Advances in Engineering Networks: Proceedings of the 12th International Conference on Industrial Engineering and Industrial Management [1st ed.] 9783030445294, 9783030445300

Citation preview

Lecture Notes in Management and Industrial Engineering

Rodolfo de Castro Gerusa Giménez   Editors

Advances in Engineering Networks Proceedings of the 12th International Conference on Industrial Engineering and Industrial Management

Lecture Notes in Management and Industrial Engineering Series Editor Adolfo López-Paredes, INSISOC, University of Valladolid, Valladolid, Spain

This bookseries provides a means for the dissemination of current theoretical and applied research in the areas of Industrial Engineering & Engineering Management. The latest methodological and computational advances that both researchers and practitioners can widely apply to solve new and classical problems in industries and organizations constitute a growing source of publications written for and by our readership. The aim of this bookseries is to facilitate the dissemination of current research in the following topics: • • • • • • • • • • • • • •

Strategy and Enterpreneurship Operations Research, Modelling and Simulation Logistics, Production and Information Systems Quality Management Product Management Sustainability and Ecoefficiency Industrial Marketing and Consumer Behavior Knowledge and Project Management Risk Management Service Systems Healthcare Management Human Factors and Ergonomics Emergencies and Disaster Management Education

More information about this series at http://www.springer.com/series/11786

Rodolfo de Castro Gerusa Giménez •

Editors

Advances in Engineering Networks Proceedings of the 12th International Conference on Industrial Engineering and Industrial Management

123

Editors Rodolfo de Castro Department of Business Administration Management and Product Design Universitat de Girona Girona, Spain

Gerusa Giménez Department of Business Administration Management and Product Design Universitat de Girona Girona, Spain

ISSN 2198-0772 ISSN 2198-0780 (electronic) Lecture Notes in Management and Industrial Engineering ISBN 978-3-030-44529-4 ISBN 978-3-030-44530-0 (eBook) https://doi.org/10.1007/978-3-030-44530-0 © Springer Nature Switzerland AG 2020 This work is subject to copyright. All rights are reserved 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 Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Organization

Program Chairs Rodolfo de Castro, Escola Politècnica Superior, Universitat de Girona, Spain Gerusa Giménez, Escola Politècnica Superior, Universitat de Girona, Spain

Program Committee Faustino Alarcón Valero, Centro de Investigación en Gestión e Ingeniería de Producción (CIGIP), Spain María del Mar Alemany Díaz, UPV, Spain Juanjo Alfaro, CIGIP Esther Alvarez, Facultad de Ingeniería, Spain Izaskun Alvarez, EHU/UPV, Spain Beatriz Andres, Universitat Politècnica de València, Spain Pablo Aparicio Ruiz, Universidad de Sevilla, Spain Anna Arbussà, Facultat de Ciències Econòmiques i Empresarials, Universitat de Girona, Spain Eugenia Babiloni, Universidad Politécnica de Valencia, Spain Frano Barbic, Universidad Politecnica de Madrid, Spain Bopaya Bidanda, University of Pittsburgh, United States Andrea Bikfalvi, Escola Politècnica Superior, Universitat de Girona, Spain Iñaki Bildosola, UPV/EHU, Spain Andrés Boza, Universitat Politècnica de València Francisco Campuzano-Bolarín, Universidad Politécnica de Cartagena, Spain Lourdes Canós Darós, Universitat Politècnica de València, Spain

v

vi

Organization

Manuel Cardós Carboneras, Universidad Politecnica de Valencia, Spain María Carmen Carnero, University of Castilla-La Mancha, Spain Javier Carrasco, UPM Ruth Carrasco-Gallego, Escuela Técnica Superior de Ingenieros Industriales. Universidad Politécnica de Madrid Ricardo Chalmeta, Universidad jaume I Ernesto Cilleruelo, UPV/EHU Vincent Clivillé, Université Savoie Mont Blanc, France Yuval Cohen, Tel-Aviv Afeka College of Engineering, Israel José Antonio Comesaña Benavides, University of Vigo Albert Corominas, UPC, Spain Pascual Cortés, UPV VALENCIA Pablo Cortés, Universidad de Sevilla, Spain Llanos Cuenca, CIGIP-UPV, Spain Rodolfo de Castro, Escola Politècnica Superior, Universitat de Girona, Spain María Victoria De la Fuente, Universidad Politécnica de Cartagena David De la Fuente, University of Oviedo Ricardo Del Olmo, Universidad de Burgos Carlos Dema, Universisdad Politécnica, Spain Alfonso Duran, UC3M Manuel Díaz-Madroñero, Universitat Politècnica de València Alejandro Escudero-Santana, Universidad de Sevilla Sofia Estellés, Universitat Politecnica de Valencia Arturo J. Fernández González, Escuela de Ingeniería Industrial (sede Campus), University of Vigo, Spain Vernadat Francois, University of Lorraine, France José Manuel Galán, Universidad de Burgos, Spain Jesus Garcia Arca, Universidad de Vigo Jose P. Garcia-Sabater, Universidad Politécnica de Valencia Julio J. Garcia-Sabater, Universidad Politécnica de Valencia, Spain Isabel García Gutiérrez, Universidad Carlos III de Madrid, Spain German Gemar, Universidad de Malaga Gerusa Giménez, Escola Politècnica Superior, Universitat de Girona, Spain Pedro Gomez-Gasquet, Universitat Politècnica de Valeència, Spain Lorenzo Gonzalez, Instituto Andaluz de Tecnología (IAT), Spain Pedro L. Gonzalez-R, School of Engineers, Spain Bernard Grabot, ENIT-INP, France Gonzalo Grau Gadea, Universitat Politècnica de València, Spain Jose Guadix, University of Seville Frederic Hauser, Laboratoire Pierre Fabre Dermo Cosmetique, France David Hermida, Universidad de Oviedo, Spain Antonio Hidalgo, Universidad Politécnica de Madrid, Spain

Organization

vii

Eloy Hontoria, Technical University of Cartagena Francisco Javier Iglesias Rodríguez, Universidad de Oviedo, Spain Josep Llach, Universitat de Girona Amaia Lusa, Universitat Politècnica de Catalunya Julien Maheut, Universidad Politécnica de Valencia, Spain MIguel Angel Manzanedo del Campo, Universidad de Burgos Juan A. Marin-Garcia, Universitat Politecnica de Valencia, Spain Paloma Maria Teresa Martinez Sanchez, Bosque University, Colombia Irene Martín-Rubio, UPM, Spain Carme Martínez, Costa Universitat Politècnica de Catalunya, Spain Manuel Mateo, UPC, Spain Cristóbal Miralles Insa, Universitat Politecnica de Valencia, Spain Guillermo Montero, Universidad de Sevilla, Spain Jesús Morcillo, Universidad Carlos III de Madrid. Escuela Politécnica Superior, Spain Josefa Mula, Universitat Politècnica de València, Spain Giovanni Mummolo, Polytechnic of Bari, Italy Jesús Muñuzuri, University of Seville, Spain Jordi Olivella Nadal, Universitat Politécnica de Catalunya Angel Ortiz, CIGIP-UPV, Spain Enrique Parra, University of Alcala, Spain Rafael Pastor, UPC David Peidro, CIGIP Raúl Poler, Universitat Politècnica de València, Spain Borja Ponte, Univesity of Oviedo, Spain David Poza, INSISOC - University of Valladolid, Spain José Carlos Prado Prado, Escuela Ingeniería Industrial Emilio Ramírez-Juidías, Universidad de Sevilla, Spain Imma Ribas, Universitat Politècnica de Catalunya, Spain Ivan Roa, Universidad Politécnica de Cataluña, Spain Jose Luis Roca Gonzalez, Centro Universitario de La Defensa en La Academia General del Aire, Spain Diego Ros McDonnell, UPCT, Spain Lorenzo Ros McDonnell, Universidad Politécnica Cartagena, Spain Juan Carlos Rubio-Romero, Universidad de Málaga, Spain Antonio Ruiz Molina, Universidad de Málaga, Spain Patxi Ruiz-de-Arbulo-López, Universidad del País Vasco (UPV/EHU), Spain Rashed Sahraeian, Shahed University, Iran Raquel Sanchis, UPV José I. Santos, Universidad de Burgos, Spain Javier Santos, Tecnun - University of Navarra, Spain Pedro Sanz Angulo, University of Valladolid, Spain

viii

Lourdes Sáiz Bárcena, Universidad de Burgos, Spain Martin Tanco, Universidad de Montevideo, Uruguay Juan Ramon Trapero, University of Castilla-La Mancha Lorna Uden, Staffordshire University, United Kingdom María José Verdecho, Universitat Politècnica de València Pilar I. Vidal-Carreras, Universidad Politécnica de Valencia, Spain Elisabeth Viles, Tecnun - University of Navarra, Spain Enara Zarrabeitia, Universidad del País Vasco, Spain

Organization

Preface

We are pleased to preface this book where you could find the selected papers of CIO 2018, XXII Congreso de Ingeniería de Organización/12th International Conference on Industrial Engineering and Industrial Management. This conference was promoted by ADINGOR (Asociación para el Desarrollo de la Ingeniería de Organización), and it was hosted at Universitat de Girona (Spain) from 12 to 13 July 2018. The CIO 2018 Conference motto was: “Advancing in Engineering Network”. The selected papers cover the most relevant research and current projects in Industrial Engineering, Management and Operations. In this book, the reader can find papers providing links between researchers and practitioners from different branches, to enhance an interdisciplinary perspective of industrial engineering and management. The contributions have been arranged in five parts: • • • • •

Strategy and Information Systems Operations Research Supply Chain Education Quality and Sustainability

We hope this book gives you the opportunity to enjoy high-quality scientific papers. Girona, Spain October 2018

Rodolfo de Castro Gerusa Giménez

ix

Contents

Strategy and Information Systems Which 4.0 Professional Competencies Should Develop Middle Managers and Operators? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Álvaro Lleo-de-Nalda, José Ignacio Terrés Goena, Elisabeth Viles Díez, and Javier Santos

3

Integrated Project Delivery: A Literature Review and Research Agenda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Iván González-Boubeta and José Carlos Prado-Prado

11

Empirical Measurement Instruments for Business Model Innovation: A Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dorleta Ibarra, Jaione Ganzarain, and Juan Ignacio Igartua

19

Scrip Dividends and Share Buyback Strategies Based on Volatility . . . . Angel Huerga and Carlos Rodríguez-Monroy

29

The Evolution of Business Intelligence with Neuroinformatics . . . . . . . . Irene Martín-Rubio, Juan Fombellida, and Diego Andina

37

A General Overview of the Industry 4.0 Concept for Production Management and Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Héctor Cañas and Josefa Mula

45

Identification and Prioritization of Industry 4.0 Projects in SMEs: A Process Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Juan Ignacio Igartua, Jaione Ganzarain, and Dorleta Ibarra

53

Blockchain for Electronic Voting Purposes . . . . . . . . . . . . . . . . . . . . . . Ricardo Chica Cepeda and Anna Arbussà Reixach

61

xi

xii

Contents

Identification of Barriers of Entry to the European Market of Medical Devices: Study of Cases in Spanish Companies . . . . . . . . . . . . . . . . . . . Yariza Chaveco Salabarria, Mª del Carmen Pardo Ferreira, Juan Carlos Rubio Romero, and Rosa Mayelín Guerra-Bretaña Application of Combinatorial Auctions to Create a 3D Printing Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Adolfo López-Paredes, Sandra Castillo, Javier Pajares, Natalia Martín, and Ricardo del Olmo SEAFRESH Project: Design and Development of an Intelligent System for Decision Support in the Chilled and Frozen Fish Sector . . . . . . . . . Antonio García Lorenzo and Joaquín Romero Rivero

71

79

85

Operations Research Improving Vegetables’ Quality in Small-Scale Farms Through Stakeholders’ Collaboration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ana Esteso, María del Mar Alemany, and Angel Ortiz

95

Assignment of Volunteers in a Sports Event: Case Restricted Fitness by Cut-off Mark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Joaquín Bautista, Manuel Mateo, and Rocío de la Torre An MILP Model for Evaluating the Impact of Strategic Decisions on Promotions in Universities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Rocío de la Torre and Manuel Mateo An Approach to Explore Historical Construction Accident Data Using Data Mining Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 María Martínez Rojas, Antonio Trillo Cabello, Mª del Carmen Pardo Ferreira, and Juan Carlos Rubio Romero A Non-parametric Enhancement of the Fill Rate Estimation . . . . . . . . . 129 Eugenia Babiloni, Ester Guijarro, and Juan Ramon Trapero FAHP Applications for Manufacturing Environments: A Contemporary Review and Classification . . . . . . . . . . . . . . . . . . . . . . 137 Victor Anaya Fons, Raúl Rodríguez Rodríguez, and Angel Ortiz A MILP Approach to Maximize Productivity in Mixed-Model Assembly Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Joaquín Bautista and Rocío Alfaro-Pozo Production Typologies in Production Scheduling: Identification and Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 Pilar I. Vidal-Carreras, Julio J. Garcia-Sabater, Angel Ruiz, and Julien Maheut

Contents

xiii

Supply Chain Conceptual Framework for the Characterization of Vegetable Breton Supply Chain Sustainability in an Uncertain Context . . . . . . . . . . . . . . 165 Vicente S. Fuertes-Miquel, Llanos Cuenca, Andrés Boza, Cécile Guyon, and María del Mar Alemany Dimensioning the Supply Chain Decision Support Systems . . . . . . . . . . 175 Julio César Puche Regaliza, Borja Ponte, José Costas Gual, Raúl Pino Diez, and David de la Fuente García What Are the Main Factors that Reduce the Efficiency of Road Transport? An Exploratory Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 Mar Fernández Vázquez-Noguerol, Andrea González-Prado, Iván González-Boubeta, and José Carlos Prado-Prado Reverse Logistics Causes and Treatment Alternatives . . . . . . . . . . . . . . 191 Pascual Cortés Pellicer and Faustino Alarcón Valero Pharmaceutical Supply Chain Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 199 Francesc Amaro-Martinez and Rodolfo de Castro Joint Price and Reorder Point Determination for Profit Maximization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Manuel Cardós Carboneras, María Victoria de la Fuente Aragón, and Lorenzo Ros-McDonnell Setting the Order-Up-to Level in a Retailer: Challenges and Empirical Comparison of Simple Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Vicent Asensio-Molina, Angel Ruiz, Jose P. Garcia-Sabater, and Julio J. Garcia-Sabater Changing Internal Logistics to Feed Production Lines . . . . . . . . . . . . . . 225 Aida Saez-Mas, Angel Ruiz, Jose P. Garcia-Sabater, and Julio J. Garcia-Sabater Proposal of a Customer-Oriented Sustainable Balanced Scorecard for Agri-Food Supply Chains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 María José Verdecho, David Pérez Perales, and Faustino Alarcón Valero Education Training in Quality Engineering Concepts and Skills: Case Study, Simulations Paper Propeller Using Six Sigma-Based Methodology . . . . . 243 José Alberto Eguren, Toni Antero Bertlin, Joel Hannes Rehunen, and Gorka Unzueta Women in STEM Education: A Longitudinal Study . . . . . . . . . . . . . . . 251 Ruth Carrasco-Gallego, Ana Moreno-Romero, and Silvia Serrano-Calle

xiv

Contents

Development of an Online Social Network for Supporting the Design, Coordination, and Following-up of Final Projects in Engineering . . . . . 261 Raúl Rodríguez Rodríguez, María José Verdecho, Juan José Alfaro-Saiz, and Pedro Gómez-Gasquet Quality and Sustainability Improving the Management of a Cultural Association by Means of Lean Office . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271 Alejandro Escudero-Santana, Pablo Aparicio-Ruiz, Elena Barbadilla-Martín, and María Rodríguez-Palero Study on Barriers and Success Factors for a Sustainable and Successful Lean Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 Néstor Gavilán and Carolina Consolación Beyond Customer Satisfaction: Are All Customers Equally Satisfied? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 Dalilis Escobar Rivera, Martí Casadesús Fa, Paulo Alexandre Costa Araújo Sampaio, and Alexandra Simon Villar Fuzzy Logic for the Improvement of Thermal Comfort and Energy Efficiency in Non-residential Buildings . . . . . . . . . . . . . . . . . . . . . . . . . . 303 Elena Barbadilla-Martín, José Guadix, Pablo Cortés, and María Rodríguez-Palero Electric and Hybrid Motorcycle Drivers at Work, How Do They Perceive the Effects of the Lack of Noise of These Vehicles? . . . . . . . . . 311 Mª del Carmen Pardo Ferreira, Juan Carlos Rubio Romero, and Manuel García Jiménez In Search After Micro-Economic Effects of Ecoinnovation Activities Among Industrial Firms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 Jabier Retegi and Bart Kamp Audiovisual Documentation as the Basis of an Occupational Health and Safety Management System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 Arturo J. Fernández-González, Iván González-Boubeta, Andrea González-Prado, and José Carlos Prado-Prado Impact of Air Quality on Urban Mobility: Analysis of a Mediterranean City . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337 Lorenzo Ros-McDonnell, María Victoria de la Fuente Aragón, Diego Ros-McDonnell, and Manuel Cardós Carboneras Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345

Strategy and Information Systems

Which 4.0 Professional Competencies Should Develop Middle Managers and Operators? Álvaro Lleo-de-Nalda, José Ignacio Terrés Goena, Elisabeth Viles Díez, and Javier Santos

Abstract There is no doubt about the impact that the 4th Industrial Revolution is having in our factories. New technologies are being developed, and huge quantities of money are being invested for modernizing our plants. Theses technologies are also impacting on operational and people management, and new professional competences are required for being able to work successfully in this environment. In this paper, combining literature review and fieldwork with 22 different companies that operate with 4.0 technologies, we identify and prioritize a list of 23 professional competencies. Keywords 4.0 professional competencies · Industry 4.0 · Talent development · Higher education · Engineering education

1 Introduction and Purpose The concept of Industry 4.0 (I4.0) is increasingly present in our society. Several facts support this idea1 : Industry 4.0 has more than 275 million results in Google in only 0.48 s. Internet of Things (IoT) is referenced over 270 million times in half a second. The United States of America will invest up to 900 billion dollars per annum until 2020 [6]. In the European Union, there are almost 30 initiatives to finance research in these technological developments. In Spain, there are two of them: Industria Conectada 4.0 and Basque Industry 4.0. Regarding the description of this concept, there are several definitions and it seems that there is no consensus about a final definition. Some of the most known definitions are:

1 Google,

10/02/2018.

Á. Lleo-de-Nalda (B) · J. I. Terrés Goena · E. Viles Díez · J. Santos Tecnun-School of Engineering, University of Navarra, Manuel de Lardizábal 15, 20018 San Sebastián, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_1

3

4

A. Lleo-de-Nalda et al.

• “Industry 4.0 is a recent concept that refers to a fourth industrial revolution, which consists in the introduction of digital technologies in industry” [12]. • McKinsey defines Industry 4.0 as “digitization of the manufacturing sector, with embedded sensors in virtually all product components and manufacturing equipment, ubiquitous cyberphysical systems, and analysis of all relevant data” [18]. • The Boston Consulting Group describes this concept as “Now, though, we are in the midst of a fourth wave of technological advancement: the rise of new digital industrial technology known as Industry 4.0, a transformation that is powered by nine foundational technology advances” [14]. • Sachon [15], professor of IESE Business School, proposes that “Industry 4.0 is a natural evolution of previous industrial revolutions, from the mechanization of manual work in the XVIIIth century to automatization, with interconnected intelligent machines which can work autonomously to obtain products in flexible and reconfigurable systems”. There are other ideas that can be added related to this concept, such as Internet of Things, additive manufacturing, or Big Data. In a report by The Boston Consulting Group, nine foundational technology advances are identified: simulation, additive fabrication, horizontal and vertical integral systems, cybersecurity, augmented reality, cloud computing, autonomous robots, Internet of Things, Big Data, and analysis [14]. Besides identifying the nine foundational technologies, some studies highlight the key aspects on which Industry 4.0 is based. In a report written by McKinsey & Company [18], they describe four pillars that group the nine technologies: • Data: This pillar contains everything related to systems’ connectivity and the gathering of huge amounts of data with sensors. For that, new technologies, such as the Internet of Things, Big Data, and cloud technology, are very helpful. • Analysis and Artificial Intelligence: In order to obtain data, it is necessary to have systems that are able to analyze information and make decisions. The following disciplines are inside this pillar: simulation, data analytics, machine learning, etc. • Human–Machine Interaction: Some tasks done by humans will be simplified with virtual reality or augmented reality systems. • Digital to Physical: This pillar refers mainly to 3D printing for prototyping and final pieces, after simulating its behavior by computer. It also includes collaborative robots for repetitive and low-value operations. With this context of technological development, it seems reasonable to assume that the new industrial paradigm is having a huge impact on operational and people management. Moreover, regarding workforce, it also seems reasonable that those 4.0 professionals would develop new skills for being able to adapt to this new environment and take advantage of it.

Which 4.0 Professional Competencies Should Develop Middle …

5

Thus, the aim of this paper is to identify what the 4.0 professional competences are that workers, operators, and middle managers should develop. Moreover, we want to prioritize the importance of each competence with respect to the present as well as in 10 years.

2 Methodology In order to obtain the goals pointed out, we defined a research methodology combining literature review with fieldwork with professionals affected by industry 4.0. In the first phase, we attempted to summarize the divergent knowledge generated around this new industrial transformation and 4.0 professional competences. Afterward, once a list of 4.0 skills was developed, we discussed it with a number of professionals that work with 4.0 technologies.

2.1 Literature Review The aim of the literature review was to understand the concept of Industry 4.0 and its impact on operational management. A wide range of reports, academic articles, studies, and interviews were analyzed to get an idea of what professionals and scientists think about Industry 4.0 and to identify the changes that this new paradigm is going to cause. The majority of the ideas about the 4.0 phenomenon were found in the European Commission, the World Economic Forum, and the reports and studies done by McKinsey, Boston Consulting Group, and IESE Business School. We identified several definitions of I4.0, and the pillars over this paradigm were built. We also identified the technologies that concretize this industrial revolution and some real cases of I4.0 implementation. Subsequently, we focused on the professional transformation that workers need to carry out in order to be able to adapt successfully to this environment [4, 5, 10, 11, 13, 15, 19]. An initial search was carried out using the keywords “Industry 4.0” and “competences”, “Industry 4.0” and “education” and “Industry 4.0” and “human” in the Web of Science. A number of them were filtered because, although they indicate the importance of new skills, they don’t specify the new 4.0 competences the new transformation is demanding. However, we found five publications that propose a number of 4.0 professional abilities [3, 5, 7, 11, 19]. Putting them all together, a list of 58 different competences was developed. Following the proposal from Gehrke et al. [5], we divided them into technical and personal skills. Moreover, we compared and grouped several of them and we defined a list of 23 different skills, 10 of them were technical competences and 13 personal.

6

A. Lleo-de-Nalda et al.

On the one hand, the technical competences and the definition of each one are: • Information technologies: Computer management, Internet, email, etc. • Information security and data protection: Awareness of the importance and confidentiality of the data. • Knowledge of legal aspects: Knowing aspects related to the responsibility of robots, redaction of norms, etc. • Programing: Generation of code and reading in different languages, such as C++, SQL, and others. • Ability to analyze data: Drawing conclusions, interpreting results, etc. • Knowledge in statistics and data visualization: Elaborating statistical studies with the information compiled. • Knowledge in organization and processes: Defining, ordering, analyzing processes and, also, continuous improvement of the above. • Ability to interact with modern interfaces: Management of different programs and the ability to learn when encountering new applications. • Maintenance and reparation of electronic equipment: Knowledge in electronics as well as the ability to fix equipment. • Implementation of 4.0 technologies: The ability to install and implement different 4.0 technologies. On the other hand, the personal competences and the definition of each one are: • Management and assignment of responsibilities: Time management, identification of priorities, team management. • Adaptability: Developing new and better methods and discarding obsolete ones. • Teamwork and cooperation: Understanding that more and better objectives are achieved, even if it generates additional costs. • Social skills: Interrelation capacity. • Networking: Professional networks, contacts, use of LinkedIn. • Communication: Oral expression, transmission of knowledge to new workers. • Trust in new technologies: Security that will work, that the data will not be stolen, etc. • Personal resilience: Recovery after a personal or group failure. • Working under pressure: Ability to manage stress and limitations. • Creativity, entrepreneurship: Contribution of new beneficial ideas. • Conflict management: Anger, reconciliation, justice. • Decision-making: Given a context, choose the best option for everyone. • Leadership capacity: Lead a group, assume granted authority. Having defined the 4.0 professional competences, we wanted to validate and prioritize them with a number of professionals that actually work within 4.0 technologies.

Which 4.0 Professional Competencies Should Develop Middle …

7

2.2 Interviews With the collaboration of Tecnun’s Department of External Relationships, we defined a list of 80 industrial companies from the North of Spain, which were invited to collaborate with this study. Finally, 22 companies were interviewed. The respondents were mostly engineers, but other profiles, such as Human Resource (HR) managers, area directors, or general directors also participated in the study. The mean duration was about 30–45 min. The authors elaborated a report, which is online and accessible to the public.2

3 Results Considering all the information compiled in the interviews, in Figs. 1 and 2, the present-day situation of operators and supervisors can be observed, as well as a forecast for the future. With regard to technical competences, on the one hand, the most important operators’ skills are the ability to interact with modern interfaces and information technologies. Moreover, in the future, respondents state the necessity of improving their knowledge related to IT, information security, and data protection. On the other hand, from the point of view of middle managers, every competence, excepting maintenance and reparation of electronic equipment, seems to be important (with values greater than 3). The ability to use modern interfaces and the knowledge in organization and processes stand out with data analysis being of the most relevance. In the

Fig. 1 Importance, the present day and in 10 years, of 4.0 technical competencies. We used a 5-point Likert scale where 1 is the minimum and 5 is the maximum 2 Terres,

J.I.; Lleó, A.; Viles, E.; Santos, J. (2018), Professional Competences 4.0. ISBN: 978-848081-595-6. https://goo.gl/u7JN1g.

8

A. Lleo-de-Nalda et al.

Fig. 2 Importance, the present day and in 10 years, of 4.0 personal competencies. We used a 5-point Likert scale where 5 is the maximum value

future, they need to develop their abilities about implementation of 4.0 solutions, information security, and data protection. With relation to personal competences, respondents stated that the most important operators’ skills in the future will be: adaptability and flexibility to change, teamwork and cooperation and creativity and entrepreneurship. In the case of supervisors or middle managers, the respondents say that, except for networking, personal competences will be even more important in the future.

4 Discussion and Conclusion Based on a two-step procedure, this paper combines a summary of professional competences, technological and personal, stated in literature that people who want to take advantage of Industry 4.0 should develop taking into account the opinion of 22 different engineers who work in well-known enterprises that have already implemented and operate with 4.0 technologies. Moreover, this paper presents a prioritization, in the present and in 10 years, about these technical and personal competences distinguishing operators from middle managers. The analysis of the information manifests that, regarding the personal competences, the amount of changes introduced requires that operators should develop the flexibility to change as well as the creativity to take advantage of this revolution. From the supervisors’ point of view, they should develop adaptation ability to the 4.0 paradigm, design new job positions, and assign responsibilities. Both supervisors and operators, due to the high level of interconnectivity caused by Industry 4.0, should increase their teamwork skills: operators for being able to work in multidisciplinary teams and supervisors, because they need to manage these groups. Interest in personal competencies or soft skills has been growing because employers are complaining about a lack of such skills in candidates [17] and the impact of

Which 4.0 Professional Competencies Should Develop Middle …

9

these skills has been well demonstrated [16]. Teaching or developing soft skills is a challenge [1]. Literature shows that different approaches have been adopted: classroom teaching [2], practical experience or other channels that include selfassessments, exposure to work situations, real-world examples, and dialog [1]. Mentoring seems to be a good resource for developing these personal competences [8, 9]. With regard to technical competences, operators need to learn more about information technologies, the new interfaces, and the data security and protection. The importance of these skills comes from the sensorization of the factories and the ability to collect information. From the supervisor’s point of view, the necessity of data analysis and process organization and management increases to be able to take advantage of the information compiled and to reorganize the factory according to the new parameters. Literature about the development of technical skills is abundant; however, future research needs to be carried out to adapt these methodologies and to design a program for developing 4.0 technical competences.

