectively. The coefficients and grade of grey relational analysis, and ranking for each and every experiment with a factorial design are illustrated in Table 6. The higher value of the grey relational grade, the closer the related experimental result is to the ideal normalised value. Table 4 Normalised value of tangential force and feed force
Exp. no
Normalisation Tangential force
Feed force
1
0.98489
0.91935
2
0.81482
0.83871
3
0
0
4
1
0.93548
5
0.64815
0.70968
6
0.16667
0.17742
7
0.87037
0.83871
8
0.61111
0.64516
9
0.51852
0.56452
10
0.85185
0.66129
11
0.87037
1
12
0.5
0.85484
13
0.96296
0.75806
14
0.72222
0.88709
15
0.85185
0.88709
16
0.90741
0.91935
17
0.85185
0.75806
18
0.70370
0.95161
Optimization of Cutting Forces in Dry Turning Process … Table 5 Value for deviation sequence
Exp. no
327
Deviation sequence Tangential force
Feed force
1
0.01511
0.08065
2
0.18518
0.16129
3
1.00000
1.00000
4
0.00000
0.06452
5
0.35185
0.29032
6
0.83333
0.82258
7
0.12963
0.16129
8
0.38889
0.35484
9
0.48148
0.43548
10
0.14815
0.33871
11
0.12963
0.00000
12
0.50000
0.14516
13
0.03704
0.24194
14
0.27778
0.11291
15
0.14815
0.11291
16
0.09259
0.08065
17
0.14815
0.24194
18
0.29630
0.04839
From Table 6 and Fig. 1, it can be visualised that experiment no. 4 has the optimum multi-performance characteristics among the18 experiments since experiment no. 4 has the largest value of grey relational grade (Fig. 2). For every level of the turning process parameters, the mean of the values of the grey relational grade was determined using the same procedure. The grey relational grade indicates the degree of correlation between the comparability and reference sequences, a higher value grey relational grade indicates that the comparability sequence is more firmly associated with the reference sequence. Furthermore, ANOVA is used to assess which process parameters are statistically important. The optimal setting combination of the process parameters can be foretold by statistical analysis of variance and grey relational analysis. Also, the main effects are plotted as shown in Fig. 3 and show the low values for depth of cut, feed rate and tool nose radius.
328
S. Verma et al.
Table 6 Grey relational coefficients, grade and ranking Exp. no
Grey relational coeff. tangential force
Grey relational coeff. feed force
Grey relational grade
Rank
1
0.97067
0.86111
0.91589
2
0.72973
0.75610
0.74291
9
3
0.33333
0.33333
0.33333
18
4
1.00000
0.88571
0.94286
1
5
0.58696
0.63265
0.60981
14
6
0.37500
0.37805
0.37652
17
7
0.79412
0.75610
0.77511
7
8
0.56250
0.58490
0.57370
15
9
0.50943
0.53448
0.52196
16
10
0.77143
0.59615
0.68379
12
11
0.79412
1.00000
0.89706
3
12
0.50000
0.77500
0.63750
13
13
0.93103
0.67391
0.80247
5
14
0.64286
0.81579
0.72932
10
15
0.77143
0.81579
0.79361
6
16
0.84375
0.86111
0.85243
4
17
0.77143
0.67391
0.72267
11
18
0.62791
0.91176
0.76984
8
2
Average grey relational grade = 0.70449
6 Results and Discussion 6.1 Optimal Parameter Setting Equations 2–5 were used to determine the rank of each experiment and the grey relational grade, and Table 7, the Response Table, shows the results obtained. Since the responses are assigned equal weighting, the distinguishing coefficient is represented by ξ and its value is chosen to 0.5 for the study. As shown in Table 6, Exp. performed at 4th No., has the largest grey relational grade, is clearly the best parameter setting for the 3 responses. For each parameter level, the average values for grades obtained in grey incidence analysis model were determined and described in the Table 7. Also, the factor effects at various levels of turning parameters are shown in Fig. 3 for the grey relational grade of tangential force as well as feed force (Fig. 4).
Optimization of Cutting Forces in Dry Turning Process …
329
Set Quality Characteristics(Responses)
Control factors & their levels
Influential factors
Choose orthogonal array
Proceed with experiment
Data analysis
S/N Ratio
Grey relational analysis
Response graph & table Estimated optimal parameters combination Analysis of variance Confirmation experiment
Optimal parameters combination
Fig. 1 Flowchart for the stepwise optimization using GRA
6.2 Optimal Parameter Combination The parameter’s output at the particular level is represented by the GRG (grey relational grade) value. As a consequence, the optimum parameter level is the one with the largest average grey relational grade. A2 B2 C1 D1 is the best process parameter configuration.
330
S. Verma et al.
Grey Relational Grade
GRG 1 0.8 0.6 0.4 0.2 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Experiment No.
Fig. 2 Grey relational grade
Fig. 3 Main effects plot for means Table 7 Table of response values for the average grey relational grade Symbol
Cutting factors
Level 1
Level 2
A
Nose radius
0.64357
0.76541
B
Cutting speed
0.70175
0.70910
Level 3 0.70262
Max–Min
Rank
0.12184
3
0.00735
4
C
Feed rate
0.82876
0.71258
0.57213
0.25663
1
D
Depth of cut
0.79901
0.72294
0.59152
0.20749
2
Optimization of Cutting Forces in Dry Turning Process …
331
Response Graph 0.64357
0.76541
1.00000
0.70910 0.70262
0.70175
0.82876
0.79901
0.71258
0.72294 0.59152
0.57213 0.80000 0.60000 0.40000 0.20000 0.00000
Nose Radius (A)
Cutting Speed (B)
Feed Rate (C)
Level 1
Level 2
Depth of Cut (D)
Level 3
Fig. 4 Graph for response values of average grey relational grade for tangential force and feed force
6.3 Most Effective Factor In Table 7, the difference of grey relational grade’s maximum and minimum values is also shown. The maximum of the difference values has the greatest influence on multi-response characteristics. The difference’s maximum value is 0.25663, which is equal to the governing factor rate of feed. The progressive system of the significance of controllable variables can be recorded as feed rate (C), depth of cut (D), tool nose radius (A) and cutting speed (B). The ANOVA was used to assess the significant study variables focusing on the grey relational grade [8]. For the experiment, the factors doc, tool nose radius and rate of feed are important as their determined values exceeded the tabular values. The ANOVA results are shown in Table 8. The feed rate had the greatest impact (36.69%) on the efficiency of turning process of the EN-8 alloyed steel, succeeded by the depth of cut (26.48%) and the nose radius (13.38%). Table 8 ANOVA Symbol
Factors
SS
DOF
MS
F-Value
Percentage Contr 13.38 0.04
A
Nose radius
0.06681
1
0.06681
6.55a
B
Cutting speed
0.00019
2
0.00009
0.01
39.69 26.48
C
Feed rate
0.19817
2
0.09908
9.72a
D
Depth of cut
0.13222
2
0.06611
6.49a
Error
0.10194
10
0.01019
Total
0.49932
17
a
Significant at 95% confidence level
332
S. Verma et al.
Table 9 Cutting performance outcome using initial and optimum conditions
Best parameter setting
Optimal conditions projected
Confirmation test
Setting
A1, B2, C1, D1
A2, B2, C1, D1
A2, B2, C1, D1
Tangential force
98.066
92.283
Feed force
107.873
101.508
Grey relational grade
0.70449
0.86235
0.92354
Grey relational grade improved by 0.21906
6.4 Confirmation Test To ensure that the analysis was accurate, this test was performed. According to Eq. 6, the value of grade obtained in grey incidence analysis (GRG) is 0.862354. To conduct the confirmation test, the grey relational grade was 0.923543. Table 9 shows the improvement made as a result of the confirmation experiment.
7 Conclusion The following are the conclusions reached as a result of the analysis: (1)
(2) (3)
The tool nose radius, depth of cut and feed rate are the most important factors influencing the turning operation of EN-8 steel, according to ANOVA table. The feed has a 39.69% influence, depth of cut having a 26.48% influence and the nose radius having a 13.38% influence. The feed rate has the strongest relationship with cutting forces of all the parameters tested. Also, when the process parameters are set to their optimum values, the predicted grey relational rating or grade improves from 0.70449 to 0.86235, indicating that the turning process is more efficient.
Optimization of Cutting Forces in Dry Turning Process …
333
Appendix Numerical example: Normalisation of S/N ratio Table 4 shows the normalised values for experimental no. 5. The normalised value for tangential force is calculated using Eq. 2. (362.846 – 191.229)/(362.846 – 98.066) = 0.64815. Similarly, the normalised value for feed force is calculated using Eq. 2. (392.266 – 176.519)/(392.266 – 88.259) = 0.70968. Deviation sequence The deviation sequence corresponding to experiment no. 5 in Table 5 is determined using Eq. 3. The deviation sequence for tangential force is determined using Eq. 3. 1 – 0.64815 = 0.35185. Similarly, the deviation sequence for feed force is determined using Eq. 3. 1 – 0.70968 = 0.29032. Grey relational coefficient (GRC) Table 6 shows the value of GRC for experiment no. 5. The GRC for tangential force is calculated using Eq. 4. [0 + 0.5(1)]/[0.35185 + 0.5(1)] = 0.58696. Similarly, the value of GRC for feed force is evaluated using Eq. 4. [0 + 0.5(1)]/[0.29032 + 0.5(1)] = 0.63265. Grey relational grade (GRG) The average of Eqs. 1, 2 in Table 6 is used to evaluate the value of GRG (for Tangential force and Feed forces). = (0.58696 + 0.63265/2. = 0.60981. The grey relational grade for experiment no. 5 (for Tangential force and Feed forces) is 0.60981 in Table 6.
334
S. Verma et al.
References 1. Nian CY, Yang WH, Tarng YS (1999) Optimization of turning operations with multiple performance characteristics. J Mater Process Technol 95(1):90–96 2. Lin CL (2004) Use of the Taguchi method and grey relational analysis to optimize turning operations with multiple performance characteristics. Mater Manuf Processes 19(2):209–220 3. Singh H, Kumar P (2005) Optimizing cutting force for turned parts by Taguchi’s parameter design approach. Indian J Eng Mater Sci 12:97 4. Aggarwal A, Singh H (2005) Optimization of machining techniques—a retrospective and literature review. Sadhana 30(6):699–711 5. Singh H, Kumar P (2006) Optimizing feed force for turned parts through the Taguchi technique. Sadhana 31(6):671–681 6. Paiva AP, Ferreira JR, Balestrassi PP (2007) A multivariate hybrid approach applied to AISI 52100 hardened steel turning optimization. J Mater Process Technol 189(1):26–35 7. Manimaran G, Pradeep Kumar M (2013) Multiresponse optimization of grinding AISI 316 stainless steel using grey relational analysis. Mater Manuf Process 28(4):418–423 8. Gopalsamy B, Mondal B, Ghosh S (2009) Optimisation of machining parameters for hard machining: grey relational theory approach and ANOVA. Int J Adv Manuf Technol 45(11– 12):1068–1086 9. Abhang LB, Hameedullah M (2012) Determination of optimum parameters for multiperformance characteristics in turning by using grey relational analysis. Int J Adv Manuf Technol 63(1–4):13–24 10. Kabra A, Agarwal A, Agarwal V, Goyal S, Bangar A (2013) Parametric optimization & modeling for surface roughness, feed and radial force of EN-19/ANSI-4140 Steel in CNC turning using taguchi and regression analysis method. 3(1):1537–1544 11. Selvaraj DP, Chandramohan P (2010) Optimization of surface roughness of AISI 304 austenitic stainless steel in dry turning operation using Taguchi design method. J Eng Sci Technol 5(3):293–301
Performance Evaluation of Cargo Warehouse Operations of an Indian Airline Using Discrete Event Simulation: A Case Study A. Tamizhinian and V. Madhusudanan Pillai
Abstract This research work primarily focusses on finding the bottleneck among the series of processes involved in the existing ‘cargo acceptance process’ in a warehouse of a company. Once pre-booking of cargo is done, the customers send their cargo through commercial vehicles to the company’s warehouse on the day of flight’s departure for acceptance and for carrying the cargo to specified destinations. The acceptance process involves booking, unloading, weighing and X-ray screening of cargo. After acceptance, cargo is stored in the warehouse and dispatched based on their flight departure timings, and the cargo gets loaded into the flight which is called ‘cargo uplifting process’. Due to congestion in the warehouse, long waiting time results. At times the customers tend to move away to the competitors to carry their cargo. So, the complete process involved in this air cargo business is studied to carryout bottleneck analysis. Discrete Event Simulation model, mimicking the cargo acceptance and cargo uplifting process being followed at warehouse, is developed to find out the bottleneck process. The analysis also helped to evaluate the current performance of the warehouse operations. The results pointed out ‘X-ray screening process’ as bottleneck and the current number of trucks used for cargo uplifting is sufficient enough to dispatch the accepted numbers. Also, alternative scenarios are analysed for a better decision-making. Keywords Bottleneck analysis · Air cargo operations · Warehouse loading and unloading · Discrete event simulation · ARENA
1 Introduction Generally, in a production environment, a bottleneck is one process in a chain of processes, whose confined capacity limits the capacity of the entire chain. Identifying bottlenecks is key for improving efficiency in the production line as it helps in locating A. Tamizhinian · V. M. Pillai (B) Department of Mechanical Engineering, National Institute of Technology, Calicut, Kozhikode, Kerala 673601, India e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_25
335
336
A. Tamizhinian and V. M. Pillai
the point of congestion in a production system. Bottlenecks can be found through: identifying the areas where accumulation occurs, throughput evaluation, assessing whether each machine is being utilised to its fullest capacity and determining the machine responsible for high waiting time. Increasing the capacity of an overall process relies on increasing the capacity of the bottleneck. The bottleneck analysis tool helps a team to identify the process steps where flow is constrained, find the root causes of those constraints and address the root causes that have been identified. It can be used when processes expectations are not met, demand is not keeping up or customers are dissatisfied. In this article, a warehouse which accepts the cargo from customers by prebooking, through a chain of activities like unloading the cargo on the warehouse floor, weighing the cargo and X-ray screening, is taken for the study. Basically, cargo found non-hazardous after screening gets accepted into the warehouse. Once accepted, cargo gets stored in the warehouse and dispatched through trucks, based on their departure timings, to get uplifted in the respective flights. It is observed that congestion in the warehouse floor restricts the smooth flow of cargo during the cargo acceptance process. Due to congestion in the warehouse, long waiting time results, leading to customers missing their booked flights and the respective cargo is pushed for the next immediate flight available, eventually causing a slight dilemma in the minds of customers who sometimes prefer the competitors, which is nothing but a ‘rejected demand’ for the company. Though the percentage of rejected demand is measured through qualitative assessment as 7% approximately, being one of the leading cargo service providers who always ensures better service to the customers, few customer’s dissatisfactions over long waiting time is the primary motivation behind this study. This research work studies and analyses the operations involved in cargo acceptance and cargo uplifting process in the warehouse to identify the bottleneck. The process flow of the cargo operations behaves the same as of a production line, instead of parts waiting for an operation, customers’ cargo waits for certain resources while flowing through the above-mentioned chain of activities. For analysis, tonnage of cargo accepted per day is considered as throughput and the operations/activities involved in acceptance and uplifting are treated as processes for building a discrete event simulation model. In essence, the performance evaluation of warehouse operations will contribute to the company with the following outcomes: (1) (2) (3)
Identification of bottleneck in cargo acceptance process Evaluation of adequacy of trucks used for dispatch in cargo uplifting process Evaluation of different ‘what-if’ scenarios in terms of percentage increase in throughput.
Performance Evaluation of Cargo Warehouse Operations …
337
2 Research Background In this section, the authors have provided a brief review of available literature related to discrete event simulation and the literature which helped in data collection using time study approach. Muhamad et al. [5] quoted ‘motion and time study’ as a scientific analysis method designed to measure the time required by an average worker for completing a given task in a stationary workplace. Hartanti [1] suggested that time study is the most widely used method in determining standard time. Lukodono and Ulfa [4] stated that the determination of the standard time is particularly important for production which still uses human labour as its main factor. Ungureanu et al. [8] said that the process of determining type of the distribution for a set of data usually involves what is known as the essence of fit test. These tests are based on some sort of comparison between the observed data distribution and a corresponding theoretical distribution. Sometimes a theoretical distribution that does make sense will be almost as good a fit. In these cases, it is to be decided by the researcher whether it makes more sense to use the best mathematical fit or a very close fit that makes sense. To get exposure in ARENA simulation software, Simulation with Arena, fourth edition by Kelton [2] is referred, as it helped the author in getting a sound knowledge in arena with chapters covering basic process in arena till the advanced level of modelling such as integrating arena with database. Liong and Loo [3] has presented the ARENA simulation model for warehouse loading and unloading and discussed how the arena could be used to optimise the residence time of any lorry in the undertaken warehouse. Shibin et al. [7] used discrete event simulation technique to solve a real time multi-criteria strategic capacity planning problem with multiple objectives such as throughput maximization, waste minimization and resource utilisation maximization. The discrete event simulation can be combined with value stream mapping to provide necessary information for decision problems encountered in lean manufacturing implementation [9].
3 Overview of the Existing Process in Warehouse An overview of the entire cargo operations being followed at the warehouse starting from booking of cargo till the cargo gets uplifted is broadly divided into two processes namely, cargo acceptance process and cargo uplifting process. In this section, those two processes are going to be discussed in detail along with a brief explanation about cargo booking.
338
A. Tamizhinian and V. M. Pillai
3.1 Cargo Booking Cargo can be pre-booked in an online portal from a week before the day of departure till a day before the day of departure. For pre-booking, details like destination, gross weight of cargo, day of departure, number of shipments should be provided in the portal. The customers can also specify the flight number in which they wish their cargo to be flown, but getting the same flight is not assured until the day of departure. In return, the customers get a unique 11-digit Air Way Bill (AWB) number for one such booking. On the day of departure, the agent from the pre-booked customer side reaches the counters with the necessary documents and gets the documents verified with a flight number assigned to the cargo. Once the paperwork is done, the vehicle carrying the cargo of the corresponding agent enters the warehouse and the process of cargo acceptance starts, which is explained in Sect. 3.2. It’s worth mentioning that the agent works in such a way that paperwork is getting finished before the cargo shows up at the dock and the counter activities never affect the cargo flow in the cargo acceptance process.
3.2 Process of Cargo Acceptance1 Once the necessary paperwork is done, the vehicle carrying the cargo enters the warehouse, which is directed to the docks available in the outbound area. The warehouse is operating with two types of cargo, completely different with respect to their handling during the process of acceptance. They are console and dense cargos. The former refers to the type of cargo whose volumetric weight is more than the gross weight or in simple words they occupy more space though they weigh less. The shipments ordered through e-commerce organisations come under console type. While the latter weighs more in terms of gross than volume. Auto parts, ready-made garments are few examples of this type. The warehouse is equipped with 6 docks for parking the vehicles, 4 weighing machines (3 for console and 1 for dense) and 4 X-ray machines (3 for console and 1 for dense). The scaled layout (see Fig. 1) can give a clear idea about the layout of resources and the location of docks, counters and resources. Once the vehicle is parked at the dock, the acceptance process starts with the unloading of cargo. In general, both types of cargo have to flow through the same set of processes such as unloading, weighing and X-ray screening to get accepted. But the way of handling is completely different, dense cargo is unloaded on the pallet and then each pallet is weighed and screened in one go. Whereas in case of console cargo, cargo is unloaded on the floor, weighed using a container and screened piece by piece. While X-ray screening, if the cargo is found hazardous according to the regulations or a chance of leakage is detected such shipment is offloaded, otherwise cargo gets accepted into the warehouse. 1
All the numbers quoted in this article are sanitized for confidentiality purpose.
Performance Evaluation of Cargo Warehouse Operations …
339
Fig. 1 Existing layout of the outbound area of the warehouse
Cargo acceptance is made sure by the cargo staff finalising the number of pieces and amount of cargo being accepted for that particular AWB, by signing the ‘Instruction for Dispatch of Goods (IDG)’. Then, the customer makes the payment with respect to the amount of cargo being accepted. The customer can skip the queue until the IDG is getting signed by the cargo staff either while waiting for docks or even before reaching the weighing station considering the congestion in the warehouse.
3.3 Process of Cargo Uplifting Upon acceptance, cargo is segregated with respect to the sector to which it has to be carried and stored in the warehouse. From the warehouse based on the departure time (D) of cargo’s corresponding flight, cargo is dispatched from the warehouse using trucks. The trucks generally uplift the outbound cargo in the departing flight then collect the cargo from the arriving flight back to the warehouse. Loading of outbound cargo into the truck is initiated at D-90 min. Then, cargo moves to the ramp area (where aircraft gets parked). At the ramp area, the space available in the aircraft for
340
A. Tamizhinian and V. M. Pillai
cargo is confirmed by the flight skipper, 45 min prior to the departure. If the volume available is less compared to cargo carried by the truck, offloading happens, the unuplifted cargo is considered as ‘offloaded cargo’ which will be planned for the next immediate flight. If the amount of cargo offloaded is a considerable part of the cargo in the truck, the truck waits till the departure time of the next flight and uplifts, else cargo will be put into trolleys and truck leaves to the parking area and wait for the arriving flight. The incoming cargo is of two categories, bonded cargo and domestic cargo. Bonded cargo is one such category which has to be delivered to international sectors via domestic sectors, e.g. HYD-CCU-AMS (Hyderabad to Amsterdam through Kolkata). Under this category, trucks collect the cargo and unload it in the TEC warehouse (dedicated for bonded transits). In the case of domestic cargo, trucks collect the cargo to the inbound area of the warehouse, for delivering it to the customer, and move to the outbound area for the next cycle of uplifting and collection. The entire process flow of cargo acceptance and cargo uplifting along with truck logistics which has been discussed in Sects. 3.2 and 3.3 is depicted in the form of flowchart below (see Fig. 2). In the flowchart, truck movements (logistics) are considered as a sub-process of cargo uplifting and collection process.