References 1. Anthony S, Garner B (2016) Teaching soft skills to business students: an analysis of multiple pedagogical methods. Bus Prof Commun Q 79(3):360–370 2. Bedwell WL, Fiore SM, Salas E (2014) Developing the future workforce: an approach for integrating interpersonal skills into the MBA classroom. Acad Manag Learn Educ 13(2):171– 186 3. Erol S, Jäger A, Hold P, Ott K, Sihn W (2016) Tangible industry 4.0: a scenario-based approach to learning for the future of production. Procedia CIRP 54:13–18 4. European Commission (2017) Implementing the Digitising European Industry actions. https:// ec.europa.eu/futurium/en/content/about-0 5. Gehrke L, Kühn AT, Rule D, Moore P, Bellmann C, Siemes S, Dawood D, Lakshmi S, Kulik J, Standley M (2015) A discussion of qualifications and skills in the factory of the future: a German and American Perspective. 15 6. Geissbauer R, Vedso J, Schrauf S (2016) Industry 4.0: building the digital enterprise, Price Waterhouse Coopers. https://www.pwc.com/gx/en/industries/industries-4.0/landingpage/industry-4.0-building-your-digital-enterprise-April-2016.pdf 7. Hecklau F, Galeitzke M, Flachs S, Kohl H (2016) Holistic approach for human resource management in Industry 4.0. Procedia CIRP 54:1–6 8. Lechuga VM (2011) Faculty-graduate student mentoring relationships: mentors perceived roles and responsibilities. High Educ 62(6):757–771 9. Lleo A, Agholor D, Serrano N, Prieto-Sandoval V (2017) A mentoring programme based on competency development at a Spanish university: an action research study, Eur J Eng Educ 1–19 10. Lorenz M, Küpper D, Rübmann M, Heidermann A, Bause A (2016) Time to accelerate in the race toward Industry 4.0. BCG Perspect 1–5 11. Lorenz M, Rübmann M, Strack R, Lueth K, Bolle M (2015) Man and machine in Industry 4.0: how will technology transform the industrial workforce through 2025. BCG Perspect 12. Ministerio de Economía y Competitividad (2018) Industria conectada 4.0, http://www. industriaconectada40.gob.es/Paginas/index.aspx 13. Rose J, Lukic V, Milon T, Cappuzzo A (2016) Sprinting to value in Industry 4.0. BCG Perspect

10

A. Lleo-de-Nalda et al.

14. Rübmann M, Lorenz M, Gerbert P, Waldner M, Justus J, Engel P, Harnisch M (2015) Industry 4.0: the future of productivity and growth in manufacturing industries. BCG Perspect 9 15. Sachon M (2017) Los cinco puntales de la cadena de valor en la industria 4.0: Cuando personas y máquinas trabajan juntos. IESE Insight 33(Second Quarter 2017), 15–22 16. Sparrow SM (2018) Teaching and assessing soft skills. J Leg Educ 67(2):553–575 17. Taylor E (2016) Investigating the perception of stakeholders on soft skills development of students: evidence from South Africa. Interdiscip J E-Ski Lifelong Learn 12(1):1–18 18. Wee D, Kelly R, Cattel J, Breunig M (2015) Industry 4.0: how to navigate digitization of the manufacturing sector. McKinsey Co 58 19. World Economic Forum (2016) The future of jobs: employment, skills and workforce strategy for the fourth industrial revolution. World Economic Forum, Geneva, Switzerland

Integrated Project Delivery: A Literature Review and Research Agenda Iván González-Boubeta and José Carlos Prado-Prado

Abstract Currently, the construction industry has found a way to improve its competitiveness in Lean Construction, particularly in its Integrated Project Delivery (IPD) tool. The IPD methodology is based on collaboration between all the participants in a project and seeks to identify and eliminate waste in all construction activities. Motivated by IPD’s potential, this paper systematically reviews the literature in order to structure concepts linked to the approach. As a result of this review, the authors develop a framework and reveal a series of gaps that could be used to define a path forward for future research work. Keywords Lean construction · Integrated project delivery (IPD) · Project management · Collaboration · Construction industry

1 Introduction Market globalization, together with the current turbulent environment, has shown that competition is also global. This has led many companies to recognize the need to draw up action plans and rationalize their resources in order to adapt what they do to the new demands of the market. In the construction industry, this has traditionally been criticized as inefficient [26], this approach has become more widespread in recent years in a bid to reduce waste and obtain a product that meets customer needs as closely as possible. One of the most interesting methodologies in response to this issue is Lean Construction. The originator of the concept, Lauri Koskela [14], developed this management philosophy by applying the principles of Lean Production to construction activities. Once created, many authors applied this way of thinking [2, 10] and obtained highly successful results.

I. González-Boubeta (B) · J. C. Prado-Prado Grupo de Ingeniería de Organización (GIO) Escuela de Ingeniería Industrial, Campus Lagoas-Marcosende C/Maxwell, 36310 Vigo, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_2

11

12

I. González-Boubeta and J. C. Prado-Prado

Within the Lean Construction framework, there are several concepts, tools, and methodologies. However, from a management point of view, the most important among all the currently existing ideas is the Integrated Project Delivery (IPD), which the American Institute of Architects (AIA) defines as “a project delivery method that integrates people, systems, business structures and practices into a process that collaboratively harnesses the talents and insights of all participants to reduce waste and optimize efficiency through all phases of design, fabrication and construction” [1], p. 4. In other words, IPD is a management methodology that allows the points of view of all the players participating in a project to be integrated in a structured way. This results in a multidisciplinary and global perspective that is vital in the world of construction, where a large number of organizations are usually immersed in a project’s subcontracting chain [21]. The literature does not have to be studied for long before references appear to another similar concept that is frequently confused with IPD: Lean Project Delivery System (LPDS). The approaches are similar, and their working methodologies are practically identical as they both operate through cooperation and partnering [4]. However, there is one fundamental difference in that IPD also includes the commercial and contractual elements that bind all the players together in the project. This means that greater importance is placed on defining the framework for cooperation so that all the participants feel comfortable and can work in harmony. The similarity between the two approaches has led to IPD being considered an evolution of LPDS as it takes many of the latter’s ideas on board as its own [23]. Furthermore, IPD stresses inclusion from the outset of all the parts beyond the basic triad (developer, designer, constructor), which further emphasizes its collaborative character. Some researchers have concentrated on understanding the advantages of Lean Construction, while others have focused on the functioning of specific elements or tools linked to the philosophy [12, 28]. However, there appears to be no academic literature centered on its methodological part (i.e., IPD) and the relation existing between that methodology and other tools that fall within the scope of Lean Construction (e.g., BIM, Last Planner® ). Therefore, despite frequent references to it in literature, IPD has seldom been dealt with as a specific topic [23]. Moreover, given the proliferation of studies on Lean Construction in recent years, there appears to be a need to structure and clarify the contents of the existing research. This article deals with these aspects by reviewing the literature on the concept of IPD in order to define a new research agenda. This is done by firstly undertaking a thematic and categorized analysis of the existing literature in order to identify what is known about IPD. That is followed by a discussion of the results based on the gaps detected and a definition of future lines for research. Finally, the most relevant aspects of the study are described in a series of conclusions.

Integrated Project Delivery: A Literature Review and Research …

13

2 Methodology The methodology employed is based on a rigorous approach taken from the guidelines of Wee and Banister [27] for literature reviews. The authors opted to carry out a systematic review using the SCOPUS database as their bibliographical source. In order that the search results were as closely linked as possible to the term IPD, an algorithm was designed that considered the terminology used for this topic. The algorithm took into account results that referred directly to IPD and also the conceptual predecessor, LPDS, from which it evolved. Likewise, the term “cooperation” was added, as it has also been used to refer to the same idea. Finally, following the recommendation of Saunders et al. [22] the search was limited to peer-review articles and conference proceedings as these bibliographical sources are more trustworthy for literature reviews. Once the results from the algorithm were obtained, the three-stage filter methodology proposed by Reim et al. [20] was used. Figure 1 shows the details of the process and the criteria followed at each stage. Finally, 46 papers related closely to IPD were obtained and thoroughly analyzed. This in-depth analysis allowed the authors to sort the studies into three groups, depending on the quality of their findings and their relevance according to the proposed objectives. Thus, 15 papers considered to be of high quality were obtained, the contents of which included, to a great extent, the subject matter of the other 31. These 15 articles (Table 1) are, therefore, the ones that are more closely related to the study objectives and they have been used as the basis for undertaking the bibliographical review.

Fig. 1 Material collection algorithm and filtering criteria

14

I. González-Boubeta and J. C. Prado-Prado

Table 1 Papers resulting from the filtering process Authors (year)

Journal or conference proceeding

Article type

Matthews and Howell [16]

Lean construction journal

Conceptual

Ballard [3]

Lean construction journal

Conceptual

Eriksson [7]

Supply chain management

Action research

Raisbeck et al. [19]

Proc. 26th annual ARCOM conference

Conceptual

Cho and Ballard [5]

Lean construction journal

Survey

Darrington [6]

Lean construction journal

Conceptual

Ghassemi and Becerik-Gerber [9]

Lean construction journal

Experts’ interviews

Kim and Dossick [13]

Lean construction journal

Case study

Smith and Rybkowski [24]

Proc. 20th annual conference of the IGLC

Literature review

Pishdad-Bozorgi et al. [18]

Proc. construction research congress 2016

Experts’ interviews

Fakhimi et al. [8]

Science and technology for the built environment

Experts’ interviews

Liu and Shi [15]

Eurasia journal of mathematics science and technology education

Case study

Mei et al. [17]

Engineering, construction, and architectural management

Modeling

Tillman et al. [25]

Proc. 25th annual conference of the IGLC

Case study

Javanmardi et al. [11]

Journal of management engineering

Modeling

3 Results After the in-depth analysis, the papers were grouped into the following categories.

3.1 Contractual Framework The contractual framework in which IPD is developed is one of the most commented issues in the literature. As a starting point, many authors [3, 15, 16, 18] have stated the need for a contract that formally lays down the basis for a project’s workings. The first work related to this subject corresponds to Matthews and Howell [16]. Their study identifies a series of problems that are typical to construction projects and proposes a contractual system to help solve them. The mechanics of this system is based on establishing the main contract between the customer and the basic developer– designer–builder triad and then a series of secondary contracts that group together, by discipline, the builder, the designers, and the various subcontractors (Primary Team

Integrated Project Delivery: A Literature Review and Research …

15

Member—PTM). In this way, it is easier to maintain the collaborative approach on which IPD is based. From this basic outline, other authors have gone on to study the most suitable type of contract. Ghassemi and Becerik-Gerber [9] propose the possibility of transactional contracts, used to exchange goods and services, or relational contracts, aimed at establishing norms of behavior for the signatories. These authors and Ballard [3] state that binding the parties by relational contract is more beneficial as this type of agreement is more in line with the basic principles of IPD. However, Darrington [6] is skeptical about these statements, pointing out that a transactional contract could also be valid as long as collaboration is applied along the whole supply chain and not because the developer expressly orders it. It is even harder to decide which contract to choose when bearing in mind that IPD supports risk and reward sharing among a project’s participants. In practice, structures for sharing can reach high levels of complexity, and several methodologies coexist for them [13]. That is why it is particularly important to put these aspects down in writing in a contract to avoid any conflicts of interest between the parties.

3.2 Culture and Organization The concept of IPD encompasses a series of cultural and organizational aspects that are crucial for understanding how this methodology works. Just like LPDS before it, IPD is committed to establishing relationships of trust between all project’s partners by employing a model based on alliancing and partnering [18, 19, 24]. Some of the most important reasons for keeping up this type of relationship are the correct adaptation of the product to meet customer requirements and the reduction of project costs from the outset. Furthermore, studies have appeared in recent years that explain the benefits of this way of working for all the parties. For example, the study by Javanmardi et al. [11] points out that a subcontractor alone cannot improve their benefits unless they are introduced into a work group. Along the same lines, Eriksson [7] states that a multidisciplinary approach can help identify and eliminate waste at all stages of a project, giving rise to win-win relationships between participants. Therefore, in order to raise awareness among everyone involved in the importance of collaborative work, prior training in the IPD methodology and its techniques is recommended [9, 13].

3.3 Main Tools and IPD-BIM Integration Within the Lean Construction framework, there are two main tools that, because of their popularity, have become intrinsically linked to the IPD methodology. They are target value design (TVD) and Last Planner. On the one hand, TVD consists of planning and monitoring a project’s cost at all stages, but particularly during the

16

I. González-Boubeta and J. C. Prado-Prado

design stage as this is when it is easier to reduce the final cost [3]. Last Planner, on the other hand, consists of reducing variability in planning by means of decision-making depending on daily performance. The reference study for this subject is by Cho and Ballard [5], who observed that applying this tool has beneficial effects even when there is a scarce implementation of an IPD culture. The success of the combined use of TVD and Last Planner has been shown in cases analyzed by Ghassemi and Becerik-Gerber [9]. Building information modeling (BIM) systems are another tool which, although not a part of IPD itself, is often linked to it. These systems are based on a virtual representation of a construction that allows all its characteristics to be seen. As they are interactive, BIM systems are considered to be an important support tool for collaboration and decision-making [13]. Integrating BIM into IPD is an issue that has gained in importance in recent years because of its practical implications. Unfortunately, there are some obstacles to achieve this integration, such as the implementation of complex software and a lack of training in its use [8]. However, these difficulties contrast with the great opportunity BIM offers the sector. According to Mei et al. [17], the path toward innovation in the construction industry requires a combination of BIM and IPD, and an effort therefore seems necessary to develop the relationship.

4 Discussion and Research Agenda Analysis of the literature shows clearly different subject categories. In general terms, the contractual framework and IPD culture are the most developed areas. This makes clear what has already been stated by Smith et al. [23] when they stressed the need to develop more connections with IPD tools and with the technological sphere (i.e., BIM). This need becomes more apparent in the case of IPD and BIM integration due to the great expectations that exist regarding the benefits that such an alliance could provide [8, 17]. The following research agenda contains the gaps detected in the cultural and contractual category: • Risk and reward sharing: Little has been written on establishing under what terms the risks and rewards should be shared, how the levels of sharing are set, or how performance will be affected. Providing more details on these aspects, included within one of the pillars of IPD [16], is crucial from a practical viewpoint. • Sustainable approach: Given the current attention being paid to sustainability, there is a remarkable absence of studies linking its three pillars (economic, environmental, and social) to IPD. The economic pillar can be said to underlie IPD owing to its potential to generate savings, but other topics, such as safety at work (social pillar) or the use of environmentally friendly materials (environmental pillar), have hardly been dealt with. There is, therefore, a lack of studies that shed light on the relationship with these areas.

Integrated Project Delivery: A Literature Review and Research …

17

• The role of geographic and socio-cultural factors: The setting and social customs play an important role in the approach to work. These factors can therefore be relevant when it comes to developing a project by implementing the collaborative culture of IPD. To this end, developing studies that could reveal barriers or impacts on performance would be of great interest from a practical point of view. By advancing along these lines of research, it would be possible to define IPD’s potential further and highlight its central role in project management.

5 Conclusions For this article, a literature review has been undertaken for the IPD concept in order to clarify what contents already exist and define new research lines. After the review and the grouping of contents in different subject categories, it has been possible to identify a series of areas that are not well developed. In consequence, a research agenda was defined. Finally, the important role that Lean Construction should have in the construction industry of the future must be highlighted. The use of IPD and other Lean Construction tools is an interesting line of activity for organizations that want to reduce costs and increase productivity. Companies should therefore take an interest in this collaborative methodology and analyze the great benefits that its application could lead to.

References 1. AIA (2014) Integrated project delivery: an updated working definition. AIA California Council’s Definitions Committee of the Integrated Project Delivery Task Force 2. Andersen B, Belay AM, Seim EA (2012) Lean construction practices and its effects: a case study at St Olav’s Integrated Hospital, Norway. Lean Constr J 122–149 3. Ballard G (2008) The lean project delivery system: an update. Lean Constr J 11–19 4. Cárdenas LFA, Armiñana EP (2009) Un nuevo enfoque en la gestión: la construcción sin pérdidas. Rev Obras Públicas 46 5. Cho S, Ballard G (2011) Last planner and integrated project delivery. Lean Constr J 67–78 6. Darrington J (2011) Using a design-build contract for lean integrated project delivery. Lean Constr J 85–91 7. Eriksson EP (2010) Improving construction supply chain collaboration and performance: a lean construction pilot project. Supply Chain Manag: Int J 15(5):394–403 8. Fakhimi A, Sardrood JM, Mazroi A, Ghoreishi SR, Azhar S (2017) Influences of building information modeling (BIM) on oil, gas, and petrochemical firms. Sci Technol Built Environ 23(6):1063–1077 9. Ghassemi R, Becerik-Gerber B (2011) Transitioning to integrated project delivery: potential barriers and lessons learned. Lean Constr J 32–52 10. Heidemann A, Gehbauer F (2011) The way towards cooperative project delivery. J Financ Manag Prop Constr 16(1):19–30

18

I. González-Boubeta and J. C. Prado-Prado

11. Javanmardi A, Abbasian-Hosseini SA, Liu M, Hsiang SM (2017) Benefit of cooperation among subcontractors in performing high-reliable planning. J Manag Eng 34(2):04017062 12. Jeong W, Chang S, Son J, Yi JS (2016) BIM-integrated construction operation simulation for just-in-time production management. Sustainability 8(11):1106 13. Kim YW, Dossick CS (2011) What makes the delivery of a project integrated? A case study of Children’s Hospital, Bellevue, WA. Lean Constr J 53–66 14. Koskela L (1992) Application of the new production philosophy to construction, vol 72. Stanford University, Stanford, CA 15. Liu J, Shi G (2017) Quality control of a complex lean construction project based on KanBIM technology. EURASIA J Math, Sci Technol Educ 13(8):5905–5919 16. Matthews O, Howell GA (2005) Integrated project delivery an example of relational contracting. Lean Constr J 2(1):46–61 17. Mei T, Wang Q, Xiao Y, Yang M (2017) Rent-seeking behavior of BIM- and IPD-based construction project in China. Eng, Constr Arch Manag 24(3):514–536 18. Pishdad-Bozorgi P, Austin R, De La Garza JM (2016) Flash track practices distilled via structured interviews from EPC projects. Constr Res Congr 2016:168–178 19. Raisbeck P, Millie R, Maher A (2010) Assessing integrated project delivery: a comparative analysis of IPD and alliance contracting procurement routes. Management 1019:1028 20. Reim W, Parida V, Örtqvist D (2015) Product-service systems (PSS) business models and tactics–a systematic literature review. J Clean Prod 97:61–75 21. Sacks R, Harel M (2006) An economic game theory model of subcontractor resource allocation behaviour. Constr Manag Econ 24(8):869–881 22. Saunders M, Lewis P, Thornhill A (2011) Research methods for business students. Pearson Education Limited 23. Smith RE, Mossman A, Emmitt S (2011) Editorial: lean and integrated project delivery special issue. Lean Constr J 1–16 24. Smith JP, Rybkowski Z (2012) Literature review on trust and current construction industry trends. In: 20th annual conference of the IGLC, 18–20 25. Tillmann PA, Do D, Ballard G (2017) A case study on the success factors of target value design. In: Proceedings of the 25th annual conference of the IGLC, 563–570 26. Tezel A, Koskela L, Aziz Z (2017) Current condition and future directions for lean construction in highways projects: a small and medium-sized enterprises (SMEs) perspective. Int J Proj Manag 36:267–286 27. Wee BV, Banister D (2016) How to write a literature review paper? Transp Rev 36(2):278–288 28. Zimina D, Ballard G, Pasquire C (2012) Target value design: using collaboration and a lean approach to reduce construction cost. Constr Manag Econ 30(5):383–398

Empirical Measurement Instruments for Business Model Innovation: A Review Dorleta Ibarra, Jaione Ganzarain, and Juan Ignacio Igartua

Abstract Recent studies have highlighted the significant impact that business model innovation (BMI) has on business strategy, performance, and competitive advantage. However, the literature is still quite fragmented; there is no commonly accepted definition of the term, and progress toward solid theoretical constructs is still under development. Thus, the need to identify empirical measurement instruments that enable the definition and measurement of BMI has been identified. The present paper conducts a systematic literature review, selecting four empirically validated BMI measurement scales for analysis. After comparing them, it was concluded that the four measurement instruments are helpful tools for managers to assess business models (BM), benchmark the competition or plan a strategy. In addition, the instruments contribute to a better understanding of BM and BMI, and establish the basis for further research in the field. Some limitations were found, however, namely, that the instruments are not generalizable to study a particular phenomenon and their reliability could be affected when applying them in a different context to the original study. Finally, some items used in the scales were found to be open to multiple interpretations and thus care must be taken to ensure alignment with the research question. Keywords Business model innovation · Construct · Empirical studies · Measurement instruments · Systematic literature review

D. Ibarra (B) · J. Ganzarain · J. I. Igartua Mechanical and Industrial Production Department, Faculty of Engineering, Mondragon Unibertsitatea, Loramendi, 4, 20500 Arrasate, Gipuzkoa, Spain e-mail: [email protected] J. Ganzarain e-mail: [email protected] J. I. Igartua e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_3

19

20

D. Ibarra et al.

1 Introduction In the last decade, the business model (BM) and business model innovation (BMI) fields have received increasing attention from both academics and practitioners [12]. Market conditions are becoming increasingly challenging [1] and as a result, companies are being forced to change their way of doing business and adapt their established BM to respond to those challenges [17]. The BM term spread in the 90s with the advent of the Internet [9]. New ways of doing business were developed, and managers and entrepreneurs were forced to explain how they would generate value [16]. Since then, several definitions of BM can be found in the literature centered in three main approaches: (i) explaining how the firm works and makes money, (ii) addressing operational aspects that provide guidance and support to managers, and (iii) supporting business strategy in the identification of opportunities and creation of competitive advantage [23]. While the first two approaches are considered descriptive and fixed, the latter is based on the idea that no company is static since it is part of an environment that is constantly changing. Thus, BM must evolve in order to survive [2], and this development of the term has favored the emergence of the BMI field. In addition, recent studies have stressed the high impact that BMI has on business strategy, performance, and sustainability [17]. Various surveys conducted in recent years [14, 26] have highlighted how CEOs favor new BMs as a source of competitive advantage, arguing that companies implementing BMI had higher operating margins than competitors carrying on product and process innovation [3]. Nevertheless, despite the relevance of BMI for competitive advantage, there is still no consensus about its definition [11]. While some authors describe it as a process driving organizational change [18], others understand it as a result [6]. Some have argued that BMI is about changing an existing BM [21], while others have focused on creating a new BM [15]. Furthermore, some authors proposed that changing one element of the BM can lead to BMI [21]. Others, in turn, argued that “one or more” [13], “two or more” [19], or even all BM components and the architecture linking them need to be reinvented [20]. Further articles explore the degree of novelty of the term maintaining that BMI should be new to the industry [20], new to the firm [7], or both [16]. Thus, the BMI literature is still quite fragmented [30]. In addition, most research is based on conceptualizations and case studies [4]. This is helpful to understand BMI but slows down the progress toward the development of solid theoretical constructs. BMI is still to large extent in need of cumulative empirics that allow a common understanding of the term [11]. Moreover, scholars are calling for more empirical research, larger samples, and replicability of the studies [8]. In the same vein, several authors emphasize the need to clarify the BMI construct through a clear definition of the unit of analysis and the delimitation of its boundaries [4, 12]. In response to the increasing interest in BMI and the lack of clarity regarding its definition, the need to identify empirical studies that define the BMI construct has been identified. To this end, the present paper conducts a systematic literature review

Empirical Measurement Instruments for Business Model Innovation …

21

of papers that propose BMI measurement instruments. Four empirically validated measurement scales are compared, and their strengths and limitations are presented in the conclusions.

2 Research Methodology To develop the research, a systematic review was conducted following the main steps proposed by Becheikh et al. [5]. Firstly, the inclusion criteria were established, and then potential studies were located and selected. The criteria used to select and assess the identified articles are the following: • Focus exclusively on BMI measurement. Studies dedicated to other issues (BMI process implementations, success BM patterns, etc.) were not included. • Be published in peer-review academic journals to ensure the content quality. Other forms such as conference proceedings or books were not retained. • Include an empirical study following a quantitative approach. Theoretical and conceptual works were not considered. Neither were case studies. For the location of articles, three referent databases from the field were selected, namely, Scopus, Web of Science, and Business Source Premier. The keyword “Business model” was searched in the title to ensure that it was treated as the main topic in the article. In order to refine the search and focus on empirical-based quantitative measurement approaches, the keywords “empiric*” AND “assess*” OR “measur*” OR “evaluat*” were included in the title, abstract, and/or full text. 318 papers were collected, and 214 were left after removing duplicates. A first quick content check was conducted by reading titles and abstracts to identify whether the content of the articles was aligned with the criteria mentioned above. Those articles that did not meet the criteria were excluded. After this process, the 23 selected articles were carefully read, identifying four that potentially responded to the purpose of the study. These works [8, 24, 27, 28] explained the process followed to define the BMI construct and proposed an empirically validated measurement instrument.

3 Results of the Review: Comparison Between BMI Measurement Instruments From the four papers identified, the purpose of three of them [8, 24, 27] was to respond to the limited knowledge developed to date about the BMI concept. They highlighted the lack of empirically validated measurement scales and they developed one. In contrast, Waldner et al. [28] followed a markedly different approach, emphasizing the impact of industry structure on the drivers of BMI. The study investigated how different stages of the life cycle of an industry and levels of industry competition

22

D. Ibarra et al.

affect BMI, and how such innovation is translated into innovation performance. In the present paper, only the definition of the BMI construct and the items defined to measure it are considered. As the purpose of the studies was different, the process followed by the authors varied too. Thus, Clauss [8] as well as Spieth and Schneider [24] firstly defined BM and BMI, and then identified BM components through a literature review to define how to measure changes in BM elements. Verma and Bashir [27] followed the same process focusing on BM components identified in a review conducted by Westerlund [29]. Waldner et al. [28], in turn, identified the European Commission Innovation Survey (CIS) questions related to product and process innovation, and mapped them according to the BM components proposed by Teece [25]. With regard to BM definitions, all the articles took into consideration previous definitions found in the literature without suggesting a new one. Thus, BM was described overall as a combination of different value domains [22, 25] that follow an activity-based or system-level perspective [3]. In addition, BM was considered the subject of innovation instead of products or processes in three of the papers [8, 24, 28]. In the case of BMI, it was argued that it responds to the modification or introduction of new modes of value creation, delivery, or capture in three of the works [8, 27, 28]. Additionally, Clauss [8] proposed that BMI requires that the three outlined dimensions are changed, while Verma and Bashir [27] argued that if any of the components in an existing business model are changed it can be referred to as BMI. When defining the BMI construct, all the authors highlighted that BMs are configurations that integrate different business elements or dimensions and identified those dimensions in the literature. Despite the fact that the selected dimensions from each author are differently named, generally they encompass value creation, capture, and delivery dimensions [25]. Thus, value creation refers to how firms create value along the value chain due to internal and external network capabilities, activities, and resources. Value delivery describes to whom and how the company offers value. Finally, value capture responds to the economic logic behind the business or how they convert the value offered into profits for the company. Table 1 shows the variables identified for measuring BM components of the BMI construct of each work. Table 1 highlights, while value delivery and capture dimensions follow a similar approach, the value creation dimension is defined through several variables. Thus, it seems that all aspects of the company value chain are integrated but different nuances can be distinguished. Overall, both internal and external activities, resources, competences, and capabilities are considered. However, the only paper which includes organizational methods such as procedures and decision-making processes is that of Waldner et al. [28]. Concerning value delivery, some authors defined variables related to the target customer, the offering of products and services, and also channels and customer relationships [8, 24, 27, 28]. In addition, Waldner et al. [28] also included activities and processes as well as organizational methods and partnerships.

Empirical Measurement Instruments for Business Model Innovation …

23

Table 1 Variables defined for measuring BM components of the BMI construct Author

Value creation

Value delivery

Value capture

[8]

– New capabilities – New technologies/equipment – New processes/structures – New partnerships

– New offerings – New customer segments/markets – New channels – New customer relationships

– New revenue models – New cost structures

[27]

– – – –

Human capital Value network Linkage with partners Assets and capabilities

– Value proposition – Channels

– Costs – Revenue sources

[24]

– – – –

Core competencies Internal value creation External value creation Distribution

– Target customer – Positioning – Product and service offering

– Logic of earnings – Logic of costs

[28]

– New/improved goods/services – Product innovations new to the market, to the firm, to the world – New or improved methods of manufacturing/producing – New or improved supporting activities for processes – Cooperation with other enterprises or institutions – New business practices for organizing procedures – New methods of organizing work responsibilities and decision-making – New methods of organizing external relationships with other firms or public institutions – Importance of organizational innovations – Importance of marketing innovations

– Product innovations new to the market – New/improved logistics, delivery, or distribution methods – New or improved supporting activities for processes – Market introduction of new or improved goods or services – Cooperation with other enterprises or institutions – New methods of organizing work responsibilities and decision-making – New methods of organizing external relationships with other firms or public institution – Importance of improved ability to develop new products or processes – New methods for product placement or sales channels – Importance of marketing innovations

– New/improved supporting activities for processes – New methods of organizing external relationships with other firms or public institutions – New methods for product placement or sales channels – New methods of pricing goods or services – Importance of marketing innovations

24

D. Ibarra et al.

On the other hand, Spieth and Schneider [24] took into account how the company differentiates its offer from competition when measuring value proposition. This has to do with the fact that several studies integrate strategy as a BM component [8]. Nevertheless, recent studies have agreed that despite the fact that strategy impacts BM, it is distinct from it [8, 9, 16]. In the same vein, some literature reviews conducted by the authors showed how some previous publications included external factors such as technology, economics, or legal issues [8, 24]. However, these external factors were not considered part of the BMI. Finally, the measurement of value capture dimension through revenue streams and cost structure seems to be commonly accepted in three of the articles [8, 24, 27]. In contrast, Waldner et al. [28] adopted a broader value approach not just focusing on the economic aspect but encompassing the value generated when improving or introducing new activities and processes. When defining items for the scale, the ones proposed by Spieth and Schneider [24] are the most general as they simply add the “have changed” indicator to the variables (e.g., target customer has changed). The approaches followed by Clauss [8] and Verma and Bashir [27] are quite similar. Both defined part of the items starting with sentences such as “we continuously…” or “we regularly” as well as “we recently…” or “we have developed”. These scales are considered helpful to obtain the opinion of managers or decisionmakers due to a Likert-type scale. However, both proposed items can be interpreted in different ways. Thus, some of the questions respond to “what are we doing” that could be related to the behavior of the firm, whereas others focused on “what have we done” that are more oriented to measure the innovation capacity or the result obtained from those behaviors. The scale developed by Waldner et al. [28] is the most different since it was adapted from the CIS survey, based on the Oslo Manual [10]. It is an adaptation of a questionnaire mainly focused on product and process innovation to measure BMI. Thus, the variables encompassed further aspects not directly covered in the other measurement instruments such as organizational methods or marketing activities. Moreover, in contrast to the other three, the latter specified a temporal frame when defining the items (e.g., during the 3 years 2008–2010, did your enterprise introduce new or significantly improved services). Overall, each of the authors based their research on the identification of BM components that can be improved or innovated in order to measure the dimensions of BMI. Although the four studies seek to measure the innovativeness of BM by developing a validated scale, the approaches differ. Thus, the most similar results are those of Clauss [8] and Spieth and Schneider [24]. In both cases, the components of the BM were identified in the literature creating a hierarchical scale of three levels responding to the value creation, delivery, and capture dimensions. Despite the similarity, Clauss [8] defined ten indicators while Spieth and Schneider [24] proposed nine. On the other hand, Verma and Bashir [27] defined eight factors without categorizing them into the three elements mentioned above.

Empirical Measurement Instruments for Business Model Innovation …

25

Finally, with regard to the samples used for each study, Clauss [8] surveyed small and medium enterprises (SME) in the manufactory sector (i.e., the electronics, engineering, automotive industries, etc.). The data was collected through personal interviews with representatives of the firms in exhibitors at leading international industry trade fairs. Spieth and Schneider [24] focused on strategy and innovation management experts from large German firms operating in various industries, having both business-to-business and business-to-customer operations. Waldner et al. [28] centered on Austrian firms surveyed in 2010 during the course of the CIS that had recently introduced product or process innovations. Finally, Verma and Bashir [27] sent a questionnaire to middle-level executives from companies across seven industries (i.e., Information technology, banking, insurance, telecommunication, publishing, media & entertainment, and academia).

4 Conclusions Even when the same objective is pursued, the diversity of approaches and perspectives used to study BMI becomes visible. Thus, the following conclusions are drawn from the present study: 1. The four measurement instruments are considered helpful tools for managers when assessing their BMs, benchmarking their competition, or planning their strategy [8]. 2. In the academic field, they contribute to a better understanding of BM and BMI. 3. The measurement scales establish the basis for further research in the BMI field since they enable conducting large-scale quantitative research [24]. 4. The measurement instruments proposed by Clauss [8] as well as Spieth and Schneider [24] can be helpful when measuring BMI since they are based on BM components that are widely accepted. 5. The questionnaire developed by Waldner et al. [28] could be appropriate when temporary range is important. In addition, it gives extra information focused on organizational factors. Nevertheless, some limitations were observed: 1. The measurement instruments proposed are useful when assessing BMI in general but are not generalizable to study a particular phenomenon related to a specific BM. 2. The reliability of the measurement instruments could be affected when applying them in a different context to the original study, since the samples used are specific to the sector, size, and activity of the firm. Finally, several items used by some authors [8, 27] could be very helpful for both replicating the study and designing a new scale. However, the present study concluded that some of the questions were open to multiple interpretations. Thus when using this approach, it would be necessary to clearly define if the purpose is

26

D. Ibarra et al.

to measure the behavior of the firm or the innovation capacity/results. Not following this step could cause some confusion when trying to define and measure BMI, since to date, the discussion as to whether BMI is a process or a result still continues.

References 1. Abdelkafi N, Makhotin S, Posselt T (2013) Business model innovations for electric mobility— what can be learned from existing business model patterns? Int J Innov Manag 17:1340003 2. Al-Debei MM, Avison D (2010) Developing a unified framework of the business model concept. Eur J Inf Syst 19:359–376 3. Amit R, Zott C (2012) Creating value through business model innovation. MIT Sloan Manag Rev 53:41 4. Andreini D, Bettinelli C (2017) Business model innovation: from systematic literature review to future research directions. Springer 5. Becheikh N, Landry R, Amara N (2006) Lessons from innovation empirical studies in the manufacturing sector: a systematic review of the literature from 1993–2003. Technovation 26:644–664 6. Berglund H, Sandström C (2013) Business model innovation from an open systems perspective: structural challenges and managerial solutions. Int J Prod Dev 18:274–285 7. Bock AJ, Opsahl T, George G (2010) Published. Business model innovations and strategic flexibility: a study of the effects of informal and formal organization. In: Proceedings of the Sumantra Ghoshal conference for managerially relevant research, London, UK, 2010 8. Clauss T (2017) Measuring business model innovation: conceptualization, scale development, and proof of performance. R&D Manag 47:385–403 9. DaSilva CM, Trkman P (2014) Business model: what it is and what it is not. Long Range Plan 47:379–389 10. De Oslo M (2005) Guía para la recogida e interpretación de datos sobre innovación. OECD, Luxembourg 11. Foss NJ, Saebi T (2017a) Business models and business model innovation: between wicked and paradigmatic problems. Long Range Plan 12. Foss NJ, Saebi T (2017b) Fifteen years of research on business model innovation: how far have we come, and where should we go? J Manag 43:200–227 13. Frankenberger K, Weiblen T, Csik M, Gassmann O (2013) The 4I-framework of business model innovation: a structured view on process phases and challenges. Int J Prod Dev 18:249–273 14. Giesen E, Riddleberger E, Christner R, Bell R (2010) When and how to innovate your business model. Strat LeadShip 38:17–26 15. Markides C (2006) Disruptive innovation: in need of better theory. J Prod Innov Manag 23:19– 25 16. Osterwalder A, Pigneur Y (2004) An ontology for e-business models. Value Creat E-Bus Model 1:65–97 17. Pölzl J (2016) Business model innovation as a business development strategy 18. Rauter R, Müller C, Vorbach S, Marko W (2012) Business model innovation and knowledge transfer. EURAM European Academy of Management, 1–35 19. Reeves M, Stalk G, Deimler M (2009) Business model innovation: when the game gets tough, change the game. Boston Consulting Group, Boston 20. Santos J, Spector B, Van der Heyden L (2009) Toward a theory of business model innovation within incumbent firms. INSEAD, Fontainebleau, France 21. Schneider S, Spieth P (2013) Business model innovation: Towards an integrated future research agenda. Int J Innov Manag 17:1340001 22. Shafer SM, Smith HJ, Linder JC (2005) The power of business models. Bus Horiz 48:199–207

Empirical Measurement Instruments for Business Model Innovation …

27

23. Spieth P, Schneckenberg D, Ricart JE (2014) Business model innovation–state of the art and future challenges for the field. R&D Manag 44:237–247 24. Spieth P, Schneider S (2016) Business model innovativeness: designing a formative measure for business model innovation. J Bus Econ 86:671–696 25. Teece DJ (2010) Business models, business strategy and innovation. Long Range Plan 43:172– 194 26. Unit EI (2005) Business 2010–embracing the challenge of change. The Economist 27. Verma R, Bashir M (2016) Business model innovation: Scale development and validation. Int J Appl Bus Econ Res 14:5057–5069 28. Waldner F, Poetz MK, Grimpe C, Eurich M (2015) Antecedents and consequences of business model innovation: the role of industry structure. Emerald Group Publishing Limited, Business models and modelling 29. Westerlund M (2009) Managing networked business models: essays in the software industry. Helsinki School of Economics 30. Wirtz BW, Pistoia A, Ullrich S, Göttel V (2016) Business models: origin, development and future research perspectives. Long range planning, 49, 36–54

Scrip Dividends and Share Buyback Strategies Based on Volatility Angel Huerga and Carlos Rodríguez-Monroy

Abstract The number of listed companies offering alternatives to cash dividends is increasing in Europe. Companies can reduce the cash outflows by giving shareholders the option to receive either shares or cash. Some investors favor scrip dividends due to the implicit-free call option attached to the scrip distributions, and recent studies confirm that the market does not react negatively, helping to reduce the agency problem. Additionally, companies can avoid dilution by repurchasing the shares offered in the scrip. Repurchase strategies using volatility and derivatives can guarantee a lower repurchase price, improving the capital ratios of the company and increasing BVPS. Keywords Scrip dividends · Dividend policies · Options · Share Buyback · Volatility · Agency problem

1 Introduction Since the financial crisis of 2007, the use of optional stock dividends of scrip dividends has significantly increased among European companies, particularly in the financial institutions, oil, energy, and telecom sectors. Tax advantages for shareholders were the main driver in the past but changes in regulation in many countries eliminated such edge. Company managers are reluctant to reduce dividends because of the potential market impact of such decision, and dividends create a constraint for managers, which conflicts with their objective of conserving cash and maintaining financial flexibility [3]. Nevertheless, recent studies show that in present years investors react favorably to scrip dividends [4] [1]. A. Huerga · C. Rodríguez-Monroy (B) Dpto. de Organización, Administración de Empresas Y Estadística, ETS de Ingenieros Industriales, Universidad Politécnica de Madrid (UPM), Madrid, Spain e-mail: [email protected] A. Huerga e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_4

29

30

A. Huerga and C. Rodríguez-Monroy

Fig. 1 Scrip dividends time frame

Scrip dividends are profit distributions to shareholders that include the option to receive new shares in lieu of cash at the shareholder discretion. At the ex-dividend date, the company gives the shareholder the right to elect to receive cash or shares for a short period of time that varies from 1 to 4 weeks typically (Fig. 1). In some cases, the scrip trades in organized markets during the election period. Scrip dividends are elected by a majority of shareholders since they grant an additional call option with the shares during the period of optionality that can be monetized. Cash dividends are one of the classical examples of the agency problem, where the conflict of interests between shareholders that want to maximize their equity value and that in general want to receive regular cash distributions and company managers that tend to prefer lower financial leverage, lower cost of debt, and therefore reduced cash dividends. Company managers have also the incentive to reduce the dilution for existing shareholders, and in several occasions in connection with scrip dividends, companies announce share repurchase plans. In this paper, we propose a share repurchase method that aims to reduce shareholder dilution and at the same time improves the capital ratio of the companies and increases net equity and shareholder value.

2 Objectives This paper is the first document of an ongoing academic research that aims to study how and why companies use scrip dividends as a distribution method. The first objective is to study the free financial options attached to scrip dividends distributions and how the market values such options. A second objective is to study is the repurchase methods that companies can use to avoid the dilution caused by scrip dividends and in particular one variation of the derivative instruments called accumulator share buyback. Using market volatility, accumulators can guarantee a minimum amount of shares repurchased at a price below the new issue price, which as a side effect improves the company’s balance sheet and increases the equity value for shareholders.

Scrip Dividends and Share Buyback Strategies Based on Volatility

31

3 Methods This paper researches the scarce academic literature about scrip dividends to assess the size of the scrip dividend market, the motivation of managers to issue scrip dividends, and the impact in share prices. Using the vanilla options pricing model [2], we assess the extra value granted to shareholders in the form of the options attached to scrip dividends. Then using numerical option pricing methods, we calculate the cost of a strategy that can allow buying back shares at a lower price.

4 Size of the Scrip Dividend Market in Europe According to Markit, the total value of dividends paid in scrip by European listed companies including UK listed companies has significantly increased over the last 10 years reaching Eur27bn in 2016. Banks, oil, and commodities companies are the sectors responsible for the growth of this form of shareholder remuneration. At least 39 major listed European companies paid scrip dividends in 2017 (Fig. 2).

5 Motivations for Scrip Dividend Distributions Some companies prefer the scrip format for dividend distributions. Part of the reasons given by company managers are: • Companies can offer greater optionality to their investors which can decide whether to receive cash or new shares, and in some jurisdictions adapt taxation to the actual disposal of the shares.

Fig. 2 Amount of scrip dividends in Europe. Source Markit

32

A. Huerga and C. Rodríguez-Monroy

• Scrip dividends offer shareholders an implicit call option for some weeks since they can choose some weeks in advance whether to receive cash or a predefine number of shares at a strike price. Investors can monetize the time value and volatility of the implicit option by selling calls/volatility in the market. • Scrip dividends can reduce the cost of cash in times of liquidity stress but at the same time increase the “agency problem” between managers and shareholders [5, 6]. Managers tend to favor scrip dividends that avoid cash flows out of the company whereas shareholders in the classical theory prefer a cash distribution. • Shareholders should be indifferent to the profit distribution policy of the company and therefore satisfied with scrip dividends in well-run profitable companies since they increase retained earnings [7]. • Companies can avoid cash outlays by paying dividends in scrip format. Cash flows from financing activities are then maximizing allowing more free cash flow for capital investment and debt repayment. • Financial institutions can improve regulatory ratios paying distributions in shares instead of cash (ratios as Common Equity Tier 1, LCR, Leverage Ratio). • Firms can obtain equity at none or smaller discount than via rights issues or other equity offering. • Better solution than a dividend cut for cash strapped companies.

6 Implicit Options Attached to Scrip Distributions Thanks to the scrip dividends issuance procedure, company shareholders are implicitly granted with options or warrants on the new issued shares. The mechanism of scrip dividends guarantees the existence of such options. The issue price is always set at the inception of the scrip dividend period, and during the election period, shareholders can take the cash dividend if the share price decreases below the issue price or share if the market price is higher than the issue price. This optionality can be considered as an extra distribution to shareholders, and in fact hedge funds and large shareholders position themselves to monetize such value. The average maturity of the implicit options attached to the scrip dividends, the scrip election period, is 19 days, in the European markets where this practice is common [4]. In some cases, companies offer scrip dividends at a discount compared to the prevailing market share prices, trying to entice existing shareholders to accept shares instead of cash. According to David and Ginglinger [4], the average discount of the scrip shares compared to market prices the week before the ex-dividend date is 8.5%. Using the Black-Scholes continuous valuation model for equity options (Black Scholes [2], the value of a European Call with a strike K = 91.5%, maturing in 19 days, with average volatility of 30%, 3% dividend yield, and market interest rates of 1% is close to 9.3%, implying an implicit value of 8.5% and a time value of 0.8%.

Scrip Dividends and Share Buyback Strategies Based on Volatility

33

In some other cases, companies are not offering any discount to shareholders, and the new shares are issued at the initial market price. But even in such case, the value of a European Call, strike at the money, maturing in 19 days, with average volatility of 30%, 3% dividend yield, and market interest rates of 1% is close to 3.2%. If the scrips fluctuate in the market, their price should follow the price of the listed options and shareholders could sell the scrip and pocket the extra value from the ex-dividend date. In other cases, shareholders can keep the scrip and sell the proportional number of call options and cash-in the premium. If investors cannot find a market for volatility, they still can lend the shares out to an investor willing to keep the call and receive the cash dividend plus the call value in the borrowing fee.

7 How to Reduce Dilution Linked to Scrip Dividends Companies can support their share price by announcing share buyback programs linked to the scrip dividend distributions. New shares in the market imply that company’s future free cash flow will be distributed among more shares, meaning a higher dilution for the existing shareholders that have chosen not to receive shares. To avoid shareholders dilution, managers can raise new debt in the market and use that new cash to repurchase the new equity issued. But this solution deteriorates the leverage ratios of the company and again can reduce the future claim of shareholders. Company managers can also issue other equity instruments that do not imply future profits distributions, like mandatory convertibles, to obtain that amount of cash. But due to market price oscillations, on both solutions, the amount cash needed for a later share repurchase is unknown in advance by company managers. Share price increases can result in a higher than forecasted financial indebtedness increase in the case of a debt issuance or an undesirable future increase in the outstanding shares at maturity, and again dilution in the case of mandatory convertibles. Both solutions imply uncertainty and a potential future reduction in the net shareholders’ equity of the company. Can it be avoided?

8 Share Repurchase Strategies with Volatility Instruments Companies can use repurchase strategies involving traded or OTC options with physical delivery can help companies to improve their capital base. Managers can sell volatility to derivatives dealers via sold puts to improve their repurchase call strike price and at the same time can purchase calls on a lower proportion. The higher sold volatility can help to reduce the call strike. Daily share repurchase accumulator strategies can smooth the delta and gamma factors of the options allowing dealers to trade higher call and put options notionals and therefore set the strike price on a single observation date or in a series of days

34

A. Huerga and C. Rodríguez-Monroy

coinciding with the scrip price set dates. Daily observation of the options can imply a higher number of shares repurchased, avoiding the European option single date observation effect and potentially increasing the number of share repurchased. The repurchase price or options strike must be set at the same time as the scrip dividend new issue price is announced. The strike K will be lower than the share price S. The theoretical price of an accumulator can be calculated as: Fair V alue Accum =

N 

x ∗ Cuo (K, H , Ti ) − y ∗ Puo (K, H , Ti )

i=1

where y > x, Cuo (K, H , Ti ) is the price of a purchased physical delivery Call option with strike K, maturity at T, and knock out barrier at H, and Puo (K, H , Ti ) is the price of a sold physical delivery Put option with strike K, maturity at T, and knock out barrier at H. The strike K of both options can be set below the spot and forward share price, thanks to a higher volatility sold but the number of shares repurchased is unknown. Calculations using numerical algorithms show that using daily observation Call and Put options, without KO barriers, and a y/x factor of 2, the repurchase price can be set between 5 and 8% below spot prices, offsetting the share discount or improving retained earnings if no discount is offered.

9 Accounting Impact of a Share Repurchase at a Lower Strike If a company can repurchase the shares offered in the scrip dividend at a lower price, this will result in a net capital gain, reflected in new Net Retained Earnings balance sheet item and with no PnL impact. The structure is always more accretive for shareholders than standard cash repurchases and increases the book value per share. Using the cost method for share repurchase accounting the cash spent for repurchasing, the stock is lower than the estimated cash outflows for dividends, and the remaining amount is accounted under retained earnings. Share BuyBack at Par Value LiabiliƟes Assets Cash Other Assets

Treasury easur Shares Other LiabilƟes

Share BuyBack at a Lower Price LiabiliƟes Assets Cash

Treasury easur Shares

Cash’

Retained Earnings

Other Assets

Other LiabilƟes

Scrip Dividends and Share Buyback Strategies Based on Volatility

35

10 Conclusions The agency problem between shareholders and management can be mitigated using scrip dividends distributions. Recent studies show that companies paying scrip dividends are committed to avoid dividend cuts and the market does not react negatively to such announce. The present paper analyzes the value of the options attached to scrip dividends that range between 9.2% for the typical discount over the market price of the company shares and 3.2% for scrip dividends issued with no discount. Such discount levels attract both arbitrage funds and large investors that find more value to scrip dividends in the present cash-rich environment, since scrip dividends offer to shareholders an extra value that can be monetized. One method to reduce the dilution associated to scrip dividends is share buyback programs performed with the help of daily strips of leveraged derivatives or share repurchase accumulators that can effectively allow the repurchase of a limited number of shares below the new issue price, which improves the book value of the company.

References 1. Bessembinder, H, Zhang, F (2015) Predictable corporate distributions and stock returns. Rev Financ Stud 28(4):1199–1241 2. Black F, Scholes M (1973) The pricing of options and corporate liabilities. J Polit Econ 81:637– 654 3. Blau, M, Fuller K (2008) Flexitility and dividends. J Corp Financ 14(2):133–152 4. David T, Ginglinger E (2016) When dividends is not bad news: the case of optional stock dividends. J Corp Financ 40(2016):174–191 5. Lasfer A (1997) Scrip dividends: the management’s view. Eur Financ Manag 3(2):237–249 6. Lasfer A (1997) On the motivation for paying scrip dividends. Eur Financ Manag 26:62–80 7. Miller M, Modigliani F (1961) Dividend policy, growth, and the valuation of shares. J Bus 34:411–433

The Evolution of Business Intelligence with Neuroinformatics Irene Martín-Rubio , Juan Fombellida , and Diego Andina

Abstract The connection among different agents provides data and information embedded in different objects and materials that can challenge the way industry and factories are been running. In this way, Business Intelligence enables Smart Manufacturing and leverage data to improve company performance. Business Intelligence facilitates new insights about product developments from a different point of views: customers, suppliers, manufacturing process, quality, innovation, etc. In twenty-first century, decision-making process in business is being not only evolved but also revolutionized by the way firms are being able to manage information, thanks to new artifacts, like sensors, and engineering methodologies like systems of system engineering (SoSE) and neuroinformatics. Keywords Business Intelligence · Neuroinformatics · Learning

1 Introduction Industry 4.0, the Fourth Industrial Revolution, or Smart Manufacturing is just here. Operations management has just merged with telecommunications information technology. The key aspect of smart manufacturing lies in the rapid and pervasive flow of information within the whole supply chain, not only inside the manufacturing plant [26].

I. Martín-Rubio (B) Dpto. Ing. Organización, Admón de Empresas y Estadística, ETSIDI - Universidad Politécnica de Madrid, Madrid, Spain e-mail: [email protected] J. Fombellida · D. Andina Group for Automation in Signals and Communications, ETSIT - Universidad Politécnica de Madrid, Madrid, Spain e-mail: [email protected] D. Andina e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_5

37

38

I. Martín-Rubio et al.

The Fourth Industrial Revolution is renovating not only operations, manufacturing, products, and logistics, but also the controlling aspects as a performance management system. The traditional methodologies and activities of performance management become smarter which involves agile process of evaluation, analysis or information, decision-making, and control in business management [17]. Business intelligence (BI) plays a key role in the analysis of the pervasive information that flows in the smart manufacturing. BI and smart manufacturing, like system networks, require powerful intelligent procedures to facilitate the integration, interaction, and process of data provided by different agents and in different materials and objects. We suggest using neuroinformatics as a powerful instrument to promote organizational learning in Industry 4.0 and its systems.

2 Business Intelligence in Industry 4.0 The term BI is commonly applied referring to new technologies, applications, and information analysis applications that allow better business informed decisions [31]. BI is a general concept for inspiring better and agile decisions at every level of organization (strategy and operations levels of firms) through relevance and reliable information at the right moment and with the collaboration of the right actor and material. In the evolution of big data, performance management analytics, and predictive modeling with different methods and programs, firms can identify areas of strength and weakness, discover new opportunities, and improve business decisions in a meaningful and actionable way [4, 8, 10]. Smart manufacturing has transformed the whole production and supply chain. It has revolutionized how organizations integrate data along the product lifecycle to assess new design, development, manufacturing, logistics, and customer needs. We can identify three critical elements in this evolution SMLC [28]: (1) Manufacturing Big Data generation and integration. Sensors, intelligent motors, automatic controls, and operations software produce data for specialized islands or cells of work. It requires a traceability system to integrate the whole supply chain and manufacturing system [25]. (2) Business and manufacturing intelligence management. Advanced computer modeling can process the Big Data and leverage flexible manufacturing management and control. BI can check production lines and entire plants. The connection of the data through the supply chain—energy, raw materials, and customer demands—will enable greater product customization. It is a great challenge to convert these data into actionable information that can be optimized at smart manufacturing. (3) Smart product development. Manufacturing intelligence and business intelligence can promote smart product development. Manufacturing intelligence and BI can promote the launch of smart product. Smart product development

The Evolution of Business Intelligence with Neuroinformatics

39

is integrated easily in the complete manufacturing process and dynamically assists their manufacturing process, controlling individual productions phases autonomously. The introduction of BI systems involves developing new engineering techniques as well as radical new paths and interactions with actors across the whole supply chain for re-engineering the constant evolution and product lifecycle management according to customer needs at real time. Customers can design the product and suggest to the factory what product to manufacture.

3 The Contribution of SoS Engineering and Neuroinformatics to BI The expansion of smart manufacturing and BI generates a variety of interconnected systems. New computational models, analytics, and engineering procedures and techniques are needed for managing high volume of data from different sources along the whole supply chain. It is a great challenge to convert these data into actionable information that can be optimized at smart manufacturing. The development of system of systems engineering (SoSE) can provide smarter methods to meet these new challenges [21–23].

3.1 Systems of Systems Engineering (SoSE) System of systems engineering (SoSE) facilitates the understanding of the collaboration of different systems in terms of costs, lead time, and quality. Advances in sensor technology, the Internet, wireless communication, and memory have all contributed to an explosion of “Big Data”. System of Systems (SoS) integrate independently operating, non-homogeneous systems to achieve synergies among different parts of the manufacturing plant. SoS are also contributing to the management of “Big Data” [2, 16]. With System of Systems Engineering Design Paradigm, new socio-technical objectives can be achieved in smart manufacturing. SoSE promotes the coordination and integration of multiple complex systems while ensuring the sustainability of the newly created system of systems. SoSE facilitates the understanding of the collaboration of different actors (different systems) and its performance indicators across product lifecycle management (quality, costs, lead time, materials). Advance in sensor electronics and Internet of Things (IoT) provides a high volume of data with multiple variables that can be integrated with SoSE. SoSE integrates distributed and independent, non-homogenous systems to achieve synergies among different parts, and agents of the manufacturing plant [2, 16]. SoSE promotes the collaboration among a wide range of systems

40

I. Martín-Rubio et al.

with new and complex socio-technical goals and performance requirements that can leverage proactive and sustainable interactions among all actors [26]. The foundation of SoS’ concept didn’t appear easily [7] but through different steps. At the beginning, Shenhar and Bonen [27] indicate that SoS is a methodology that includes a combination of systems operating at the same time to get collective objectives. Later, Maier [19] provides an insight closer to the actual comprehension of SOS. For Maier, SoS is formed by elements or systems. Every individual system enjoys operational and managerial independence. As these definitions don’t separate the system from a systems of system, Bar-Yam et al. [3] studied disciplines related to behavior from a natural (biology) and sociological point of view, and conclude that SoSE has properties of emergent behavior, adaptation, self-organization, complex system individual specialization, and synergy. Consequently, Jamshidi [14, 15] presents the following features of SoS: operational and managerial independence, emergent behavior, geographic distribution, and evolutionary improvement. Grimm and Railsback [13] suggest the following elements to record the design of models SoSE: emergence, adaptation, prediction, and interaction. In conclusion, every system promotes the behavior analysis of the agents and the self-organization of the whole structure. Therefore, with this model, it is possible to understand the skills of every system (agent) to adapt to every interaction according to the environment changes and their goals [3]. Cormen, 2009, indicates not only the relevance of learning capability of systems, but also its memory and the management of the uncertainty of environment. The interaction, between two systems, has two possibilities: competition or collaboration. Advanced sensor applications give signals of environmental changes in their surrounding area that provides source of interactions with other agents. The sensor gauge what happens and the switching possibility could be the following instructions for collaboration and sharing resources with other agents or competing.