4 Solution Methodology In order to evaluate the performance of operations in the warehouse as well as adequacy of trucks used for dispatch, Discrete Event Simulation (DES) is chosen as it can even model the complex system’s behaviour in an ordered sequence of well-defined events [6]. Also, DES helps to explore opportunities for new scenarios without disrupting the current system.
4.1 System Definition (see Footnote 1) The warehouse which processes domestic outbound cargo is treated as the system for model building. It is modelled in a way that DES takes care of the flow that comes into the system and goes out of the system or in other words the conservation of mass with respect to the warehouse. So, the model starts with the arrival of pre-booked cargocarrying vehicle to the warehouse, followed by dock allocation, unloading, weighing, X-ray screening of cargo, accumulation of cargo in the warehouse, dispatch of cargo from warehouse, uplifting of cargo, collection of cargo and ends with unloading the collected cargo in the inbound area of the warehouse. While defining the system, below are the few assumptions as well as system information being considered for model building,
Fig. 2 Process map of cargo acceptance and cargo uplifting
Performance Evaluation of Cargo Warehouse Operations … 341
342
A. Tamizhinian and V. M. Pillai
(1)
Only three vehicles are allowed to be in queue if all the docks are occupied (followed in real time) Shift break of 20 min for X-ray screening process (followed in real time) Space constraint at the dock area and warehouse is not considered—for checking the maximum accumulation happening in each area with respect to time of the day First-In-First-Out (FIFO) priority is used, when cargo is dispatched from warehouse The variation in the processing speed among the X-ray machines is considered (console and dense handling) Trucks are dispatched only for those flights which are carrying more than 300 kg of cargo Clubbing multiple sector cargo in same truck only if time difference between departures of flights is around 30 min and departing from same terminal The actual case of carrying bonded cargo (~18%) to TEC warehouse is incorporated with time spent by truck at TEC warehouse as 15 min Infinite number of trolleys are available for cargo uplifting and collection process to manage offload The delivery of incoming cargo to the customers is not considered.
(2) (3)
(4) (5) (6) (7) (8) (9) (10)
4.2 Input Data Collection (see Footnote 1) The collection of input data is often considered the most difficult part in carrying out a simulation modelling. According to [8], the input data may be either obtained from historical records or collected in real time by direct observation of a task. The flowchart explaining the various inputs that have been extracted and observed to drive the simulation model is shown in Fig. 3. The input data such as cargo acceptance rate in terms of tons per hour, departure and arrival schedule of flights and the flight capacity are extracted from the historical records. The input data such as time taken for various activities in cargo operations is collected through time study using stopwatch, whereas the time taken for activities in cargo uplifting process is collected through GPS application as raw data and cleansed using Alteryx tool. The observed data is fed into Arena input analyser to obtain the best fitted theoretical distribution to drive the simulation model. Table 1 shows the time taken for the activities involved in cargo acceptance and cargo uplifting processes in the form of a probabilistic distribution, e.g. activity 1 is the inter-arrival time of the freight forwarders, which follows gamma distribution with shape and scale parameter, 8.55 and 0.96, respectively.
Performance Evaluation of Cargo Warehouse Operations …
343
Fig. 3 Flow chart of various inputs collected to drive the simulation model
Table 1 The best fitted theoretical distribution for the activities involved in cargo operations S. no
Activity vehiclesb
1
Inter arrival time of
2
Parking in the docka
Best fit
Expression
Gamma
(−0.5 + GAMM(8.55,0.96)
Normal
NORM(128,22.3)
3
Unloading of console
Weibull
11 + WEIB(76.9,0.699)
4
Unloading of dense cargoa
Normal
NORM(96.9,45.9)
5
Weighing of console cargoa
Normal
NORM(84,24)
Normal
NORM(30,12)
cargoa
cargoa
6
Weighing of dense
7
Screening of console cargoa
Beta
73 + 480 × BETA(1.04,2.2)
8
Screening of dense cargoa
Weibull
46 + WEIB(114,1.04)
9
Dispatch of cargob
Weibull
20 + WEIB(27.9,1.01)
10
Uplifting of cargo from 1 truckb
Exponential
45 + EXPO(23.8)
11
Collection of cargo from 1 flightb
Exponential
30 + EXPO(18.2)
Weibull
20 + WEIB(26,1.01)
12 a
Unloading at inbound
areab
b
time in seconds, time in minutes
4.3 Discrete Event Simulation (DES) Model Using ARENA Discrete event simulation model for the cargo operations being followed in the warehouse is developed using ARENA simulation software. The reason for choosing ARENA to model a discrete event simulation is that it offers a complete range of statistical distribution options to model process variability with utmost accuracy as
344
A. Tamizhinian and V. M. Pillai
well as its flowchart modelling methodology which does not require any programming knowledge. The complete process explained in Sects. 3.2 and 3.3 driven by the inputs discussed in Sect. 4.2 considering few assumptions mentioned in Sect. 4.1 is structured into two sub-models. Apart from the inputs discussed in Sect. 4.2, a few important parameters required to drive the simulation model is given below. (1) (2) (3)
To overcome sampling error and to achieve desired half-width, model is set to run for 50 replications Operating hours of warehouse as replication length is set as 1440 min Set a warm-up period of 360 min to let the system reach a steady state.
The screenshot of the model which depicts the vehicle entry and unloading process is shown in Fig. 4. The screenshot of the model which depicts the weighing and X-ray screening process of console and dense cargo is shown in Fig. 5. The first sub-model mimics the cargo acceptance process while the uplifting process is modelled as the second sub-model. The screenshots of the model mimicking the cargo uplifting and collection process happening at city side and air side are shown in Figs. 6 and 7, respectively.
Fig. 4 Screenshot of ARENA simulation model for vehicle entry and unloading of cargo
Fig. 5 Screenshot of ARENA model for weighing and screening processes
Performance Evaluation of Cargo Warehouse Operations …
345
Fig. 6 Screenshot of ARENA model for dispatch, unloading at TEC warehouse and unloading at inbound
Fig. 7 Screenshot of ARENA model for uplifting and collection processes
346
A. Tamizhinian and V. M. Pillai
Fig. 8 Screenshot of ARENA model for truck load during dispatch
Fig. 9 Screenshot of ARENA model for incoming cargo arrival
The screenshot of the model which depicts the logic for determining the truck load during dispatch, which is integrated as a sub-model in the dispatch process as ‘D-90 signal’ is shown in Fig. 8. The screenshot of the model depicting the workflow within the sub-model ‘Incoming cargo’, which takes care of extracting the arrival schedule and cargo category of incoming cargo from an excel file is shown in Fig. 9.
5 Results and Discussion (see Footnote 1) 5.1 Evaluation of Current Performance of the Warehouse In this section, the performance measures such as turnaround time of cargo, turnaround time of vehicle, waiting time of cargo and amount of cargo in queue to get screened, warehouse accumulation and average utilisation of resources are analysed based on the results obtained from the ARENA simulation model. Below are the few inferences drawn from the results, which helped in achieving few objectives of this research work.
Performance Evaluation of Cargo Warehouse Operations …
347
Table 2 Performance measures of the existing cargo operations at the warehouse S. no
Performance measures
Current scenario
1
Average turnaround time of vehicle
36 min
2
Average time taken for accepting 1 ton of cargo
105 min
3
Average utilisation of X-ray machines
90%
4
Average utilisation of weighing machines
33%
5
Waiting time of 1 ton of cargo for X-ray screening
30 min
6
Average amount of cargo in queue for X-ray screening
10 tons
7
Average utilisation of cargo carrying trucks
89%
8
Average accumulation in warehouse
13 tons
9
Average holding time of cargo in warehouse
28 min
• X-ray screening process is found out to be the bottleneck being utilised at full capacity with an average utilisation of 90% • The waiting time of cargo for X-ray screening is completely a non-value-added time accounting for 25% of acceptance time of cargo • Queue for screening process is the reason for congestion at warehouse • With current number of trucks, dispatch & uplifting of current accepted tonnage is manageable with an average truck utilisation of 89%. The other key performance measures which play a crucial role in explaining the performance of existing cargo operations in the warehouse are shown in Table 2.
5.2 Evaluation of Proposed Scenarios It is known that X-ray machines are being utilised to its fullest capacity, installation of one more X-ray machine is proposed to ensure the smooth flow of cargo during the acceptance process. Without disrupting the current system, scenarios of adding one more console cargo handling X-ray machine to the current system and adding one more dense cargo handling X-ray machine to the current scenario are evaluated using the DES model. Table 3 shows the performance measures obtained for the proposed scenarios. Under the above two scenarios, the waiting time of cargo for screening can be reduced by 23% and 33% respectively. Also, the results suggested that the queue for X-ray screening can be reduced by 10% if the warehouse deploys one console cargo handling X-ray machine and 18% in case of addition of dense-handling X-ray machine. Eventually, congestion in the warehouse can be relieved and the customers can be assured with an improved service as the proposed scenarios can reduce the acceptance time considerably which is clearly evident from the results shown in
348
A. Tamizhinian and V. M. Pillai
Table 3 Performance measures of the warehouse under proposed scenarios S. no
Performance measures
+1 x-ray m/c (console)
+1 x-ray m/c (dense)
1
Turnaround time of vehicle
36 min
36 min
2
Turnaround time of cargo per ton
95 min
92 min
3
Utilisation of X-ray machines
77%
76%
4
Utilisation of weighing machines
33%
33%
5
Waiting time of cargo per ton for X-ray screening
23 min
20 min
6
Average amount of cargo in queue for X-ray screening
9 tons
8.2 tons
Fig. 10 Comparison of average utilisation of x-ray machines among the scenarios
Fig. 11. Also, the overall throughput of the warehouse can also be increased since Xray machines’ utilisation has come down under two scenarios by 13.3% and 16.7%, respectively (see Fig. 10).
6 Conclusion (see Footnote 1) The discrete event simulation model helped in evaluating the current performance of the cargo operations in a warehouse. It is identified that the X-ray screening process is the bottleneck responsible for the rejected demand. Also, it is evaluated that the current number of trucks used for dispatch is adequate. Ultimately the DES model has served its best-known purpose of estimating the performance of the system under proposed scenarios. It is graphically shown in Fig. 12 that under proposed scenarios there is a considerable increase in percentage of throughput which could generate revenue for the company.
Performance Evaluation of Cargo Warehouse Operations …
349
Fig. 11 Comparison of acceptance time among the scenarios
Fig. 12 Percentage of throughput increase under proposed scenarios
Further, there is a scope for improvement, in future the model can be updated by considering the cargo delivery process at the inbound area to evaluate the turnaround time of incoming cargo. Also, models can be developed to evaluate the adequacy of trolleys used to manage the offload during the process of cargo uplifting and collection.
References 1. Hartanti LPS (2016) Work measurement approach to determine standard time in assembly line. Int J Manage Appl Sci 2(1):192–195 2. Kelton WD (2002) Simulation with ARENA. McGraw-hill 3. Liong CY, Loo CS (2009) A simulation study of warehouse loading and unloading systems using Arena. J Qual Meas Anal 5(2):45–56
350
A. Tamizhinian and V. M. Pillai
4. Lukodono RP, Ulfa SK (2018) Determination of standard time in packaging processing using stopwatch time study to find output standard. J Eng Manage Ind Syst 5(2):87–94 5. Muhamad MR, Mahmood W, Hasrulnizzam W (2005) Productivity improvement through motion and time study 6. Rusca F, Popa M, Rosca E, Rusca A (2018) Simulation model for maritime container terminal. Trans Prob 13 7. Shibin KT, Gunasekaran A, Papadopoulos T, Childe SJ, Dubey R, Singh T (2016) Energy sustainability in operations: an optimization study. Int J Adv Manuf Technol 86(9–12):2873– 2884 8. Ungureanu D, Sisak F, Kristaly DM, Moraru S (2005) Simulation modeling. Input data collection and analysis. In: The 14th international scientific and applied science conference ELECTRONICS ET, pp 43–50 9. Xia W, Sun J (2013) Simulation guided value stream mapping and lean improvement: a case study of a tubular machining facility. J Ind Eng Manage (JIEM) 6(2):456–476
Blockchain: A Makeover to Supply Chain Management Justin Sunny, Kenil Shah, Prajwal Ghoradkar, Manu Jose, Malhar Shirswar, Hiran V. Nath, and V. Madhusudanan Pillai
Abstract Blockchain, the distributed ledger technology with phenomenal capabilities, is gaining the attention of supply chain managers. This paper is all about how blockchain renovates the current supply chain management practices. It discusses the major shortcomings of existing supply chains and the ways blockchain halts them. Shortcomings are identified from the literature and mapping is done with the characteristics of blockchain technology with proper justification. Blockchain has numerous possibilities in the domain of supply chain management and this paper presents them from different perspectives. Based on the evidences from the literature, this paper investigates how blockchain can influence the flow of money, material and information in a supply chain along with important applications. The possible combinations of blockchain with other technologies are identified and their scope in supply chain management is briefly discussed. Ultimately, this paper lists out the core attributes of blockchain-enabled supply chains. A comprehensive view of some of the real-world use-cases is also presented in this paper. In short, this paper provides a short reference for the readers, to understand the way how blockchain is being a makeover to supply chain management. Keywords Blockchain technology · Distributed ledger · Supply chain management
1 Introduction We are in the middle of the fourth industrial revolution that is radically changing our lives through facilitating the best products and services at a lower price [19]. Supply chain management has a crucial role in this. The quality and availability of J. Sunny · V. M. Pillai (B) Department of Mechanical Engineering, National Institute of Technology Calicut, Kozhikode, India e-mail: [email protected] K. Shah · P. Ghoradkar · M. Jose · M. Shirswar · H. V. Nath Department of Computer Science and Engineering, National Institute of Technology Calicut, Kozhikode, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_26
351
352
J. Sunny et al.
the products or services, price of products and services, customers’ purchasing experience, and most importantly, the profitability of an organization are all affected by supply chain management [16]. In the simplest form, supply chain management is the combined and coordinated flows of goods from origin to the final destination and also the information flows and cash flows that are linked with it [18]. Modern supply chains are big in size and it is being very complex due to many reasons. With the continuously increasing globalization and networking, cross-border trade rose drastically. Due to a high level of competition, organizations started giving more choices to customers, and now customers’ expectations are far beyond the predictions. Technological advancements transformed the traditional way of trading products and services. Finally, the large number of intermediaries made supply chains more fragmented. In short, nowadays, supply chain management has a lot more to it than just where and when. Though modern SCs are effective in terms of many aspects like product variety, distribution and customer services, still they are subjected to many issues. Lack of transparency, lack of information security, lack of trust, etc. are a few of them and managers are struggling to tackle these issues. In 2008, Satoshi Nakamoto introduced a radical concept called blockchain, a distributed digital ledger, which can provide transparency, trust and immutability in a highly decentralized environment [26]. Blockchain has a wide range of applications in many fields, including finance, health, agriculture and real estate. Supply chain management is one of the important areas where blockchain has created a huge stir. With the increasing complexity of the supply chains, the need for efficient management increases and blockchain will be a feasible solution for the same. This paper primarily takes a look at the shortcomings of existing supply chains and discusses how the characteristics of blockchain technology can help managers in overcoming this. The rest of this paper is organized as follows: Sect. 2 gives an introduction to blockchain technology. Section 3 highlights the shortcomings of existing supply chains. Advantages, peculiarities and possibilities of blockchain-enabled supply chains are discussed in Sect. 4, and some of the compelling use-cases are pointed out in Sect. 5. Section 6 concludes the paper.
2 Basics of Blockchain Blockchain is essentially a public ledger that holds lists of transactions in a cryptographically linked sequence of blocks, in a decentralized manner [7]. Each block contains a set of transactions along with the block header. The block header contains information like block version, nonce, timestamp, the hash of the parent block, the hash of the Merkle tree root, etc. [42]. The number of transactions in a block depends upon the size of each transaction and the maximum allowed block size. The first block of the chain has no parent and is known as the genesis block. The general structure of a blockchain is given in Fig. 1. Decentralized/distributed
Blockchain: A Makeover to Supply Chain Management
353
Fig. 1 Structure of blockchain
nature, immutability, consensus-driven governance and transparency are the core characteristics of blockchain technology [35]. 1.
2.
3.
4.
Distributed/Decentralized: Blockchain is fundamentally a decentralized database that runs on a peer-to-peer network of nodes (devices that act as communication points in the network, for example, computers). The data in a blockchain is stored in all the nodes of the network. Hence, one party cannot take the whole control of the network. This eliminates the need for a trusted third party. Immutability: Transactions recorded in a blockchain are made immutable through cryptography. Each block in the chain is linked to the previous block by storing the hash of the previous block. Hash functions are such that even a small change in input leads to an altogether different output. Thus, a change in any transaction will change the hash of the block. Since every block stores the hash of the previous block, these changes will propagate all the way to the end of the chain, thus making it almost impossible to change the contents. Consensus-Driven: Blockchains have dedicated consensus mechanisms to govern the creation of blocks. Different blockchains work with different consensus mechanisms. For example, Bitcoin blockchain runs with proof of work (PoW) consensus and Quarum works with proof of authority (PoA). The consensus mechanism always paves the opportunity to verify each and every block in a blockchain independently. Also, each consensus model provides a set of rules for the validation of blocks. Transparency: Generally, information appended in a blockchain is visible to every member of the network and hence, it can be easily verified and audited by anyone. Each user in the network has a pair of keys, namely private key and public key. The public key can share with others in the network, whereas the private key is to be kept safe. All transactions in the blockchain are digitally signed by the sender using the private key and will be broadcasted throughout
354
J. Sunny et al.
a) Public blockchain
b) Private blockchain
c) Consortium blockchain
Fig. 2 Schematic diagram of different types of blockchain [31]
the network [23]. The nodes in the network can verify the transaction by using the public key of the sender. There are three different types of blockchains, namely public blockchain, private blockchain and consortium blockchain [25]. Figure 2 shows their schematic diagrams. Public or open blockchains allow everyone to enter the network and interact with any other node in the network. Transactions made in a public blockchain will be visible to all the participants in the network. Most of the public blockchains follow the proof of work (PoW) consensus mechanism; i.e., making the miners (nodes that create new blocks) solve a complex, computationally expensive, yet easily verifiable cryptographic problem as proof for successfully creating a block. Examples of public blockchain are Bitcoin, Litecoin and Ethereum. Private or closed blockchains allow only predetermined individuals to view and enter data in the chain. As the name indicates private blockchain can ensure privacy as it can put restrictions. Private blockchains are usually preferred for better privacy, higher scalability and increased throughput. Ripple and Hyperledger Fabric are examples of private blockchains. A consortium blockchain is a blend of a public and private blockchain but is more similar to a private blockchain. Control over a consortium blockchain is granted to a group of approved organizations or individuals. Here, the existing members can decide the entry of new members into the chain. An example of a consortium blockchain is Quorum.
3 Shortcomings of Existing Supply Chain Systems Due to the vast network across national boundaries covered by a supply chain, even simple transactions turn into long multi-step procedures. This increasing complexity of the supply chains has resulted in a large number of issues and challenges. Transparency and traceability in supply chains are some of the most significant challenges in the existing supply chain management system. Customers are confused with the
Blockchain: A Makeover to Supply Chain Management
355
genuineness of products and services they receive just because of the lack of transparency. For example, it may not be possible for a consumer to differentiate between an authentic Fossil watch and a duplicate one. As of now, customers cannot trace the complete history of products they purchase. Here, complete history refers to each and every detail of the product including the origin and the processes they have undergone. Exploiting this fact, duplicate and substandard products are simply entering the market. Lack of traceability can cause severe damages in case of defective (substandard products) or duplicate products (counterfeited products). For example, a large number of people have fallen ill with Salmonella infections linked to raw turkey products, which has led to a recall of more than 150 tonnes of raw turkey products [4]; all because of the lack of traceability in the supply chains associated with the products. Once they enter the market, it will be very hard to trace them. This possibility reduces the customer confidence in purchasing products or services. Lack of traceability can have devastating effects on the food supply chains. Unsafe food causes sickness to around 600 million people every year [37]. Mattel, a toy-making company, had to recall about a million products in the United States due to the toys being covered by poisonous lead paints [34]. In 2009, Toyota had to recall around 4 million vehicles due to faulty gas pedals [14]. The company was receiving pedals from many different sources and found it extremely difficult to trace the party responsible for the faulty pedals, which caused an estimated loss of about US$2 billion. Thus, the need for traceability in supply chains is critical. Lack of transparency in supply chain operations still remains a critical issue. It constrains the opportunity for collaboration between supply chain members and affects the overall performance. The best example is the bullwhip effect (BWE). BWE is the increasing swings or variations in demand as it transmits toward the upstream members in a supply chain [11]. Besides transparency and traceability, existing supply chains suffer from many other shortcomings. Most supply chains still use centralized systems to store important data. This way of keeping records is not secure and is always subjected to severe threats. Data hiding and tampering are major concerns for centralized data storage. Organizations that are using centralized information systems will have the sole power to hide or manipulate the real information from the outside world. Also, centralized data management is more prone to frauds and data breaches, and could cause a lot of damage. Moreover, since a single entity has the control over the data in centralized systems, the existing supply chains lack trust among the members. As the number of intermediaries in a supply chain rises, this issue will get amplified. Wikipedia says about another issue called single point of failure (SPOF), which is part of centralized data management systems. SPOF is totally undesirable as it can stop the entire functionalities of any supply chains or industrial systems [38]. Unfortunately, even in this digital era, a good percentage of supply chains are totally relying on paper-based documentations (agreements, specific contracts, transaction receipts, etc.) and this causes unnecessary delays in the production and distribution phases.