3.2 Neuroinformatics Over the past century, we have seen a dramatic evolution in advanced methods for constructing and programming computing and manufacturing machines. However, these methodologies are an evolution of existing technologies, based on human designers and programmers. Artificial learning provides the basis for artificial modeling in this radically different approach [9]. New algorithmic self-programming and construction from datasets deeply influence smart manufacturing technologies. Bio-inspired systems use an entirely different concept based on biological evolution, rather than depending on human programmers and designers. Biological systems build themselves by means of a process that involves systematic spatiotemporal generation from the interaction of units within a massive distributed parallel complex processing system. One key issue in the analysis of the biological nervous system

The Evolution of Business Intelligence with Neuroinformatics

41

is explaining the coding, storage, and decoding process of neural signals [31]. This is not just an important item on the research agenda but a major challenge for computer science and industry. The evolution of neuroinformatics can be applied to non-nervous technology. BI and smart manufacturing can benefit from it. Coding processes may take place at different levels of abstraction and in different ways. These processes are characterized by the implementation of the analysis of brain decisions and disorders with accurate scientific data that structure and scale multiple dimensions of different variables [11]. Current advanced bio-inspired BP models such as artificial metaplasticity MLP (AMMLP) [1, 20] use artificial neural networks (ANNs), which are closer to biological learning, improving their performance and information learning capability through experience an arbitrary dataset of patterns. When an unknown pattern input is applied to the structure of the system, it learns from previous behaviors and produces an appropriate performance. Data patterns are thus necessary to properly develop neuroinformatics applications, but in neuroscience they are difficult to integrate and often not amenable to computation. Today, neuroinformatics facilitates high-throughput data processing and modeling, as web platforms for sharing time-aligned, brain-behavior data in order to identify statistical models that explain the correlation between nervous systems and their performance from big complex datasets [18, 29].

3.3 BI Meets SoSE and Neuroinformatics BI maturity is evolving from BI database management (BI 1.0) to Web search engines (BI 2.0) that suggest bringing together new engineering techniques for improving the analysis of continuously enriched databases. A new research opportunity is emerging as BI 3.0 with the use of mobile phones and tablets. Mobile devices and their entire ecosystem—with the use of sensor-based, radio frequency identification (RFID) tags for inventory tracking, email, query logs for websites, etc.—are transforming different facets of society with exciting new streams of innovative applications, products, and services [6]. Business are leveraging their data assets [5]. Mwilu et al. [24] have presented a systematic literature review of present research related to BI in the cloud. The new architectures should combine the conventional data warehouse platform with the utilities provided by the cloud. A classic example of a human artifact inspired by biological nervous systems is a backpropagation (BP) neural network or multilayer perceptron (MLP), currently used in building deep networks capable of learning multiple levels of representation to accomplish a complex task, with higher levels capturing more abstract concepts through deeper learning from datasets, with the aim of adjusting the system parameters and layers [4, 32]. The purpose is to design the backpropagation neural performance system. Yan et al. provide an example of application of this objective with the introduction of a new score evaluation index applied to the requirements of BI architectures.

42

I. Martín-Rubio et al.

The evolution of Industry 4.0 must also address the critical issue of traceability [12]. Traceability in smart materials is based on the analysis of interactions among software engineering artifacts, thanks to the advance in sensor technologies [30, p. 1]. Through proper nonlinear model, neuroinformatics has the potential to provide complex yet rigorous traceability. The application of powerful distributed processing can also reduce the time and communication needed for product and process traceability. The application of neural systems to BI systems boosts the analysis of organizational behavior at companies. The analysis of the path of creating value with a high source of information is revolutionizing the smart manufacturing environment. Managers are aware of the strategic and operational challenges that innovation in the potentials of BI entails for increasing quality, profitability, and customer satisfaction. Fink et al. [10] encourage the strategic value of new investment in BI infrastructure, but it is needed an effort to recognize the organizational routines and organizational learning capabilities that promote radical innovation entails. Today, metaplasticity and neuroinformatics can enhance organizational learning through intensive arbitrary data experience patterns and artificial management of information learning capability patterns. Crucial implications for practice emerge for our findings. Managers can start deploying BI architectures with neuroinformatics and SoS engineering. A critical team of smart people and engineers in these areas are needed for increasing competitiveness in Industry 4.0.

4 Conclusion As a sophisticated method based on mature models that have been previously tested in neuroscience, neuroinformatics provides a new research agenda for designing BI and Industry 4.0 systems. Neuroinformatics not only can integrate a huge amount of relevant data and combine them with advanced algorithms but also promote organizational learning in Industry 4.0. A new understanding of key business processes can improve many value drivers, such as sales forecasting, fraud detection, credit risk management, customer segmentation, warehouse efficiency, predictive maintenance, and new product development. New pervasive applications of SoSE and neuroinformatics can develop the Industry 4.0 while transmitting, gauging, and switching behaviors and its indicators. The introduction of SoSE and neuroinformatics in the development of Industry 4.0 means not only technological evolution, but also a revolution in the learning capabilities of the firm to adapt in a more agile way to the challenges of the environment. The key tasks for neuroinformatics are structuring the complex and vast amounts of information and transforming data into knowledge in order to obtain a functional understanding of Industry 4.0, while maintaining confidentiality and information security. System of systems engineering, as a systematic way to integrate a wide variety of different systems, will help the scientific community to fulfill these requirements.

The Evolution of Business Intelligence with Neuroinformatics

43

References 1. Andina D, Alvarez-Vellisco A, Jevtic A, Fombellida J (2009) Artificial metaplasticity can improve artificial neural network learning. Intl Autom Soft Comput Spec Iss Signal Process Soft Comput 15(4):681–694 2. Barnabas K, Tannahill A, Jamshidi M (2014) System of systems and big data analytics— bridging the gap. Comput Electr Eng 40(2014):2–15 3. Bar-Yam Y, Allison MA, Batdorf R, Chen H, Generazio H, Singh H, Tucker S (2004) The characteristics and emerging behaviors system of systems. NECSI: Complex Physics, Biol Soc Syst Project 4. Brichni M, Dupuy-Chessa S, Gzara L, Mandran N, Jeannet C (2017) BI4BI: a continuous evaluation system for business intelligence systems. Expert Syst Appl 76(2017):97–112 5. Chaduri S, Dayal U, Nqarasayya V (2011) An overview of business intelligence technology. Commun ACM 54(8) 6. Chen H, Chiang RHL, Storey VC (2012) Business intelligence and analytics: from big data to big impact. MIS Q 36(4) 7. Darabi HR, Mansouri M (2013) The role of competition and collaboration in influencing the level of autonomy and belonging in system of systems. IEEE Syst J 7(4) 8. Davenport TH (2006) Competing on analytics. Harv Bus Rev 84(1):98–107 (Retrieve 28th March 2017) 9. Douglas R (2011) Constructive cortical computation. Procedia Comput Sci 7. In: The European future technologies conference and exhibition 2011, pp 18–19 10. Fink L, Yogev N, Even A (2017) Business intelligence and organizational learning: an empirical investigation of value creation processes. Inf Manag 54:38–56 11. Fletcher M, Liang B, Smith L, Knowlesc A, Jackson T, Jessop M, Austin J (2008) Neural network based pattern matching and spike detection tools and services in the CARMEN neuroinformatics project. Neural Netw 21:1076–1084 12. Fombellida J, Martin-Rubio I, Torres-Alegre S, Andina D (2018) Tackling business intelligence with bioinspired deep learning. Neural Comput Appl 13. Grimm V, Railsback SF (2005) Individual-based modeling and ecology. Princeton Univ. Press, Oxfordshire 14. Jamshidi M (2005) Theme of the IEEE/SMC. Technical report, Waikoloa, Hawaii, USA 15. Jamshidi M (ed) (2009) System of systems engineering—innovations for the 21st century. Wiley, New York, NY 16. Jamshidi M (2010) From large-scale systems to system of systems—control challenges for the 21st century. In: IFAC large-scale systems symposium, Lille, France, 11–14 July 2010 17. Kibira D, Morris KC, Kumaraguru S (2016) Methods and tools for performance assurance of smart manufacturing systems. J Res Natl Inst Stand Technol 121 18. Koslow SH, Subramanian S (eds) (2005) Databaing the brain: from data to knowledge (Neuroinformatics). Wisley, 523 pp 19. Maier MW (1998) Architecting principles for systems-of-systems. Syst Eng 1(4):267–284 20. Marcano-Cedeño A, Marín-de-la-Bárcena A, Jiménez-Trillo J, Piñuela JA, Andina D (2009) Artificial metaplasticity neural network applied to credit scoring. Int J Neural Syst 21(4):311– 317. https://doi.org/10.1142/s0129065711002857 21. Martín-Rubio I, Andina D (2018) Smart manufacturing in a SoSE perspective. In: Yahyaoui I (ed) Advances in renewable energies and power technologies, vol 2, pp 479–507 22. Martín-Rubio I, Florence-Sandoval AE, Jiménez-Trillo J, Andina D (2015) From smart grids to business intelligence, a challenge for bioinspired systems. In: International work-conference on the interplay between natural and artificial computation IWINAC 2015: bioinspired computation in artificial systems, pp 439–450 23. Martín-Rubio I, Andina D, Tarquis AM (2016) Business intelligence: new products development and supply chain systems in a SoSE perspective. In: WAC (World automation conference), Puerto Rico, 31 July–4 Aug. https://doi.org/10.1109/wac.2016.7582998

44

I. Martín-Rubio et al.

24. Mwilu OS, Comyn-Wattiau I, Prat N (2016) Design science research contribution to business intelligence in the cloud—a systematic literature review. Futur Gener Comput Syst 63:108–122 25. Olsen P, Borit M (2013) How to define traceability. Trends in food science & technology 26. Romero D, Noran O (2015) Green virtual enterprises and their breeding environments: engineering their sustainability as systems of systems for the circular economy. IFFFACCCPapersOnLine 48–3:2258–2265 27. Shenhar AJ, Bonen Z (1997) The new taxonomy of systems: toward an adaptive systems engineering framework. IEEE Trans Syst Man Cybern A Syst Hum 27(2):137–145 28. SMLC (Smart Manufacturing Leadership Coalition) (2010) What is smart manufacturing, 2010. https://smartmanufacturingcoalition.org/sites/default/files/what_is_smart_manufacturing_-_ time_magazine.pdf. Accessed 2 Aug 2017 29. Takemiya M, Majima K, Tsukamoto M, Kamitani Y (2016) BrainLiner: a neuroinformatics platform for sharing time-aligned brain-behavior data. Front Neuroinformatics 26 30. Vale T, Santana de Almeida E, Alves V, Kulesza U, Niu N, de Lima R (2017) Software product lines traceability: a systematic mapping study. Inf Softw Technol 84:1–18 31. Watson P, Jackson T, Pitsilis G, Phillip L et al (2007) The CARMEN neuroinformatics server. UK e-science. All hands meeting 32. Yan SL, Wang Y, Liu JC (2012) Research on the comprehensive evaluation of business intelligence system based on BP neural network. Syst Eng Procedia 4:275–281

A General Overview of the Industry 4.0 Concept for Production Management and Engineering Héctor Cañas and Josefa Mula

Abstract This article presents an overview of the Industry 4.0 concept in the production management and engineering context with a view to identifying the most relevant aspects that comprise it. To this end, the different available definitions about it, the considered support elements, and the proposed main differences between conceptual frameworks are analyzed. Finally, some relevant contributions made to production planning in an Industry 4.0 context are covered. Keywords Industry 4.0 · Production engineering and management · Production planning

1 Introduction The fact that no clear definition exists for the term Industry 4.0 is noteworthy. Most authors refer to the idea that emerged in 2011 when the notion of reinforcing manufacturing in Germany was supported and the term Industry 4.0 was used (Kagermann et al. [13]). The Industry 4.0 concept is generally classified as the fourth industrial revolution in which the Internet of Things (IoT) and a cyberphysical system (CPS) are included in processes. CPS is characterized by three phases: the first generation involves identifying with RFID labels; the second generation is equipped with actuator sensors that perform a limited number of tasks; and the third CPS generation can store and analyze data, and is equipped with many sensors and actuators compatible with networks [14]. Wang et al. [25] stress that the Industry 4.0 objective is to integrate business processes and process engineering, and the IoT/services to operate using resources efficiently, flexibly (by adapting to market demands), and ecologically with high quality and at low cost. Bücker et al. [5] propose a reference Industry 4.0 H. Cañas · J. Mula (B) Research Centre on Production Management and Engineering (CIGIP), Universitat Politècnica de València (UPV), Calle Alarcón, 03801 Alcoy, Alicante, Spain e-mail: [email protected] H. Cañas e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_6

45

46

H. Cañas and J. Mula

framework. Among the points that form this frame, the efficiency achieved through the human–technology interaction stands out, which allows high levels of autonomy to be accomplished to make processes more efficient. Hermann et al. [8] compile the basic technologies that enable Industry 4.0 scenarios in different companies and stress the efficient use of resources through self-optimization. Qin et al. [21] state that there are two design principles: the first, interoperability, which involves different subconcepts, including digitalization, communication, standardization and flexibility, a real-time response, and personalization; the second principle is awareness of the predictive maintenance, decision-making, smart presentation, self-awareness, selfoptimization, and self-configuration subconcepts. Hermann et al. [8] propose other Industry 4.0 design subconcepts, which include interconnection, where all organization members (machines, products, and devices) are interconnected by the IoT; transparent information, decentralized decisions where all objects and people can access relevant information; and technical assistance. Cohen et al. [6] propose four other principles for Industry 4.0 design: intelligence, knowledge, information, and connectivity. The rest of the article is as follows: Sect. 2 identifies the main definitions, concepts, and frameworks that encompass Industry 4.0. Section 3 deals with some relevant works on production management in the Industry 4.0 context. Finally, the conclusions and future works lines are presented.

2 Literature Review 2.1 Definitions and Concepts A factory is one of the main components of Industry 4.0 where sensors, machines, conveyor belts, and robots are connected to automatically exchange information and acquire sufficient intelligence to predict and maintain machines. According to the Industry 4.0 concept, smart factories are integrated into three ways: vertically, horizontally, and end-to-end [18]. About smart products, Qin et al. [21] indicate that they are equipped with sensors, identifiable components, and processors with information/knowledge to provide consumers with guidelines and also feedback about using the production system. Smart products know the production background, its current status, and objective, and can instruct machines to perform a production task and order conveyor belts to transport them to their next production stage [2]. According to this smart factory concept, questions arise as to the role that humans will play in work centers, and about workers’ mentality and competences to provide different technologies with assistance. Table 1 summarizes the main Industry 4.0 principles (Table 1).

A General Overview of the Industry 4.0 Concept …

47

Table 1 Comparison of the Industry 4.0 principles Principle

Hermann et al. [8]

Decentralized decisions

x

Interconnection/Connectivity

x

Transparent information

x

Intelligence/Awareness

Qin et al. [21]

x x x

Knowledge Interoperability

Cohen et al. [6]

x x

x

2.2 Conceptual Frameworks It is necessary to develop, understand, and evaluate reference frameworks, business processes, reference models, and maturity models to implement Industry 4.0 according to an approach based on technology, the human being, and processes [19]. Bücker et al. [5] propose a reference Industry 4.0 framework and employ HTO concepts (human, technological, and organizational) to represent the framework’s organizational dimension. These authors also use the Industry 4.0 design principles (interconnection, transparent information, decentralized decisions, and technical assistance) of Hermann et al. [8]; this framework seeks efficiency through the human–technology interaction to achieve high levels of autonomy to ensure more efficient processes [4]. Qin et al. [21] use the 5C architecture proposed by Lee et al. [16] to present a reference framework that combines the level of intelligence with engineering by creating nine intelligence applications; the lowest level corresponds to a low intelligence and simple automation level, while the highest level implies a high level of intelligence and complex automation. Lu [17] proposes a reference framework for the interoperability of Industry 4.0 with four levels: operational (organizational), systematic (applicable), technical, and interoperable semantic. Wang et al. [24, 25] propose a smart factory reference framework composed of physical resources, industrial networks, information cloud, and supervising/control terminals. Lee et al. [16] suggest a reference framework for CPS systems that they call a 5C architecture formed by five levels: smart connection, converting data into information, cybernetic, cognitive, and configuration; it is worth mentioning the attributes identified for the smart factory, namely, self-awareness, self-predictability, self-comparison, self-configuration, self-maintenance, and self-organization. Abersfelder et al. [1] propose a reference framework for personalized mass production based on Industry 4.0 concepts with different layers: networks, the Internet of Services (IoS), a management system for warehouses, CPS, enterprise information systems (EIS), and a production performance system. Adeyeri et al. [3] put forward a reference framework by integrating technological agents to reconfigure a production system for Industry 4.0 and a reconfigurable manufacturing system (RMS) platform that intends to help Industry 4.0 to be put into practice and carried out. Cohen et al. [6] present a reference framework to implement an assembly system based on the principles of connectivity, information,

48

H. Cañas and J. Mula

knowledge, and intelligence of Industry 4.0; the reference framework is validated with the case study about a manufacturer of industrial refrigerators. Zheng et al. [29] propose a conceptual framework of smart Industry 4.0 production systems, and they also present scenarios with demos and conceptual frameworks for designing, machining, monitoring, controlling, and smart machine programming; a 3D industrial scanning implementation was carried out to automate quality inspections. Table 2 summarizes the main related principles and terms identified within each analyzed reference framework. Table 3 compares the proposed frameworks related to Industry 4.0. Table 2 Principles and terms related to the Industry 4.0 reference frameworks Principle

Related term

Interconnection/Connectivity

Level of connection, feedback, database, industrial network, Big Data

Transparent information Decentralized decisions Technical assistance Human factor

Operator

Technology

Automation, sensors, cybernetic level, terminals for supervising and control

Organization Intelligence/Awareness

Self-configuration, decision-making, self-optimization, self-awareness, knowledge, cognitive level, smart factory, smart transport, smart building, smart home, smart plant, smart production, smart factory, smart self-organize network, smart self-adaptable smart system (SASS), smart design, smart monitoring, smart machining, smart control, smart scheduling

Interoperability

Digitalization, real-time response, communication, standardization, personalization, vertical integration, horizontal integration, end-to-end integration, integration (things, data, services, people)

Table 3 Industry 4.0-related frameworks Principle

Adeyeri et al. [3]

Interconnection/Connectivity

x

Transparent information

Bücker et al. [5]

Lee et al. [16]

Wang et al. [25]

x

x

x

x

x

Lu [17]

Cohen et al. [6]

Zheng et al. [29]

x

x

x

Technical assistance

x

Human factor

x

Technology

Qin et al. [21]

x

Organization

x

x x

x

x

Intelligence/Awareness/Knowledge

x

x

x

Interoperability

x

x

x

x

x

x x

x

x x

A General Overview of the Industry 4.0 Concept …

49

3 Production Planning in the Industry 4.0 Context Industry 4.0 allows flexibility, mass product personalization, increased productivity, quality, and speedy production to be obtained, and facilitates the possibility of facing challenges, like an increase in tailor-made products, short lead times, and high-quality products [29]. Schuh et al. [22] propose designing a production network with optimum migration routes according to an Industry 4.0 approach which identifies the routes on which the costs of migrating from one production network to another, plus the total costs of future network scenarios, are compensated. Sokolov and Ivanov [23] develop a dynamic control structure (DCS) and a dynamic programming model of services for supply networks in the Industry 4.0 context; the DCS that these authors propose is supplemented with optimum control program and mathematical programming theories. All this is integrated into a decentralized system by dynamic object modeling; these authors stress the importance for Industry 4.0 smart factories for short-term planning and that smart factories based on collaboration with CPS will be the future form that industrial networks will take. Waschneck et al. [26] present a literature review of the various existing control methods for complex job shops according to the Industry 4.0 context. These authors identify that the exchange of ideas between job shop scheduling and Industry 4.0 is poor. Ivanov et al. [9 propose a dynamic model and algorithm to program/schedule and design the supply networks of Industry 4.0 smart factories based on Sokolov and Ivanov [23]. Ivanov et al. 10] propose a model and a solution algorithm for programming a flexible flow workshop with continuous flows and discrete assignments; running operations and the availability of machines are considered dynamic and are distributed on a rolling horizon. Guizzi et al. [7] propose a reference architecture to plan Industry 4.0 CPS production. Zhang et al. [27] deal with the flow workshop planning problem in the Industry 4.0 context. To this end, they build a reference framework by including the key technologies that prepare Industry 4.0. Klement et al. [15] develop a tool based on a metaheuristics and heuristics hybrid to support decision-making. Zhang et al. [28] deal with the state of the art of self-organizable production systems. Ivanov et al. [11 develop an optimum control model and an algorithm to plan a flow workshop based on a flexible Industry 4.0 assembly line. To do so, they consider two principles: use the results obtained when modeling decision scheduling in an OPC; follow the computations procedure based on the Hamiltonian maximum and maximization principle. Ivanov et al. 12] propose a mathematical model that reconfigures a dynamic supply network.

4 Conclusions From a literature review, the concepts, terms, and reference frameworks that compose Industry 4.0 were established. It was found that a definition of the term Industry 4.0 which encompasses its different technologies and design principle does not exist. The

50

H. Cañas and J. Mula

comparison made of the different conceptual frameworks indicated that this subject in academic and research fields is confusing. Therefore, a continuous and up-todate definition is necessary by collecting and analyzing principles and conceptual frameworks, as is the need to standardize the terms that comprise Industry 4.0. Finally, we found that more works have been published in the Industry 4.0-based production management and engineering context to program the production of flow workshops and for supply chain planning. Further research intends to extend this literature review in order to consider works on the matter that have been published in the last 6 years. Additionally, a new definition, reference framework, and analytical models oriented to industry 4.0 tactical production planning (aggregate planning, material requirement planning, master production schedule, etc.) are welcome.

References 1. Abersfelder S, Heyder A, Franke J (2015) Optimization of a servo motor manufacturing value stream by use of ‘Industrie 4.0.’ In: 2015 5th international conference on electric drives production, EDPC 2015—proceedings, pp 2–6 2. Abramovici M, Stark R (eds) (2013) Smart product engineering 3. Adeyeri MK, Mpofu K, Adenuga Olukorede T (2015) Integration of agent technology into manufacturing enterprise: a review and platform for industry 4.0. In: IEOM 2015—5th international conference on industrial engineering and operations management, proceeding. https:// doi.org/10.1109/IEOM.2015.7093910 4. Bauer W, Schlund S, Marrenbach D, Ganschar O (2014) Industrie 4.0 – Volkswirtschaftliches Potenzial für Deutschland, Berlin 5. Bücker I, Hermann M, Pentek T, Boris O (2016) Towards a methodology for Industrie 4.0 transformation. Inf Syst 1:209–221 6. Cohen Y et al (2017) Assembly system configuration through Industry 4.0 principles: the expected change in the actual paradigms. IFAC-PapersOnLine 50(1):14958–14963. http:// linkinghub.elsevier.com/retrieve/pii/S2405896317334754 7. Guizzi G, Vespoli S, Santini S (2017) On the architecture scheduling problem of Industry 4.0. In: CEUR workshop proceedings, 2010, pp 94–100 8. Hermann M, Pentek T, Otto B (2016) Design principles for Industrie 4.0 scenarios. In: Proceedings of the annual Hawaii international conference on system sciences, 2016 March, pp 3928–3937 9. Ivanov D, Dolgui A, Sokolov B, Werner F, Ivanova M (2016) A dynamic model and an algorithm for short-term supply chain scheduling in the smart factory industry 4.0. Int J Prod Res 54(2):386–402. https://doi.org/10.1080/00207543.2014.999958 10. Ivanov D, Sokolov B, Ivanova M (2016) Schedule coordination in cyber-physical supply networks Industry 4.0. IFAC-PapersOnLine 49(12):839–844. https://doi.org/10.1016/j.ifacol. 2016.07.879 11. Ivanov D, Dolgui A, Sokolov B (2017) A dynamic approach to multi-stage job shop scheduling in an Industry 4.0-based flexible assembly system. IFIP Adv Inf Commun Technol 513(March 2018):475–482. https://doi.org/10.1007/978-3-319-66923-6_56 12. Ivanov D, Dolgui A, Sokolov B, Ivanova M (2017) Optimal control representation of the mathematical programming model for supply chain dynamic reconfiguration. IFAC-PapersOnLine 50(1):4994–4999. https://doi.org/10.1016/j.ifacol.2017.08.900 13. Kagermann, H., Lukas, W.D., Wahlster, W. (2011). Industrie 4.0: Mit dem Internet der Dinge auf dem Weg zur 4. industriellen Revolution. VDI nachrichten, 13(1).

A General Overview of the Industry 4.0 Concept …

51

14. Keller M, Rosenberg M, Brettel M, Friederichsen N (2014) How virtualization, decentralization and network building change the manufacturing landscape: an Industry 4.0 perspective. Int J Mech Aerosp Ind Mechatron Manuf Eng 8(1):37–44 15. Klement N, Silva C, Gibaru O (2017) A generic decision support tool to planning and assignment problems: industrial application & Industry 4.0. Procedia Manuf 11(June):1684–1691. https://doi.org/10.1016/j.promfg.2017.07.293 16. Lee J, Bagheri B, Kao HA (2015) A cyber-physical systems architecture for Industry 4.0-based manufacturing systems. Manuf Lett 3:18–23. https://doi.org/10.1016/j.mfglet.2014.12.001 17. Lu Y (2017) Industry 4.0: a survey on technologies, applications and open research issues. J Ind Inf Integr 6:1–10. https://doi.org/10.1016/j.jii.2017.04.005 18. Lucke D, Constantinescu C, Westkämper E (2008) Smart factory—a step towards the next generation of manufacturing. In: Mitsuishi M, Ueda K, Kimura F (eds) Manufacturing systems and technologies for the new frontier. Springer, London, pp 115–118 19. Oesterreich TD, Teuteberg F (2016) Understanding the implications of digitisation and automation in the context of Industry 4.0: a triangulation approach and elements of a research agenda for the construction industry. Comput Ind 83:121–139. https://doi.org/10.1016/j.compind.2016. 09.006 20. Posada J et al (2015) Visual computing as a key enabling technology for Industrie 4.0 and industrial internet. IEEE Comput Graph Appl 35(2):26–40 21. Qin J, Liu Y, Grosvenor R (2016) A categorical framework of manufacturing for Industry 4.0 and beyond. Procedia CIRP 52:173–178. https://doi.org/10.1016/j.procir.2016.08.005 22. Schuh G, Potente T, Varandani R, Schmitz T (2014) Global footprint design based on genetic algorithms—an “industry 4.0” perspective. CIRP Ann Manuf Technol 63(1):433–436. https:// doi.org/10.1016/j.cirp.2014.03.121 23. Sokolov B, Ivanov D (2015) Integrated scheduling of material flows and information services in industry 4.0 supply networks. IFAC-PapersOnLine 28(3):1533–1538. https://doi.org/10.1016/ j.ifacol.2015.06.304 24. Wang S et al (2015) Towards smart factory for Industry 4.0: a self-organized multi-agent system with big data based feedback and coordination. Comput Netw 101:158–168 25. Wang S, Wan J, Li D, Zhang C (2016) Implementing smart factory of Industrie 4. Int J Distrib Sens Netw 2016(4):1–10 26. Waschneck B, Altenmüller T, Bauernhansl T, Kyek A (2016) Production scheduling in complex job shops from an Industrie 4.0 perspective: a review and challenges in the semiconductor industry. In: Proceedings of SamI40 workshop at I-KNOW ’16 27. Zhang J, Ding G, Zou Y, Qin S, Fu J (2017) Review of job shop scheduling research and its new perspectives under Industry 4.0. J Intell Manuf 1–22. https://doi.org/10.1007/s10845-0171350-2 28. Zhang J, Yao X, Zhou J, Jiang J, Chen X (2017) Self-organizing manufacturing: current status and prospect for Industry 4.0. In: 2017 5th international conference on enterprise systems (ES), pp 319–326. https://doi.org/10.1109/ES.2017.59 29. Zheng P, Wang H, Sang Z, Zhong RY, Liu Y, Liu C, Xu X (2018) Smart manufacturing systems for Industry 4.0: conceptual framework, scenarios, and future perspectives. https://doi.org/10. 1007/s11465-018-0499-5

Identification and Prioritization of Industry 4.0 Projects in SMEs: A Process Approach Juan Ignacio Igartua, Jaione Ganzarain, and Dorleta Ibarra

Abstract Industry 4.0 promises a new and disruptive innovative approach to businesses, their business models, and operations. Nevertheless, SMEs approach to this new paradigm shows a real need for a guided support in the implementation of Industry 4.0 projects, and the selection of initiatives and technological investment projects in an environment full of technological offer and jargon. In this context, we propose a process model to guide SMEs in the implementation and prioritization of Industry 4.0 projects. By systematically running the stages proposed in the process, companies will determine their needs, challenges, and opportunities regarding the implementation of Industry 4.0 from a vertical and horizontal integration approach of their value chain and business processes. This methodology involves the analysis of the operations, the diagnosis and evaluation of Industry 4.0 impacts, the portfolio management, and the implementation of the selected projects. This methodology helps companies evaluate impacts and make decisions based on an integral approach. This unique and integrated decision-making process has been tested in one industrial service small company with very high management satisfaction results, due to its structured decision-making approach. Keywords Industry 4.0 · Technology management · Process approach · Production operations · Production technologies · Decision-making · SMEs

J. I. Igartua (B) · J. Ganzarain · D. Ibarra Mechanical and Industrial Production Department, Faculty of Engineering, Mondragon University, Loramendi 4, 20500 Mondragón, Spain e-mail: [email protected] J. Ganzarain e-mail: [email protected] D. Ibarra e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_7

53

54

J. I. Igartua et al.

1 Introduction Among other innovation forces, there is no doubt that digital innovation is a profound force in our business environment and society. The digital innovation and transformation is not just about improved communications across stakeholders, the reconsideration of time and distance frameworks, or the challenge over the future skills of company personnel; it is about changing the very nature of business, consumption, competition, and how markets operate. Digital technologies and the related innovations are powerful, deep, and have multiple, direct and secondary impacts. Digital technologies reduce barriers to new business entries, distort products’ limits, and foster innovative activities. In this context, companies already in the market face significant pressures in their business models. The innovations that digital disruption fosters will change businesses’ core elements in the coming years, as well as the way companies manage their processes. The potential of those changes could lead to big opportunities companies should be aware of. “Digital disruption” refers to changes, both positive and threatening, and will in Grebe’s words affect industries in three ways [4]: • Customer insights, along with the capacity to reach customers more efficiently. • Operating models—how operations and processes are managed. • Business models—the way value is created, delivered, and captured. All this requires a logic guidance starting from the assessment of companies’ situation, as well as the definition of specific actions that will help to shape business Industry 4.0 transformation on a process-based approach with a horizontal and vertical perspective [10]. In this context, it is where a process-based methodology plays its role.