356
J. Sunny et al.
4 Blockchain-Enabled Supply Chain Systems Shortcomings in the existing way of managing the supply chains can be tackled with the help of blockchain technology. Blockchain technology provides the opportunity to track and trace the supply chain events and hence, it creates transparency in the entire supply chain network. The consensus-driven administration of blockchain always guarantees trust and makes the network politically decentralized. Different from a centralized database, manipulation of data is not possible in blockchains as the blocks are cryptographically connected to each other. Supply chains can conveniently select a particular type of blockchain depending on the nature of trade. However, private or consortium blockchain is more preferable for supply chains as it can ensure a certain level of privacy which is essential for running a business [40]. Table 1 presents the mapping of shortcomings of existing supply chains with characteristics of blockchain technology. Smart contracts and the Internet of Things (IoT) have a significant role in making blockchains favorable for applications in supply chain management. A smart contract is a piece of program that can be used as an automatable and enforceable agreement [8]. Smart contracts can be deployed in blockchains to effectively execute business logics. A smart contract gets triggered once a predefined condition occurs. For example, payment gets automatically released once a buyer acknowledges the delivery of a product. IoT devices can capture genuine data that can be stored in blockchains. With the assimilation of IoT devices, blockchain allows a wide range of application scenarios in supply chain management, including inventory management, warehouse operations, production and manufacturing operations, and transportation operations [30]. Merging smart contracts and IoT devices with blockchain technology will impart a substantial level of transparency in a supply chain. Next-generation supply chains will be fully driven by technologies. Sometimes, a combination of multiple technologies can have better applications, just like IoT uplifted the opportunities of blockchain technology. Combinations of blockchain with other technologies can significantly renovate the supply chains. 1.
2.
Blockchain with Artificial Intelligence (AI): AI refers to the ability of machines to replicate human intelligence. Learning is an important aspect of AI and that requires data and self-learning algorithms. AI uses various algorithms to learn the patterns from a large amount of input data and acts accordingly. The success of AI depended on the quality of data. As an immutable ledger, blockchain can always function as a source for genuine data. A combination of blockchain and AI has crucial applications in supply chain management. For instance, it can be used to accurately forecast the demand for making optimal order decisions. AI algorithms can be deployed in blockchains to make such predictions faster. Blockchain with Digital Twin Technology: In simple terms, a digital twin is a virtual replica of an entity or a process. A digital drawing of a car will be its digital twin. While integrating this technology in the context of supply chains, all the assets, including machines, employees, and all sorts of inventories will have
Blockchain: A Makeover to Supply Chain Management
357
Table 1 Mapping of shortcomings of existing supply chains with characteristics of blockchain technology Shortcomings of existing supply chains
Characteristics of blockchain technology to tackle the shortcomings
Remarks
Chance for data manipulation Immutability
A blockchain is a permanent record of transactions. Once a block is added, it cannot be altered
Single point of failure (Example: server breakdowns)
Distributed
The data in blockchain is decentralized, i.e., it is stored in many parts of the network. Thus, making it less prone to the single point of failure
Less opportunity for collaboration and related issues (Example: Bullwhip effect)
Transparency
Lack of transparency causes distortion in the supply chains. Since the data in blockchain can be made visible to everyone in the network, it helps to achieve transparency. Transparency creates an opportunity for collaborations
Inability to track origin or Transparency products leading to counterfeiting and inefficient recalls
Using blockchain, all the relevant details related to a product or service can be stored at every point in the supply chain and can be easily traced back whenever required
Lack of trust among participants
Consensus
The consensus mechanism makes the network politically decentralized. It ensures validation and verification of blocks
Requirement of intermediaries cost and speed
Consensus
Because of the consensus-driven administration, intermediaries are not at all required in blockchain networks
Unnecessary documentation (Example: in cross border trades)
Distributed
In blockchain technology, documents can be digitally saved in a distributed ledger. Authorities (geographically dispersed) can authorize documents with a digital signature using their private keys. This will reduce unnecessary costs and delays (continued)
358
J. Sunny et al.
Table 1 (continued) Shortcomings of existing supply chains
Characteristics of blockchain technology to tackle the shortcomings
Remarks
Difficulty in auditing
Transparency
Digitally signed transactions can be easily verified and audited by anyone in the network
3.
4.
a distinct digital identity. As a distributed ledger where data can be immutably stored, blockchain can keep these identities safely. Along with the physical movement of assets, corresponding details can be updated in blockchains using their digital identities. This will help supply chains to restrict the entry of counterfeit products into the market. The digital twin of documents like warranty proofs, contracts, receipts, etc. can be securely stored in blockchains. Digital twin technology has been used with blockchain for traceability applications. Blockchain with Cloud of Technology (CoT): IoT is basically a network of devices that communicates with each other without the need for any human intervention. The possibility of IoT in the blockchain is a widely researched area. Now, the use of IoT devices increased drastically and it is difficult to locally store and process the large amount of data they capture. Here is the significance of cloud computing. A combination of IoT and cloud computing is generally referred to as CoT. CoT can effectively integrate the processes in a complex supply chain [41]. Blockchain can be combined with CoT to make its governance decentralized [27]. CoT supports blockchains with benefits like higher scalability, additional security and fault tolerance. Blockchain with RFID technology: RFID stands for radio frequency identification. RFID technology is useful for capturing information. It uses radio waves to establish communication between an electronic tag and a reader. An electronic tag can be attached with sources like a machine or a device. As of now, this technology is widely used with blockchain as part of traceability applications. Near Field Communication (NFC) is a subset of RFID technology. NFC devices can act both as a tag and reader. In supply chains, RFID or NFC can be attached to various entities to track and trace their movements, and related information can be collected using readers. The RFID technology helps supply chain members quickly capture data from various sources.
Definitions of supply chain management highlight that the existence of supply chains is relied on the flow of money, material and information [22]. In fact, types of blockchains presented in Sect. 2, namely public, private and consortium blockchains can serve supply chains in these dimensions. Flow of money: Flow of money in supply chains is going to get transformed with the public blockchain-based cryptocurrencies. In fact, many supply chains already started accepting cryptocurrency for trade. Big companies like Microsoft, Lamborghini, Alibaba, etc. are planning to include cryptocurrencies into their
Blockchain: A Makeover to Supply Chain Management
359
operations [17]. Cryptocurrency platforms that support smart contracts can offer several attractive features for the supply chains in managing financial operations, for example, Ether of Ethereum public blockchain. Public blockchain platforms are completely decentralized and a third-party intermediary such as a bank is not at all required for monitoring the transactions and resolving disputes. Smart contracts can be devised for automating financial activities. For instance, smart contract-based automatic payment release will significantly speed up the procurement process. With this, money can be automatically released to the vendor’s account once the buyer acknowledges the delivery of products or fulfillment of services. Supply chains can apply blockchain technology to track the flow of money. In fact, blockchain-based traceability solutions are already proposed in humanitarian supply chains to keep monitoring the flow of financial contributions from people across the world and their utilization [43]. Flow of material: Blockchains are highly preferable for managing the flow of materials in supply chains, especially private and consortium blockchains. Flow of material can be tracked from its origin to destination and the same can be traced back. This will help the supply chain actors to monitor the distribution of products and services. Customers can ultimately ensure the genuineness of the goods they are purchasing. The quality of products is an important aspect in supply chain management. Smart contract-enabled blockchains can be used with IoT, to check whether products are in good condition or not. Advanced blockchain applications, such as automatic ordering of raw materials [9], can transform the existing supply chains into a completely digitalized form. Blockchain has applications in reverse supply chains as well. Customer returns can be tracked using blockchain traceability solutions. Flow of information: Proper flow of information is very critical in supply chain management, as its asymmetry can affect the supply chain performance [6]. Blockchain, the distributed ledger, can act as a central source of genuine information as every member has the information otherwise available centrally or unorganized way, hence will do the needs of centralization and decentralization. Lack of centralized information in supply chains can be mitigated with private or consortium blockchains. It can broadcast supply chain-related information in real-time. Blockchain ledger can be made open to appropriate supply chain members to read and write trade-related information, based on predefined terms and conditions. For instance, retailers, wholesalers, distributors and manufacturers can access the demand data in real-time, to make accurate order decisions. This will reduce supply chain issues such as the bullwhip effect [13]. Organizations can set up their own private blockchains with other relevant parties to ensure proper distribution and allocation of products.
360
J. Sunny et al.
4.1 Attributes of a Blockchain-Enabled Supply Chain Following are the important attributes of a blockchain-enabled supply chain: 1.
2.
3.
4.
5.
6.
Better connectivity: Connectivity in a supply chain ensures constant communication between the managers of different organizations involved in various functions [10]. Blockchain is a peer-to-peer network that enables real-time interactions between the nodes. Better possibilities for collaboration: Collaboration is a partnership process in which independent organizations are working together to plan and execute supply chain operations in order to fulfill customer demand. Blockchain is a decentralized network, where independent organizations, having no mutual trust, can share relevant data securely on a distributed ledger. Organizations in a blockchain-enabled supply chain can make use of this data to improve their performance and that of the entire supply chain. Better visibility: Visibility refers to the ability of supply chain members to access or share relevant data, which is useful for mutual benefit [20]. If the visibility is higher, a supply chain is said to be transparent. With blockchainbased traceability solutions, participants can monitor the movements of physical goods. A higher level of visibility will allow the end customers to instantly check the authenticity of the product they purchase. Better sustainability: Economy, ecology and society are the three pillars of sustainability. A sustainable supply chain works with goals from all these three dimensions [1]. Organizations in a blockchain-enabled supply chain can monitor and control their impact on ecology. For instance, IoT devices can measure the carbon emission rate and the same can be appended to a blockchain network deployed by the pollution control board. Smart contracts can be appropriately framed to alert the organizations, once the emission level goes beyond the permission limits. Better agility: Agility is essentially needed for a supply chain to cope with frequent changes in market trends [36]. If agility is there in a supply chain, it can transform unexpected business disruptions into valuable opportunities. In order to become agile, a supply chain needs to possess advanced sensing, flexibility, cooperation and velocity. Information sharing via blockchain will help the managers to accurately predict the upcoming market trends. The incorporation of blockchain smart contracts can drastically reduce the time consumed for documentations. Blockchain-enabled supply chains can quickly execute underlying operations. Better security: Blockchain-enabled supply chains will be protected against cyber hacks and other security issues. In modern supply chains, information security is a big concern, because the loss of relevant information may reveal the competitive strategies, causing severe loss to the firms.
Blockchain: A Makeover to Supply Chain Management
361
5 Use-Cases Blockchain has been adopted by a lot of top companies to revamp the working of their supply chains. Walmart and IBM partnered in 2016 to implement a system to track the origin of food items [15]. Bumble Bee Foods have started using SAP’s blockchain to track the complete supply flow of yellow fin tuna [32]. In order to define and manage the business rules and regulations between parties in the supply chain, American food service company Golden State Foods (GSF) has partnered with IBM to generate a solution that uses RFID tags, IoT devices and blockchain technology. The main purpose of this venture is to track the movement of fresh beef by monitoring its temperature. Department of Treasury, USA, plans to implement a real-time asset tracking system using blockchain [24]. In 2015, Maersk, a Copenhagen-based shipping company, joined with IBM to implement a system for connecting the vast network of shippers, carriers, ports and customs [33]. The Abu Dhabi National Oil Company (ADNOC) has implemented a blockchain-based automated system to integrate oil and gas production across the entire value chain [39], to reduce the time taken for executing transactions with its operating companies and to increase operational efficiencies throughout the supply chain. Walmart along with IBM and Tsinghua University have formed the “blockchain food safety alliance”, which will seek to find solutions for improving food tracking and safety in China [12]. Companies like De Beers, Everledger, TrustChain, etc. have started using blockchain for tracking diamonds and other precious gems [3]. Provenance, a British tech startup has developed several blockchain-based solutions for companies like Unilever, Cult Beauty and Grass Roots [29]. British Standards Institution (BSI) has partnered with OriginTrail to prevent counterfeiting and foul use of training certificates and business standards [2]. Binance, the world’s biggest trading platform for cryptocurrencies and digital assets, also partnered with OriginTrail to track the movement of funds [28]. The success of BCautoSCF, the blockchain-driven supply chain finance platform, clearly points out the caliber of blockchain technology in managing financial flow [5]. LeewayHertz developed a blockchain-based solution for TraceRX to enable endto-end traceability across the supply chains of medicines [21]. Blockchain technology is in its growing phase and its use-cases are being exponentially increased.
6 Conclusion Blockchain is truly a makeover for the existing way of managing supply chains. Decentralization, immutability, consensus-driven governance, and transparency are the core characteristics of blockchain technology. These characteristics are feasible for tackling the key issues in the existing supply chains. Blockchain technology transforms existing supply chains by enabling features like auditability, authenticity, trust and security. In this paper, we reviewed the shortcomings of the existing supply
362
J. Sunny et al.
chains and discussed how blockchain can be used to tackle them. We also looked at some of the compelling use-cases of blockchain in supply chain management.
References 1. Beske P, Seuring S (2014) Putting sustainability into supply chain management. Supply Chain Manage Int J 19(3):322–331 2. BSI. BSI partners with OriginTrail to develop blockchain-enabled solutions. Accessed February 29, 2020, https://www.bsigroup.com/en-IN/About-BSI/Media-Centre/press-rel eases/2019-press-releases/january-2019/bsi-partners-with-origintrail-to-develop-blockchainenabled-solutions/ 3. Cartier LE, Saleem HA, Krzemnicki MS (2018) Blockchain, chain of custody and trace elements: an overview of tracking and traceability opportunities in the gem industry. J Gemmol 36(3) 4. CDC. Outbreak of multidrug-resistant salmonella infections linked to raw turkey products. Accessed February 29, 2020, https://www.cdc.gov/salmonella/reading-07-18/index.html 5. Chen J, Cai T, He W, Chen L, Zhao G, Zou W, Guo L (2020) A blockchain-driven supply chain finance application for auto retail industry. Entropy 22(1) 6. Chopra S, Meindel P, Kalra DV (2016) Supply chain management, 6th ed. Pearson Education India 7. Chuen DLK (Ed) (2015) Handbook of digital currency: bitcoin, innovation, financial instruments, and big data. Academic Press 8. Clack CD, Bakshi VA, Braine L (2016) Smart contract templates: foundations, design landscape and research directions 9. Cohn JM, Finn PG, Nair SP, Panikkar SB, Pureswaran VS (2019) Autonomous decentralized peer-to-peer telemetry, U.S. Patent 10,257,270 10. Croom S, Fawcett SE, Osterhaus P, Magnan GM, Brau JC, McCarter MW (2007) Information sharing and supply chain performance: the role of connectivity and willingness. Supply Chain Manage Int J 11. Croson R, Donohue K (2006) Behavioral causes of the bullwhip effect and the observed value of inventory information. Manage Sci 52(3):323–336 12. Dignan L (2017) IBM, Walmart, JD.com and Tsinghua University eye blockchain for food supply chain safety in china. Accessed February 25, 2020, https://www.zdnet.com/article/ ibm-walmart-jd-com-and-tsinghua-university-eye-blockchain-for-food-supply-chain-safetyin-china/ 13. Engelenburg VS, Janssen M, Klievink B (2018) A blockchain architecture for reducing the bullwhip effect. In: Shishkov B (ed) Business modeling and software design. Springer, Cham, pp 69–82 14. Forbes (2010) Toyota expects recall to cost $2 billion. Accessed February 29, 2020, https:// www.forbes.com/2010/02/04/toyota-earnings-recall-markets-equities-prius.html#f94319 656680 15. Hackius N, Petersen M (2017) Blockchain in logistics and supply chain: trick or treat? In Wolfgang K, Blecker T, Ringle C (Eds) Digitalization in supply chain management and logistics: smart and digital solutions for an industry 4.0 environment. proceedings of the Hamburg International Conference of Logistics (HICL). epubli GmbH, Berlin, pp 3–18 16. IBM. Supply chain management. Accessed October 11, 2019, https://www.ibm.com/topics/ supply-chain-management 17. Icoholder. 250+ places that accept bitcoin payment (online and physical companies). Accessed April 8, 2020, https://icoholder.com/blog/places-accept-bitcoin/ 18. Janvier-James AM (2012) A new introduction to supply chains and supply chain management: definitions and theories perspective. Int Bus Res 5(1):194–207
Blockchain: A Makeover to Supply Chain Management
363
19. Lasi H, Fettke P, Kemper HG, Thomas F, Michael H (2014) Industry 4.0. Bus Inf Syst Eng 6(4):239–242 20. Lee H, Kim MS, Kim KK (2014) Interorganizational information systems visibility and supply chain performance. Int J Inf Manage 34(2):285–295 21. LeewayHertz. Case study: enabling traceability across supply chain with blockchain. Accessed April 5, 2020, https://leewayhertz.com/wp-content/case-study/TraceRX-CaseStudy.pdf 22. Leng FL, Zailani S (2012) Effects of information, material and financial flows on supply chain performance: a study of manufacturing companies in Malaysia. Int J Manag 29(1):293 23. Merkle RC (1990) A certified digital signature. In: Advances in cryptology—CRYPTO’ 89 proceedings, pp 218–238 24. Min H (2019) Blockchain technology for enhancing supply chain resilience. Bus Horiz 62(1):35–45 25. Morkunas VJ, Paschen J, Boon E (2019) How blockchain technologies impact your business model. Bus Horiz 62(3):295–306 26. Nakamoto S (2008) Bitcoin: a peer-to-peer electronic cash system 27. Nguyen DC, Pathirana PN, Ding M, Seneviratne A (2019) Integration of blockchain and cloud of things: architecture, applications and challenges 28. OriginTrail. 2019. $40 million worth of bitcoin stolen from binance now traced using OriginTrail. Accessed February 29, 2020, https://medium.com/origintrail/40-million-worth-of-bit coin-stolen-from-binance-now-traced-using-origintrail-b938bd179d2e 29. Provenance.org. Case studies. Accessed February 29, 2020, https://www.provenance.org/casestudies 30. Rejeb A, Keogh JG, Treiblmaier H (2019) Leveraging the internet of things and blockchain technology in supply chain management. Future Internet 11(7):161 31. Salha RA, El-Hallaq MA, Alastal AI (2019) Blockchain in smart cities: exploring possibilities in terms of opportunities and challenges. J Data Anal Inf Process 7(3):118–139 32. SAP (2019) Bumble bee foods and SAP create blockchain to track fresh fish from ocean to table. Accessed February 29, 2020, https://news.sap.com/2019/03/bumble-bee-foods-sap-cre ate-blockchain-track-fish/ 33. Scott T (2018) TradeLens: how IBM and Maersk are sharing blockchain to build a global trade platform. Accessed February 24, 2020, https://www.ibm.com/blogs/think/2018/11/tradelenshow-ibm-and-maersk-are-sharing-blockchain-to-build-a-global-trade-platform/ 34. Story L (2007) Lead paint prompts mattel to recall 967,000 toys. Accessed February 29, 2020, https://www.nytimes.com/2007/08/02/business/02toy.html 35. Sultan K, Ruhi U, Lakhani R (2018) Conceptualizing blockchains : characteristics & applications. In: 11th IADIS international conference information systems 2018, pp 49–57 36. Swafford PMSG, Murthy N (2008) Achieving supply chain agility through IT integration and flexibility. Int J Prod Econ 116(2):288–297 37. WHO. Food safety. Accessed February 29, 2020, https://www.who.int/news-room/fact-sheets/ detail/food-safety 38. Wikipedia. Single point of failure. Accessed February 25, 2020, https://en.wikipedia.org/wiki/ Single_point_of_failure 39. WorldOil (2018) ADNOC has implemented IBM blockchain technology to streamline daily transactions. Accessed February 29, 2020, https://www.worldoil.com/news/2018/12/10/adnochas-implemented-ibm-blockchain-technology-to-streamline-daily-transactions 40. Wüst K, Gervais A (2018). Do you need a blockchain? In: 2018 crypto valley conference on blockchain technology (CVCBT) 41. Yan J, Xin S, Liu Q, Xu W, Yang L, Fan L, Chen B, Wang Q (2014) Intelligent supply chain integration and management based on cloud of things. Int J Distrib Sens Netw 10(3) 42. Zibin Z, Xie S, Dai H-N, Chen X, Wang H (2018) Blockchain challenges and opportunities: a survey. Int J Web Grid Serv 14(4):352–375 43. Zwitter A, Boisse-Despiaux M (2018) Blockchain for humanitarian action and development aid. J Int Humanit Action 3(16)
Assessment of the Challenges Obstructing Performance of Indian Food Supply Chain Dynamics Janpriy Sharma, Mohit Tyagi, and Arvind Bhardwaj
Abstract Food is a necessity of human life and is a pillar of life physical, emotional and spiritual virtue. In this era of globalisation and increasing awareness among the consumers, there is a demand for ample food items availability along with diversified food items. Hence the need of the hours is to strengthen the food supply chain (FSC) and ensure a smooth flow of demand and supply patterns. FSC refers to an interconnected sequence of activities, aimed to fulfil consumer demands. FSCs have multiple partner’s alliances working on various tiers of their operationality. FSCs efficient flow is necessary because of its importance linked to the lives of people along with food and social security of all. But, in reality, FSC faces many hindrances which deaccelerate the pace of the chain and these hindrances act as challenges that must be overcome. Lagging supply chains causes reduced profitability and dissatisfaction at the consumer end. This study is aimed to identify the challenges which hamper the FSCs by gaining insights into research literature and brainstorming session with the experts of the Indian food industry. For the identified six challenges expert views were grabbed and to check for the interdependence between them and to differentiate well between the challenges in causal group and effect group, a mathematical tool named Decision Making and Evaluation Laboratory (DEMATEL) is implied. For checking the reliability and robustness of the results, sensitivity analysis is carried out. The finding of this study withstands to check out which challenges affect severely the performance of FSC and accordingly operations can be refined to have smooth product flow with improved profitability. Keywords Food supply chain · Food supply chain management · Challenges · MCDM · DEMATEL · Indian food industry
J. Sharma · M. Tyagi (B) · A. Bhardwaj Department of Industrial and Production Engineering, Dr. B.R. Ambedkar National Institute of Technology, Jalandhar, Punjab, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_27
365
366
J. Sharma et al.
1 Introduction and Background In the global food trade, it is important to be able to ensure the quality and safety of the products. It becomes a challenge to deliver food products after covering a sequence of operations and steps in various regional, national and international markets depending on customer demand. As per the report from Food and Agriculture Organisation (FAO), roughly one-third of global food production, valued at more than $750 billion annually, was lost or wasted annually [1]. Retailing of food belongs to past history of 1000 years or even older when ancient markets and agoras used to sell food commodities [2]. Food industries are deployed for the procurement process, manufacturing process and their distribution, where these processes are backed up by various primary, secondary and tertiary activities and logistics for transportation of products [3]. FSC has its own network of interacting stakeholders comprising millions of industries on small scale, retail shops, unit shops, government and non-government organisations. Food supply chain management (FSCM) glimpse those activities and functions beginning with production, its distribution to consumption end to maintain safety and quality with food products in an efficient and effective manner [4, 5]. One sight view of the FSC can be taken from Fig. 1. In India, the food supply chain and food processing industries play an important role in the financial and social share. The Indian food industry market ranks sixth largest in the world, and the Indian food processing industry has 32% of share in country’s total share of the food market. It is one of the largest industries and stands at the fifth position in terms of production rate, consumption rate, overseas trading and profit volumes (Ministry of Food Processing Industries, MOFPI, Government of India). Indian food supply chain is at the developing phase of its development and still possesses a lot of challenges that constraint its growth [6]. The present research focuses on the challenges being faced in the successful implementation of food supply chain. The identified challenges are enough to answer abnormalities in the adoption and handling of food supply chain management (FSCM) practices. To achieve this, various challenges in lieu of research literature and experts talk were considered and DEMATEL was used as a tool to find causal and effect parameters. Hence, the proposed tool can be utilised for interdependence and to differentiate factors into causal and effect groups, enabling decision-makers to plan accordingly to minimise the effects of these challenges of these factors.
Supplier
Farmer
Food Industry
Fig. 1 Food supply chain at a glance
Wholesaler
Retailer
Consumer
Assessment of the Challenges Obstructing Performance of Indian …
367
2 Literature Review FSCM has its vast perspective which requires collaboration between producers, suppliers and consumers. FSC sums up the variety of products as per the needs of customers, covering up diverse markets. FSCM is trending as a novel discipline of research at various regional, national and international levels. Joshi et al. [7] identified critical factors for the food supply chain of fruits, vegetables and dairy products and used fuzzy interpretive structure modelling (ISM) approach for ranking of factors identified. Tyagi et al. [8] analysed the critical enablers related to a flexible supply chain performance system using the fuzzy DEMATEL approach. Kazancoglu et al. [9] studied the challenges faced by the milk supply chain and used the grey prediction method for loss estimation. Siddh et al. [10] explored deployment of perishable food supply chain quality (PFSCQ) practice and categorised PFSCQ into various categories of suppliers, customers and quality. Tyagi et al. [11] developed a framework to analyse the impact of CSR practices on the supply chain performance system using an extension of the DEMATEL approach under a fuzzy environment. Gomiero [12] considered the sustainability aspect of agricultural food supply chains and also explored deployment issues of FSC within the domain. Balaji et al. [13] identified the causes behind the deployment of an effective perishable food supply chain in context to food and vegetables and used ISM and fuzzy MIMAC for modelling the inhibitors of deployment. Marsden et al. [14] explored the dates of the supply chain of Oman and Tunisia and identified key parameters which governed the supply chain. Sharma et al. [15] highlighted the key perspectives associated with the FSC during the pandemic times of COVID-19. In light of the above discussion to bridge the gap between supply and demand patterns in the supply chain, various challenges are confronted. This paper comprises six factors which are challenges to food supply chain deployment and its strategies deployment (Table 1).
3 Methodology In this work, DEMATEL is used to extract out solutions. DEMATEL is used to differentiate causal and effect group elements out of total elements under consideration. In reference to the literature presented in this paper, it is clear that the evaluation of real challenges by application of DEMATEL is not found in the research literature, which is the objective of this study. The methodology opted for this study is sequenced in Fig. 2. The tool used here for evaluation was DEMATEL laid by the Science and Human Affairs Program of the Battelle Memorial Institute of Geneva in time of 1972–1976 [33, 34]. DEMATEL represents structured modelling flow and ends with a feasible readable solution [11, 33, 35] Steps in the methodology of DEMATEL method are given as follows:
368
J. Sharma et al.
Table 1 Identified challenges to food supply chain management S. no
Identified challenges
Explanation
References
1
Lack of traceability
Consensus about product history, its operationality and stages
[12, 16–18]
2
Perishability
Virtue by which produce gets spoiled, decayed or rated as unfit for consumption if exposed to refrigerated conditions
[10, 19, 20]
3
Quality variation
Variation within the composition of product because of varying storage conditions, improper post and pre-harvest scenarios
[10, 13, 14, 21, 22]
4
Seasonality
Influence on product because of seasonal variability
[23–25]
5
Environmental and climate variability
Variations within climatic and surrounding environmental conditions
[9, 14, 20, 26, 27]
6
Social performance
Safe fresh, healthy and nutritious food in response to market demand along with safe working condition
[24, 28–32]
Identification of challenges in FSC using literature review Compilation of the assessments
Evaluation of FSC challenges using DEMATEL approach based on expert reviews
Sensitivity Analysis
Results and discussion Fig. 2 Research methodology layout
Step 1: Computation of average direct relationship matrix The impact of a factor on other factors is checked and calculated as if to check the impact of jth criteria over kth criteria. When j = k, the diagonal elements are set to zero. For each respondent, a non-negative matrix can be given as X r = [x z jk ]n×n
Assessment of the Challenges Obstructing Performance of Indian …
369
where ‘z’ denotes the number of respondents, i.e., (1 ≤ z ≤ m) and x jk denotes the extent of respondents relying on the relation of jth criterion over kth criterion. It compiles the opinion of ‘m’ respondents and finds out the average direct relation matrix as given below: A = a jk a jk =
(1)
m 1 z x m z=1 jk
Step 2: Exploration of the normalised direct relation matrix ‘N’ N=K×A where K =
n
1
max j=1 a jk 1≤ j≤n
(2)
j, k = 1, 2, 3 . . . n
Step 3: Estimation of total relation matrix ‘T’ using Eq. 3 T = N (I − N)−1
(3)
where I is an identity matrix. Step 4: Generation of the causal diagram In total, the relation matrix, the sum of rows (R) and the sum of columns (C) are calculated. To visualise the relative importance, the values of ‘Prominence’ and ‘Relation’ are calculated with the help of R and C values as (R + C) and (R − C), respectively. Based on ‘Prominence’ and ‘Relation’ values, the categorisation of each criterion into cause and effect groups has been done. For the cause group, positive values of relations are considered, while negative values are considered for the effect group. To model the causal diagram, mapping of (R + C, R − C) dataset has been done. T = t jk n×n ⎡ R=⎣
n (k=1)
⎤
⎡
t jk ⎦
= [t j ]n×1 ; C = ⎣ k×1
n ( j=1)
⎤ t jk ⎦
= [t j ]1×n 1×k
This study encompasses some of the challenges in the food supply chain. Here C1 stands for ‘Lack of Traceability’, C2 stands for ‘Perishability’, C3 stands for ‘Quality Variation’, and C4 stands for ‘Seasonality’, C5 stands for ‘Environmental and Climate Variability’, C6 stands for Social performance. Data was collected from
370
J. Sharma et al.
experts belonging to food industries; it was summarised; and the average direct relationship method was developed. Tables 2, 3, 4 and 5 account for the illustration using the DEMATEL methodology. For mapping of values on ‘Prominence’ and ‘Relation’-based values, causal diagram is made, as shown in Fig. 3. Causal diagram represents relation as cause and effect groups. From Fig. 3 it can be visualised that challenges C3 , C5 , C4 are among the causal group, whereas challenges C1 , C2 , C6 are part of the effect group. Table 2 Average direct relation matrix computation C1
C2
C3
C4
C5
C6
C1
0.0000
1.8182
1.7273
1.4545
1.1818
1.4545
C2
2.1818
0.0000
1.6364
1.1818
1.4545
1.2727
C3
2.0909
2.0000
0.0000
2.0000
1.3636
1.7273
C4
1.6364
2.5455
2.1818
0.0000
2.3636
2.6364
C5
1.9091
1.9091
1.9091
2.1818
0.0000
2.4545
C6
1.4545
1.5455
1.6364
1.7273
1.7273
0.0000
Table 3 Computed direct normalised direct relation matrix C1
C2
C3
C4
C5
C6
C1
0.0000
0.1600
0.1520
0.1280
0.1040
0.1280
C2
0.1920
0.0000
0.1440
0.1040
0.1280
0.1120
C3
0.1840
0.1760
0.0000
0.1760
0.1200
0.1520
C4
0.1440
0.2240
0.1920
0.0000
0.2080
0.2320
C5
0.1680
0.1680
0.1680
0.1920
0.0000
0.2160
C6
0.1280
0.1360
0.1440
0.1520
0.1520
0.0000
Table 4 Computed total relation matrix C1
C2
C3
C4
C5
C6
C1
0.4617
0.6196
0.5821
0.5380
0.5009
0.5780
C2
0.6261
0.4832
0.5779
0.5220
0.5191
0.5673
C3
0.6985
0.7176
0.5306
0.6494
0.5866
0.6798
C4
0.7810
0.8682
0.8014
0.6055
0.7523
0.8540
C5
0.7488
0.7776
0.7371
0.7223
0.5368
0.7945
C6
0.6054
0.6345
0.6073
0.5872
0.5671
0.4996
Assessment of the Challenges Obstructing Performance of Indian …
371
Table 5 Calculated sum of influences assumed and received on criteria D
R
D+R
D−R
C1
3.2803
3.9214
7.2017
−0.6412
C2
3.2956
4.1006
7.3962
−0.8050
C3
3.8624
3.8363
7.6987
0.0261
C4
4.6624
3.6244
8.2867
1.0380
C5
4.3170
3.4629
7.7799
0.8542
C6
3.5011
3.9732
7.4743
−0.4720
Causal diagram for the identified challeneges 1.50
D-R
1.00
C4
C5
0.50 0.00 7.00 -0.50
7.20
7.40 C1
-1.00
C6
7.60
C3 7.80
8.00
8.20
8.40
C2
D+R
Fig. 3 Causal diagram
4 Sensitivity Analysis Sensitivity analysis is done with the aim to check whether the presented results are enough robust and depict feasibility [36]. Sensitivity analysis also gives the idea of the reliability of the perception given by various decision-makers. In this analysis, weights are provided to the experts based upon their designation and experience they pose, and then further calculations of the chosen methodology are accomplished. In this study group, five experts having good knowledge of practical aspects of FSC were chosen. In the first stage of this analysis, equal weights were given to all the five experts and then calculation was made and this was named Scenario 1. In the second stage and third stage, named as Scenario 2 and Scenario 3, the overall weight was kept constant but based on the designation and experience of experts, adjustments were made in the individual weights of experts and then the calculations were made. Table 6 shows the details of weights selected for each scenario of sensitivity analysis. Based upon the weights imparted to the rating providers for different scenarios, calculation was performed accordingly and the obtained result for each scenario is mentioned in Table 7. Based upon the above calculations for better visualisation of the cause and effect parameters for each scenario, causal diagram is appended in Figs. 4, 5 and 6, respectively. After plotting Figs. 4, 5 and 6 for respective Scenarios 1, 2 and 3, a consensus can be derived that the plotted values for various challenges and their ranking remains
372
J. Sharma et al.
Table 6 Details of weights provided to assessments in sensitivity analysis Experts
Scenario 1
Scenario 2
Scenario 3
Perception 1
0.2
0.4
0.3
Perception 2
0.2
0.25
0.2
Perception 3
0.2
0.05
0.15
Perception 4
0.2
0.1
0.25
Perception 5
0.2
0.2
0.1
Table 7 Cause and effect parameters value for each scenario obtained after sensitivity analysis Challenges
Scenario 1
Scenario 2
Scenario 3
D+R
D−R
D+R
D−R
D+R
D−R
C1
7.2017
−0.6412
7.1842
−0.6289
7.4999
−0.6360
C2
7.3962
−0.8050
7.2786
−0.8011
7.5543
−0.8293
C3
7.6987
0.0261
7.4581
0.0192
7.7387
0.0199
C4
8.2867
1.0380
8.086
0.9805
8.3038
0.9848
C5
7.7799
0.8542
7.4994
0.8174
7.9031
0.8346
C6
7.4743
−0.4720
7.2399
−0.4602
7.4446
−0.4839
Causal diagram for Scenario 1 of sensitivity analysis 1.5
D-R
1.038
0.8542
1 0.5
0.0261 0 7 -0.5 -1
7.2 -0.6412
7.4 -0.472 -0.805
7.6
7.8
8
8.2
8.4
D+R
Fig. 4 Causal diagram of Scenario 1 considered for sensitivity analysis
unaffected for various scenarios. Sensitivity analysis depicts that the generated results are close to the perception provided by experts. Hence, assessments grabbed for this study are genuine and impart the true notion behind this study. In Fig. 7, a diagrammed is given to underpin all scenarios under consideration.
Assessment of the Challenges Obstructing Performance of Indian …
373
D-R
Causal diagram for scenario 2 sensitivity analysis 1.2 1 0.8 0.6 0.4 0.2 0 7.2 7.3 -0.2 7.1 -0.4602 -0.4 -0.6289 -0.6 -0.8011 -0.8 -1
0.9805 0.8174
0.0192 7.4
7.5
7.6
7.7
7.8
7.9
8
8.1
8.2
D+R
Fig. 5 Causal diagram for Scenario 2 considered for sensitivity analysis
D-R
Causal diagram for scenario 3 sensitivity analysis 1.2 1 0.8 0.6 0.4 0.2 0 7.5 7.6 -0.2 7.4 -0.4839 -0.4 -0.636 -0.6 -0.8293 -0.8 -1
0.9848 0.8346
0.0199 7.7
7.8
7.9
8
8.1
8.2
8.3
8.4
D+R
Fig. 6 Causal diagram for Scenario 3 considered for sensitivity analysis
5 Results and Discussions The presented work uses analysis done by the DEMATEL approach, depicting mutual importance-based rating for the considered challenges in the food supply chain. The results showed that challenge ‘seasonality’ [C4] has the highest value 8.2867 which is followed secondly by the ‘environmental and climate variability’ [C5] with a value of 0.8542 and thirdly by ‘quality variation’ [C3] by 0.0261. Whereas the challenge ‘social performance’ [C6] outranks fourthly with −0.4720, the challenge ‘lack of traceability’ [C7] stands fifth with −0.6412 and ‘perishability’ ranks least by the value of −0.8050. Hence in the prominence group, the challenges outranking follows the order C4 > C5 > C3 > C6 > C2 > C1, whereas in the effect group it is sequenced as C4 > C5 > C3 > C6 > C1 > C2. In the effect group positive values are placed in
374
J. Sharma et al.
Sensitivity Analysis for all scenarios
8.2867
Scenario 1 D+R 7.6987 7.7799 9 7.3962 8 7.2017 7 6 5 Scenario 1 D-R 4 0.9848 1.038 -0.4839 3 -0.472 0.0199 2 0.0261 0.8346 1 0.8542 8.086 -0.8293 7.2399 0 -0.805 -0.636 -0.6412 -1 7.4581 7.4994 7.2786 7.1842
7.4743
Scenario 3 D-R
8.3038 7.4446 7.7387 7.9031 7.5543 7.4999 Scenario 3 D+R 0.9805 -0.4602
0.0192
-0.6289 -0.8011 0.8174
Series1 Series2 Series3 Series4 Series5 Series6
Scenario 2 D+R
Scenario 2 D-R
Fig. 7 Sensitivity analysis for all three scenarios considered for the study
the causal group whereas negative ones are placed in the effect group to underpin the mutual interaction between the challenges under consideration. In reference to Fig. 3 the causal group and effect group can also be differentiated as challenges ‘Seasonality (C4)’, ‘environmental and climate variability (C5)’ and ‘quality variation (C3)’ belong to the causal group because of positive ‘D − R’ value. However, challenges ‘lack of traceability (C1)’, ‘perishability (C2)’ and ‘social preferences (C6)’ have negative D − R value, which represents that they belong to the effect group. Causal group elements have higher mutual interaction than the effect group elements.
6 Managerial Implications and Future Scope The presented work cluster the various challenges recognised from the core of research literature and after materialising the industrial visits. For the analysis of the challenges, the tool DEMATEL is implied to establish the priority, based upon the mutual interaction. Results portrayed that the challenge ‘Seasonality (C4)’ within the food supply chain is dominating challenge as consistent variability in the conditions affects the food at most. Seasonality refers to the availability of demand and consumption pattern surges for a certain period of time. Moreover, seasonality causes situations that cause serious demand forecasting issues, inability to fulfil consumer
Assessment of the Challenges Obstructing Performance of Indian …
375
demand. Seasonal availability of the product enforces the industries to extract more working capital for procurement and storage. Managers must ensure that aspect of seasonality is part of the framework governing the policy formulation process. Tackling the seasonality during the course of actions of FSC can improve the potencies of venues associated with it. Managers should effectively and efficiently manage the seasonal availability of the various key ingredients to ensure smooth flow of the product through the channels of supply chain. Government should ensure the timely availability of the food commodities which have seasonal production cycles at genuine costs, to ensure hassle-free fulfilment of product demands. The presented work can be more expedited in the direction of product-specific food chains with more number of challenges and covering the wider range of the industries, to capture the regional trends over vivid geographical regimes. Furthermore, other mathematical tools and extension modules of fuzzy, grey sets can be implied to enrich the study.