2 SME’s Competitiveness Through Industry 4.0 Regardless of the consideration of which is or is not included under the “Industry 4.0” paradigm, the vast majority of definitions consider that it’s nature is related to the introduction of digital technologies and other non-digital technologies (but supported in digital environments) in business environments, especially factories, that will lead them to the fourth industrial revolution [12]. This disembarkation of the digital in the factory actually affects absolutely all the links of the value chain of the company and its networks, its suppliers, customers, services, and therefore all the activities concerning manufacturing environments [13]. The activities of technology development, design, purchasing, logistics, marketing, sales, after-sales services and, of course, manufacturing itself in the midst of all of them are to a greater or lesser extent affected by the effects of this industrial paradigm.

Identification and Prioritization of Industry 4.0 …

55

2.1 Competitiveness in SMEs Globalization has reduced the competitive advantage of SMEs in traditional moderate technology industries. In this context, SMEs struggle to find an alternative that involves shifting economic activity toward a knowledge-based economic activity characterized by a global demand for innovative products. Due to the rapid development of technologies and specialized technologies related to Internet, industrial Internet of Things, and ICTs, many SMEs look at this strategy as a key one for their competitiveness in order to become dynamic companies for the future. Due to this strategic approach, the fourth industrial revolution or Industry 4.0 has grasped the interest of businesses, as well as numerous academics, pursuing research on how those technologies will impact companies’ competitiveness, and therefore their business innovation and performance. Industry 4.0 and associated technologies will both pull and push applications and technological solutions making companies more integrated, sustainable, and product interconnected [6, 7]. Thus, Industry 4.0 helps to solve challenges related to activities, key resources and overall efficiency, production, and demographic challenges, enabling continuous resource productivity and efficiency [6]. New or improved innovations of service-based activities will be developed by companies based on these technologies, thanks to their in-cloud and on-device capabilities. In order to succeed, SMEs will need new capabilities, people, and investment in a sustainable approach (both economic and environmental). Moreover, Industry 4.0 will allow companies to roadmap future products and services, based on the market future needs as well as the technological evolutions foreseen [5].

2.2 Introducing Industry 4.0 in SMEs Companies and managers remain cautious despite the promising potential of Industry 4.0, especially in SMEs. Thus, a study from Schroder [9] reflects that “around 5 per cent of companies are thoroughly networked, while a third of them are taking the first steps in that direction, or at least have plans to do so”. What seems clear is that company size plays a significant role in the adoption of these technologies [9]. Economy of scale is an approach that large companies use, which often leads to highly automated production and management of processes. On the other hand, SMEs are more based on manual and hybrid activities which give them more flexibility. They are more focused on niche markets with medium to high degrees of specialization. In this context, large companies seem to appreciate more advantages and gains from the technologies associated to Industry 4.0. Many SME managers do not have a structured approach toward the implementation of Industry 4.0. Some studies [9] indicate that “four out of ten SMEs do not have an Industry 4.0 strategy, while on the other hand two out of ten of large companies do have a strategy”. Despite these data, SMEs will have to embrace the technologies

56

J. I. Igartua et al.

in order to maintain their competitiveness, even in a constantly evolving technological environment. The biggest challenges that SMEs have to come across are the promotion of an appropriate strategy and the process of decision-making based on an “efficient cost-benefit analysis of the relevant technologies, and their approach towards data security and standards” [11]. One of the main conditions regarding the implementation of Industry 4.0 technologies in SMEs is related to both vertical and horizontal integration capabilities [1]. Vertical integration seeks to put together processes and data coming from different managerial functions and systems (engineering, operations, sales, etc.) into a unified solution; assuring the compatibility between the various systems, business processes, and data platforms of the firm. On the other hand, horizontal integration seeks the integration of several activities associated to key processes, based on the flows of value chains of materials, energy, or information [9]. As a result, SMEs when implementing Industry 4.0 technologies need to be aware of the data flows, both internal and external, that will have to be managed with customers and suppliers (horizontal integration) as well as with company management functions (vertical integration). This vertical and horizontal integration is an important challenge for SMEs, due to their limited economic resources, their low absorptive capacity, and restricted know-how. Moreover, most of the SMEs do not have their own IT department or specialized people in these technologies, which forces managers to be the ones that will assess and evaluate the nature and impact of the Industry 4.0 technologies in their business. Managers in this context are the ones involved in the analysis of business impacts, as well as the definition of the roadmap of the company regarding the implementation of technologies. These differences are for Schroder [9] the reason why “SMEs frequently encounter difficulties in selecting the right solution and complain of a lack of user transparency”. Given this scenario, our project aims to guide SMEs in the implementation of Industry 4.0 projects based on a process approach that takes into account the vertical and horizontal integrations, the client/worker/manager perspective of technology implementation, and the effort and complexity of implementation (Fig. 1).

3 A Process Approach Toward Industry 4.0 As stated in the previous section, many SMEs have a need to strategically plan their approach toward Industry 4.0. As the number and complexity of available technological alternatives grows, managers of SMEs need a methodical approach to the implementation of Industry 4.0 strategy based on efficient cost–benefit analysis of the relevant technologies, and the identification, prioritization, and development of Industry 4.0 projects. Therefore, many companies comprehend that they need a more systematic approach toward the implementation of Industry 4.0. Despite this, some managers

Identification and Prioritization of Industry 4.0 …

57

Fig. 1 Industry 4.0 implementation assessment portfolio

feel cautious and consider that finding a solution is rather difficult due to the “darkness” of technologies. In this context, this paper describes a systematic process-based approach toward the detection and prioritization of Industry 4.0 projects in SMEs. To address the vertical and horizontal integration discussed beforehand and the challenges regarding the adoption of Industry 4.0, we propose a process model as a guide for Industry 4.0 implementation. This process model is based on the technological management process-based approach [3] and the Industry 4.0 stage maturity model [2]. This approach will foster the identification, prioritization, and development of Industry 4.0 projects. The process approach developed helps managers in SMEs to make explicit, operational, and integrate Industry 4.0 issues into the “normal” operations and management activities of the firm. This way, Industry 4.0 and associated technologies may be seen as a flow through the business, not dissimilar to the way in which operational information and materials flow. The Industry 4.0 practices may now be identified, and their continuity and contribution assessed, using this business process approach. The aim of the model is to guide and instruct companies in the methodical implementation of Industry 4.0. Systematically carrying out the three proposed stages will take an SME to the “Operations analysis” of their processes from the Industry 4.0 approach (Stage 1), the “Diagnosis and evaluation of Industry 4.0 impacts” on business integration (Stage 2), and the development of an Industry 4.0 project portfolio (Stage 3), which will support a comprehensive strategy for implementing Industry 4.0. The proposed process defines four stages (Fig. 2). The proposed process is similar to the three-phase methodology for Industry 4.0 Collaborative Diversification [2], but focusing its effort in operations activities. Thus, the operations analysis toward Industry 4.0 stage is dedicated to analyze each one

58

J. I. Igartua et al.

Fig. 2 Industry 4.0 implementation: a process approach

of the activities in the manufacturing process, as well as in the identification of each one of the key parameters of the process from an integrated vies (materials, energy, water, waste, etc.). The analysis is based on the value stream mapping approach [8], which assures a systematic review of each activity in the operations of the company, as well as an integrated process view. At this stage, company’s managers and operations experts are involved to identify key aspects and variables of the activity related to the horizontal and vertical integrations to be supported by the Industry 4.0 implementation. The outcome of this phase is a detailed analysis of the variables of each activity in the process, inputs, process parameters, and outputs, as well as the links with previous and following activities. Within the diagnosis and evaluation of Industry 4.0 impacts stage, the company quantifies (scale 1–9) each one of the variables, inputs, outputs, and process controls of all the operations in the value chain their impact against three factors (quality, cost, and process under control), as well as identifies for each one of the variables, inputs, outputs, and process, and controls the technologies Industry 4.0 that better support the vertical and horizontal integration managements needed by the SME. The output of this stage is a prioritize matrix of all variables, inputs, outputs, and process controls of all the operations regarding their business-managerial importance and the Industry 4.0 technologies that could be implemented in order to foster the vertical and horizontal integrations.

Identification and Prioritization of Industry 4.0 …

59

The third stage, previous to the implementation stage, refers to the definition of an Industry 4.0 technology portfolio that analyzes from customer, manager, and workers approach the different possible Industry 4.0 value propositions that could be implemented based on the Industry 4.0 technologies identified. This roadmap will facilitate the selection of the value propositions from a business and managerial perspective (based on the vertical–horizontal approach), as well as the Industry 4.0 projects behind them. The value propositions are established for the three perspectives (customer, manager, and workers), and for each one three levels are determined (initial, defined, and advanced) increasing the value, cost, and technological implementation difficulty as the level rises. The result of this phase is a map of all the Industry 4.0 value propositions and technological projects to be implemented in the company with a vertical and horizontal integration implementation strategies. This map shows from different perspectives the diverse initiatives the company could develop.

4 Conclusions This unique and integrated decision-making process has been tested in one industrial service small company with very high management satisfaction results, due to its structured decision-making approach. Moreover, the use of management techniques like value stream mapping approach [8] and the horizontal and vertical perspectives [10] are considered to be a key in the implementation of Industry 4.0, and the establishment of cost–benefit analysis for decision-making. This ongoing project will be applied in different SMEs in the Basque country due to their urgent need to define an Industry 4.0 implementation approach and choose the projects and investment they need to consider for their short-, medium-, and long-term investment decisions. SMEs need to consider Industry 4.0 implementation in an integrated and holistic way, with clear directions and goals in an environment with limited resources. The proposed process model, to guide SMEs in the implementation and prioritization of Industry 4.0 projects, has shown in its limited application, a practical and well-structured process that suits the needs of managers, workers, and clients in an environment with a great offer of technological solutions, offering, and promises. Acknowledgements We would like to thank the Regional Government of Gipuzkoa (GFA) for their support in the development of this project, as well as for participating company.

60

J. I. Igartua et al.

References 1. Chromjakova F, Bobak R Hrusecka D (2017) Production process stability—core assumption of INDUSTRY 4.0 concept. IOP Conf Ser: Mater Sci Eng 215(1):012024–012024 2. Ganzarain J, Errasti N (2016) Three stage maturity model in SME’s towards industry 4.0. J Ind Eng Manag 9(5):1119–1128 3. Gregory MJ (1995) Technology management: a process approach. Proc Inst Mech Eng Part B: J Eng Manuf 209(5):347–356 4. Grube D, Malik AA, Bilberg A (2017) Generic challenges and automation solutions in manufacturing SMEs. In: Katalinic B (ed) 28th DAAAM international symposium on intelligent manufacturing and automation, DAAAM 2017. Danube Adria Association for Automation and Manufacturing, DAAAM, pp 1161–1169 5. Hermann C, Schmidt D, Kurle S, Thiede S (2014) Sustainability in manufacturing and factories of the future. Int J Precis Eng Manuf-Green Technol 1(4):283–292 6. Kaggermann H (2015) Change through digitization-value creation in the age of the Industry 4.0. Manag Perm Chang 23–45 7. Lasi H, Privatdozent PF, Kemper HG, Feld T, Hoffmann M (2014) Industry 4.0. Bus Inf Syst Eng 6(4):239–242 8. Meudt T, Metternich J, Abele E (2017) Value stream mapping 4.0: holistic examination of value stream and information logistics in production. CIRP Ann Manuf Technol 66(1):413–416 9. Schroder Ch (2016) The challenges of Industry 4.0 for small and medium-sized enterprises 10. Sommer L (2015) Industrial revolution—Industry 4.0: are German manufacturing SMEs the first victims of this revolution? J Ind Eng Manag 8(5):1512–1532 11. Wang S, Wan J, Li D et al (2016) Implementing smart factory of Industrie 4.0: an outlook. Int J Distrib Sens Netw 12. Zezulka F, Marcon P, Vesely I et al (2016) Industry 4.0—an introduction in the phenomenon. IFAC-PapersOnLine 49(25):8–12 13. Zuehlke D (2010) SmartFactory-towards a factory-of-things. Annu Rev Control 34(1):129–138

Blockchain for Electronic Voting Purposes Ricardo Chica Cepeda and Anna Arbussà Reixach

Abstract The aim of this document is to describe the potential opportunities of implementing blockchain technology for electronic voting purposes, which may help reduce the costs of traditional voting systems (on-site, off-site), as well as the potential effects that it may have in electoral outcomes. This work also describes the initiatives adopted by some countries that have used this technology in electoral processes seeking to improve quality, confidence, transparency, and security. Keywords Blockchain · Decentralization · E-voting · Innovation technology · Internet of Things

1 Introduction The growth and development of ICT in all fields of society has produced a new mechanism that allows improving the offer of services by public institutions, in a manner that enhances direct interaction with citizens. One of the main aspects of the so-called New Economy of Technological Change is related to the Internet of Things (IoT), which has generated a great impact in some markets the last few years, mainly with the use of blockchain for cryptocurrencies. This disruptive technology created by Satoshi Nakamoto in 2008 has been notably recognized in attributes of security, reduction of transaction cost, and decentralization.

R. Chica Cepeda Departamento de Organización, Gestión Empresarial y Diseño de Producto, Doctoral School, Universitat de Girona, Campus Montilivi, Despatx: 123, 17003 Girona, Spain e-mail: [email protected] A. Arbussà Reixach (B) Grupo de Investigación Avanzada sobre Dinámica Empresarial e Impacto de las Nuevas Tecnologías en las Organizaciones, GRADIENT, Universitat de Girona, Campus Montilivi, Despatx: 308, 17003 Girona, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_8

61

62

R. Chica Cepeda and A. Arbussà Reixach

Up to today, blockchain has been adopted in a very subtle and cautious way by few countries, mainly in Europe and North America, despite the multiple benefits that can offer in economic and social aspects. Focusing on public institutions, this new technology challenges the idea of the traditional role of the state and centralized public institutions. A centralized model needs trusted intermediaries. In many cases, this leads to security problems. With blockchain centralization, a traditional role of the state is not required. The blockchain is more than an ICT innovation; it is, in fact, a new type of economic organization and a new type of governance [5]. Voting is a crucial event in any democracy where voting is used to choose representatives for public positions, to take decisions (referendum), or to reach a large-scale agreement. One of the main problems for electoral processes in places where the vote is not mandatory is the low number of participants among the potential voters. Some examples are Colombian referendum in 2016, to support or not the negotiations between the government and the guerillas’ group known as FARC, with a 62% of abstention [10], the United Kingdom Brexit election also in 2016, with more than 28% of non-participation [14], and the Catalonia 2017 referendum with a 53% abstention [3]. This document is divided into six sections. Section two briefly explains the blockchain architecture, section three describes some characteristics of Internet voting systems and comments on the benefits of upscaling them to a decentralized blockchain platform, section four comments briefly on voting paradoxes and how blockchain-based voting systems may have effects on the participation of citizens in voting, section five presents some ongoing blockchain voting initiatives, and section six concludes.

2 Blockchain Architecture Blockchain was proposed as a peer-to-peer payment system that allows cash transactions through the Internet without relying on the trust of the need for a financial institution [11]. The purpose of this technology is to eliminate all intermediaries during the information exchange through a decentralized network. The blockchain technology potentially allows individuals and communities to redesign their interactions in politics, business, and society, with an unprecedented process of disintermediation on large scale, based on automated and trustless transactions [1]. The main characteristic of the blockchain protocol is the possibility of storing copies of the information that has been added to the chain of each one of the nodes or user, and this data is upgraded on real time; in addition, none of the information blocks that have been validated could be modified (Fig. 1). As a digital platform, blockchain stores and verifies the entire history of transactions between users across the network [7]. The platform is made of a sequence of

Blockchain for Electronic Voting Purposes

63

Fig. 1 Centralized and decentralized network [2]. (Reproduced with permission from Followmyvote 2019)

blocks, each one holding a complete list of transaction records like a conventional public ledger [4]. Each block points to the immediately previous block via a reference that is essentially a hash value of the previous block, which is called the parent. This kind of system opens a new way of managing information as a distributed ledger, where the list of all transactions is shared among all users, rather than being stored on a central server.

2.1 Decentralized Protocol To be distributed, an electronic voting system needs to decentralize one of the main operations, namely, the validation of the transactions exchanging value. This problem is solved with cryptography. The idea with blockchain is to shift the user’s trust from a human-controlled central entity to a few reliable cryptography functions. Blockchain uses security methods like asymmetric cryptographic keys, which have two keys (public key and private key) to encrypt and decrypt data. The public key may be disseminated widely, while the private key is known only by the owner and accomplishes two functions: authentication, where the public key is used to verify that a holder of the paired private key, for instance, casts the vote, and encryption, whereby only the holder of the paired private key can decrypt the message encrypted with the public key [6] (Fig. 2). In the case of blockchain, when a legitimate user casts his vote, the system broadcasts a transaction to all the nodes that compromise the peer-to-peer network. Taking into consideration that a variety of users are broadcasting the transaction to the network, the nodes must agree on exactly which transaction was broadcasted and the order in which these transactions happened. This will result in a single, global ledger for the system and, at any given point, all the nodes in the peer-to-peer network have

64

R. Chica Cepeda and A. Arbussà Reixach

Fig. 2 Asymmetric encryption scheme [6] (Reproduced with permission from Ssl2buy, 2017)

Fig. 3 Blockchain working scheme [16] (Reproduced with permission from Blockgeeks.com (Rosic A, 2018))

a ledger consisting of a sequence of blocks, each containing a list of transactions, that they have reached consensus on [8] (Fig. 3).

2.2 Consensus Blockchains use distributed ledgers to record all the information. In the case of bitcoin and other cryptocurrencies, it records the amount of value that is transferred.

Blockchain for Electronic Voting Purposes

65

This approach can be extended to a different kind of information, like votes. The idea of this network is that all users or nodes should collectively agree on the ledger information, instead of an authority keeping a centralized record all the information. This protocol requires that the network maintains consensus around the information recorded on the blockchain. This mechanism impacts the security parameters of the protocol. The most common way to create consensus is the proof of work [6].

2.2.1

Proof of Work

Proof of work (PoW) is a distributed consensus mechanism for blockchain technology. In this mechanism, some users also known as miners create the blocks throughout mathematical functions and process the hash (takes input data, performs operations, and returns output data). Miners collectively verify the entire blockchain, and transactions are not considered to be complete until several new blocks have been added on top of them (Fig. 4).

Fig. 4 Level and use of blockchain implementation in countries

66

R. Chica Cepeda and A. Arbussà Reixach

3 Internet Voting Systems There are two types of Internet voting: On-site, which is conducted at controlled places, where election officials can authenticate eligible voters and the electronic infrastructure that must be used. The second type allows voters to transmit their votes from any Internet connection to which they have access using a computer or smartphone; in any case, the votes are transmitted through a network of communications, either in a centralized or decentralized protocol, from the place where it has been issued up to a remote digital urn or central server. Systems for Internet voting have been increasingly used in many countries; recent examples are Estonia in 2005 with a system based on the national ID card that is given to all Estonian citizens. These cards contain encrypted files that identify the owner and allow them to carry out several online and electronic activities including online banking services, digitally signing documents, access their information on government databases, and i-voting [13].

3.1 Requirements for Internet Voting Protocol All the voting protocols have a similar set of security requirements: privacy of the votes; the result must be totally secret until the election is completed and it must be verifiable (Table 1) [17]. Table 1 General security requirement for electronic voting protocols Security requirements

Description

Privacy

Is not revealed to anyone the way an eligible user voted

Authentication of voters

To ensure that only eligible voters can vote and only one vote per person is counted

Accuracy

Valid votes cannot be removed or manipulated. No invalid votes can be added

Secrecy of intermediate results

All results are kept secret until the election is completed

No-coercion

The system must not enable the selling of votes or the coercion of voters

Verifiability

Voters must be assured of the correct treatment of their votes and have means to irrefutably prove of any fraud

Blockchain for Electronic Voting Purposes

67

Table 2 Risks of internet voting protocols Election

Description

Authentication

– There is not a physical probe that the person voting is really the authorized voter – Possibility of stolen voter packages or identification cards – Misuse of elector’s ID card and personal information voting by others without the knowledge of the elector

Voting

– – – – – –

Validation

– Internet signal cut-off – Attacking the web application

Storage

– Hacker – Manipulation of the algorithm of the voting counting program in the server (The company that installed can decide also who win) – Replacement of the voting counting software

Decryption

– Remove or replace the cryptography parameters

Unable access to election website Network saturation Internet signal cut-off Dissociation of the instructions for user verification and voting options Phishing Malware

3.2 Risks of Internet Voting Systems During elections, there are many security parameters that can compromise the results, the software that could be altered or hacked to benefit a specific option, the validation, and some other potential risks in the final stages of the voting protocol (Table 2) [18].

3.3 Blockchain Voting Blockchains protocols for most of the cryptocurrencies are known as public blockchain, on a way that this type of network is open to the public; this condition offers a transparency parameter that increases the security, but when we talk about the use of this technology for purposes that can be regulated by a specific entity like a corporation, country, or city for electronic voting, then it will be called a private blockchain because it needs to apply some constraints like age, city, citizenship, etc. Every user with access to the voting software can see the same results, be able to audit each vote in real time, and traced it to its source without sacrificing the condition of vote’s anonymity. One of the main characteristics of blockchain technology regarding electronic voting is the aspect of transparency, which allows access to the whole history and verifies the transactions without modifying them in any way.

68

R. Chica Cepeda and A. Arbussà Reixach

The possibility to cast votes over the Internet anywhere, with laptops, PC or smartphone at any time during the established schedule, with proper registration is also a great achievement of blockchain, among the option to audit the process. Costs of elections with blockchain technology will decrease noticeably considering that it will not be necessary to implement and conditioning spaces in local common areas, as well the mobilization for both voters and juries during the electoral journey.

4 Blockchain and Voting Paradoxes Voting paradoxes typically suggest that something is wrong with the way individual opinions are being expressed or processed in voting. Voting paradoxes have an important role in the history of social choice theory [12]. The usual justification for referendum voting is to provide an answer to the question “What do people want?” When people vote on a single issue with three or more alternatives, the answer to this question may be unclear given the voting paradoxes, in particular, the Condorcet winner, but an issue has only two alternatives like in Colombian 2016 plebiscite question: “Do you support the final agreement for the end of the conflict and the construction of a stable and lasting peace” or the Catalonia 2017 Referendum: “Do you want Catalonia to be an independent state in the form of a republic?” Majority voting on a binary issue eliminates Condorcet’s paradox since the vote is decisive, and the outcome is preferred to the alternative by at least half of the voters [9]. Taking into consideration this situation, these cases share a common problem, the low participation of the citizens during the election. This factor is reflected by the lack of credibility that the citizens have in the electoral processes, and in Colombia’s case, due to the great distances that must be traveled to access the electoral precinct. Blockchain as a decentralized protocol allows remote Internet voting, offering new security levels, and compares it to traditional electronic voting systems, by counting each vote that is introduced to the chain only once, without being modified by third parties, and the consensus mechanism validates all the information among the participants, keeping anonymity as the main principle for any electoral process.

5 Current Blockchain Voting Initiatives Estonia is one of the first countries with an integrated digital government platform that offers more than 90% services online, with the use of an electronic personal identifier. In 2005, the country ran the first nation-wide election using Internet voting and is developing its platform for 2019 blockchain elections. The Western Australian is developing a blockchain system to run March 2017 election. Also, the Australian Securities Exchange (ASX) will adopt blockchain

Blockchain for Electronic Voting Purposes

69

technology for financial market to clear and settle trades; this system has been in development and testing for the past 3 years and aims to cut the cost of transactions and increase security. The Rock and Roll Hall of fame and Museum, located in Ohio, USA ran an annual election to honor musicians and artists for 2017 processed 1.8 million votes over a blockchain platform; this election covered 50 states and many countries, with votes cast by smartphones [15]. Ohio State Bar Association, USA for their 2017 president election used a blockchain platform with 20,000 users. Some other countries in Europe are developing their own blockchain systems for 2019 and 2020 elections, among Swiss, Italia, and France. Private initiatives like follow my vote from USA, Sctyl from Spain, and Smartmatic (USA, Venezuela) are also researching and developing products for electoral purposes.

6 Conclusions It is important to highlight that the use of the blockchain for cryptocurrencies’ development like bitcoin has allowed its use in other fields mainly commercial, logistics, health, and recently for e-government. In public organizations, the electoral voting process will increase the security parameters, reduce costs, and generate innovative mechanisms for auditing the system. The blockchain disruptive technology will not only create many changes in the relationship between citizens and their public entities but also in the way of managing multiple services that until today needed intermediaries. The cooperation of the public sector with private local initiatives will allow a deeper research that improves the blockchain protocols from different ambits and its use on a larger scale, making it accepted in more countries. Another aspect to consider is the lack of legislation applicable to blockchain technologies, since most of the countries are awaiting to see the results of ongoing pilot tests.