References 1. Gustafsson J, Cederberg C, Sonesson U, Emanuelsson A (2011) The methodology of the FAO study: global food losses and food waste-extent, causes and prevention Food agriculture organisation, United Nations 2. Stanton LJ (2018) A brief history of food retail. Br Food J 120(1):172–181 3. Manzini R, Accorsi R (2013) The new conceptual framework for food supply chain assessment. J Food Eng 115(2):251–263 4. Blandon J, Henson S, Cranfield J (2009) Small-scale farmer participation in new agri-food supply chains: case of the supermarket supply chain for fruit and vegetables in Honduras. J Int Dev 21(7):971–984 5. Marsden T, Banks J, Bristow G (2000) Food supply chain approaches: exploring their role in rural development. Sociol Rural 40(4):424–438 6. Sharma J, Tyagi M, Bhardwaj A (2020) Parametric review of food supply chain performance implications under different aspects. J Adv Manage Res 17(3):421–453 7. Joshi R, Banwet DK, Shankar R (2009) Indian cold chain: modeling the inhibitors. Br Food J 111(11):1260–1283 8. Tyagi M, Kumar P, Kumar D (2015) Assessment of critical enablers for flexible supply chain performance measurement system using fuzzy DEMATEL approach. Glob J Flex Syst Manag 16(2):115–132 9. Kazancoglu Y, Ozkan-Ozen YD, Ozbiltekin M (2018) Minimizing losses in milk supply chain with sustainability: an example from an emerging economy. Resour Conserv Recycl 139:270– 279 10. Siddh MM, Soni G, Jain R, Sharma MK (2018) Structural model of perishable food supply chain quality (PFSCQ) to improve sustainable organizational performance. Benchmarking Int J 25(7):2272–2317 11. Tyagi M, Kumar D, Kumar P (2015) Assessing CSR practices for supply chain performance system using fuzzy DEMATEL approach. Int J Logis Syst Manage 22(1):77–102 12. Gomiero T (2017) Food quality assessment in organic vs. conventional agricultural produce: findings and issues. Appl Soil Ecol 123:714–728 13. Balaji M, Arshinder K (2016) Modeling the causes of food wastage in Indian perishable food supply chain. Resour Conserv Recycl 114:153–167
376
J. Sharma et al.
14. Mbaga M, Suleiman Rashid Al-Shabibi M, Boughanmi H, Mohamed Zekri S (2011) A comparative study of dates export supply chain performance: the case of Oman and Tunisia. Benchmarking Int J 18(3):386–408 15. Sharma J, Tyagi M, Bhardwaj A (2021) Exploration of COVID-19 impact on the dimensions of food safety and security: a perspective of societal issues with relief measures. J Agribus Dev Emerg Econ. https://doi.org/10.1108/JADEE-09-2020-0194 16. Montanari R (2008) Cold chain tracking: a managerial perspective. Trends Food Sci Technol 19(8):425–431 17. Van Rijswijk W, Frewer LJ, Menozzi D, Faioli G (2008) Consumer perceptions of traceability: a cross-national comparison of the associated benefits. Food Qual Prefer 19(5):452–464 18. Wilson TP, Clarke WR (1998) Food safety and traceability in the agricultural supply chain: using the Internet to deliver traceability. Supply Chain Manage Int J 3(3):127–133 19. Ndraha N, Sung WC, Hsiao HI (2019) Evaluation of the cold chain management options to preserve the shelf life of frozen shrimps: a case study in the home delivery services in Taiwan. J Food Eng 242:21–30 20. Tamplin ML (2018) Integrating predictive models and sensors to manage food stability in supply chains. Food Microbiol 75:90–94 21. Hodges RJ, Buzby JC, Bennett B (2011) Postharvest losses and waste in developed and less developed countries: opportunities to improve resource use. J Agric Sci 149(S1):37–45 22. Muriana C (2015) Effectiveness of the food recovery at the retailing stage under shelf life uncertainity; an application to Italian food chains. Waste Manage 41:159–168 23. Gokarn S, Kuthambalayan TS (2017) Analysis of challenges inhibiting the reduction of waste in food supply chain. J Clean Prod 168:595–604 24. Rohmer SUK, Gerdessen JC, Claassen GDH (2018) Sustainable supply chain design in the food system with dietary considerations: a multi-objective analysis. Eur J Oper Res 273(3):1149– 1164 25. Van der Vorst JG, van Kooten O, Marcelis WJ, Luning PA, Beulens AJ (2007) Quality controlled logistics in food supply chain networks: integrated decision-making on quality and logistics to meet advanced customer demands 26. Borodin V, Bourtembourg J, Hnaien F, Labadie N (2016) Handling uncertainty in agricultural supply chain management: a state of the art. Eur J Oper Res 254(2):348–359 27. Utomo DS, Onggo BS, Eldridge S (2018) Applications of agent-based modelling and simulation in the agri-food supply chains. Eur J Oper Res 269:794–805 28. Aggarwal S, Srivastava MK (2016) Towards a grounded view of collaboration in Indian agrifood supply chains: a qualitative investigation. Br Food J 118(5):1085–1106 29. Beer S, Lemmer C (2011) A critical review of “green” procurement: life cycle analysis of food products within the supply chain. Worldwide Hosp Tourism Themes 3(3):229–244 30. Carter CR, Rogers DS (2008) A framework of sustainable supply chain management: moving toward new theory. Int J Phys Distrib Logis Manage 38(5):360–387 31. Govindan K (2018) Sustainable consumption and production in the food supply chain: a conceptual framework. Int J Prod Econ 195:419–431 32. Sage C (2003) Social embeddedness and relations of regard: alternative ‘good food networks in south-west Ireland. J Rural Stud 19(1):47–60 33. Tzeng G-H, Chiang C-H, Li C-W (2007) Evaluating intertwined effects in e-learning programs: a novel hybrid MCDM model based on factor analysis and DEMATEL. Expert Syst Appl 32(4):1028–1044 34. Wu W-W (2008) Choosing knowledge management strategies by using a combined ANP and DEMATEL approach. Expert Syst Appl 35(3):828–835 35. Mangla SK, Luthra S, Jakhar SK, Tyagi M, Narkhede BE (2018) Benchmarking the logistics management implementation using Delphi and fuzzy DEMATEL. Benchmarking Int J 25(6):1795–1828 36. Tyagi M, Kumar P, Kumar D (2018) Assessment of CSR based supply chain performance system using an integrated fuzzy AHP-TOPSIS approach. Int J Log Res Appl 21(4):378–406
Farm Mechanization Through Custom Hiring Centres in Punjab: A Survey Manik Rakhra and Ramandeep Singh
Abstract By improving the Internet of Things (IoT), a vast amount of sensor knowledge sources will be conveniently accessed using different large-scale IoT systems. Frameworks are used for the processing and interpretation of mathematical evidence and for the decision help in real time and intelligent solutions. Recently, many companies have substituted robotics for human work. Any agricultural robots conduct big tasks such as irrigation, planting and harvesting in agriculture. Agriculture production and efficiency must be improved by the farm power and machinery of the farmer, but it is not easy for any farmer to hit this machinery. Customer hire services provide better scope for consumers by creating different service centers in India to deal with this issue. The strongest is the Argo Support Center situated in Punjab. In agriculture, intelligent farming is mainly concerned with enhancing agriculture quality and production. This output tends to enhance the farmer’s way of life by reducing hard labor and challenging jobs. We are focused on equipment sharing and leasing to establish intelligent farming. In the ownership of machinery concerned with the production of consumer rental facilities, economic feasibility has a significant role. In this regard, engineers and farmers are proposing a technical approach that plays a leading role in exchanging and renting intelligent machinery. Keywords Agriculture · Internet of things · Hiring services · Machinery
1 Introduction The demand of human beings is focused on agriculture since it is known as the principal source of food grains. The addition of emerging technology is one of the biggest obstacles for the creation of countries for the growth of the agricultural sector. We have been confronted with serious problems in recent years, the primary explanation being the lack of awareness of recent methods of intelligent farming. The Food and Agriculture Organization (FAO) anticipates that by 2050 the world’s M. Rakhra (B) · R. Singh Department of Computer Science and Engineering, Lovely Professional University Phagwara, Punjab 14411, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_28
377
378 80
No.of Publications per year
Fig. 1 Count of “IOT in agriculture” publications
M. Rakhra and R. Singh
70 60 50 40 30 20 10
2010
2011
2012
2013
2014
2015
2016
Years
population will increase between 6 billion and 10–11 billion [1] so that it aims to increase efficiency in food production by 70%. It is necessary to enhance the quality of the usage of water in order to improve this productivity. The internet plays a wonderful function in the world. In 1999, a British pioneer assessed the word Internet of Things (IOT) for the first time. IOT gives technical universe awareness. One of the main sectors affected by IOT is agriculture. At the highest stage, knowledge obtained from many heterogeneous sources may be arranged in the form of intelligent algorithms to notify the existing method innovatively. This enables distributed data to run through the wireless sensor network node in real time (WSN). The rise in IOT along with agriculture is shown in Fig. 1. IOT helps provide a package of knowledge such as smart farm sensor data collection, tracking internal operations and growing output hazards, waste control and cost savings, enhancement in market results and productivity. The key goal of this paper is to share the forum for the problems of smart farming. The causes for farmers’ suicides include debt, misuse of alcohol, environment, poor production rates, burden and family obligations, fatalism, lack of water, higher farm prices, private money lenders, usage of agrochemicals and biodiversity declines. Tiny and marginal farm staff are mostly farmers who died by suicide due to various seasonal rains. With no alternate source of revenue besides agriculture, poor farmers must rely on funding to make the significant amount of money required for agricultural infrastructure and its daily spending. Most regional peasants have thus ended their lives by consuming toxins or hanging etc. Farmers’ suicide in the province is not a recent occurrence. Suicide is a relevant topic, where more than one lakh die every year in the Indian context. In Punjab between 2002 and 2011, 4876 farmers and farm employees committed suicide due to debt. The reason is not the transient pattern of civilization but the deep-seated agricultural and rural suffering. Various studies have shown that the pattern of indebtedness has driven farmers to commit suicide, but the social parameter that is responsible for debt is more significant. Farm machinery is necessary in order for farming operations to be timely and effective because of intensive crop growth and the restricted duration between crop and next seed harvest and manpower shortage. Increased crop scale, evolved crop
Farm Mechanization Through Custom Hiring Centres in Punjab …
379
patterns, the usage of large crop yields, sound planting practices etc. involve field mechanization at manageable prices. Effective, cost-efficient and efficient agricultural machinery has also been built and individually made available to farmers through cooperatives. These agro-service centers will be set up across the whole world, one in a community of 4–5 townspeople each. No machines for rice transplantation, sugarcane grinding and collection of cotton, potato and vegetable production are required for Punjab. Job shortages are on the rise and farmers face timely obstacles. This machinery must be adopted, established and promoted by Punjab. The burning of paddy and other crops causes significant environmental concerns. The available machinery could, where necessary, be updated to suit local requirements and encouraged for the management of crop residues. The preservation of crop residues will improve organic soil materials, minimize micro-nutrient scarcity and also increase the potential of soil water retention. The mechanization of dairy practices should be granted high priority in order to improve the rural economy. Service centers for dairy equipment would be developed to provide dairy farmers with costly machinery on a custom rental basis. The upgrading and standardization of effective repair and servicing of functional installations would be preserved. The State shall set down and produce the specifications of various instruments accordingly.
1.1 The Role of Agricultural Mechanization in Emerging Regions This section is aimed at highlighting a broad range of automation in agriculture which includes agricultural system modernization in crop production activities. The present paper highlights the significant roles that precision farming plays in influencing the boost in labor and farm productivity and production with its wide use in agricultural activities. The ownership of agricultural machinery is very limited in many developing countries because of the small farms held by the majority of farmers. In addition, owning of equipment is defined solely by its economic aspects for many farmers in developing countries [2]. It also needs huge investment and depends on the amount of farmers’ finances [3]. This also leads to farmers or individuals collectively using farm machinery in groups or operational ways. In general, all types of organizations are distributed to various groups. In developing countries the following are the ways of owning farming equipment: (1) public hire, (2) private hire, (3) private owner-user with overproduction hire, (4) private landlord-user-exclusive, (5) cooperative acquisition and (6) unofficial joint venture. Such types include hiring machine facilities, which can be coordinated by the public and private sectors.
380
M. Rakhra and R. Singh
2 Research Background Sharma et al. [4] reported that it was very difficult to purchase the machinery due to high price consideration. [5] opined that the performance of the farm operations during the peak periods helps in expeditious, which further helps to save time and increase productivity [5]. In 2000 Aggarwal and Yadav [6] concluded after conducting a meeting in three districts of Haryana State that the trend of tractors is in demand. The use of the bfarm tractor is a very tillage operation. Bhatia [7] reported that a tractor was not a technology like fertilizers and seeds. The use of a tractor is only justified for a particular time period. [8] emphasized that in future agricultural machinery is in demand which is beneficial for the high capacity of crop production [8]. Sarkar [9] reported that in India the initial problem of the farmers is lack of awareness in farm holding and financial support. Kulkarni [10] identified that the lack of knowledge to the farmers is the main concern that dominates the efficiency and productivity [9]. Goyal et al. [11] illustrated that almost 120 agro-service centers have been developed in the district of Bathinda to date for specific agricultural needs [10]. Collective research conducted by Sidhu and Vatta [12] which is further useful in evaluating the contribution of Cooperative Agro Machinery Service Centers (AMSCs) to visualizing the change in Punjab and concluded that hiring machinery services from machinery centers was comparatively cheaper than private operators. In the nutshell, we can say that this success is beneficial for the state to reduce the burden of the farmers which furthermore helps to improve the economy of the state. Custom recruiting has been used as an alternative to buying costly farm equipment. Although it is not just a single report that is accountable for evaluating the state and economic feasibility of recruiting facilities, we may claim on the flip side that such surveys have provided the correct course for the actual position of the enquiry [11].
3 Research Methods This section is linked with detailed methods and the principles used for sample collection. Selection of study districts Table 1 depicts the district-wise list of the number of Primary Agricultural Cooperative Societies in Punjab (Table 2). Table 1 provides the district-wise number of custom-hiring services. There are two divisions in the present paper, i.e., Moga and Ludhiana were randomly picked for further analysis.
Farm Mechanization Through Custom Hiring Centres in Punjab …
381
Table 1 District-wise numbers of primary agricultural cooperative societies Name of the district
Number of PACS (A)
Number of PACS with machines (B)
Percentage of (B) w.r.t. (A)
Amritsar
385
131
Moga
164
145
34.02 88.41
Hoshiarpur
288
105
36.45
Jalandhar
249
155
62.24
Gurdaspur
217
35
16.12
Kapurthala
104
64
61.53
Ferozepur
254
194
76.37
SBS Nagar
144
122
84.72
Faridkot
79
64
81.01
Muktsar
142
105
73.94
Ludhiana
369
366
99.18
Bathinda
187
179
95.72
Mansa
109
84
77.06
Patiala
267
89
33.33
Sangrur
289
274
94.80
Ropar
174
85
48.85
Fatehgarh Sahib
113
105
92.92
3534
2302
65.14
Punjab
Table 2 The farmers face issues
Farmers’ problem
Farmers’ percentage
1. Issue in availability of machine
54
2. Concept of favoritism
11
3. Bad condition of machine
31
4. Cost, if any
7
5. Problem creation in machine
12
6. Non-availability of machine
17
7. Unexpert operators
45
8. Lack of self confidence
62
3.1 Data Collection This collection of data is based upon primary as well as secondary data. The secondary data is obtained from the office of the Registrar Cooperative Societies, Chandigarh and from the office of the Deputy Registrar. The primary data were gathered using a well-prepared method by interviewing the interviewees [12]. This information is collected for the year 2009–2010 which further covers the following aspects.
382
M. Rakhra and R. Singh
• Concern of Farm Machines: This heading included the model of the machine, cost, year of purchase etc. • Running Costs: This includes the expenditure like the cost of fuel, repair, service charges, maintenance, wages and salary paid to staff and drivers etc. • Gross Returns: Gross returns are basically a concern with the custom hire out of the tractors by multiplying the annual use of machines for the specific operations paid per unit (area or hour). • Net Returns: A positive return is derived by subtracting gross returns from annual spending. Primary data was also noted from the selected persons; this data is collected by personal visit. The information is gathered regarding the socio-economic details, kind of machinery used, how much knowledge to operate the machines and whether the farmer needs an operator or not. This kind of data is helpful to improve productivity and efficiency [13, 14].
3.2 Data Tabulation and Analysis The PACS data collected with AMSC were tabulated using the software Excel. To achieve the study’s objectives, the procedure mentioned below was used to interpret the findings using the following formula: Xv = KBt
(1)
where Xv = System number/implement for the non-year, t = Time vector (1, 2, …, n) for each cycle and K = Constant. Log transformation of the above function is ln (Xv ) = ln K + t(ln B)
(2)
4 Results and Discussion Recently, mechanization of the farms is responsible for the incentive cultivation in Punjab. This farm machinery is held responsible to become the smart state [15]. There were so many issues to the farmers that may cause the lack of productivity and efficiency and these problems can be rectified by the provision provided by custom hiring services given by the state government through AMSC established at PACS. However, the AMSC’s economic position is very necessary for this consideration to improve productivity.
Farm Mechanization Through Custom Hiring Centres in Punjab …
383
4.1 Condition of Economic Feasibility of PACS Custom Job Services In the longer run, PACS were kept accountable for maintaining their economic viability. The segment concerned deals with the AMSC’s economic feasibility by assessing the expense of its equipment, maintenance, wages and salaries during the 2009–2010 fiscal year.
4.2 Impact of Custom Hiring Services on Farm Machines For the development of agriculture, mechanization plays a pivotal role. The state agricultural sector has witnessed a decline in innovation agriculture in the recent past. The primary reason for this is yield inflation, which is attributed to the farmers’ lack of awareness. According to research, more than 30% of small farmers hold fewer than 2 ha [16–18]. Custom hiring services are beneficial for the small farming household. This overall study will impact the farm economy like operational holdings, cropping pattern, socio-economic pattern etc. The income and expenditure on both the categories (custom hiring and non-hiring system) were studied and it has been found that custom hiring services are more beneficial than non-custom hiring one. Even then multiple problems are faced by farmers in custom hiring services farm machines. In the nineteenth and twentieth centuries, the major demand of Indian agriculture is manpower, but with the continuous growth, this demand changed from manpower to mechanized power. In the way of agricultural productivity, manpower and mechanized power have been always held main hurdles in the same field. In order to resolve these issues, joint ownership is the most feasible and a great option. This mechanization in Indian agriculture started with the introduction of the green revolution model. In comparison with other states, Punjab has the highest mechanical power in the order of 3.5 kW/ha, whereas mechanical power in the states of Bihar, Jharkhand and Orissa is less than 1 kW/ha. In Punjab, approximately 10 lakh persons are in the agriculture profession, out of which about three lakhs are small. Recent data reported that two lakh small farmers are left farming, the major reason for the uneconomical condition. To promote small farmers custom hiring is a beneficial policy for agricultural mechanization. The recent paper is concerned with the average size of operating landholding in J&K, which is approximately half than average operating holding size in India (the table is mentioned in this paper). During 2005–2006, it has reported that 51.2% animate power and 38.4% farm power is the requirement for the total power [19–22]. It means that animate power still plays a major role in sources of farm power, so it is not possible to replace animal power with mechanical power. To maintain green revolution it is necessary to select, test animal drawn improved implements of pudding, serving and interactual operations [3, 23–26]. This will increase the efficiency of animals and reduce the drudgery of the farmers. This paper basically
384
M. Rakhra and R. Singh
tells about setting up of custom hiring center under NICRA in Wakherwan, district Pulwama. This custom hiring model is helpful to use machines that are unaffordable for the farmers [27, 28]. Therefore this service is helpful to increase the annual use of latest equipments, which is further useful to enhance the productivity.
5 Conclusions In the end, various sectors substitute for robotics for human labor. Several farm robots, like the drainage, planting and harvesting, carry out large agricultural tasks. Agriculture production and efficiency must be improved by the farm power and machinery of the farmer, but it is not easy for any farmer to hit this machinery. Customer recruiting services have greater customer reach to solve this problem through the growth of different service centers in India. The best place in Punjab is Argo Service Centre. In agriculture, intelligent farming is primarily concerned with enhancing agriculture’s quality and productivity. This would continue to strengthen the farmers’ way of life by reducing challenging jobs and hard labor. Our key focus is on the sharing and renting of equipment to develop smart agriculture. The above mentioned are the reasons which are held responsible for the downfall of the yield. In the present research authors are doing their innovative efforts by making an app through which they can enter the data accordingly. This data can be entered in case the peoples are illiterate. The authors collected data by personal visit and found that it is very strange that very few farmers have this information that how they can get a loan for agriculture purpose, what are the necessary equipments, how can be used, how to maintain etc. To the best of the authors’ knowledge, this kind of app has never been developed before. This will definitely help in the field of agriculture.
References 1. FAO. Feeding the world. World summit on food security, 16–18 Nov. 2009. ftp://ftp.fao.org/ docrep/fao/meeting/018/k6021e.pdf 2. Rakhra M, Singh R (2021) Economic and social survey on renting and hiring of agricultural equipment of farmers in Punjab. In 2021 9th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO). pp. 1–5. https://doi. org/10.1109/ICRITO51393.2021.9596343 3. Kaur M (2020) Calcification detection in breast cancer. Int J Psychosoc Rehabil 24(4):5723– 5732. https://doi.org/10.37200/ijpr/v24i4/pr2020377 4. Sharma JL (1974) An analytical study into custom hiring services vis-à-vis agricultural resource productivity. MSc Thesis (unpublished), Punjab Agricultural University, Ludhiana, India 5. Kaur K (1988) Economics of custom hiring of agricultural machinery in the Punjab state. MSc Thesis (unpublished), Punjab Agricultural University, Ludhiana, India 6. Yadav S, Aggarwal S (2000) Economic analysis of utilization of farm tractors in selected districts of Harayana. Agril Engg Today 24(1):14–21 7. Bhatia BS (2000) Tractor—a boon or bane for Punjab farmers. The Tribune, Chandigarh Edition, Chandigarh, October 16, 2011, p 4
Farm Mechanization Through Custom Hiring Centres in Punjab …
385
8. Nagarajan SS (2004) Role of farm implements in crop production. Agro India 8(5):12–14 9. Sarkar A (2007) Farm mechanization—have we progressed. Agril Today 10(11):29–32 10. Kulkarni SD (2009) Mechanization of agricultural—Indian scenario. In: Proceedings of 5th APCAEM TC meeting & expert group meeting on application of agricultural machinery for sustainable agriculture, Philippines 11. Goyal S (2011) 118 agro service centres lend a helping hand to farmers. The Tribune, Chandigarh Edition, Chandigarh, April 1, 2001, p 3 12. Sidhu RS, Vatta K (2012) Improving economic viability of farming: a study of cooperative agro machinery service centres in Punjab. Agril Econ Res Rev 25:427–434 13. Singh K, Sukhpal S, Kaur M, Kingra HS (2007) Flow of funds to farmers and indebtness in Punjab: a study conducted on behalf of the Punjab State Farmers Commission. Punjab Agricultural University, Ludhiana 14. Rakhra M, Singh R (2020) A study of machinery and equipment used by farmers to develop an uberized model for renting and sharing. Mater Today Proc. https://doi.org/10.1016/j.matpr. 2020.11.784 15. Takeshima H, Edeh H, Lawal A, Isiaka M (2014) Tractor owner-operators in Nigeria: Insights from a small survey in Kaduna and Nasarawa states, IFPRI—discussion papers 16. Rakhra M, Singh R (2020) Internet based resource sharing platform development for agriculture machinery and tools in Punjab, India. 978-1-7281-7016-9/20/$31.00 ©2020 IEEE 17. Takeshima H (2015) Drivers of growth in agricultural returns to scale: the hiring in of tractor services in the Terai of Nepal. IFPRI—discussion papers 18. Ahmed A (2013) Farm mechanization in Bangladesh: evidence from IFPRI national household survey. In: Rural mechanization: policy and technology lessons from Bangladesh and other Asian Countries, pp 7–8 19. Soni P, Ou Y (2010) Agricultural mechanization at a glance selected country studies in Asia on agricultural machinery development. In: United Nations Asian and Pacific Centre for agricultural engineering and machinery 20. Bagheri A, Ghorbani A (2011) Adoption and non-adoption of sprinkler irrigation technology in Ardabil Province of lran. Afr J Agric Res 6(5):1085–1089 21. Biggs S, Justice S, Lewis D (2011) Patterns of rural mechanization, energy and employment in South Asia: reopening the debate. Econ Pol Wkly 46(9):78–82 22. Biggs S, Justice S (2015) Rural and agricultural mechanization: a history of the spread of small engines in selected Asian countries. IFPRI—discussion papers. http://csisa.org/wp%0Acont enthwloads/sites/2/201412006/BiggsJustic 23. Bigot Y, Binswanger H (1987) Agricultural mechanization and the evolution of fanning systems in sub-Saharan Africa. Johns Hopkins University Press 24. Clarke LJ (2000) Strategies for agricultural mechanization development: the roles of the private sector and the government. Agric Eng Int CIGR J 2 25. Clarke LJ (1997) ‘No Title’. Agricultural engineering branch, agricultural support systems division. FAO, Rome, Italy, (September) 26. Diao X, Cossar F, Houssou N, Kolavalli S, Jimah K, Aboagye P (2012) Mechanization in Ghana searching for sustainable service supply models. FPRJ—Discussion Papers. http://ebr arv.ifuri.org/cdmfref/collectionlplS738colll57 27. Rakhra M, Singh R (2021) Smart data in innovative farming. Mater Today Proc. ISSN 22147853.https://doi.org/10.1016/j.matpr.2021.01.237 28. Rakhra M, Singh R, Lohani TK, Shabaz M (2021) Metaheuristic and machine learningbased smart engine for renting and sharing of agriculture equipment. Math Prob Eng 2021(5561065):13. https://doi.org/10.1155/2021/5561065
Hospitals’ Selection Under Ayushman Bharat Scheme with Heuristic Search Method Using A** Algorithm Manimay Dev and Dinesh Kumar
Abstract Healthcare in India is diversifying and it has become one of the largest sectors in terms of both revenue and employment generation. Based on the criteria laid down by the Government of India under its Ayushman Bharat-National Health Protection Scheme (AB-NHPS), this paper identifies six elite hospitals among the network of 11 hospitals across the city of Jamshedpur (India) using A** heuristic search algorithm. For this purpose, 15 criteria have been identified, which are both quantitative and qualitative in nature. The edge cost value g(n) and heuristic cost value f(n) for the hospitals have been assigned on the basis of the criteria under AB-NHPS. A** heuristic search has been used to select the set of hospitals and has been proposed in the form of the algorithm to be used for the purpose of the study. The final result identifies six elite hospitals in the city, with one hospital in each region. With the growing population, hospital denseness is also increasing across the country, intending to ameliorate the existing facilities and foster state-of-theart medical technologies. This paper helps in identifying the hospitals with the best medical facilities in a region based on criteria established by the Government of India. The study is based in the city of Jamshedpur in Jharkhand (India), but the hospital identification technique used for the study can be further extended and generalized to be used for the larger number of hospitals. Keywords Ayushman Bharat · A** algorithm · Edge cost value · Heuristic function value
1 Introduction The healthiness of the people and the economic growth of a region are natural concomitance. In today’s world, the number of people suffering from chronic diseases is on the rise along with the rise in the cost of the treatment. As the living standards of M. Dev · D. Kumar (B) Department of Production and Industrial Engineering, National Institute of Technology, Jamshedpur, India e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_29
387
388
M. Dev and D. Kumar
people have soared up, and people are becoming more health-conscious, the healthcare industry is going through a phase of mutation. Of all the intricacies surrounding the healthcare industry, quality of service accounts for the primary concern [32]. As per the WHO report of 2011, hospital density in India per 10,000 people is around 1.8 compared to 39 per 10,000 people in Australia [41]. Also, there is a considerable shortage of support services like a blood bank, ambulance and proper bio-medical waste management [30]. The number of beds available in the hospitals in India per 10,000 population is 7 as compared to 29 in the USA [43]. The condition of public health centres (PHCs) and community health centres (CHCs) are doleful as 10.7% of them are running without regular water supply, about 8% of them are without regular electricity supply; there is a shortfall of more than 18% of pharmacists, 23.4% of nursing staff and 43.3% of laboratory technicians [2]. The hospitals which provide these basic amenities have a very high cost of treatment, which has been increasing at a very rapid pace over the last few decades [30]. The poor service provided by the hospitals and due to the rapid increase in the demand for better facilities had led to a demand–supply gap, which has resulted in the rapid increase in the number of healthcare centres in the country, compromising the service quality. Thus, it has generated the need for the identification of hospitals that satisfies the benchmark criteria for its functioning as specified by the health regulations of the country. The government along with other regulatory authorities should ensure that the hospitals operating in the country should meet the minimum requirements and standards prescribed to them. Despite the presence of numerous factors leading to unsatisfactory conditions at healthcare centres in India, there has been a sign of constant improvement in the services and framework of the healthcare industry, and it is flourishing at the rate of 15% annually [1]. Medical tourism is also growing at a rate of 30% per year and has become a USD 1 billion market [15]. The role of government has been vital in giving regular impetus to this waning and fragile sector. Since the report of the Health Survey and Development Committee of 1946 (also known as Bhore Committee), there have been far-reaching changes in the healthcare industry. In 2018, the Government of India launched the Ayushman Bharat healthcare insurance scheme, also known as Pradhan Mantri Jan Arogya Yojana (PMJAY), which aimed to provide medical coverage and financial assistantship to the vulnerable 100 million families across the country based on Socio-Economic Caste Census (SECC) 2011 data, without any cap on the family size and age. This accounted for more than 40% of the country’s population. In public and empanelled private hospitals, this scheme will work cashless and paperless for secondary and tertiary healthcare facilities through the use of an e-card anywhere in the country. For the empanelment of private hospitals and healthcare centres, the government has set some sturdy and staunched criteria that genuinely cover all basic amenities and medical procedures. Thus, the empanelled hospitals under this scheme must satisfy the minimum requirement for the daily functioning of the hospitals specified by the government. If none of the hospitals in a particular region satisfies the benchmark criteria laid, the model identifies the best among them which provides superior healthcare facilities compared to other hospitals in the region. Hence, to achieve this objective, the paper uses the criteria which include both qualitative and quantitative factors that are in
Hospitals’ Selection Under Ayushman Bharat Scheme with Heuristic …
389
concordance with these specified criteria for selection and ranking of the hospitals. In the present study, 11 top-notch hospitals have been selected across the different regions of the city of Jamshedpur (India) whose names have been anonymized to maintain their privacy, starting from hospital H1 to hospital H11, as shown in Fig. 1. The aim of this work is to identify a network of hospitals in such a way that one hospital is selected from each region of the city which is superlative to all other hospitals in that region. For this purpose, the heuristic search technique using A** algorithm has been put into use. A** algorithm has been used in the past for solving optimal path-finding problems. The path-finding problems are guided by directed arc-weighted graphs and are also called as representation graph which finds the path explicitly from a start node to a goal node. This algorithm is a simple modification of A* algorithm and follows the same working procedure except that the A** algorithm is used in the case where maximization is the objective of node selection during path finding, [16] instead of minimization as in the case of A* algorithm. This paper has been written and organized in the following sequence: Sect. 2 contains the review of previous pieces of literature, which include different books and journals; it is followed by Sect. 3, which details the methodology used in the study; Sect. 4 gives the results and discussions; Sect. 5 provides the research limitations and future scope, and is followed by Sect. 6 which includes the references used for the study.
Fig. 1 Location of the hospitals on the map
390
M. Dev and D. Kumar
2 Literature Review The preliminary objective of the heuristic search is to iteratively improve and find an optimized solution for a search problem using a heuristic function or cost measure [27]. To cut down search time, informed best-first search had been one of the most popular search strategies. A* algorithm has been used to find the shortest solution to the tile puzzle solving [3]. It has been used in 2-D route planning using spatial routing with digitalized maps [23] to solve the mover’s problem for moving objects to find a continuous and collision-free path for six degrees of freedom environment containing multiple obstacles [11]. There are several other informed heuristic search methods commonly used which include greedy best-first search, beam search, algorithm A*, simplified memory-bound A* (SMA*) and iterative deepening A* [33]. The heuristic search approach has been used for the wide range of applications in diverse fields such as finding applications in resource-constrained project scheduling problem (RCPSP) using various exact and meta-heuristic procedures [39], in scenarios to avoid transformer overload and abnormal voltage fluctuations for distribution feeder reconfiguration [37], in dynamic window stage search algorithm (DWS*) in solving flexible manufacturing system (FMS) problems [31], in extending clique initialization scope to extend the subsets of examination [6], in assembly line balancing problem [38], in developing optimum water distribution system (WDS) using genetic algorithm (GA) [25]; in path planning of robot using multi-neuron heuristic search (MNHS) which is a modification of A* algorithm [24], in optimization of the size of test suit during generation of program by software tester [8], in improving web searching using semantic search [17]. Some of the recent works where heuristic search methods have been found applicable are in the mining of the protein complexes in a PPI (protein–protein interaction) models in the study of bioinformatics [42], in solving watchman route problem (WRP) on a grid map using the shortest path [35], in parallel and disk-based search using a two modified variants of A* in the form of new algorithm called PEDAL and PE2A* [21], in mobile tower localization to increase the throughput and service quality using HGLA and MIN–MAX, MAX– MIN, and HEFT comparison [34], in controlling autonomous AI systems through the mental attitude of the users having a brain control interface (BCI) mechanism using the weighted A* (WA*) algorithm [7], in load balancing in an unbalanced electrical network using the number of contractor for swapping to control imbalance index [20], in design on digital IIR filters with different mutation strategies which are used to attenuate signal frequencies through differential evolution (DE) method using a hybrid heuristic search approach and evolutionary binary approximation method for global and local search, respectively [36], in identification of the position and extent of damage on a big structural systems using echolocation found in bats and dolphins, through heuristic echolocation search algorithm (ESA) [28], in solving the matching problem for stable marriage (SM) using heuristic search algorithm MCS [40], in solving cutting and packing problem in a three-dimensional sphere packing inside a parallelepiped [22]; in solving containerization problem for the third-party logistics [18]; in optimizing the design of intelligent control in type-2 fuzzy systems [19];
Hospitals’ Selection Under Ayushman Bharat Scheme with Heuristic …
391
in optimizing the locations of distributed power generating systems and sizing of grids [4]; for optimizing two-sided assembly line balancing problem [26]; in solving home healthcare problems involving services such as nursing, physiotherapy and housekeeping [13], among many others. Further reviews show that the beam search algorithm has been used to identify different industrial objects and to track seam trajectory for an arc welding robot working in a 3-D work envelope [29]. In isomorphism of the directed graph, a backtracking procedure has been used for information retrieval [5]. Tabu search algorithm has been used in developing the heterogeneous fleet vehicle routing problem (HVRP) [14].
3 Methodology 3.1 Selection of Hospitals Using A** Algorithm For path finding and graph traversal, A* algorithm has been widely used as a bestfirst heuristic search algorithm. A* algorithm was invented by Nils Nilsson, Bertram Raphael and E. Hart in the year 1968 [10]. It uses additive cost measure, i.e., the cost of the path is the sum of the costs of its arc. It is expressed in terms of evaluation function f (n) as estimated cost of the path from the start node (S) to the goal node (G) running through some intermediate node n, and is given as f (n) = g(n) + h(n)
(1)
where g(n) is the edge cost value of the currently assessed path from node S to node n. h(n) is the heuristic function value from node n to the goal node G. Let the optimum value of g(n) from node S to node n be denoted by g(n)∗ and the optimum value of h(n) from the node S to goal node G be denoted by h(n)∗ . The admissibility of A* [12] depends on the fact that the value of g(n) ≥ g(n)∗ , i.e., the value of g(n) is upper bound to the value of g(n)∗ and the value of h(n) ≤ h(n)∗ , i.e., the value of h(n) is lower bound to the value of h(n)∗ . The A** algorithm [9] uses the same procedure to conduct the search as done by A* with one anomaly that instead of f (n) = g(n) + h(n), A** uses the evaluation function as f (n) = max g n + h n |n on the current path to n
(2)
392
M. Dev and D. Kumar
Additionally, the selection break ties arbitrarily but in favour of the goal nodes. If A** selects the node m after node n for expansion, then f (n) ≤ f (m) [16]. In the present work, the search is driven by a set of criteria under the cost category and the benefit category, which transcripts the value of g(n) and h(n), respectively. Specifically, a total of six cost criteria (D1–D6) and nine benefit criteria (D7–D15) were selected, which are concomitance to the criteria under the Ayushman Bharat hospital empanelment scheme. Among these criteria, D1, D2 and D7 are quantitative by nature, while the others are qualitative. The list of cost criteria and benefit criteria has been shown in Tables 1 and 2, respectively. A questionnaire was prepared to comprise these factors for the study to understand the degree of inclination of the hospitals and their current status and future scope towards the empanelment under the scheme. Eleven hospitals (H1–H11) were selected for the study in the city of Jamshedpur (India), which is one of the most important and well-known industrialhub of the country, and was subsequently visited. The locations of the hospitals have been shown in Fig. 1 and have been anonymized for the purpose of privacy. At the beginning of the survey, the general manager (GM) and the chief medical officer Table 1 List of cost criteria Notation Criteria D1
Number of beds in the hospital
D2
Number of doctors and specialist surgeons in the hospital
D3
Basic amenities like drinking water, clean sanitation, cafeteria, wheelchair, trolley and waiting lounge with real-time display of prescription status in OPD
D4
Bio-medical waste management facility
D5
Availability of electricity, OTs with air conditioning facility, annual maintenance contract for equipment
D6
Extent of safety measures for fire, flood and earthquake
Table 2 List of benefit criteria Notation
Criteria
D7
Distance of the hospital from the nearest airport (km)
D8
Possibility of getting online appointment
D9
Level of implementation of PACS (Picture Archiving and Communication System) and digital radiology
D10
Availability and spectrum of the in-house pathology (ISO-certified and NABL Accredited)
D11
Level of NABH Accreditation and its ensuing possibility
D12
In-house facilities of ambulance, blood bank unit, pharmacy, kitchen and laundry
D13
Level of integration of services to MIS, SAP and CCTV system at public locations
D14
Availability of beds in ICU, NICU, HDU and Emergency Care
D15
Availability and spectrum of in-house radiology facility
Hospitals’ Selection Under Ayushman Bharat Scheme with Heuristic … Table 3 Linguistic term for rating the alternatives on a 10-point scale for qualitative criteria
393
Lingual terms
Rating (Q)
Exceptionally good (EG)/exceptionally high (EH) 10 Very good (VG)/very high (VH)
9
Moderately good (MG)/moderately high (MH)
7
Fair (F)/moderate (M)
5
Moderately poor (MP)/moderately low (ML)
3
Very poor (VP)/very low (VL)
2
Very very poor (VVP)/very very low (VVL)
1
(CMO) of the hospitals were contacted through email and over phone calls; and meetings were arranged for a formal visit to the hospitals. The insights provided by the managers and medical officers were very valuable and were extremely helpful to get a better understanding of the functioning of the hospitals. Also, the annual reports provided by the hospitals were of great help in data collection along with the information available on their official website. After the satisfactory level of data was available for the analysis, the edge cost value, i.e., g(n), and heuristic function value, i.e., h(n), was assigned to each node (hospital) using the linguistic term ratings on a 10-point scale shown in Table 3 for qualitative criteria and Table 4 for quantitative criteria, respectively. The final representation showing the nodes and paths along with the heuristic values have been shown in Fig. 2. Here the value of g(n) located between two nodes is the average of the values obtained under cost criteria (see Table 1) for each node. Hence for the two nodes ‘a’ and ‘b’, the value of g(n) can be calculated as Qa + Qb (3) g(n)a−b = 2 where Q a is the value of g(n) for the node ‘a’ and Q b is the value of g(n) for node ‘b’. Table 4 Rating of the alternatives on a 10-point scale for quantitative criteria S. No
D1
D2
D7
Rating (Q)
1
≥501
≥100
≤3.9
10
2
301–500
80–99
4–7.9
9
3
101–300
60–79
8–11.9
7
4
51–100
40–59
12–15.9
5
5
21–50
20–39
16–19.9
3
6
11–20
10–19
20–23.9
2
7
4–10
≤9
≥24
1
394
M. Dev and D. Kumar
On arrow: g(n)
START
On node: h(n)
Region 1: Golmuri
H1
Region 2: Adityapur
Region 3: Bistupur
Region 4: Telco
49
55
H2
H4
H3
H5
77
46
Region 6: Tamolia
43
H6
42
H8
H7
32
28
39
Region 5: Sakchi
23
H10
H9
H11
0
GOAL
Fig. 2 Search graph for the selection of hospitals using A** algorithm
50
82
Hospitals’ Selection Under Ayushman Bharat Scheme with Heuristic …
3.2 Algorithm for the A** Search for Hospital Selection
395
396
M. Dev and D. Kumar
4 Results and Discussions This paper presents the use of a heuristic-based search algorithm to find the series of eminent hospitals among a group of hospitals that provides the best medical facilities in the locality. To achieve this, 11 hospitals (H1–H11) were identified for the study from the city of Jamshedpur (India). These hospitals were evaluated on the basis of the criteria led by the Government of India under its prestigious Ayushman Bharat scheme launched in 2018 for the empanelment of the hospitals with the aim to provide free healthcare insurance to around 100 million socio-economically weaker families in India. The search is based on heuristic algorithm A**, which has been used in the selection of the hospitals in such a way that at least one hospital is selected from each region of the city to provide at least the minimum coverage of hospitals for treatment. The search heuristics are based on the data collected through interviews, surveys, annual reports of the hospitals and from the hospital’s website. The selected hospitals identified as H1-H2-H6-H7-H10-H11, as shown in Fig. 2, are the top-notch hospitals in the city fulfilling the requirements of the basic amenities and are equipped with advanced medical facilities and day-to-day functioning requirements as specified under the Ayushman Bharat scheme. For every region, the node with the highest value of f (n) can be taken as the best hospital among all other hospitals present in the region. Hence, the methodology can be used for a large set of hospitals spread over the big regions covering the state and at the country level. At a time when there is growing health complications among the people and there is always a high demand for emergency medical services, there is a need for the hospitals to improve their medical facilities and provide quality health services. There is a need to identify the hospitals and healthcare centres that are functioning at par with the international standards, which can be trusted by the citizens and could be easily accessible at times when emergency services are required.
5 Limitation and Future Scope The study has been carried out in the city of Jamshedpur, which is the first planned industrial city of India. Being an important and one of the oldest industrial towns in the country, the healthcare facilities available in the city are above par with the healthcare facilities available in the neighbouring areas. In the small cities where healthcare facilities are substandard and critical life support systems are not available, the study with the present set of criteria might be difficult. The hospitals in small towns and villages may not be able to meet the standards set under Ayushman Bharat. Therefore, the criteria used in the study might have to be more lenient in order to ensure more coverage and proper reachability of the benefits of this insurance scheme to the poor people who are living in remote parts of the country. The study here presents a small working model of this search algorithm; hence it can be used for a large network of hospitals covering the entire state or the country. Further, this study used the A**
Hospitals’ Selection Under Ayushman Bharat Scheme with Heuristic …
397
algorithm in order to find the best path for the hospital selection criteria, but the work can also be carried out using other heuristic and meta-heuristic search algorithms, which can give better time complexity and more accurate results with the given constraint. This study can also be used by the private health insurance companies for the empanelment of hospitals with the criteria that best suit their operational strategy. Acknowledgements The authors would like to express thanks of gratitude to the medical officers, consultants and healthcare experts who participated in the survey and gave their valuable time and feedback. They are also thankful to the hospitals that participated in the study (their names have been kept confidential). The authors would also like to extend their sincere thanks to members of our institute’s department for their support in carrying out the study.