References 1. Atzori M (2017) Blockchain technology and decentralized governance: is the state still necessary? J Gov Regul 6(1):1–37. https://doi.org/10.22495/jgr_v6_i1_p5 2. Followmyvote. (2019, 2 octubre). [The Online Voting Platform of The Future - Follow My Vote] [Foto]. Recuperado 7 abril, 2020, de https://followmyvote.com/ 3. Catalan Government (2017) Referèndum d’autodeterminació DE Catalunya Resultats definitius, 41. http://estaticos.elperiodico.com/resources/pdf/4/3/1507302086634.pdf 4. Chuen DL (2015) Handbook of digital currency, vol 1 5. Davidson S, De Filippi P, Potts J (2016) Economics of blockchain. SSRN Electron J 1–23. https://doi.org/10.2139/ssrn.2744751

70

R. Chica Cepeda and A. Arbussà Reixach

6. Ssl2buy (2017) Symmetric vs. Asymmetric encryption—what are differences? Retrieved from https://www.ssl2buy.com/wiki/symmetric-vs-asymmetricencryption-what-aredifferences 7. Kakavand H, Kost De Sevres N, Chilton B (2017) The blockchain revolution: an analysis of regulation and technology related to distributed ledger technologies. SSRN Electron J. https:// doi.org/10.2139/ssrn.2849251 8. Kibin Lee JI (2016) Electronic voting service using. J Digit Forensics Secur Law 11(2) 9. Lacy D, Niou EMS (2000) A problem with referendums. J Theor Polit 12(1):5–31. https://doi. org/10.1177/0951692800012001001 10. Mundo B (s.f.) bbc.com. Obtenido de http://www.bbc.com/mundo/noticias-america-latina37539590 11. Nakamoto S (2008) Bitcoin: a peer-to-peer electronic cash system. Www.Bitcoin.Org 9. https:// doi.org/10.1007/s10838-008-9062-0 12. Nurmi H (1999) Voting Paradoxes, and how to deal with them. Springer 13. Paper P (Diciembre de 2011) Introducing electronic voting—essential considerations. Obtenido de International Idea: http://www.eods.eu/library/IDEA.Introducing-ElectronicVoting-Essential-Considerations.pdf 14. Results ER (2016) Electoral commission. Obtenido de http://www.electoralcommission.org. uk/find-information-by-subject/elections-and-referendums/past-elections-and-referendums/ eu-referendum/electorate-and-count-information 15. Votem (s.f.) Case study—Rock and Roll Hall of fame. Obtenido de https://votem.com/wpcontent/uploads/2017/04/RRHoF-Case-Study.pdf 16. Rosic A (s.f.-b) [What is Blockchain Technology? A Step-by-Step Guide For Beginners]. Retrieved 3 mai, 2018, de http://blockgeeks.com/guides/what-is-blockchain-technology 17. Communications-Electronics Security Group (2002) Comments on the Report “e-Voting Security Study”. Recuperado de https://www.scytl.com/wp-content/uploads/2013/05/Comments_ on_the_report_e-Voting_Security_Study_UK.pdf 18. National Academies of Sciences EM, Sciences DEP, Board CST, Affairs PG, Committee on Science TL, Committee on the Future of Voting: Accessible RVT (2018) Securing the Vote: Protecting American Democracy. Washington DC, USA: National Academies Press

Identification of Barriers of Entry to the European Market of Medical Devices: Study of Cases in Spanish Companies Yariza Chaveco Salabarria, Mª del Carmen Pardo Ferreira, Juan Carlos Rubio Romero, and Rosa Mayelín Guerra-Bretaña Abstract This work aims to know if there are barriers to introduce medical devices in the European market. It integrates a research project that studies the same topic from two different perspectives: developed and developing countries. It consists of an exploratory case study with Spanish companies, whose results are presented in this article. It will be complemented by a descriptive case study with Cuban companies in the sector, the latter in progress. Preliminarily, some disagreements are shown regarding technical barriers, and the internal limitations of companies are highlighted as their main difficulty in accessing this demanding market. The final results are not yet available. Keywords Entry barriers · Medical devices · Study of cases

1 Introduction Market entry barriers are used as a mechanism to restrict access to products imported into international trade. Both tariff and non-tariff barriers are used. The Agreement on Technical Barriers to Trade establishes that technical regulations, standards, and conformity assessment procedures are neither discriminatory nor create unnecessary Y. Chaveco Salabarria (B) Escuela Técnica Superior de Ingeniería Industrial, Universidad de Málaga, C/Dr. Ortiz Ramos S/N, 29071 Málaga, Spain e-mail: [email protected] M. C. Pardo Ferreira · J. C. Rubio Romero Dpto. de Economía y Administración de Empresas, Escuela Técnica Superior de Ingeniería Industrial, Universidad de Málaga, C/Dr. Ortiz Ramos S/N, 29071 Málaga, Spain e-mail: [email protected] J. C. Rubio Romero e-mail: [email protected] R. Mayelín Guerra-Bretaña Centro de Biomateriales, Universidad de la Habana, La Habana, Cuba e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_9

71

72

Y. Chaveco Salabarria et al.

obstacles to trade, recommending member states to base their measures on international standards as a means of to facilitate trade and create a predictable trade environment. In the sector of medical devices, in particular, entry to the market is determined primarily by compliance with essential safety and efficacy requirements. Ignore this premise may constitute a barrier to the commercialization of these technologies in the medium term, especially in the European Union (EU) whose conformity assessment procedure (CE marking) is included in its EU directives with highly demanding requirements [7]. These requirements are considered by many as barriers to entry, so it is convenient to analyze if they really are, since this concept is often used incorrectly [3]. Some authors point out that there is also a lack of understanding of the concept of technical barriers, since these are usually associated with the difficulties that exporters have, in the technical, financial, and/or cultural order, to meet the established requirements, especially when it comes to small- and medium-sized companies with lower technological capacity [9]. The adoption in 2016 of the new European regulation for medical devices could represent the beginning of a new stage of regional commercialization in this sector. These regulations, in addition to seeking greater harmonization within the EU, completely modify the legislative framework that contained previous directives that increased the level of demand in terms of its application, traceability, transparency, and definition of the responsibilities of all actors or stakeholders of the sector [1]. It is anticipated that European market entry requirements could increase, as well as the internal difficulties of companies (especially small and medium size) to adapt their products and processes to the new regulations. In order to know the perception of Spanish companies of medical devices about the existence and impact of barriers to entry in this sector in the EU market, an exploratory case study was carried out. This work in particular shows the preliminary results of the study, which will serve as a basis to generate hypotheses about the manifestation of these barriers in practice and propose measures to overcome them.

2 Methods We used the case study based on the methodology described by two of its main authors, Yin and Eisenhardt [6]. Cases were selected by theoretical, non-random sampling [6] that offer a greater learning opportunity [12] and that allow an analytical (not statistical) generalization of the results [11, 14]. Likewise, it started with a multiple design (of more than one case) by virtue of reinforcing its internal and external validity [14]. The number of cases included meets the recommendation of [6], which suggests multiple management of not less than four (4) cases and no more than ten (10). In addition, considering that the generalization of the conclusions will be of higher quality, the greater the number of cases investigated [2].

Identification of Barriers of Entry to the European Market …

73

The units of analysis are companies registered as manufacturers, importers, or distributors, which have experience in the sector (specifically in medical equipment and devices). The selection was also based on other criteria such as the diversity of size, the innovative and exporting effort they make as well as the technological content of their products, seeking plurality in the sample so that it can be known if there is an analogy of opinions between the companies despite their organizational differences. The methodological design of the case study is according to the model proposed by Villareal and Landeta [13], which is shown in Fig. 1. Multiple sources of evidence were used [8], and triangulation of data sources was applied [5]. Thus, several sources of information were used: the scientific literature related to the methodology used and the medical equipment sector, related to barriers to trade, sanitary regulations, regulations and directives for the European Conformity Marking (CE), including databases Web of Science, original articles and references from identified magazines, and materials published in institutional repositories such as World Trade Organization (WTO) and World Health Organization (WHO). Second is the institutional documentation of the participating companies (web pages, press releases, articles, promotional materials) as well as visits and semistructured interviews with key employees of these companies (commercial representatives, quality management, technical advisors and business directors), interviewed who are all of the superior levels and have more than 10 years of experience in the sector. Third is an empirical research that served as a reference. In this case, a specific questionnaire was adapted from a similar investigation carried out in Brazil by Grecco and Moraes in 2008 (University of Sao Paulo) [4]. It was applied to nine Spanish companies as the initial sample. This questionnaire contains five dimensions that group variables of interest related to the export of medical devices to Europe. It includes questions on the following topics: characterization of the company and its commercialized products, comparison of requirements between the Spanish and European market in terms of regulations and applicable standards (quality, risk management, and safety), identification and impact of the difficulties encountered in the process of commercialization of their products, level of investment in research–development– innovation (R+D+i) carried out by the companies, as well as determining the factors they consider decisive to overcome these barriers.

3 Results The total population of the study was 25 Spanish companies, of which 13 responded to the request for a response rate of 52%. Finally, the initial sample was formed with the six manufacturing companies that best covered the study profile (Table 1) and three other important consultants in the sector in Spain. The questionnaire adapted from Grecco Moraes was applied to these nine companies.

74

Y. Chaveco Salabarria et al.

Fig. 1 Villareal and Landeta model [13] (published with permission of the editor. Original Source: González C, Curbelo Rodríguez R, Torre-Alonso JC et al. (2010). El estudio de casos como metodología de investigación científica en dirección y economía de la empresa. Una aplicación a la internacionalización. Investigaciones Europeas de Dirección y Economía de la Empresa, 16 (3), 31–52 (figura 1). Copyright © 2010, AEDEM. All rights reserved)

Identification of Barriers of Entry to the European Market …

75

Table 1 Characterization of the manufacturing companies Company

Size (by number of employees)a

Type of medical device (MD)b

Age of export (years)

Location

A

Small

In vitro diagnostic medical devices (IVDD)

>10

Barcelona

B

Small

Medical devices (MDD)

10

Zaragoza

E

Large

MDD

>10

Madrid

F

Small

MDD

>10

Madrid

a According b According

to European Commission (Official Journal No. L124, May 2003) to current EC Directives

From the analysis of the results of the questionnaire and the interviews conducted, it was verified that there are certain contrasts in some cases. It is inferred that these contrasts could be conditioned by the differences in the type of products, the export capacity of the companies, the personal experience of the respondents, or other causes. Even when the study has not been completely completed, the following can be commented. With regard to the demands of the European market, all manufacturing companies stated that they complied with the European directives and the Spanish regulations that apply to them, as well as testing their products in accredited laboratories. 80% of them have been certified by a European notified body. The other 20% have obtained the CE mark for self-certification because their products are low risk. Regarding investment in R+D+i, most of the companies stated that they dedicate their efforts and resources to this aspect, seeking to improve their products and production processes for the customer. Only one company declared itself to be not very innovative, although it has remained in the market for more than 10 years. We will have to follow its evolution in the future with the new legislative changes announced. The capacity for innovation in this sector is extremely important for survival. The weighting of the impact of the difficulties encountered by the companies for the commercialization of their products contemplates divided criteria. 50% considered that there are technical barriers, since they classified as very or extremely important the obstacles related to regulations and technical standards, while the other 50% did not consider it important. The technical impediments faced during the commercialization in general were also assessed unevenly. The 87.5% coincided in pointing out the subject of demonstration of conformity (certifications, tests, and product analysis) within the greater

76

Y. Chaveco Salabarria et al.

range of importance. In this same range, 75% highlighted the aspects related to the certification of the quality management system, the risk analysis, and the validation of the software (in the appropriate cases), while other aspects such as access to regulations and technical standards, the adequacy of raw materials, and inputs for production and equipment and productive infrastructure were weighted in a range of low importance for these purposes. On the other hand, 100% of the cases indicated with a high degree of importance the price of the product. Other technical difficulties evaluated were also considered important. Among them, they highlighted the administrative demands implicit in the commercialization process (87.5%), as well as the guarantee of representation, distribution, marketing and technical assistance of the company, the influence of the culture, and habits of the consumers, perceived as important by 75% of the respondents. One relevant point is that all the respondents agreed that there are no tariff barriers in the EU. The majority considered that it has no importance for commercialization within the region considering that they operate in their own market. Finally, the respondents expressed some aspects that, in their opinion, can be considered significant to overcome the entry barriers to the European market of medical devices. Among these, several regulatory factors were mentioned, such as the constant updating of current legislation, financial and investment regulations to undertake innovative projects, environmental factors that take into account the global and local crisis, factors related to marketing and after-sales for that the products are adapted to the needs and expectations of the different clients, factors related to internal processes of the company, and factors related to human capital.

4 Preliminary Findings Among the Spanish companies included in the sample, small discrepancies of criteria have been found with respect to the possible entry barriers to the European market of medical devices. However, it seems to confirm that tariff barriers are non-existent or reduced and non-tariff barriers (especially technical barriers) are recognized only by the minority of respondents. The main difficulties perceived by companies are more related to their own internal limitations to demonstrate the adequacy of their products or processes to the requirements of the European market. The four main aspects most commented by the companies to reach the marketing objectives in the region are: compliance with international standards and European regulations to achieve the CE mark; innovation in the design and development of products; strong commercial networks for representation and after-sales assistance; and technically qualified personnel with knowledge of the sector. This study is part of a wider investigation. The last phase of the study is currently in progress. It will be complemented by another case study that is carried out in

Identification of Barriers of Entry to the European Market …

77

Cuban companies in the sector. So, these conclusions are not considered definitive at the level that this research is found. Acknowledgements The authors express their gratitude to Agencia Española de Cooperación Internacional (AECID), Asociación Universitaria Iberoamericana de Posgrado (AUIP) and Universidad de Málaga (UMA), for all support provided for the development of this project. Also thank the participated Spanish companies for their collaboration. Conflict of Interest This work was approved by the Málaga University Ethical committee (CEUMA) under approval number 83-2019-H. The authors declare that they have no conflict of interest, and the participants in this study were informed and consented the use of their responses.

References 1. Agencia Española de Medicamentos y Productos sanitarios (AEMPS) (2016) Jornada de actualización sobre los nuevos reglamentos europeos de productos sanitarios. www.aemps. gob.es 2. Arias M (2003) Metodologías de investigación emergentes en economía de la empresa. Papers proceedings 2003, XVII congreso nacional, XIII congreso hispano-francés AEDEM, Université Montesquieu Bordeaux IV, Bordeaux, pp 19–28 3. Carlton DW (2004) Why barriers to entry are barriers to understanding (No. w10577). National Bureau of Economic Research 4. D’elia MAG, Zouain DM (2008) Superação das barreiras técnicas ao comércio internacional pelas pequenas e médias empresas de base tecnológica-a exportação de produtos eletromédicos para a união européia. Revista de administração e inovação-rai 5(1) 5. Denzin N (1984) The research act. Prentice Hall, Englewood Cliffs, NJ 6. Eisenhardt KM (1989) Building theories from case study research. Acad Manag Rev 14(4):532– 550 7. French-Mowat E, Burnett J (2012) How are medical devices regulated in the European Union? J R Soc Med 105(suppl 1):S22–S28 8. Fong C (2002) Rol que juegan los activos intangibles en la construcción de ventaja competitiva sustentable en la PYME. Un estudio de casos con empresas de Cataluña y Jalisco, Tesis Doctoral, Universidad Autónoma de Barcelona, Barcelona 9. Garrido AE (2004) As barreiras técnicas ao comércio internacional. Instituto Nacional de Metrologia, Normalização e Qualidade Industrial, Rio de Janeiro. Accessed 31 March 2005 10. Organización Mundial del Comercio (OMC). https://www.wto.org/spanish/tratop_s/tbt_s/tbt_ info_s.htm (consultado Abril 2016) 11. Ragin C, Becker R (1992) What is a case? Exploring the foundations of social enquiry. Cambridge University Press, Cambridge 12. Stake RE (1994) Case studies. In: Denzin NKY, Lincoln YS (eds) Handbook of qualitative research. Sage Publications, Thousand Oaks, CA, pp 236–247 13. Villarreal O, Landeta J (2010) El estudio de casos como metodología de investigación científica en dirección y economía de la empresa. Una aplicación a la internacionalización. Investigaciones Europeas de Dirección y Economía de la Empresa 16(3), 31–52. ISSN: 1135-2523 14. Yin R (1994) Case study research: design and methods. Beverly Hills

Application of Combinatorial Auctions to Create a 3D Printing Market Adolfo López-Paredes, Sandra Castillo, Javier Pajares, Natalia Martín, and Ricardo del Olmo

Abstract This work presents the bases of a managed market, “Lonja”, using combinatorial auctions. This is used to purchase products made with 3D printing (additive manufacturing technologies). In this market, the organization and coordination of collaborative offers will be facilitated between the customers that will receive the bids from the manufacturers or providers of 3D printing services. This “Lonja” or market will enable customers to obtain better prices from the manufacturers. On the other hand, the manufacturers can optimize their installed production capacity, and they can reduce operating costs in each case according to the technology. Keywords Auctions · Combinatorial · 3D printing · Industry 4.0 · Simulation

1 Object Currently, the market for 3D printing products is not adequately developed. As an individual or a company is interested in obtaining a product by 3D printing, they should seek potential suppliers in repositories (e.g., www.3dhubs.com), to select between different suppliers (e.g., www.shapeways.com, www.sculpteo.com), or to request to marketplaces (e.g., www.i.materialise.com). In addition to the referred links, there are also digital printing local services that do not always have the needed technology nor are they competitive, in order to request offers. In all cases, the buyers must submit technical documentation through means that do not guarantee confidentiality as well as the intellectual property of their products. This limits the number of chances of obtaining good prices, as well as the competitiveness of the manufacturers, the advantages of the final users of additive manufacturing technologies. The market developed to provide reliable and more efficient 3D printing services has been called LONJA3D.

A. López-Paredes (B) · S. Castillo · J. Pajares · N. Martín · R. del Olmo INSISOC (UIC086), Escuela de Ingenierías Industriales, University of Valladolid, C/Paseo el Cauce 59, 47011 Valladolid, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_10

79

80

A. López-Paredes et al.

In many markets, the allocation of lots and the corresponding prices are determined as a process of sequential proposals as well as acceptances of prices by buyers and sellers. This type of process is called an “auction” Anandalingan et al. [1]. Of all the applications that have proliferated in recent years, those related to electronic purchases have focused the attention of the largest marketplaces and electronic commerce companies Bichler [2], He et al. [14], Chandrashekar et al. [7], Zaman and Grosu [26]. Furthermore, there has been a large growth of applications in different fields as radio frequency spectrum Cramton [9], Bichler and Goeree [3], optimization of project portfolios Villafáñez et al. [24], Song et al. [22], load containers on transport routes Schwind et al. [21], Lindsey and Mahmassani [17], buses’ routes Cantillon and Pesendorfer [6], Buer and Haass [5], railway transport slots, Borndörfer et al. [4], yield management Gu and Zhu [13], Marinescu et al. [18], and service management Olivares et al. [19], Fujii et al. [12]. The 3D printing allocation problem is able to consider as a combinatorial allocation problem in which a group of bundles must be allocated to various printers seeking the maximum utility. The aim of the LONJA3D is to integrate a set of functionalities and combinatorial auction algorithms that allow the final users to obtain the same advantages and services of 3D printing: joint purchases using combinatorial auctions, protection of the intellectual property of 3D designs, low prices, and shorter delivery times. The expected result should be an increase in the penetration of 3D printing technologies, as a consequence of a larger competitiveness of companies that offer 3D printing services as well as an improvement in the productivity by reducing the marketing and the operating costs.

2 Combinatorial Auctions Combinatorial auctions are a special type of “managed market” (smart market) in which people can bid not only on individual items but on combinations of them [11]. Vernon Smith asserts that these auctions are especially interesting as the participants in the exchange can be benefited from the complementarity in the bids and as the buyers as well as the sellers have financial or production constraints (see Pekeˇc and Rothkopf [20]): “This type of auction can be useful when participants’ values are complementary or when participants have production and financial constraints. However, combinatorial auctions are currently rare in practice”. Auction theory has attracted attention due to which it has practical implications and it finds out the behavioral complexity of trade. The complexity should be considered if an auction needs to be designed suitably. The advantages of auctions versus commerce are that these promote competition between buyers in one way that raises the seller’s revenue and diminishes the transaction costs, so this improves the efficiency of the market. Combinatorial auctions have been used successfully in complex markets Cramton et al. [10]; however, these have not yet been applied to mass commercialization,

Application of Combinatorial Auctions …

81

Fig. 1 Type of auctions by the number of sellers and buyers

while the allocation mechanisms require addressing NP-hard problems and the computational demands are high. Pekeˇc and Rothkopf [20] summarize the limitations that have been applied to this type of markets: “The main problems confronted in implementing these auctions are that they have computational uncertainty (i.e., there is no guarantee that the winning bids for such an auction can be found in a ‘reasonable’ amount of time when the number of bidders and items becomes larger) and that the auction is cognitively complex and can lead participants to pursue perverse bidding strategies”. Combinatorial auctions are those auctions in which bidders are able to place bids on packages rather than just individual merchandise, see Fig. 1. To figure out solutions to the problem of 3D printing allocation, one has to implement auctions in which bidders are able to place bids on combinations of items. The field of combinatorial auctions requires the contribution of economics, operational research, and experimental economics, mainly agent-based models. Auctions are decentralized mechanisms that divide the complexity of the problem. The design of an auction influences its outcome. Each auction is different in the manner information is released such as the rules to bid, whether the auction is over one good or a bundle, etc. Auction models should be designed according to the determined problem at hand. In fact, many auctions are linked to the performance. A seller will often request an auction format that maximizes the revenue from the sale of the product. Otherwise, a buyer cumulating bids from a set of potential sellers will desire an auction that minimizes the cost of the product. A combinatorial double auction mixes the mechanism of double auction with combinatorial auction to establish a proper trade between the buyers and the sellers, see Fig. 2. Combinatorial double auctions enable the buyers and the sellers to deliver bids, and are much more efficient than various one-sided auctions combined Hsieh and Liao [15]. Different complex decision problems in the real applications are tricky to solve due to computational complexity. The winner determination problem in combinatorial double auctions is one of them. The winner determination problem in combinatorial

82

A. López-Paredes et al.

Fig. 2 Double auction by the number of sellers and buyers

double auctions is cumbersome to solve from a computational perspective. Multiagent systems are able to generate through their interactions an approach in which several agents are to jointly work out a problem Hsieh and Liao [16]. Combinatorial auctions present two characteristics that influence their design: computational complexity of the winner determination and the opportunities for cooperation between competitors Pekeˇc and Rothkopf [20]. Some properties of the auctions’ mechanism should be dealt with such as the allocative efficiency. It is a recommended property of an auction. It is reached as one maximizes the total value to the winners of the units being auctioned. A combinatorial auction is demanding to get the allocative efficiency, even in a theoretical approach. Cost minimization or revenue maximization is usually the other aim for auctions. Companies carry out procurement auctions often submit cost minimization as a primary goal. In government auctions, revenue maximization is questionable, due to the difficulty to find the equilibrium bidding strategies. In addition to this, a combinatorial auction designer has to cope with a more essential obstacle: to calculate the cost minimizing or revenue maximizing allocation for a given group of bids. Another goal for auction designers is the low transaction costs. Both the bid taker and the bidders must attend the costs of participating in the auction. Auctions with lower participation costs are recommended. Delay in concluding the auction is an additional transaction cost. Thereby, high auction speed is advantageous. The issue of computational complexity can be neglected as a few items are sold. However, if combinatorial bidding is to become a normal practice, the auction design should present scalability, i.e., it should be feasible for sales of many items Pekeˇc and Rothkopf [20]. Suginouchi et al. [23] proposed a preliminary combinatorial auction model to optimize the production of a manufacturer that offers customized manufacturing of pairs of shoes using components made by 3D printing. LONJA3D market uses combinatorial auctions as collaborative mechanisms for purchasing, according to

Application of Combinatorial Auctions …

83

Olivares et al. [19], Choi and Han [8], and Wang and Song [25]. The underlying logic in implemented auctions follows the proposal of Villafáñez et al. [24].

3 Results and Conclusions 3D printing presents features that others do not show. On the basis of the costs provided by the different manufacturers for the printing of parts, LONJA3D develops at every moment with the demands of the consumers the best possible offers and allows to consolidate the order to the most efficient manufacturer for each technology. 3D printing manufacturing process shows some limitations as the limited surface of the build platform and printing time, among others. In this way, the combinatorial auctions become an adequate instrument to optimize the effective participation of the productive capacity of the manufacturers to the effective demand at each moment of time, also increasing the competition between all the producers and improving the benefits for the clients. Combinatorial auction examples are the sale of airport time slots, the auction of Federal Communications Commission’s radio spectrum licenses as well as the allocation of delivery routes. Acknowledgements This research has been partially financed by the projects: “Lonja de Impresión 3D para la Industria 4.0 y la Empresa Digital (LONJA3D)” funded by the Regional Government of Castile and Leon and the European Regional Development Fund (ERDF, FEDER) with grant VA049P17; the project ABARNET (Agent-Based Algorithms for Railway NETworks optimization) financed by the Spanish Ministry of Economy, Industry and Competitiveness with grant DPI2016-78902-P; and the project “Nuevos Modelos Computacionales en Ingeniería de Organización: Sistemas de Soporte a la decisión basados en Agentes (SSDBA)” with reference INSISOC-FUNGE/063/160011.

References 1. Anandalingan G, Day RW, Raghavan S (2005) The landscape of electronic market design. Manage Sci 51(3):316–327 2. Bichler M (2000) An experimental analysis of multi-attribute auctions. Decis Support Syst 29:249–268 3. Bichler M, Goeree J (eds) (2016) Handbook of spectrum auction design. Cambridge University Press, Cambridge 4. Borndörfer R, Klug T, Schlechte T, Fügenschuh A, Schang T, Schülldorf H (2016) The freight train routing problem for congested railway networks with mixed traffic. Transp Sci 50(2):408– 423 5. Buer T, Haass R (2016) Cooperative liner shipping network design by means of a combinatorial auction. Flex Serv Manuf J. https://www.econstor.eu/handle/10419/145380 6. Cantillon E, Pesendorfer M (2006) Auctioning bus routes: the london experience. In: Combinatorial auctions. MIT Press 7. Chandrashekar TS, Narahari Y, Rosa CH, Kulkarni DM, Tew JD, Dayama P (2007) Auctionbased mechanisms for electronic procurement. IEEE Trans Autom Sci Eng 4(3):297–321. https://doi.org/10.1109/TASE.2006.885126

84

A. López-Paredes et al.

8. Choi JH, Han I (2007) Combinatorial auction based collaborative procurement. J Comput Inf Syst 47(3):118–127 9. Cramton P (2013) Spectrum auction design. Rev Ind Organ 42(2):161–190 10. Cramton P, Shoham Y, Steinberg R (2006) Combinatorial auctions. MIT Press. ISBN 0-26203342-9 11. de Vries S, Vohra R (2003) Combinatorial auctions: a survey. Informs J Comput 15(3):284–309. ISSN 1526-5528. https://doi.org/10.1287/ijoc.15.3.284.16077 12. Fujii N, Oda J, Kaihara T, Shimmura T (2017) A combinatorial auction-based approach to staff shift scheduling in restaurant business. In: Serviceology for smart service system, Selected papers of the 3rd international conference of serviceology. Springer, pp 131–138 13. Gu Y, Zhu J (2017) Capacity allocation and revenue sharing in airline alliances: a combinatorial auction-based modeling. Math Probl Eng. Article ID 3178650. https://doi.org/10.1155/2017/ 3178650 14. He M, Jennings NR, Leung H (2003) On agent-mediated electronic commerce. IEEE Trans Knowl Data Eng 15(4):985–1003 15. Hsieh F-S, Liao C-S (2015) Schemes to reward winners in combinatorial double auctions based on optimization of surplus. Electron Commer Res Appl 14:405–417 16. Hsieh F-S, Liao, C-S (2015) Scalable multi-agent learning algorithms to determine winners in combinatorial double auctions. Appl Intell 43:308–324 17. Lindsey C, Mahmassani HS (2017) Sourcing truckload capacity in the transportation spot market: a framework for third party providers. Transp Res Part A Policy Pract 102:261–273 18. Marinescu DC, Paya A, Morrison JP (2017) A cloud reservation system for big data applications. IEEE Trans Parallel Distrib Comput 28(3):606–618. https://doi.org/10.1109/TPDS. 2016.2594783 19. Olivares M, Weintraub GY, Epstein R, Yung D (2012) Combinatorial auctions for procurement: an empirical study of the Chilean school meals auction. Manage Sci 58(8):1458–1481 20. Pekeˇc A, Rothkopf MH (2003) Combinatorial auction design. Manage Sci 49(11):1485–1503 21. Schwind M, Gujo O, Vykoukal J (2009) A combinatorial intra-enterprise exchange for logistics services. IseB 7(4):447–471 22. Song W, Kang D, Zhang J, Xi H (2017) A multi-unit combinatorial auction based approach for decentralized multi-project scheduling. Auton Agent Multi-Agent Syst 31(6):1548–1577 23. Suginouchi S, Kaihara T, Kokuryo D, Kuik S (2016) A research on optimization method for integrating component selection and production scheduling under mass customization. Procedia CIRP 57:527–532 24. Villafáñez F, López-Paredes A, Pajares J, Fuente D (2014) From the RCPSP to the DRCMPSP: methodological foundations. In: Proceedings of the 2014 world congress in computer science, computer engineering, and applied computing 25. Wang S, Song H (2009) A multi-agent based combinational auction model for collaborative e-procurement. In: IEEE International Industrial Engineering and Engineering Management, https://doi.org/10.1109/IEEM.2008.4738042 26. Zaman S, Grosu D (2013) Combinatorial auction-based allocation of virtual machine instances in clouds. J Parallel Distrib Comput 73(4):495–508

SEAFRESH Project: Design and Development of an Intelligent System for Decision Support in the Chilled and Frozen Fish Sector Antonio García Lorenzo and Joaquín Romero Rivero

Abstract This paper contains the main aspects of a research project that, between its objectives, sought to redefine the logistics–energetic process of companies in the chilled and frozen fish sector, relying on an intelligent software system. This project was co-financed by the European Fund for Regional Development (ERDF) within the framework of the operational program Feder Galicia 2007–2013, call CONECTA PEME (IN852A 2013/19-0). Keywords Information systems · Genetic algorithms · Chilled and frozen fish sector

1 Introduction The SEAFRESH project—“Efficiency in storage and transportation of chilled and frozen fish”—comes as a response to the high energy consumption of sea products processing sector. Within this energy consumption, the production and management of cold for freezing and refrigeration stores occupy an important role. Although the sector has greatly improved in thermal insulation, cold production machinery, and automatic doors, control systems designs are made for static methods, that they mostly do not have intelligence to analyze the information and get to act early and automatically based on the results obtained. Thus, this paper focuses on the main aspects of the new proposed logistics– energetic process, and it relies on an intelligent decision system. This action falls within the addressed within this ambitious project, and it involved an engineering A. García Lorenzo (B) Departamento de Organización de Empresas y Marketing, Escuela de Ingeniería Industrial - Sede Campus, Universidad de Vigo, C/Maxwell S/N, 36310 Vigo (Pontevedra), Spain e-mail: [email protected] J. Romero Rivero Imatia Innovation, S.L. Edificio CITEXVI, Fonte das Abelleiras S/N, 36310 Vigo (Pontevedra), Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_11

85

86

A. García Lorenzo and J. Romero Rivero

expert in energy efficiency, a software company, a firm in the sector by way of demonstrator, and two research organizations.