References 1. Acharyulu GVRK, Raja Shekbar B (2012) Role of value chain strategy in healthcare supply chain management: an empirical study in India. Int J Manag 29(1):91 2. Bajpai V (2014) The challenges confronting public hospitals in India, their origins, and possible solutions. Research article. Adv Public Health 3. Banerji RB (1982) Theory of problem solving: a branch of artificial intelligence. Proc IEEE 70(12):1428–1448 4. Banhidarah AK, Al-Sumaiti AS (2018) Heuristic search algorithms for optimal locations and sizing of distributed generators in the grid: a brief recent review. In: 2018 advances in science and engineering technology international conferences (ASET), pp 1–5. 5. Berztiss AT (1973) A backtrack procedure for isomorphism of directed graphs. J ACM 20(3):365–377 6. Carter MW, Johnson DG (2001) Extended clique initialisation in examination timetabling. J Oper Res Soc 52(5):538–544 7. Cavazza M, Aranyi G, Charles F (2017) BCI control of heuristic search algorithms. Front Neuroinform 11:6 8. Chen TY, Lau MF (1995) Heuristics towards the optimisation of the size of a test suite. WIT Trans Inf Commun Technol 14(10) 9. Dechter R, Pearl J (1985) Generalised best-first search strategies and the optimality of A*. J ACM 32(3):505–536 10. Dechter R, Pearl J (1988) The optimality of A*. In: Kanal L, Kumar V (Eds) Search in artificial intelligence. Symbolic computation. Springer, New York, pp 166–99 11. Donald BR (1987) A search algorithm for motion planning with six degrees of freedom. Artif Intell 31(3):295–353 12. Doosun K, Kevin L (2012) Revisiting optimal water-distribution system design: issues and a heuristic hierarchical approach. J Water Resour Plan Manage 138(3):208–217 13. Ebendt R, Drechsler R (2009) Weighted A∗ search—unifying view and application. Artif Intell 173(14):1310–1342 14. Fathollahi-Fard AM, Hajiaghaei-Keshteli M, Mirjalili S (2019) A set of efficient heuristics for a home healthcare problem. Neural Comput Appl, March 15. Gendreau M, Laporte G, Musaraganyi C, Taillard ÉD (1999) A tabu search heuristic for the heterogeneous fleet vehicle routing problem. Comput Oper Res 26(12):1153–1173 16. Govindarajan V, Ramamurti R (2013) Delivering world-class health care, affordably. Harvard Business Review, November 1, 2013 17. Gregorics T (2014) Remarks on the A** algorithm. Acta Univ Sapientiae Informatica 6:190– 205
398
M. Dev and D. Kumar
18. Guha R, McCool R, Miller E (2003) Semantic search. In: Proceedings of the 12th international conference on world wide web. WWW ’03. ACM, New York, NY, USA, pp 700–709 19. Hajiaghaei-Keshteli M, Afshari AJ, Nasiri E (2017) Addressing the freight consolidation and containerization problem by recent and hybridized meta-heuristic algorithms. Int J Eng 30(3):403–410 20. Hamza MF, Yap HJ, Choudhury IA (2017) Recent advances on the use of meta-heuristic optimization algorithms to optimise the type-2 fuzzy logic systems in intelligent control. Neural Comput Appl 28(5):979–999 21. Handhal FK, Rashid AT (2018) Load balancing in distribution system using heuristic search algorithm. In: International conference on advances in sustainable engineering and applications, ICASEA 2018—Proceedings, pp 48–53 22. Hatem M, Burns E, Ruml W (2018) Solving large problems with heuristic search: generalpurpose parallel external-memory search. J Artif Intell Res 62:233–268 23. Hifi M, Yousef L (2015) A dichotomous search-based heuristic for the three-dimensional sphere packing problem. Cogent Eng 2(1):994257 24. Holmes PD, Jungert ERA (1992) Symbolic and geometric connectivity graph methods for route planning in digitized maps. IEEE Trans Pattern Anal Mach Intell 14(5):549–565 25. Kala R, Shukla A, Tiwari R (2011) Robotic path planning in static environment using hierarchical multi-neuron heuristic search and probability based fitness. Neurocomputing 74(14):2314–2335 26. Li Z, Kucukkoc I, Mukund Nilakantan J (2017) Comprehensive review and evaluation of heuristics and meta-heuristics for two-sided assembly line balancing problem. Comput Oper Res 84(August):146–161 27. Lu JJ, Zhang M (2013) Heuristic search. In: Dubitzky W, Wolkenhauer O, Cho K-H, Yokota H (Eds) Encyclopedia of systems biology. Springer, New York, NY, pp 885–886 28. Nobahari M, Ghasemi MR, Shabakhty N (2017) A novel heuristic search algorithm for optimisation with application to structural damage identification. Smart Struct Syst 19(4):449–461 29. Presern S, Gyergyek L (1983) An intelligent tactile sensor-an on-line hierarchical object and seam analyzer. IEEE Trans Pattern Anal Mach Intell PAMI 5(2):217–220 30. Ramani KV, Mavalankar D. (2006) Health system in India: opportunities and challenges for improvements. J Health Organ Manag;20(6):560–572 31. Reyes A, Yu H, Kelleher G, Lloyd S (2002) Integrating petri nets and hybrid heuristic search for the scheduling of FMS. Comput Ind 47(1):123–138 32. Roncarolo F, Boivin A, Denis J-L, Hébert R, Lehoux P (2017) What do we know about the needs and challenges of health systems? A scoping review of the international literature. BMC Health Serv Res 17(1):636 33. Russell SJ, Norvig P (2003) Artificial intelligence: a modern approach, 2nd ed. Pearson Education 34. Sahu DP, Singh K, Manju M, Taniar D, Tuan LM, Son LH, Abdel-Basset M, Long HV (2019) Heuristic search based localisation in mobile computational grid. IEEE Access 7:78652–78664 35. Seiref S, Jaffey T, Lopatin M, Felner A (2020) Optimally solving the watchman rout problem on a grid with heuristic search. In: Proceedings of the international conference on automated planning and scheduling, Vol 2020 36. Sidhu DS, Dhillon JS, Kaur D (2017) Hybrid heuristic search method for design of digital IIR filter with conflicting objectives. Soft Comput 21(12):3461–3476 37. Taylor T, Lubkeman D (1990) Implementation of heuristic search strategies for distribution feeder reconfiguration. IEEE Trans Power Delivery 5(1):239–246 38. Tonge FM (1960) A heuristic program for assembly line balancing. RAND Corporation, Santa Monica, CA. Retrieved from https://www.rand.org/pubs/papers/P1993.html 39. Ul N, Wani H, Taneja K, Adlakha N (2013) Health system in India: opportunities and challenges for enhancements. IOSR J Bus Manage. p-ISSN 2319-7668 40. Vanhoucke M (2010) A scatter search heuristic for maximising the net present value of a resource-constrained project with fixed activity cash flows. Int J Prod Res 48(7):1983–2001
Hospitals’ Selection Under Ayushman Bharat Scheme with Heuristic …
399
41. Viet HH, Trang LH, Tuyen LP, Chung T (2020) A shortlist-based bidirectional local search for the stable marriage problem. J Exp Theor Artif Intell 32(1):147–163 42. Wang R, Wang C, Liu G (2020) A novel graph clustering method with a greedy heuristic search algorithm for mining protein complexes from dynamic and static PPI networks. Inf Sci 522:275–298 43. WHO (2011) GHO | By Country | India—statistics summary (2002—Present). WHO. 2011. Retrieved from http://apps.who.int/gho/data/node.country.country-IND?lang=en
Role of Blockchain-Oriented Smart Contract in Supply Chain Abhishek Rajput, D. J. Ghode, and Rakesh Jain
Abstract Supply chain visibility and optimisation were always a concern of utmost importance due to the increasing complexity of the network of trading parties in the chain which result in inefficient processes. This has led to a higher cost and poor customer satisfaction. A novel elucidation for this old-age problem of trust can be given by blockchain technology (BT). It is an evolving technology that comforts the supply chain by providing means to track and manage the goods movement in the chain using a decentralised network that is secured enough and provides the benefit of removing the intermediaries. With the introduction of the blockchain, the smart contract has gotten one of the most looked for after advancements on account of the high customisation they supplement to transactions. In this paper, we will focus on smart contracts as one of the probable future research areas of BT. This paper is a literature review-based research of smart contracts in supply chain management, which includes analyses of the technological principles, benefits and challenges and how it works. The paper also deals with the historical background of the smart contracts to make the readers familiar with the subject which eventually will help in a better understanding. This work may provide a platform for further research with a wider scope on this topic. Looking at the current scenario in the supply chain and logistics industry, this paper can also guide the business processes to implement blockchain-based applications in the industry. Keywords Supply chain management · Smart contract · Blockchain technology · Literature review
A. Rajput · D. J. Ghode (B) · R. Jain Department of Mechanical Engineering, Malaviya National Institute of Technology, Jaipur 302017, India © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_30
401
402
A. Rajput et al.
1 Introduction With the growing complexity of networks in the supply chain, it has become cumbersome to efficiently manage the chain. The traditional ways of managing the supply chain pose some problems, such as coordination of different stakeholders and troubles related to finance and lead-time. Global businesses are concerned with a lack of trust between the parties, authenticity of the products and the real ownership issues [13]. The next main challenge is opaqueness in the supply chain as hardly a few companies know about their second- or third-tier suppliers. This sometimes leads to improper planning of resources and hampering their performance. The customer is unaware of the origin of the food that is being consumed and whether that product has followed a correct path in the chain. Transparency and traceability in the chain will ensure the efficient flow of products from raw material to finished form, will help to mitigate risks and provide analytics to improve operations. These issues can be resolved with the help of disruptive technology, which we have in hand is to use blockchain-oriented smart contract in the supply chain. Smart contract is a software code following some rules, which once implemented is immutable and stored in a decentralised way, hence making sure that the data of transaction is authentic and visible to every stakeholder in the supply chain on a real-time basis. When closely looked into the application of smart contract, it is found that the majority of them are linked to supply chain [11]. Thus, this paper will discuss the research question as to what could be the extent and role of smart contract implementation in supply chain. Section 2 will give a basic understanding of smart contract and blockchain technology and will also link it to supply chain management with the help of available literature. Section 3 will provide the research methodology, and define research questions and show the process of how the papers were screened for reading. In Sect. 4 answers to the research question is given and further reviewed in Sect. 5. In the last section conclusion of the paper is given.
2 Research Background Before diving into the principles of smart contract, let us discuss the underlying technology behind it, blockchain. It is an immutable distributed ledger technology. In blockchain, the user creates a transaction that is sent to the network for mining. After mining, which is basically a process of verification; the transaction is stored in the block form. A single block consists of many such transactions. These blocks are then connected to each other by a cryptographic hash which is immutable in a unidirectional way [14]. Using the BT platform like Ethereum, a decentralised application (Dapps) can be developed, and at the core of Dapps there is smart contract. They are pieces of self-executable code that codify business logic and works on the principle of “If this then that”; i.e., when it is implemented it checks for some
Role of Blockchain-Oriented Smart Contract in Supply Chain
403
conditions and if conditions are met then the desired set of action takes place as per the code [7]. They facilitate three functions: (1) they store rules, (2) they verify rules and (3) they are self-executable. The best example to understand smart contract is with vending machine at the airport or metro stations. Every vending machine stores a rule that if you insert a dollar you may get a snack of choice and then when you really insert a dollar it verifies the amount and after the correct verification you get a snack of your choice. This is how smart contract works. Smart contract is a binding agreement between the stakeholders that they follow as per the rules written in it [1, 18]. And because they are blockchain-oriented there is no intermediary, so it executes automatically. The world of business is in the process of embracing smart contract having numerous applications in IoT, financial sector, supply chain, healthcare etc. How Blockchain is linked with SCM? As it has been already discussed about the problems that traditional supply chains are facing of transparency and traceability, these can only be reduced with the help of technology that promises to collect the data accurately and store it properly so that it cannot be tampered [1]. For efficiently tracking the products we need to implement some type of barcode, QR code or RFID tag system in which data can be easily stored and read with the help of readers. But in the current times, the rising problem of cloning this tag system is becoming a major concern which is hampering the reputation of the firm as counterfeited products are bound to be used by end customers [17]. The supply chain currently uses an ERP system to store their secure data, but the problem lies in whether that data stored in the cloud system is free from hacking or is that data safe, as it is stored in a centralised way. Here BT comes to the rescue, as it will store the data in a decentralised manner where there are multi nodes that store the data. Therefore, if somehow the data is erased from one node, it can be easily recovered from the other nodes, as it is almost impossible to destruct the complete data on the network [6]. In Fig. 1 how the traditional supply chain will be transformed is shown. Based on the above literature, it becomes necessary for researchers to dive further and to know the positive side of this technology, and also to understand the barriers that are faced in the implementation.
3 Research Methods In this paper systematic mapping study is used to sort the articles and to answer our research questions. This study is quite helpful in answering questions related to software engineering field which is smart contract in our case. The aim of systematic mapping process is to illustrate the state of information available in research articles. It basically analyses the results, which takes into account the frequency of published articles; hence providing wide coverage to the research area [7, 10]. The procedure for the systematic mapping is given in Fig. 2.
404
A. Rajput et al.
Traditional Supply Chain Raw Materials
Supplier
Manufactuer
Distributor
Retailer
Consumer
Smart-contract Based Supply Chain
Raw material
Supplier
Consumer
Smart Contract
Retailer
Manufacturer
Distributor
Fig. 1 Traditional supply chain process transformation into smart supply chain Process Steps Definition of research
Conduct Search
Screening of papers
Keywording using abstracts
Data extraction and mapping
Review Scope
All Papers
Relevant Papers
Classification Scheme
Systematic Map
Outcomes
Fig. 2 The systematic mapping process [10]
Role of Blockchain-Oriented Smart Contract in Supply Chain
405
The first step was to finalise the research question (RQ) as to what motivated us to go in this research. As a result, two RQs were finalised. RQ 1: What are the benefits of blockchain-oriented smart contract supply chain? RQ 2: What are the challenges that relate to the application of smart contracts in supply chain management? In the second step, a search was conducted between August 2019 and March 2020 using different combinations of keywords such as smart contract, supply chain management, distributed ledger, blockchain etc. After this, we were having the most tedious job of screening the papers. Screening was done on the basis of title, abstract reading and by also giving importance to citation of articles. The articles which suited to answer the research questions were finalised for the second phase of screening. In the second phase, inclusion criteria for further screening were used, which was to finalise a paper that contains the benefits and challenges of smart contract for full-text reading. After going through all the articles, the quest for answers is discussed in the next section.
4 Results After the second phase of screening, 18 papers were thoroughly read so that we can answer the research question.
4.1 RQ 1: What are the Benefits of Blockchain-Oriented Smart Contract Supply Chain? After going through the articles, finally, we were able to identify the benefits of smart contract in supply chain management, which are further discussed in the points form.
4.1.1
Trust
Trust came out to be the prominent benefit, which most of the researchers have talked about. After implementing smart contract, every stakeholder is pretty confident that they will first get the accurate data about the transaction and secondly, a sense of trust will be developed when they know that these transactions are performed on the basis of some pre-written rules, which is known to everyone and is immutable once the smart contract is implemented in the supply chain [9, 18].
406
4.1.2
A. Rajput et al.
Disintermediation
It is the process of removing the intermediary in the supply chain, as earlier the stakeholder’s bank plays the role of intermediary in financial transactions; but after smart contract, there is no need for banks, as financial transactions can be selfexecutable after using smart contract once approved by all the parties [19]. Due to trust among parties of supply chain, there is no hindrance in implementing the smart contract in complete supply chain. This has manifold benefits like lowering the cost of supply chain transactions and saving lead-time in the chain due to disintermediation. Disintermediation also eliminates the single node of failure and distributes the risks associated, therefore making it more durable [12, 16].
4.1.3
Product Provenance
Product provenance is basically to know the origin of the product and a record of ownership transfer takes place when it moves downstream in supply chain. This record becomes very crucial while dealing with the luxury items like diamond, antiques, and authentic products like aeroplane spare parts or organic food chain [19]. With the application of smart contract every transaction is stored in the block form which consists of timestamp and transaction hash of the previous block, which is quite helpful in tracking all the transactions to find the previous owner and the origin of the product [8, 14]. Due to this tracking and traceability, the linear economy can be converted into the circular economy as explained by Casado-vara et al. [2].
4.1.4
Operational Improvements
When data collected is secure enough, everyone looks for operational improvements in the business. As there is a large volume of accurate data collected, the companies can effectively monitor their processes and be able to spot the errors in the processes. Smart contracts are basically a written code that works on the principle “If this then that”; therefore, it would be quite easy to provide mass customization in the products offered to the end customer with the help of mass data that can tell which product is generating more revenue or which model is giving more production [2, 19]. Inventory can be managed more effectively, and new products and designs can be developed by this data [3].
Role of Blockchain-Oriented Smart Contract in Supply Chain
407
4.2 RQ 2: What are the Challenges that Relate to the Application of Smart Contracts in Supply Chain Management? After talking about the benefits of smart contract, it sounds appealing to use them everywhere but unfortunately, we do not see many cases around implementing this technology in businesses. Hence, there might be some challenges that are faced by smart contract application, which is going to be discussed in the next segment.
4.2.1
Privacy concerns
Smart contract is for sure going to increase the customer experience in terms of quality and reliability, but if we look deeper, we also see that it can also have a negative impact on the manufacturer. Suppose a firm implements this technology and gains profit in terms of customer satisfaction through transparency and traceability. At the same time, there may be a risk to trade secrets, designs, sales data that could be used by their competitor [1]. It poses a serious challenge as to how the information disclosure could be safe and not used for unethical means [5, 14].
4.2.2
Compromise Performance
To successfully adopt this technology, firms have to trade-off their performance for achieving scalability as new hardware and software brings the cost for implementation, also in terms of higher cost of practitioners [7]. There is also an additional cost associated with smart contract, i.e., mining cost, which is a compulsory cost that has to be provided by the stakeholders to get their transaction mined and stored in blockchain [5]. Hence, this scalability issue is holding back the implementation of smart contract.
4.2.3
Legal issues
Another significant aspect that stole the attention of supply chain stakeholders was the legality issues of smart contract. As per the Berne Convention, which is followed in many parts of the world, the copywritten transfer should be in written form. However, smart contract is basically a computer program. Therefore, it arises the question of the validity of these smart contracts [11]. Secondly, the global supply chain is dealing with diverse laws and regulations which have different ownership transfer rules throughout the world. It can result in conflicts among the stakeholders [6, 15].
408
4.2.4
A. Rajput et al.
Technological Complexity
Immutability is seen as one of the main advantages of this technology, but this is also becoming a hindrance in the application as if by mistake the data entered by the party is erroneous; it is practically not possible to change it. So, data entry and transaction should be done in a very sound manner [4, 19]. Second is the lack of business models and the practises from which new firms can implement this technology. This high degree of computerisation is also a problem for the stakeholders from the developing nations and least developed nations, as they are not capable to pay for this technology. Some researchers also have thought that the introduction of this technology may overcomplicate the existing supply chain system [14].
5 Discussions In this paper, we have presented a foundation of smart contract working on the principle of blockchain and have shown how it can be beneficial if implemented in the existing supply chain. After a literature review, we came with our research question which was answered in Sect. 4. Due to its impressive benefits, many businesses are looking forward to adopting this technology, but it also poses some challenges. Now we will discuss what could be the solution and future research work in smart contract. The first challenge, which we found of privacy, can be solved to some extent with the help of the use of private blockchain. In this beforehand, the contract rules will be made in such a way that the crucial information of an organisation is not at stake and also to limit the visibility of information disclosure. The second challenge faced in the adoption is to sacrifice performance which can be solved by developing some incentive mechanisms for small firms and larger firms should look at the bigger picture. An incentive should also be provided in transaction cost which is a compulsory cost and is borne by every stakeholder. The next barrier was the legality issues due to variability in rules and regulations as per the geographical location. To reduce this, various governments should come forward and form mechanisms to provide the same legal positions to smart contract independent of the location worldwide. And in the last, the technological barriers can be eliminated by performing more and more research in this field so that the business model can be optimized further. Developed nations should help the developing nation to adopt smart contract in supply chain. As our whole global business is interconnected to become profitable, every stakeholder should benefit from this technology.