2 Background The process of the demonstrator company, in the general case, begins with the acquisition of the fish in a fish market. From there, the merchandise is transported to the plant by road. Upon arrival, it is frozen by immersion (brine) and, after leaving the pre-cooling and dried tunnels, it is palletized and transferred using forklifts to cold stores, where it is stored until its later picking and dispatch. In this sense, as it is known, there are different methods of picking on the basis of the means used, and the complexity of this varies considerably according to the different cases [1]. At the time of locating and displacing, the transportation of the goods to location often means the largest percentage of time spent [5] and for this reason also conditions the layout of the warehouse, as well as the storage system to use [1]. Once decided on the layout of the warehouse, there are also in turn different alternatives, methods, or algorithms to perform the picking [3]. Essentially, these algorithms are based on optimization methods or heuristics, being these last ones mainly used. Strategies can become more complicated if, for example, they consider multiple cross aisles or order batching with two cross aisles [2, 4]. After this introduction, note that the problem of the demonstrator company, as well as the sector is generally and significantly more simple. The company works with batches and has a high turnover of them. This takes it to locate these batches in a grouped way and in a compact storage system. Due to which it has sufficient capacity and high turnover mentioned, there is no risk that the FIFO (First In, First Out) is not fulfilled. In addition, picking involves also the extraction of much of each batch, not isolated pallets. Therefore, the algorithm, from the logistical point of view, carries implied rules of positioning that locate the set of free holes nearer from the starting point and that allows to place a full batch. The algorithm also includes compact storage restrictions; in other words, the holes must be filled, from the end to the beginning in each row and from the bottom to the top in each position. In addition to these logistical aspects, obviously, energy type ones were taken into account. Of these last, just to mention that it was decided to divide into two, the equations govern the balance of all the variables of energy, considering, on the one hand, the one corresponding to the process in the brine and, on the other hand, the cold stores one.

SEAFRESH Project: Design and Development …

87

3 Design of the System Once the equations that link the different logistics and energy variables were developed and concreted, it was established as a criterion that the last purpose of decision system was to ensure a thorn of product temperature below −18 °C at any time. All of these, at the same time, this was carried out with the lower energy consumption, whereas to this end the entire system, the times of entry of goods with their characteristics, as well as internal demand in each cold store according to the pallets entered. On these premises, the development of a control algorithm took place to assist decision-making on the storage of goods and switched the cold stores on for the optimization of energy consumption and maintenance of temperature of the fish within established margins. So, on the one hand, the algorithm inputs were defined as: • Temporary periods: Start time and end time of the temporary processes to simulate. • Brine: For its characterization, the following variables were contemplated: – Initial temperature of the brine: It allows the algorithm to calculate the time of switching the brine on, so that it reaches the target temperature of −19 °C and the resulting energy consumption. – Speed of processing, expressed in kg/h: Time that merchandise is in brine, and the output temperature of the goods prior to packaging and palletizing can be estimated by the processing speed. • Merchandise: The incoming to cooling load is characterized by: – Total weight in kg: Algorithm separates this load in a set of pallets with 900 kg of fish each (configurable). – Species and size of the incoming goods: According to these values, it is possible to estimate the brine output temperature of merchandise. – Check in time: It allows the algorithm to calculate different temporal events derived from the simulation process (brine switched on time prior to entry of the merchandise, switched on/off cycles of cold stores, etc.). • Cold stores: For each of them, it is defined as follows: – Capacity and organization: Organization of storage space is available in each of the cold stores, indicating their number of rows, capacity in number of pallets for each row, and the different heights on which these pallets can be stacked. Each row and hole available within the cold store is assigned to a numeric value that indicates the time in seconds to place a pallet in the said space. Thus, the algorithm can rate the suitability of placing the pallets in rows nearer the door, for the reduction of operation times and, therefore, energy losses. – Current state of occupation: It is necessary to know the current state of occupation of each cold store, in particular, which rows/holes are occupied by goods and the temperature to be able to post properly within the calculation energy

88

A. García Lorenzo and J. Romero Rivero

(considering that all not incoming charge with a temperature above −18 °C, also demand cooling power). – Initial temperature of each cold store. – Systems of cooling: The number of available compressors and the power in KW in each one is indicated for the correct calculation of the energy balance of the cold stores. – Charges: Intrinsic and operating charges of each cold store for the implementation of the abovementioned energy equations. On the other hand, as obtained from the algorithm, outputs are taken into account all information that helps to make decisions when deciding where are placed the goods and when turned different refrigeration systems on: • Placement of the charge: The cold stores and row will be stored, and the incoming goods grouped in pallets are indicated. • Schedule power on/off of the cooling systems: A moment list when it is necessary to turn on or off the brine and the different compressors associated with cold stores to maintain the temperature of the load in a range of desirable, minimizing the energy balance of loads, is provided. • The incoming load and cold store temperature: The simulated temporal evolution of different temperatures to take into account is presented, that is, each incoming pallet and the temperature of the cold stores according to the associated energy equation. • Total energy consumption: Consumption is shown in kWh, associated with the switched on of different compressors of cold stores. Modeling the algorithm was carried out, starting from the characterization of its inputs and outputs. As a first step, using Artificial Intelligence techniques was evaluated as part of the control algorithms to develop, so that an implementation which allowed the goal of minimize energy consumption was obtained, at a time that the temperature of the incoming goods in the desired margins was kept. In this sense, the possibility of using expert systems, neural networks, or genetic algorithms was evaluated. Finally, it was decided to make use of a genetic algorithm, mainly due to the lack of need of memorization and prior training of the algorithm (what guaranteed the use of the algorithm from the first moment and for any cooling system that was defined) and to obtain suitable results in very short periods of time (facilitating have results for decision-making before the goods arrival for processing). In this way, each of the individuals belonging to the population of the genetic algorithm consists in a temporal simulation of temperatures and energy consumption, using the system current conditions of refrigeration as initial state of the process to be simulated (brine and cold stores temperatures, arrival time, weight and species of the goods, etc.). Each of the individuals differs genetically from others in the following information: • Brine on and off schedule: It allows to predict the temperature of the goods at the outlet of the brine line.

SEAFRESH Project: Design and Development …

89

• Compressors on and off schedule: It is taken into account for the calculation of total energy consumption of the system and for temporal simulation of temperatures. • Placement of the incoming goods into cold stores: It allows to simulate the temporal processes associated with placement of the goods in the cold stores (doors opening, lighting turning on, etc.) and the loads of energy associated to these processes (operating and incoming goods loads).

4 Development of the System Once designed the algorithm, a computer system was developed. It should be noted that to run the algorithm efficiently it is needed to have as soon as possible the information on the goods going into the plant. To resolve this issue, a specific application for mobile devices that allows to transmit this information to the system was developed (Fig. 1). It was also necessary to monitor brine, cold stores, and their compressors (Fig. 2). In this way, the process was redesigned, including these aspects (Fig. 3).

Fig. 1 Mobile application of the goods progress

90

A. García Lorenzo and J. Romero Rivero

Fig. 2 Plant monitoring

Fig. 3 The redesigned process scheme

The developed system uses a genetic algorithm running 500 iterations of simulation on a population of 100 individuals. In this case, in particular, each individual is configured with the information of switching on and off of the different compressors. This configuration is that it mutates and combines in each iteration to achieve the result that manages to maintain cold stores at the target temperature, minimizing energy consumption. Operations that are executed on each individual consist of performing a simulation of the process of brine, input, and storage into the cold stores. In summary, the algorithm works predicting every minute the fish, brine and cold stores temperature, as well as the compressors energy consumption and its corresponding cost.

SEAFRESH Project: Design and Development …

91

Fig. 4 Simulation control user interface 1

Likewise, you can tell how the system selects a cold store hole to accommodate a pallet. As mentioned above, the demonstrator company works with batches and presents a high rotation of them. This takes it to locate these batches of grouped shape, following a compact storage system. Thus, the algorithm aims to place in the same empty row the pallets of the same batch, trying to minimize the opening of the door. Finally, note that all the information that the simulation algorithm is capable of extracting is presented graphically to the user (Fig. 4). From this interface, parameters of the simulation are configurable, such as the speed of the brine, the ranges of working valid for the cold store, the type of entry of the goods into the cold stores, electricity tariff, or the date of completion of the simulation. While the simulation is running, user displayed the temporal evolution of all the parameters of the individual with the best assessment so far. Finally, the system has a second display of results (Fig. 5) collecting the information where the pallets must be located with the incoming goods, when and how many fan coil units should be turned on or turned off and a summary of the energy and the cost consumed in the process. In addition, it is presented in a three-dimensional view of the cold store and the place where the new pallets should be, in such a way that the user can easily confirm their entry into the cold store.

5 Results and Conclusions Once developed the intelligent software, we proceeded to test in the demonstrator company that verifies the correct modeling and simulation of the system, noting that its predictive behavior was correct.

92

A. García Lorenzo and J. Romero Rivero

Fig. 5 Simulation control user interface 2

The analyzed results showed that the software is able to adapt to different situations that arise for cold stores of chilled and frozen fish sector. The main challenge of implementation resides in the target installation, where compressors of last generation, frequency variation devices, and regulation capacity through data connections for its execution are necessary. On the other hand, the fan coil units must have the capacity for management in the same way. Finally, a monitoring system as installed in this project would serve as vital sustenance. Lastly, system decision criteria will depend on the product to be preserved, being the energy equations similar to those referred to.

References 1. De Koster R, Le-Duc T, Roodbergen KJ (2007) Design and control of warehouse order picking: a literature review. Eur J of Oper Res 182:481–501 2. Ho YC, Su TS, Shi ZB (2008) Order-batching methods for an order-picking warehouse with two cross aisles. Com Ind Eng 55:321–347 3. Petersen CG II (1997) An evaluation of order picking routeing policies. Int J of Oper Prod Man 17(11):1098–1111 4. Roodbergen KJ, De Koster R (2001) Routing methods for warehouses with multiple cross aisles. Int J Prod Res 39(9):1865–1883 5. Tompkins JA, White JA, Bozer YA, Frazelle EH, Tanchoco JMA (2003) Facilities planning. Wiley, NJ

Operations Research

Improving Vegetables’ Quality in Small-Scale Farms Through Stakeholders’ Collaboration Ana Esteso, María del Mar Alemany, and Angel Ortiz

Abstract Small farms are responsible for 80% of the world’s agricultural production although they have difficulties to meet the market quality requirements. Corporate social responsibility (CSR) programs where modern retailers invest in empowering small farmers have been implemented obtaining an increase of the supply chain (SC) profits in cases where supply and demand are balanced. In this paper, a MILP model based on Wahyudin et al. (In: Proceedings of the international multiconference of engineers and computer scientists, Hong Kong, pp. 877–882, [1]) to select the investments to carry out by modern retailers, and the product flow through the SC in situations of supply and demand imbalance is proposed. Its objective is to find out if collaboration programs have a positive impact on SC profits when supply and demand are not balanced. This model allows for the rejection of demand and product wastes. Results show that collaboration programs positively impact on the SC profits and consumer satisfaction level when there is an imbalance between demand and supply. Keywords Agri-food supply chain · Small farm · Farmer skills · Mixed integer linear programming · Food quality

A. Esteso (B) · M. M. Alemany · A. Ortiz Research Centre on Production Management and Engineering (CIGIP), Universitat Politècnica de València, Camino de Vera S/N, 46022 Valencia, Spain e-mail: [email protected] M. M. Alemany e-mail: [email protected] A. Ortiz e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_12

95

96

A. Esteso et al.

1 Introduction Approximately, 85% of farms in the world are small farms (less than 2 ha in size) that are responsible for 80% of the world’s agricultural production [2]. In general, small farmers show weaknesses in accessing market, adopting new technology, upgrading skills in managing business, and improving the vegetables’ safety and quality [3, 4]. Simultaneously, consumers demand high-quality vegetables (HQV) for which they are willing to pay a high price. The problem arises when demand cannot be fulfilled by farmers since they are not harvesting enough HQV. To reduce the wastes produced by not selling the non-quality vegetables (NQV), small farmers can sell them at a very low price to alternative markets. However, if the quality of products were improved at small farms, the whole agri-food supply chain (AFSC) profits would increase, wastes would be reduced, and consumers’ demand would be fulfilled. To increase the capabilities of small farmers and to provide the funds to adopt new technology or machinery impacts on the vegetables quality improvement [1, 3, 5, 6], there are also models for helping AFSC operative decisions while considering the quality of products [7]. An Indonesian research group has proposed to include these activities in corporate social responsibility programs employed for empowering farmers, and consequently, increasing the vegetables’ quality [1, 3–8]. However, none of them study if the results obtained can be extrapolated to environments in which demand is not equal to supply. The objective of this paper is to fill this gap by answering the research question: Is the implementation of a collaboration program (CP) appropriate to empower farmers when the demand of vegetables is higher/lower than supply? Presumably, it is, since collaboration is useful to reduce AFSC costs, ensure product’s quality, and reach consumers’ trust while reducing uncertainty of the chain [9]. For analyzing it, an extension of the MILP model proposed by Wahyudin et al. [1] is presented and solved for three scenarios: (i) demand < supply, (ii) demand = supply, and (iii) demand > supply. The rest of the paper is structured as follows. Section 2 exposes the problem under study and the assumptions made. Section 3 formulates the MILP model. Section 4 discusses the results achieved after the model resolution. Finally, Sect. 5 draws a set of conclusions and possible future research lines.

2 Problem Description The AFSC under study is involved in the production and distribution of vegetables. The AFSC is made up of small farmers, farmer cooperatives (aggrupation of close farmers), modern retailers, and consumer markets. Since most small farmers are not able to produce HQV, farmer cooperatives (FC) classify the whole harvest into HQV and NQV. Vegetables that meet the quality requirements imposed by consumers are transported from FC to modern retailers.

Improving Vegetables’ Quality in Small-Scale Farms …

97

These products will be after transported to consumer markets to be sold to end customers at a high price. To reduce the wastes of vegetables, FCs directly sell the NQV in consumer markets at a lower price than HQV. Modern retailer’s benefits directly depend on the quantity of HQV sold in the consumer market. Since the supply of HQV is lower than demand, modern retailers establish a CP to increase the proportion of HQV to be obtained at each farm. By this program, modern retailers assign a skill level to each farmer in function of the proportion of HQV to be obtained in their farms. For example, skill level 0 corresponds to less than 70% of HQV, skill level 1 corresponds to a proportion between 70 and 80%, skill level 2 to proportions between 80 and 90%, and finally, skill level 3 to proportions between 90 and 100%. Then, retailers can invest to take a farmer to the next skill level. This improvement increases the ability of farmers to buy the latest technology, to apply latest agriculture system, and to provide other supporting utilities [1], increasing the proportion of harvest to be of high quality. It is demonstrated that this CP has a positive impact on all members of the AFSC when demand is balanced with supply [1]. However, it is unknown if these conclusions are applicable to environments in which there is an imbalance between supply and demand. To find out, we propose a MILP model based on Wahyudin et al. [1] to select the investments to carry out by modern retailers and the product flow through the SC with the following assumptions: • The quantity of vegetables to be harvested by farmers is known in advance, as well as the proportion of HQV to be obtained at each farm. • End customers demand HQV. If only NQV are available, we assume that consumers buy them at a lower price. • If demand is higher than supply, some demand will be rejected. On the contrary, if demand is lower than supply, some product will be wasted. • Initially, all farmers are at the skill level 0 of the CP. It is the proportion of HQV to obtain by each farmer if it remains in skill level 0. The improvement of such proportion with each skill level is known. • The objective of the model is to maximize the profits obtained by the whole AFSC. Economic data for each period, such as distribution costs, production costs, training costs, penalty costs, and vegetable selling price, are known. Investments in the CP cannot exceed the available budget.

3 MILP Model Formulation The model aims to determine if CP positively impacts on SC profits when supply and demand are not balanced. Since Wahyudin et al. [1] assumed that all harvest was sold, their model is extended by including harvest and demand data and by quantifying wastes and rejected demand due to supply and demand imbalances. The nomenclature used to formulate the model is presented in Table 1 where

98

A. Esteso et al.

Table 1 Model nomenclature Parameters sivt

Quantity of vegetable v produced by farmer i at period t

demvt m

Demand of vegetable v in market m at period t

vct pijm

Price of selling one unit of vegetable v of quality c from farmer i through FC j at market m at period t

pcvt

Penalty cost for overproducing/underproducing one unit of vegetable v at period t

dijijvt

Cost of distributing one unit of vegetable v from farmer i to FC j at period t

rijvt

Cost of producing one unit of vegetable v by farmer i in FC j at period t

djkjkvt

Cost of distributing one unit of vegetable v from FC j to modern retailer k at period t

htij

Cost of training the farmer i in FC j at period t for increasing one level in the CP program

djmvt jm

Cost of distributing one unit of vegetable v from FC j to consumer market m at period t

dkmvt km

Cost of distributing one unit of vegetable v from modern retailer k to consumer market m at period t

gijt

Vegetable’s worth when being produced by farmer i in FC j at period t

α

Percentage of quality improvement with each CP skill level

L

Maximum skill level of the CP

lij

Initial skill level of farmer i in FC j

CPB

Available budget for CP investments

Decision variables qijvct

Quantity of vegetables v of quality c transported to FC j facilities from farmer i at period t

vct qkijk

Quantity of vegetables v of quality c sold to modern retailer k from farmer i in FC j at period t

qmvct ijm

Quantity of vegetables v of quality c sold to consumer market m from farmer i in FC j at period t

vct Qijkm

Quantity of vegetables v of quality c coming from farmer i in FC j sold by retailer k to market m at period t

wivt

Quantity of vegetables v wasted in farmer i at period t due to overproduction

rdemvt m

Quantity of rejected demand in market m at period t due to scarcity of vegetables v

SLtij

CP program current skill level of farmer i in FC j at period t

Fijt

CP program levels improved by farmer i in FC j at period t

i refers to farmers, j to FC, k to modern retailers, m to consumer markets, v to vegetables, c to product’s quality, t to time periods, and FCi to the set of farmers i that belong to FC j. The MILP model of the addressed problem can be presented as follows: max Z =

     v

−

v

−

c

i

j∈FCi m

t

   c

i

j∈FCi k

c

i

j∈FCi k

m

· djkjkvt

−

t

c

   v

t

   v

k vct qkijk

   v

 vct vct Qijkm + qmvct ijm · cpijm −

vct Qijkm · dkmvt km −

c

j∈FCi m

i

   v

t

i

i

j∈FCi

qmvct ijm

t

· djmvt jm

t

wivt +

 m

  qijvct · dijijvt + rijvt

 rdemvt m

· pcvt

Improving Vegetables’ Quality in Small-Scale Farms … −

   i

j∈FCi

Fijt · htij

99

(1)

t

Subject to: sivt =



qijvct + wivt ∀i, v, t

(2)

c

j∈FCi

  qijvct ≤ sivt · gijt + α · SLtij ∀i, j ∈ FCi , v, c = 1, t

(3)

  qijvct ≤ sivt · 1 − gijt − α · SLtij ∀i, j ∈ FCi , v, c = 2, t

(4)



qijvct =

vct qkijk ∀i, j ∈ FCi , v, c = 1, t

(5)

k

qmvct ijm = 0 ∀i, j ∈ FCi , k, v, c = 1, t 

qijvct =

qmvct ijm ∀i, j ∈ FCi , v, c = 2, t

(6) (7)

m vct qkijk = 0 ∀i, j ∈ FCi , k, v, c = 2, t vct qkijk =



(8)

vct Qijkm ∀i, j ∈ FCi , k, v, c, t

(9)

vct vct Qijkm ≤ qkijk ∀i, j ∈ FCi , k, m, v, c, t

(10)

m

   i

j∈FCi

k

vct Qijkm +

c

 i



j∈FCi

Fijt · htij ≤ CPB

(12)

t

 gijt + α · SLtij ≤ 1 ∀i, j ∈ FCi , t

SLtij = lij +

(11)

c

j∈FCi

 i

vt vt qmvct ijm + rdemm = demm ∀m, v, t

t 

Fijt1 ∀i, j ∈ FCi , t

(13)

(14)

t1 =0

SLtij ≤ L ∀i, j ∈ FCi , t

(15)

Fijt , SLtij INTEGER, vct vct vt vt qijvct , qkijk , qmvct ijm , Qijkm , wi , rdemm CONTINUOUS

(16)

100

A. Esteso et al.

The objective of the model (1) is to maximize the profits obtained by the AFSC. The first term of the equation represents the profit obtained when selling HQV or NQV. The rest of the equation expresses the different costs of the AFSC: production and distribution costs, penalty costs (rejected demand and vegetable wastes), and investments in the CP. Constraint (2) states the vegetables’ balance at farmers. Constraints (3) and (4) establish the minimum quantity of HQV or NQV being transported from farms to FC. Constraints (5) and (6) indicate the flow to be followed by HQV, allowing them to be transported from FC to modern retailers but not allowing them to be directly transported to consumer markets. Similarly, constraints (7) and (8) establish the flow of NQV, only allowing them to be transported directly from FC to consumer markets. Constraints (9) and (10) state the vegetables’ balance at modern retailers. Constraint (11) defines the vegetables’ balance in consumer markets. Constraint (12) ensures that the total investment made for training farmers do not exceed the budget. Constraint (13) establishes that the maximum proportion of vegetables being HQV is equal to 1. Constraint (14) calculates the current skill level of a farmer at each period. Constraint (15) ensures that the skill level of each farmer does not exceed the maximum number of levels available at the CP program. Finally, constraint (16) states the definition of variables.

4 Discussion of Results The model was implemented in MPL® 5.0 and solved by using the solver GurobiTM 6.0.4. Input data and values that decision variables acquire once solved the model were stored in a Microsoft Access Database. The computer used for solving different scenarios has an Intel® Xeon® CPU E5-1620 v2(C) @ 3.70 GHz processor, with an installed capacity of 32.0 GB and a 64-bit operating system. First, the model is solved for the instance proposed by Wahyudin et al. [1] for three scenarios: (i) demand < supply, (ii) demand = supply, and (iii) demand > supply. For scenarios in which demand is higher or lower than supply, the initial demand is augmented or reduced a 40%, respectively. Results (Fig. 1) show that the same quantity of HQV is sold before and after implementing CP in all scenarios. This is because the temporal horizon is not broad enough to obtain a return of the retailers’ investments. In cases in which demand is higher than supply, some demand will be rejected. When demand is lower than supply, only NQV will be wasted. Since the previous instance considers two periods of time horizon, the model is solved for a new instance with 120 periods of time horizon. For that, data used in the previous instance is replicated for the 120 periods of time. Results presented in Fig. 2 show that the quantity of HQV sold after implementing CP increases in all scenarios, while the quantity of NQV sold decreases. When wasting vegetables, only NQV are thrown away. Demand is only rejected in the scenario in which demand is higher than supply.

Improving Vegetables’ Quality in Small-Scale Farms …

101

Fig. 1 Results for the two periods of time instance

Fig. 2 Results for the 120 periods of time instance

Total AFSC profit has increased in all scenarios for the second instance: 2% when demand is lower than supply, 2.4% when demand is equal to supply, and 30.2% when demand is higher than supply. Similarly, the quantity of HQV sold has increased by 4.1% when demand is lower than demand, and 40.7% when demand is equal to or higher than supply. Modern retailers have invested in increasing three skill levels of a farmer when demand is lower than supply and spent all the CP budget when demand is equal to or higher than supply.

5 Conclusions and Future Research Lines An extension of the MILP model proposed by Wahyudin et al. [1] is presented to prove the validity of their conclusions for cases where demand and supply are not balanced. Results show that it is profitable to invest in farmer’s empowerment in situations with an imbalance between supply and demand provided that the considered time horizon allows the return of investments made.

102

A. Esteso et al.

The model can be employed by modern retailers as a tool to select which investments to carry out depending on the increase of profit that such investments would produce. It is also useful to determine the flow of products among the AFSC actors to optimize the whole AFSC profits. The proposed model could be more extended for contemplating in a more realistic way the AFSC behavior. In real AFSCs, not all consumers are willing to buy NQV, even if it has a low price. In such cases, some demand will be rejected, while NQV will be wasted. This situation will reinforce the need of improving the skills of farmers to increase the quality of harvested products. The benefits of employing CP for that could be analyzed by extending the proposed model. Finally, several parameters could be modeled as uncertain to represent the real behavior of agri-food sector, namely, the consumer demand, economic data, the quantity of harvested vegetables, the HQV proportion to be obtained during harvest at each farm, and the improvement of such proportion with each skill level. Acknowledgements The first author acknowledges the partial support of the Programme of Formation of University Professors of the Spanish Ministry of Education, Culture, and Sport (FPU15/03595), and the partial support of Project “Development of an integrated maturity model for agility, resilience and gender perspective in supply chains (MoMARGE). Application to the agricultural sector”. Ref. GV/2017/025, funded by the Generalitat Valenciana. The other authors acknowledge the partial support of Project 691249, “RUC-APS: Enhancing and implementing Knowledge based ICT solutions within high Risk and Uncertain Conditions for Agriculture Production Systems”, funded by the EU under its funding scheme H2020-MCSA-RISE-2015.