6 Conclusion Blockchain technology has proved to be disruptive in each field whether that may be the financial sector, healthcare, food safety or real estate. But it has shown that
Role of Blockchain-Oriented Smart Contract in Supply Chain
409
supply chain would be the most transformed sector with blockchain. It has emerged as the future of supply chain or we can say that smart contract is transforming existing chains into smart supply chain. With this technology, people will know whether the products they are using is authentic or not. To conclude, this paper has provided with the remarkable benefits of blockchain-oriented smart contract in supply chain such as improved trust between the stakeholders, removal of intermediary, tracking the product and improved operations and also shown the barriers in its adoptions like privacy concerns, legality issues, compromising with performance and complexity in technology. Now it is for the researchers to investigate further to help the businesses and devise models for the successful implementation of this technology in supply chain.
References 1. Azzi R, Chamoun RK, Sokhn M (2019) The power of a blockchain-based supply chain. Comput Ind Eng 135(June):582–592 2. Casado-vara R, Prieto J, Corchado M (2018) How blockchain improves the supply chain. Proc Comput Sci 134:393–398 3. Cole R, Stevenson M, Aitken J (2019) Blockchain technology: implications for operations and supply chain management 4(January):469–483 4. Ghode D, Yadav V, Jain R, Soni G (2020) Adoption of blockchain in supply chain: an analysis of influencing factors. J Enterp Inf Manag 33(3):437–456 5. Helo P, Hao Y (2019) Blockchains in operations and supply chains: a model and reference implementation. Comput Ind Eng 136(July):242–251 6. Kshetri N (2018) Blockchain’s roles in meeting key supply chain management objectives. Int J Inf Manage 39(December 2017):80–89 7. Macrinici D, Cartofeanu C, Gao S (2018) Smart contract applications within blockchain technology: a systematic mapping study. Telematics Inform 35(8):2337–2354 8. Montecchi M, Plangger K, Etter M (2019) It’s real, trust me! Establishing supply chain provenance using blockchain. Bus Horiz 62(3):283–293 9. Nawari NO, Ravindran S (2019) Blockchain and the built environment: potentials and limitations. J Build Eng 25(October 2018):1–16 10. Petersen K, Feldt R, Mujtaba S, Mattsson M (2008) Systematic mapping studies in software engineering. In 12th international conference on evaluation and assessment in software engineering, EASE 2008 (February 2015), pp 1–10 11. Prause G (2019) Smart contracts for smart supply chains. IFAC-Pap Line 52(13):2501–2506 12. Queiroz MM, Telles R, Bonilla SH (2019) Blockchain and supply chain management integration: a systematic review of the literature. Supply Chain Manage Int J 24(1):62–84 13. Saberi S, Kouhizadeh M, Sarkis J, Shen L (2019) Blockchain technology and its relationships to sustainable supply chain management. Int J Prod Res 57(7):2117–2136 14. Saveen A, Monfared RP (2016). Blockchain ready manufacturing supply chain using distributed ledger. Int J Res Eng Technol 05(09):1–10 15. Savelyev A (2017) Contract law 2.0: ‘Smart’ contracts as the beginning of the end of classic contract law. Inf Commun Technol Law 26(2):116–134 16. Tönnissen S, Teuteberg F (2019) Analysing the impact of blockchain-technology for operations and supply chain management: an explanatory model drawn from multiple case studies. Int J Inf Manage (April):1–10 17. Toyoda K, Takis Mathiopoulos P, Sasase I, Ohtsuki T (2017) A novel blockchain-based product ownership management system (POMS) for anti-counterfeits in the post supply chain. IEEE Access 5:17465–17477
410
A. Rajput et al.
18. Wang Y, Han JH, Beynon-Davies P (2019) Understanding blockchain technology for future supply chains: a systematic literature review and research agenda. Supply Chain Manage 24(1):62–84 19. Wang Y, Singgih M, Wang J, Rit M (2019) Making sense of blockchain technology: how will it transform supply chains? Int J Prod Econ 211(November 2018):221–236
Propeller Design and Optimization for Drones Vyom Patel, Keval Nikam, Shantanu Dikshit, Manav Agarwalla, and Chinmay Zagade
Abstract The current research work seeks to propose a design of a propeller that can be used for powering drones. The drone industry is currently in scope to move forward immensely. The research would provide insight in designing a propeller and optimizing it for the generation of maximum lift while considering the drag forces to be encountered. The propeller is designed for an application of a two-rotor drone which can be used for other configurations by some minor changes and obtaining the lift required for the application of the particular configuration. The behaviour of propellers may differ from configuration as the distance between the propeller disks, and the number of blades on the propeller disk plays an important role in deciding the final outcome from the blades. The aim is to design a propeller blade and change the blade profile according to the analysis results, which is an important aspect of the research. Keywords Propeller design · Lift · Drag · Airfoil · Optimization
1 Introduction The propeller design is a very broad concept since we use a propeller for ample applications, which counts from a simple fan to heavy application in aircrafts. Previously, there have been developments in the drone industry in some of the conventional designs; the quad design is most common nowadays. The design for two rotors is generally avoided due to stability issues. The proper balance between the structure is important to keep the drone in the air. The lift/thrust force needs to be constant in order to generate a uniform lift force and to obtain a proper balance. There are many other designs used according to their payloads and application specialities [5]. The octa and hexa configurations are also in use in some applications. In any configuration, the force is obtained by the propeller,therefore it is important that the V. Patel · K. Nikam (B) · S. Dikshit · M. Agarwalla · C. Zagade Department of Mechanical Engineering, Dr. D.Y. Patil Institute of Engineering Management & Research, Pune, India e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 A. Sachdeva et al. (eds.), Recent Advances in Operations Management Applications, Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-16-7059-6_31
411
412
V. Patel et al.
propeller is designed for generating a uniform and constant force to ensure the efficiency and stability of the drone. The attitude control of the drone is another aspect for obtaining stability, but it requires costly equipment, while on the other hand, it can also be achieved by designing it accurately. The structure of the propeller from the design aspect must be in proper balance. Any imbalance in these may lead to a hindrance in the flight which may impact the two major parameters: the flight time and the stability of the drone. In the years we have seen drones aiding in assisting and assessing various inaccessible areas and providing information for the situation at hand. Drones when equipped with certain devices are much more versatile and can be applied to a much broader area [4]. It has been seen earlier that drones have assisted in the construction field, mapping the vegetation in an area, land surveying for dam construction and for locating minerals. In the military field also, we have seen various drones which have been used for surveillance. And the approach here can be even applied in designing the wings of a fixed-wing-type drone, which is commonly used in military due to its longer flight time. According to the application, we have the configurations like the octa and hexa configurations which are mostly used for obtaining high payload capacities. The quads are much more versatile as they can be designed for medium payload to low payload and has a well balance between the flight time and the authority it provides over the control than many other configurations. The approach here is to design a propeller for drones and a specific lift is to be generated for the propeller to fulfil the criteria [6]. The design proposed here is for a two-rotor drone and by the criteria of lift provided it can be applied for all the configurations [17]. The design approach is similar to that of a wing since it can be observed that a propeller is a continuously rotating wing [1].
2 Methodology The method applied here is based on a lift criterion to be fulfilled and the criteria were to obtain a total lift of 500 g as a normal weight of any two-rotor and four-rotor drones. The steps for the design are listed in Table 1. Table 1 Steps for propeller design Steps
Applied method
Important data used
References
Selection of airfoil
Comparison of airfoils
Reynolds no. of 3 × 105
[14]
Optimization
Lift and drag forces
Lift and drag equations
[2]
Modelling and simulation
Thrust and power
Rotational speed of 18,000 rpm
[7]
Propeller Design and Optimization for Drones
413
3 Propeller Design The propeller design, as listed above, is based on important steps, but the major aim is to obtain the desired lift force. The lift and drag forces for any wing is given by the lift and drag equations [8, 13]. The equations for lift and drag are as follows: L=
1 Cl ρ AV 2 2
(1)
D=
1 Cd ρ AV 2 2
(2)
The lift and drag equation will aid in the parameters we need to obtain for the generation of lift and the design of a propeller. The coefficient of lift (Cl ) and coefficient of drag (Cd ) are the parameters that will be obtained by the airfoil selection [3, 9]. The density of the medium (air in our application) is 1.2754 kg/m3 . The area will be further broken down as we follow the steps and the velocity of the wing is fixed as 18,000 rpm of a normally available BLDC motor.
3.1 Selection of Airfoil The selection of airfoil is important since it will aid in obtaining the coefficient of lift and drag. The NACA Airfoil database is used for this purpose [10]. The airfoil listed for the Reynolds no. of 3 × 105 as our application [14]. The airfoils selected for comparison are NACA 2412, 4412, 4415, 6409, 6412. The comparison of the airfoils was done on QBlade Software [15]. The main criterion of comparison is to obtain the maximum coefficient of lift and minimum coefficient of drag. The comparison graph is shown in Fig. 1. The NACA 6412 airfoil yields a maximum coefficient of lift of 1.375 and a coefficient of drag of 0.015. The other characteristics curves of NACA 6412 are given in Fig. 2. The boundary layer diagram is another important factor for determining the stability of airfoil when the propeller profile when obtained would be less (Fig. 3). The pressure distribution over the airfoil profile is necessary to learn the generation of the forces (Fig. 4).
3.2 Optimization for Lift and Drag Optimization is a need for developing a proper and optimum blade structure. One of the parameters of the lift and drag equations is now broken down according to the ease of the design. The area is broken down to the radius multiplied by the chord
414
V. Patel et al.
Fig. 1 Comparison of airfoils
Fig. 2 Coefficient of lift vs angle of attack for NACA 6412
length [16]. The radius defines the length of the blade/wing. The chord defines as the width of the blade/wing. The lift equation is now transformed as follows: L=
1 Cl ρ(r × c)V 2 2
(3)
Propeller Design and Optimization for Drones
415
Fig. 3 Boundary layer diagram for NACA 6412
Fig. 4 Pressure distribution over NACA 6412
D=
1 Cd ρ(r × c)V 2 2
(4)
where ‘r’ determines the radius or the length and ‘c’ determines the chord length. As we know we are designing a propeller which is a rotating wing. Thus, the velocity is not constant through the length of the blade/wing. The speed variation is given in Fig. 5.
Speed (m/s)
416
V. Patel et al. 50 40 30 20 10 0
6.91271 0
0.01
13.82542
0.02
20.73813
0.03
27.65084
0.04
34.56355
0.05
41.47626
0.06
0.07
Radius (m)
Fig. 5 Speed variation with radius
3.3 Lift and Drag Variation Plots The only parameters we can vary now are the radius and chord length. Thus the variation is needed to design the propeller to obtain the dimension or the variation we need to use for the design of the propeller. To examine the variation, the plots of lift to radius and lift to chord length were plotted and the same for the drag as well. The plots are given in Figs. 6 and 7. The plots show that the lift varies exponentially with radius and linearly with chord length. The drag plots are shown in Figs. 8 and 9. As we can see, the drag varies almost similar to the lift just in significantly lower values of forces. Thus, this gives us the idea that the impact of drag forces isn’t much of a hindrance and the lift can be obtained as desired according to the analytically obtained variations. The variations in lift and drag aid us in designing the propeller by providing the information that as we increase the chord length at the hub, i.e., at low radius and keep reducing it to an extent the drag is minimum and the length is of standard propellers. 1.6 y = 0.0031e102.82x
1.4
Lift (N)
1.2 1 0.8 0.6 0.4 0.2 0 0
0.01
0.02
0.03
0.04
Radius (m) Fig. 6 Lift variation with radius
0.05
0.06
0.07
Propeller Design and Optimization for Drones
417
2 y = 90.504x - 8E-15
1.8 1.6
Lift (N)
1.4 1.2 1 0.8 0.6 0.4 0.2 0 0
0.005
0.01
0.015
0.02
0.025
Chord Length (m) Fig. 7 Lift variation with chord length
Drag 0.018 y = 3E-05e102.82x R² = 0.9363
0.016
Drag (N)
0.014 0.012
0.009873204
Drag
0.01
Expon. (Drag)
0.008
0.00571366
0.006 0.004 0.002 0
0.002925394 0.00123415 4.57093E-05 0.000365674 0
0.02
0.04
0.06
0.08
Radius (m)
Fig. 8 Drag variation with radius
4 Propeller Modelling The propeller is now to be designed according to the data we have obtained analytically and then simulated to verify whether the criterion is fulfilled (Stopforth 2017). The propeller is designed in QBlade and the design data is listed in Table 2. The propeller optimization is carried by applying a twist to obtain a uniform generation of thrust throughout the length of the blade/wing. This twist is apart from the pitch as the pitch is the same for all the elements along the length of the blade, as the twist may vary through the length of the blade (Fig. 10).
418
V. Patel et al.
Drag 0.025 y = 0.9873x - 1E-16 R² = 1
Drag (N)
0.02
Drag
0.015
Linear (Drag) 0.01 0.005 0 0
0.005
0.01
0.015
0.02
0.025
Chord Length (m)
Fig. 9 Drag variation with chord length
Table 2 Propeller dimensions
Position (mm)
Chord length (mm)
0
5
3.75
15
7.5
20
11.25
19
15
18.5
18.75
18
22.5
17.5
26.25
17
30
16.5
33.75
16
37.5
15.5
41.25
15
45
14.5
48.75
14
52.5
13.5
56.25
13
60
10
5 Propeller Analysis The propeller designed earlier is now to be analysed using a two-blade configuration with a wind speed of 0.1 m/s. The speed of the rotor is 18,800 rpm. The speed of the rotor is decided by selecting the configuration of the drone. The drone configuration selected is a two-rotor drone. The electric motor selected yielded a maximum
Propeller Design and Optimization for Drones
419
Fig. 10 Propeller design
rotational speed of 18,800 rpm. The total thrust obtained would be sufficient enough to carry its own payload so that the drone is sufficient enough to carry our various surveillance operations. The propeller disk would contain two blades and there are two propeller disks as we are testing for a two-rotor drone. The analysis is done for a single propeller disk as the results can be coupled together for the thrust calculation [12]. The propeller disk is shown in Fig. 11. The propeller disk is further simulated in Q-blade with the above-discussed parameter as an indoor drone with a relatively small wind speed [11]. The simulation results for the propeller disk are shown in Fig. 12. The result shown above shows the pressure distribution over the profile of the propeller disk. The result determines that the pressure is very low at the centre of the hub and then goes on increasing till the tip of the blade, on both sides. This confirms our design as the velocity of the blades increases with the length/radius of the propeller.
Fig. 11 Propeller disk
420
V. Patel et al.
Fig. 12 Pressure distribution over the propeller disk
After the pressure distribution the other important criterion for analysis is the direction of the airflow as it would be important for proper orientation of the propeller for the drone mounting. The propeller analysis for the direction of airflow is shown in Fig. 13.
Fig. 13 Airflow for rotating propeller disk
Propeller Design and Optimization for Drones
421
The simulation results determine that the airflow is in the downward direction. This further leads to the thrust generation, which consequently fulfils our initial criteria for the generation of considerable thrust for the drone to carry a payload. Until now the flow analysis determined the pressure and the direction of the flow which leaves us with the mechanical integrity of the blade. The blade must be able to handle the drag forces without failing. The beam bending analysis is done using the drag forces as the input. The material used for the analysis is PLA (poly lactic acid) with Young’s modulus of 3.01e+06 Pa. The static force analysis is shown in Fig. 14. The drag forces try to bend the propeller blades. As visible in the result, the impact is particularly seen near the hub of the rotor, i.e., where the propeller blades are joined with the hub. The rest of the blade is fine as the bending profile is negligible with respect to the length/radius of the blade. The maximum bending stress is 0.18 MPa which the material can sustain. This determines that the blades are strong enough to sustain the drag forces. The propeller analysis verifies many key points prior to the thrust force calculation. The major key points verified here are pressure distribution which results in the
Fig. 14 Bending stresses over the propeller disk
422
V. Patel et al.
determination of the forces generation, airflow direction determines the direction in which the force acts by using Newton’s Third Law of Motion and the static analysis determines the mechanical strength of the blades with respect to the drag forces.
6 Results and Discussion The propeller modelled above is to be analysed on fulfilling of the earlier criterion and the simulations were run on QBlade. The results include the plots, mainly focusing on the parameters like thrust, rotational speed, pitch and power (Figs. 15 and 16).
Fig. 15 Thrust variation with rotational speed
Fig. 16 Power variation with thrust
Propeller Design and Optimization for Drones
423
The results state that the lift generated is maximum when the rotational speed is maximum and the pitch angle is around 26º. The maximum lift which can be generated through the propeller blade designed is around 9.6° N. Thus the propeller clearly satisfies the earlier criterion of lift. The power it would consume is around 333.5° W. As the software used for the simulation is basically for the windmill blade design, the simulation results obtained are inverted (graph is obtained in the downward direction).
7 Conclusion The results obtained yield the lift force of the 9.6° N as mentioned earlier. The combined lift would be 19.2° N. For the propeller, we consider it a two-blade propeller and will keep increasing as we increase the number of blades. The lift would even increase as we increase the number of propellers on the drone. The propeller can be used for different drone configurations by learning the weight to be carried and the speed of mobility needed for the particular applications. Thus, by learning these parameters the desired lift can be calculated just by changing the number of blades and the number of propellers according to the configuration being used. This verifies that the method used is fruitful for the design of the propeller and satisfies the desired criterion. There are various parameters listed in the steps which can be changed according to the ease of the application and design. For designing the propeller for different configurations of drones, there is more analysis to be carried out for determining how the propeller behave typically for a particular configuration. This is needed before carrying out the physical prototype because the distance at which the propeller disks are, over a particular configuration, results in the difference of the airflow direction which changes the propeller disk behaviour. If two or more propeller disks interact with each other, the change in direction of air may result in a change in the thrust force generated. The coaxial configuration is one good example of the case discussed. Even on the conventionally used quad-copters the distance between the propeller disks is important as it impacts the thrust generation as well as the stability of the drone.
References 1. Aref P, Ghoreyshi M, Jirasek A et al (2018) Computational study of propeller-wing aerodynamic interaction. Aerospace 5:1–20. https://doi.org/10.3390/aerospace5030079 2. Durai A (2014) Experimental investigation of lift and drag characteristics of a typical MAV under propeller induced flow. Int J Micro Air Veh 6:63–72. https://doi.org/10.1260/1756-8293. 6.1.63 3. Jathar LD, Borse KCNSL (2015) Study of flow over sedan car body by changing different parameters using open foam 2:342–348
424
V. Patel et al.
4. Kawasaki K, Motegi Y, Zhao M et al (2015) Dual connected Bi-Copter with new wall trace locomotion feasibility that can fly at arbitrary tilt angle. IEEE Int Conf Intell Robot Syst December:524–531. https://doi.org/10.1109/IROS.2015.7353422 5. Mairaj A, Baba AI, Javaid AY (2019) Application specific drone simulators: recent advances and challenges. Simul Model Pract Theory 94:100–117. https://doi.org/10.1016/j.simpat.2019. 01.004 6. Mallick TC, Bhuyan MAI, Munna MS (2017) Design & implementation of an UAV (Drone) with Flight Data Record. In: 2016 international conference on innovations in engineering and technology ICISET 2016. https://doi.org/10.1109/ICISET.2016.7856519 7. Marten D, Wendler J (2013) QBLADE: an open source tool for design and simulation of horizontal and vertical axis wind turbines. Int J Emerg Technol Adv Eng 3:264–269 8. Nikam K, Kumar R (2019) Simulation of air flow through bus for different position of two top openings by using open foam package. Think India J 22:4070–4080 9. Nikam KC, Borse SL (2014) Study of air flow through a open windows bus. Int J Fluids Eng 6:53–64 10. Prasad KS, Krishna V, Kumar BBA (2013) Aerofoil profile analysis and design optimisation. J Aerosp Eng Technol 3:1–10 11. Serrano D, Ren M, Qureshi AJ, Ghaemi S (2019) Effect of disk angle-of-attack on aerodynamic performance of small propellers. Aerosp Sci Technol 92:901–914. https://doi.org/10.1016/j.ast. 2019.07.022 12. Vecchia PD, Malgieri D, Nicolosi F, De Marco A (2018) Numerical analysis of propeller effects on wing aerodynamic: tip mounted and distributed propulsion. Transp Res Proc 29:106–115. https://doi.org/10.1016/j.trpro.2018.02.010 13. Wald QR (2006) The aerodynamics of propellers. Prog Aerosp Sci 42:85–128. https://doi.org/ 10.1016/j.paerosci.2006.04.001 14. Wang K, Zhou Z, Zhu X, Xu X (2019) Aerodynamic design of multi-propeller/wing integration at low Reynolds numbers. Aerosp Sci Technol 84:1–17. https://doi.org/10.1016/j.ast.2018. 07.023 15. Xie X, Haberland C (1999) New numerical design tool for concept evaluation of propeller aircraft. Aircr Des 2:147–165. https://doi.org/10.1016/S1369-8869(99)00012-9 16. Yu P, Peng J, Bai J et al (2020) Aeroacoustic and aerodynamic optimization of propeller blades. Chin J Aeronaut 33:826–839. https://doi.org/10.1016/j.cja.2019.11.005 17. Zhang Q, Liu Z, Zhao J, Zhang S (2016) Modeling and attitude control of Bi-copter. In: AUS 2016—2016 IEEE/CSAA International conference on aircraft utility systems, vol 100191, pp 172–176. https://doi.org/10.1109/AUS.2016.7748042