References 1. Wahyudin RS, Hisjam M, Yuniaristanto, Kurniawan B (2015) An agri-food supply chain model for cultivating the capabilities of farmers in accessing capital using corporate social responsibility program. In: Proceedings of the international multiconference of engineers and computer scientists, Hong Kong, pp 877–882 2. Lowder SK, Skoet J, Raney T (2016) The number, size, and distribution of farms, smallholder farms, and family farms worldwide. World Dev 87:16–29 3. Sutopo W, Hisjam M, Yuniaristanto (2011) An agri-food supply chain model for cultivating the capabilities of farmers accessing market using social responsibility program. Int Sch Sci Res Innov 5(11):1588–1592 4. Sutopo W, Hisjam M, Yuniaristanto (2012) An agri-food supply chain model to enhance the business skills of small-scale farmers using corporate social responsibility. Makara J Technol 16(1):43–50 5. Sutopo W et al (2013a) A goal programming approach for assessing the financial risk of corporate social responsibility programs in agri-food supply chain network. Proc World Congr Eng 2013:732–736 6. Sutopo W, Hisjam M, Yuniaristanto (2013b) An agri-food supply chain model to empower farmers for supplying deteriorated product to modern retailer. In: IAENG transactions on engineering technologies: special issue of the international multiconference of engineers and computer scientists 2012. Springer Netherlands, Dordrecht, 189–202 7. Grillo H, Alemany MME, Ortiz A, Fuertes-Miquel VS (2017) Mathematical modelling of the order-promising process for fruit supply chains considering the perishability and subtypes of products. Appl Math Model 49:255–278

Improving Vegetables’ Quality in Small-Scale Farms …

103

8. Sutopo W, Hisjam M, Yuniaristanto (2013c) Developing an agri-food supply chain application for determining the priority of CSR program to empower farmers as a qualified supplier of modern retailer. In: 2013 World Congress on Engineering and Computer Science, pp 1180–1184 9. Esteso A, Alemany MME, Ortiz A (2017) Conceptual framework for managing uncertainty in a collaborative agri-food supply chain context. Working conference on virtual enterprises. Springer, Cham, pp 715–724

Assignment of Volunteers in a Sports Event: Case Restricted Fitness by Cut-off Mark Joaquín Bautista, Manuel Mateo, and Rocío de la Torre

Abstract This paper presents a mixed integer linear programming model (MILP) for optimizing the volunteers’ assignment to types of tasks required by the organizers of a sports event. The model takes into account a set of volunteers and their profiles, a set of required tasks, and a set of venues of the event. The main objective is to maximize the global aptitude (Fitness) of the volunteers to perform the tasks on the venues, considering a cut-off mark as lower bound for the selection. The study case is related to the 17th FIBA Basketball World Cup that took place in Spain in 2014. By using the CPLEX solver, the model is exploited and the proposed method can solve instances with nearly 15,000 volunteers, 2148 job positions as a result of 14 types of tasks, and 6 venues, in a CPU time less than 6 s. Keywords Assignment problems · MILP · Sports event · Cut-off mark · 17th FIBA basketball world cup

1 Introduction The majority of the sports events need, for its proper development, a big staff that work and collaborate in its preparation and accomplishment. However, not all the staff has a contract and a salary. It is usual that most of them are volunteers, which means they do not receive any salary for their work. Many times, these volunteers J. Bautista (B) · M. Mateo Dpto. de Organización de Empresas. Institut Organització I Control, Escuela Técnica Superior Ingeniería Industrial, Universitat Politècnica de Catalunya, Av. Diagonal, 647, 7º, 08028 Barcelona, Spain e-mail: [email protected] M. Mateo e-mail: [email protected] R. de la Torre Dpto. de Organización de Empresas, Escuela Técnica Superior Ingeniería Industrial, Universitat Politècnica de Catalunya, Av. Diagonal, 647, 7º, 08028 Barcelona, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_13

105

106

J. Bautista et al.

develop the majority of the required tasks in the organization, and their number and profile can vary according to the type of the event [11]. The profile of these volunteers is usually very wide (for instance, some of them have got sports knowledge or others not). Moreover, they can have different abilities and experiences that can be more or less useful for the event. Thus, sometimes it is necessary to select among the volunteers which one is the best option for organizational purposes. This is the reason why the volunteers’ management is currently a hot topic [14]. So far, the main decision in the sports event management is the tournament scheduling [13], especially those related with the logistic or profit optimization problems [7]. Other authors such as Drexl and Knust [3] consider the sports event management problem as a match schedule problem. However, just a few authors refer to the sports event management as a human resources (HR) management problem [10], and none of them, under the assumption of the volunteers’ management. Therefore, the problem developed in this paper is an assignment problem [9] in the particular case of linear programming [2] and with the cut-off mark as a constraint. This problem has a lot of variants. Among them, the generalized assignment problem GAP has been studied by several authors, such as Öncan [12] or Krumke and Thielen [8]. The determination of a cut-off mark is an extended practice in HEIs [5], where the enrolment and the course assignment can depend on it. These cut-off marks can be used as a parameter in inspections [4], or even in the vehicle routing problem [6]. So, to the best of our knowledge, there are no other works that deal with the volunteers’ management as an assignment problem taking into account a cut-off mark. On the following, the information required for the assignment of the candidates is presented: personal data (age, studies, nationality, driving license, job situation, among others); abilities in specific areas (for instance, medical assistance or teamwork); expertise in the sports field; geographical proximity (the event will not always be emplaced in the same location); language knowledge (local language and English are a must); the candidate’s experience in similar events is an extra, and it has to be taken into account; and finally, it is necessary to consider the candidate’s preference for each one of the different working areas in which he/she can participate. These areas can go from security, marketing, press, and entertainment to health assistance. Considering all of these items, each of the candidates obtains a mark, which is mandatory for his/her possible selection. The organization of the rest of the paper is as follows: Sect. 2 includes a mathematical program for the problem; in Sect. 3, the case study about the 17th FIBA Basketball World Cup is presented; and finally, conclusions and references are detailed in Sect. 4 and References, respectively.

2 MILP Model for the Volunteer Assignment To formalize this purpose, a mathematical model adapted to mixed integer linear programming (MILP) is presented here.

Assignment of Volunteers in a Sports Event …

107

Parameters: I J K n j,k

N

Set of volunteers: i = 1, . . . , n(n ≡ |I |). Set of tasks: j = 1, . . . , m(m ≡ |J |). Set of venues (or matches): k = 1, . . . , l(l ≡ |K |). Maximum number of volunteers that can develop the task j ∈ J in the venue k ∈ K . It is equivalent to the number of job positions available to develop the task j ∈ J in the venue k ∈ K . Maximum number of volunteers that can participate in the event. It is equivalent to the number of job positions available in the event: N =   n j,k . j∈J k∈K

pi, j

pˆ j

ai, j

bi,k

f i, j,k

Mark (or score) obtained by the volunteer i ∈ I to develop the task j ∈ J based on his/her skills and aptitudes. It is equivalent to the aptitude or fitness of volunteer i ∈ I to develop the task j ∈ J . In this work, we suppose that the values pi, j are integers between 0 and 100: pi, j ∈ [0, 100]Z . Cut-off mark associated to the type of task j ∈ J according to the requirements of the organizers. It is equivalent to the threshold score in which any volunteer i ∈ I must accredit to be selected for a job position with type of tasks j ∈ J . In this work, we suppose that the values pˆ j are homogeneous and integer between 0 and 100: pˆ = pˆ j ∈ [0, 100]Z . Degree of satisfaction obtained by the volunteer i ∈ I when he/she develops the task j ∈ J . It also expresses the volunteer’s wish to realize a certain task. Generally, it is stated that: ai, j ∈ [0, 10]Z . In this work, we will suppose that values ai, j are binary: ai, j ∈ {0, 1}. Degree of satisfaction obtained by the volunteer i ∈ I when he/she is assigned to the venue k ∈ K . It also expresses the volunteer’s wish to be present in a venue, or match. Generally, it is stated that bi,k ∈ [0, 10]Z . In this work, we will suppose that values bi,k are binary: bi,k ∈ {0, 1}. Aptitude (Fitness) of volunteer i ∈ I to develop the type of task j ∈ J in the venue k ∈ K . It is calculated as: f i, j,k = ai, j bi,k pi, j

Variables:   F pˆ Global aptitude (Fitness) of all the volunteers to the tasks and the venues depending on a homogeneous cut-off mark p. ˆ xi, j,k Binary variable that is equal to 1 if the volunteer i ∈ I is assigned to the task   j ∈ J in the venue (or match) k ∈ K , and equal to 0 otherwise. X pˆ Number of volunteers that participate in the event depending on the cut-off mark pˆ MILP model: MILP-2 (v.1) : ai, j ∈ {0, 1}; bi,k ∈ {0, 1}; pˆ = pˆ j ∈ [0, 100]Z n  m  l    maxF pˆ = f i, j,k xi, j,k i=1 j=1 k=1

(1)

108

J. Bautista et al.

Subject to: l 

xi, j,k ≤ ai, j ∀i = 1, . . . , n∀ j = 1, . . . , m

(2)

k=1 m 

xi, j,k ≤ bi,k ∀i = 1, . . . , n∀k = 1, . . . , l

(3)

xi, j,k ≤ n j,k ∀ j = 1, . . . , m∀k = 1, . . . , l

(4)

j=1 n  i=1

pˆ j

l 

xi, j,k ≤ pi, j ∀i = 1, . . . , n∀ j = 1, . . . , m

(5)

k=1 n  m  l    X pˆ − xi, j,k = 0

(6)

i=1 j=1 k=1

xi, j,k ∈ {0, 1} ∀i = 1, . . . , n∀ j = 1, . . . , m∀k = 1, . . . , l

(7)

In the model (MILP-2 (v.1): ai, j ∈ {0, 1}; bi,k ∈ {0, 1}; pˆ = pˆ j ∈ [0, 100]Z ), the objective function (1) represents the maximization of the aptitude for all the volunteers to the tasks and venues depending on the cut-off mark p; ˆ the constraints (2) guarantee that none of the volunteers is assigned to a task that he/she does not want, while constraints (3) guarantees that none of the volunteers must go to a venue or match not preferred by him/her; the constraints (4) limit the assignment of volunteers to the available job positions, according to tasks and venues; the constraints (5) are used to rule out volunteers who do not exceed the cut-off mark p; ˆ on the other hand, equality (6) determines the number of participants in the event depending on the cut-off mark; finally, the constraints (7) impose that the decision variables (xi, j,k ) are binary. Another way to model the problem that we deal (version: MILP-2 (v.2)) consists in substitute the constraints (2) and (3) by the constraints (8). l m  

xi, j,k ≤ 1 ∀i = 1, . . . , n

(8)

j=1 k=1

where the set of constraints (8) guarantees that any volunteer will occupy at most a job position at some venue.

Assignment of Volunteers in a Sports Event …

109

3 Case Study About the 17th FIBA Basketball World Cup The computational experiment proposed is focused on analyzing the behavior of Mixed Integer Linear Programming (MILP) for problems of volunteer assignment in a sports event with restricted fitness by cut-off mark. The studied event is the 17th FIBA Basketball World Cup. Data Source The values for the parameters have been obtained by deploying stages 1–5 of the methodology used for the preparation of the FIBA-2014BWC-SPAIN database [11]. These stages are: Stage-1: First, the profiles of the volunteers are designed related to the set of tasks that are going to be developed during the event, what defines the set J . Second, the set of venues of the event is fixed, which defines the set K . Stage-2: Determine the number of job positions associated with each type of task j ∈ J and in each venue k ∈ K , which defines the values of the parameters n j,k . Stage-3: The list of volunteers is prepared with their skills to participate in the event, which defines the set I . In addition, thanks to the information provided by each volunteer through the application to access any job position at any venue, the values ai, j (volunteer task) and bi,k (volunteer venue) are recorded. Stage-4: All the volunteers (elements of set I ) are evaluated according to their aptitude to perform the tasks (elements of set J ) that the event requires. This fact is summarized in a mark, pi, j , for each volunteer i ∈ I and each type of task j ∈ J , based on the information provided by him/her in the application to be selected. Stage-5: For each volunteer i ∈ I to perform a type of task j ∈ J in the venue k ∈ K , the aptitude ( f i, j,k ), also known as fitness, is obtained from the three parametric values (ai, j , bi,k , and pi, j ) determined in stages 3 and 4. Definitively: f i, j,k = ai, j bi,k pi, j ∀i ∈ I, ∀ j ∈ I, ∀k ∈ K . Data and Model Dimensions For this work, the instances #1.0 a #1.10, taken from the FIBA-2014BWC-Spain database [1], have been solved once adapted to the model MILP-2 (v.2). Briefly, the general data are: – – – – – – –

Number of candidate volunteers: n = 14, 774 Number of type of tasks: m = 14 Number of venues: l = 6 Number of job positions: 2148 Number of variables: 1,241,016 binary and 2 real. Number of constraints: 221,695 Number of scenarios: |E| = 11(ε = 0, . . . , 10). Each scenario corresponds to a cut-off mark: pˆ ≡ pˆ j = {0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100}.

As it is explained in Mateo et al. [11], the number of job positions versus the number of applications for each type of task j is not very homogeneous. It ranges

110

J. Bautista et al.

from 143.9 applications per position in the most demanded task (6907 applications per 48 positions in task 1), followed by 28.74 applications per position in the second place (2184 per 76 in task 9) to 2.89 in the least demanded task (762 per 264 in task 13). Similarly, the number of job positions versus the number of applications for each venue k ranges from 11.98 applications per position in the most demanded venue (3907 applications per 326 positions in venue 5) to 2.19 in the least demanded venue (868 per 397 in venue 1). Results The compiled codes for the procedure involved were executed on a DELL Inspiron-13 (Intel(R) Core(TM) i7-7500U @ 2.70 GHz CPU 2.90 GHz, 16 GB of RAM, × 64 Windows 10 Pro). The characteristics of the procedure are: – Model: MILP-2 (v.2): ai, j ∈ {0, 1}; bi,k ∈ {0,  1}; pˆ = pˆ j ∈ [0, 100]Z . – Objective function for minimizing the F pˆ value. – Implementation for IBM ILOG CPLEX solver (Optimization Studio v.12.2, win× 86–64). – Maximum CPU time of 300 s allowed for solving each instance #1.ε (11 instances: #1.0 to #1.10). Table 1 shows the results of the experiment obtained by CPLEX for the model implemented (MILP-2 (v.2)). In Table 1, the column headers represent the following characteristics: ε∈E p ≡ pˆ j

Identification number of instances for scenarios ε = 0, . . . , 10. Cut-off mark applied to the selection of volunteers in the 11 scenarios

Table 1 Results for the Instances #1.ε (#1.0 to #1.10) from database FIBA-2014BWC-Spain using MILP-2 (v.2) procedure (300 s. CPU max). See details in Bautista et al. [1]       ε∈E p = pˆ j N F ∗ pˆ X pˆ X pˆ /N C PU (s) 0

0

2148

176912

2148

1.000

2.61

1

10

2148

176912

2148

1.000

4.20

2

20

2148

176912

2148

1.000

5.28

3

30

2148

176912

2148

1.000

4.55

4

40

2148

176912

2148

1.000

3.48

5

50

2148

176544

2140

0.996

3.00

6

60

2148

176274

2135

0.994

2.52

7

70

2148

162471

1930

0.899

1.91

8

80

2148

130083

1503

0.700

1.50

9

90

2148

24349

262

0.122

1.23

10

100

2148

1000

10

0.005

0.88

50

2148

141389

1702

0.792

2.86

Max

100

2148

176912

2148

1.000

5.28

Min

0

2148

1000

10

0.005

0.88

Average

Assignment of Volunteers in a Sports Event …

N   F ∗ pˆ X  pˆ  X pˆ /N C PU (s)

111

Number of job positions   available in the event. Optimal value of F pˆ depending on p. ˆ Number of selected volunteers for the event depending on p. ˆ Ratio (occupation level) of position jobs in the event depending on p. ˆ CPU time in seconds.

An analysis of Table 1 reveals the following: – Procedure MILP-2 (v.2) obtains and ensures optimal solutions in all instances (#1.0 to #1.10). The CPU time to solve the 11 instances ranges between 0.88 s and 5.28 s, and CPLEX uses a mean time of 2.86 s. – Referred to the occupied type of tasks, the range is between 10 and 2148, achieving a mean value of 1702. This implies a mean occupation level of 79.2% (considering all the scenarios). With a cut-off mark equal to or lower than 40, the occupation level is 100%. – For the 10 job positions with a cut-off mark equal to 100, these are distributed among the venues in the following way: 2 in the venue 1 for the tasks type 3 and 9; 1 in the venue 2 assigned to the type of task 14; 1 in the venue 3 assigned to the type of task 2; 3 in the venue 4 assigned to the tasks 2, 4, and 9, respectively; 2 in the venue 5 assigned to the type of task 2 and 9, and finally 1 in the venue 6 assigned to the type of task 5. – For the 262 job positions with a cut-off mark equal to 90: 37 volunteers were assigned to venue 1; 25 to venue 2; 61 to venue 3; 57 to venue 4; 52 to venue 5; and 30 to the venue 6. The type of task with less number of volunteers is 13 and the minimum number of volunteers is 1; the one with more volunteers is 2, and the maximum number of volunteers is 48; the mean number of volunteers per type of task is 18.71.

4 Conclusions We conclude that MILP is a highly competitive technique for problems of assignment of volunteers to sports events, since it can obtain optimal solutions for instances extracted from the FIBA-2014BWC-SPAIN database with nearly 15,000 volunteers and more than 2,000 job positions. Regarding future research, we will continue with two lines: Analyze the impact generated by the partial or total exclusion of volunteers, when their degrees of satisfaction adopt integer values; Design and exploit new models with time constraints (work shifts), spatial ones (capacity of the venues), and contingency (occupational and ergonomic risks). Acknowledgements This research was subsidized by the Ministry of Economy and Competitiveness of the Government of Spain through project OPTHEUS (ref. PGC2018-095080-B-I00), including European Regional Development Funds (ERDF).

112

J. Bautista et al.

References 1. Bautista J, Mateo M, De La Torre R (2018) Anexos para “Assignment of Volunteers in a sport event. Case Restricted Fitness by Cut-off mark”. https://doi.org/10.13140/RG.2.2.32569.67684 2. Dantzig GB, Thapa MN (1997) Linear programming 1: introduction. Springer, New York. https://doi.org/10.1007/b97672 3. Drexl A, Knust S (2007) Sports league scheduling: graph- and resource-based models. Omega 35:465–471. https://doi.org/10.1016/j.omega.2005.08.002 4. Duffuaa SO, Siddiqui AW (2003) Process targeting with multi-class screening and measurement error. Int J Prod Res 41(7):1373–1391. https://doi.org/10.1080/0020754021000049853 5. Hervás A, Guàrdia Olmos J, Peró Cebollero M, Capilla Llafró R, Soriano Jiménez JJ (2013) A structural equation model for analysis of factors associated with the choice of engineering degrees in a technical university. Abstr Appl Anal 2013:1–13. https://doi.org/10.1155/2013/ 368529 6. Kagaya S, Kikuchi S, Donnelly RA (1994) Use of a fuzzy theory technique for grouping of trips in the vehicle-routing and scheduling problem. Eur J Oper Res 76(1):143–154. https:// doi.org/10.1016/0377-2217(94)90012-4 7. Kendall G, Knust S, Ribeiro CC, Urrutia S (2010) Scheduling in sports: An annotated bibliography. Comput Oper Res 37:1–19. https://doi.org/10.1016/j.cor.2009.05.013 8. Krumke SO, Thielen C (2013) The generalized assignment problem with minimum quantities. Eur J Oper Res 228:46–55. https://doi.org/10.1016/j.ejor.2013.01.027 9. Kuhn HW (1955) The Hungarian method for the assignment problem. Nav Res Logist 2:83–97. https://doi.org/10.1002/nav.3800020109 10. Llovet I (2016) Diseño de una herramienta de soporte en la gestión de voluntarios en una competición deportiva global. Proyecto Final de Carrera, ETS Enginyeria Industrial Barcelona, Barcelona. http://hdl.handle.net/2117/104436 11. Mateo Doll M, Bautista Valhondo J, de la Torre Martínez R (2019) Managing volunteer assignment in a sport event. In: Ortiz Á, Andrés Romano C, Poler R, García-Sabater JP (eds) Engineering digital transformation. Lecture notes in management and industrial engineering. Springer, Cham. https://doi.org/10.1007/978-3-319-96005-0_31 12. Öncan (2007) A Survey of the generalized assignment problem and its applications. In: INFOR: information systems and operational research, vol 45, pp 123–141. https://doi.org/10.3138/ infor.45.3.123 13. Pinedo ML (2009) Planning and scheduling in manufacturing and services. Springer, New York. https://doi.org/10.1007/978-1-4419-0910-7 14. Troilo M, Bouchet A, Urban TL, Sutton WA (2016) Perception, reality, and the adoption of business analytics: Evidence from North American professional sport organizations. Omega-Int J Manag Sci 59:72–83. https://doi.org/10.1016/j.omega.2015.05.011

An MILP Model for Evaluating the Impact of Strategic Decisions on Promotions in Universities Rocío de la Torre and Manuel Mateo

Abstract European universities have been suffering several changes in their regulatory frameworks during the last decade. As a result, these universities are adopting new strategic policies with regard to their workforce planning. This paper addresses the strategic staff planning problem in public universities. Besides aspects considered in previous works, such as worker’s promotion rules, hiring and laying off, workforce diversity, as well as the available economic resources, here it is addressed the possibility of strategic decisions to reach higher rates in promotions. The problem is formulated as a mixed integer linear program. The objective is not only economic but also the achievement of a preferable staff composition. Keyword MILP · Quantitative methods · Strategic staff planning

1 Introduction Strategic capacity planning is a fundamental long-term decision in determining the required resources for an organization. These required resources vary according to the core business and activity of the organization, where the main resource in the service sector is usually people. University is a typical service organization for which the principal resource is a group of highly skilled and difficult to replace professionals. Historically, implementations of strategic capacity planning in service organizations began in early 80s. But a large majority of universities adopted strategic planning only in late 90s [9]. Regardless of the resistances of universities to start developing and implementing their strategic plans, there have been several changes in the regulatory framework of the European universities (i.e., the Bologna process). According R. de la Torre (B) INARBE Institute, Public University of Navarre, Campus de Arrosadia, 31006 Pamplona, Spain e-mail: [email protected] M. Mateo Department of Management. Institut Organització i Control. ETSEIB, Universitat Politècnica de Catalunya, Av. Diagonal 647, 7th Floor, 08028 Barcelona, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_14

113

114

R. de la Torre and M. Mateo

to Mckelvet and Holmen [11], it is this change that was the main incentive for European universities to adopt new strategic policies to survive in the new paradigm. Nevertheless, as noted by Machuca et al. [10], Roth and Menor [12], and Ernst et al. [6], there is a noticeable gap between the increasing importance of strategic planning and the actual implementation. Ugboro et al. [15] proposed a guide for the adoption of strategic planning practices in public service organizations, taking into account aspects such as the personnel organized in units (according to the expertise field) and their localization. Corominas et al. [3] depicted a mathematical model for aggregate workforce planning of a company considering workers’ learning curves as well as hiring and firing rules. However, aspects such as workers’ internal promotions and the achievement of an ideal workforce composition are not considered in these models. In the same line, Song and Huang [14] formulate a model addressing hiring and firing rules for workers; however, the optimization criteria for workforce planning are based on purely economic metrics. It is important to note that the considered staff is homogenous, i.e., all workers present the same capacity and skills. On the other hand, Ahn et al. [2] and Heimerl and Kolish [7] deal with the heterogeneity of workforce for staff planning, but at the tactical level (for a short term). Also considering workforce heterogeneity, Kim and Newbhard [8] address the question of how organizations can function better with highly skilled workers through strategic plans. However, the problem of determining the workforce planning for universities is treated by few researchers, such as Shattock [13], Agasisti et al. [1], and de la Torre et al. [5]. But only the work of de la Torre et al. [5] considers the achievement of an ideal staff composition through optimization. In their paper, however, the proportion of people from a category that reach the merits to be promoted to the upper category is considered as a given data, while in the current research paper we consider that the university can make this ratio grow by means of strategic actions (such as investing in training and research). To the best of our knowledge, this paper is the first research work in literature aiming at solving the strategic capacity planning problem in universities, while considering several specific issues such as: hiring, firing and worker’s promotion rules, workforce heterogeneity, the achievement of an ideal workforce composition (in expertise and size), the required service level, the minimum cost, and the possibility of changing the probability of promotion of the academic workforce by means of strategic investments. All of the aforementioned aspects for determining the strategic staff planning in universities are supported by the development of a mixed integer linear programming (MILP) model for dealing with the problem. The organization of the rest of the paper is as follows: Sect. 2 contains a brief description of the problem; in Sect. 3 a methodology for promotions in universities is presented; Sect. 4 includes a mathematical program for the problem 5; and finally, conclusions and references are detailed in Sects. 5 and “References”, respectively.

An MILP Model for Evaluating the Impact of Strategic Decisions …

115

2 Problem Description The strategic capacity planning in the university consists in determining the staff size and composition of the workforce in a long-term horizon (the length is indicated by T ). The main decisions regarding workforce are related to hiring, firing, and promotions. This problem is challenging for universities, because of the several additional restrictions (a part from those mentioned above) to take into account: workforce heterogeneity, promotion rules, the achievement of a preferable staff composition, all of this, considering a required service level, and a minimum cost as optimization criteria. Some of these restrictions are briefly introduced in the following. The determination of the staff composition at a strategic level is a dynamic problem, as boundary conditions and resources can vary over time. Unlike other organizations, university workforce is heterogeneous, i.e., workers usually perform in different knowledge fields and have different expertise levels. Specifically, workers are organized into units or departments (the number of units is denoted by U) considering the different knowledge fields; and for each unit workers are classified in K different categories according to their expertise level. The promotion rules and type of labor contract/salary conditions vary depending on the respective personnel category. The two main types of categories are permanent and temporary. On the one hand, the permanent categories, workers sometimes can follow two different professional career pathways: the contractual pathway (KC) and the public/tenure pathway (KP). Usually, progression in public pathway is harder than in contractual one, since although in both pathways workers can be promoted in case a spot in an upper category is available, the exam is harder in the public pathway. On the other hand, the temporary categories (KT ) are mainly composed of young researchers, with less experience and, therefore, less skilled than those in permanent categories. The contract is periodically renewed, often on a yearly basis, as long as the required academic merits are progressively satisfied. Furthermore, productivity and salary are mostly different for workers in different categories, adding even more complexity to the problem. University workers can perform different tasks; teaching, research, and technology transfer. Workforce composition could be determined mostly based on available budget and teaching demand, the latter being the core activity for the universities. However, the rest of personnel duties (i.e., research and technology transfer) should be also assigned and performed. So, giving the nature of public universities, we intend that, besides the teaching demand and costs, it is necessary to achieve a proper service level and a preferable configuration of a workforce composition that can permit the performance of the other tasks. One of the key strategic decisions to achieve this goal is the personnel promotions. For the staff, the path through categories is a long process challenged by the necessity of progressively accomplishing the required academic merits. To reach the academic merits of promotion, the university might provide mechanisms or economic resources. In case the university does not incur additional expenditures or economic

116

R. de la Torre and M. Mateo

provisions in personnel budget, and due to the nature of the ordinary tasks displayed by workers, there still exist a certain number of workers that actually can obtain the merits for promoting. However, it is possible to increase this proportion of workers by planning resources for this purpose. These resources can be related to training courses, grants, research, or dissemination activities. Hence, a dynamic and heterogeneous workforce over a considered time horizon is modeled. The problem aims to configure a preferable staff composition by considering the characteristics and promotional rules for different categories.

3 Methodology In de la Torre et al. [4], a methodology is proposed to help universities in their strategic capacity planning, with the objectives to define the long-term workforce planning, considering uncertainties, and evaluate the impact of strategic decisions, as the implementation of specific personnel policies (in this case, policies to help a bigger rate of success in promotions). The phases that make up the methodology are briefly the following: • Phase I: Problem’s characterization. According to different characteristics of the case to be dealt with (organizational structure, demand, service level, costs, uncertainty, planning horizon, and evaluation criteria), the problem should be classified. • Phase II: Model’s formulation. Next step consists of designing a mathematical optimization model, considering the most suitable variables and constraints. • Phase III: Data collection and pre-analysis. The sources of information for each type of data and process of the information are to be defined as the model needs them. • Phase IV: Model solving. The mathematical model is solved. • Phase V: Add uncertainty and refine results. If uncertainty is relevant, for any of the parameters, this final step will be necessary. In that case and later in de la Torre et al. [5], one of the parameters is the proportion of workers in a unit that can promote, as maximum, from one category to next possible one in a time period (ruskt ). With the objective of encouraging promotions, the university may introduce mechanisms to help workers in his/her daily work to gain merits for promotions (for instance, additional economic resources for scholarships abroad, for attending conferences, etc.). Such monetary amounts are considered proportional to workers’ salary (a percentage in the cost of the objective function). They will be incurred proportionally to the difference between the resultant promotional ratio, above cited and here simplified as rukt , and a predetermined proportion of workers that can promote without need of additional expenditures rukt_min (see objective function in Sect. 4). Finally, that rukt is also bounded in terms of a maximum value rukt_max (constraints 5 in Sect. 4).

An MILP Model for Evaluating the Impact of Strategic Decisions …

117

4 Model’s Formulation The mathematical model for the problem described in Sect. 2 is presented (Table 1, 2). [M I N ]z =

   ∀u,t

+

 ∀t

 (ckt ·ukt ) + vt · Aut + c f ·

∀

μt · δt + 



   +    − + δkt w− λkt · δkt ukt + ∀u,k∈K C,t

  θk · ckt · rukt − rukt_min

∀k,t

(1)

∀u,k,t

wukt · h kt + Au ≥ (1 + αut ) · Cut [∀u, t]

(2)

∀k

Table 1 Database and parameters’ description Data

Description

T ; U; K

Set of periods (t = 1,…,T), units (u = 1,…,U), categories (k = 1,…,K) respect

K T ; K P; K C

Set of temporary, permanent, and contractual categories, respectively

\Gammak+

Set of categories to which it is possible to access from the category k [∀k]

\Gammak−

Set of categories from which it is possible to access to the category k [∀k]

ckt

Cost in [mu/worker] associated to the category k in period t [∀t; ∀k]

cf

Average cost associated to the dismissal staff, in [mu/worker]

vt

Cost in [mu/hour] associated to part-time lecturers in period t [∀t]

Cut

Required teaching hours for the unit u, in period t [∀t; ∀u]

h kt

Teaching capacity in hours, associated to each worker in the category k in period t [∀t; ∀k]

L ukt

Expected personnel layoffs in the unit u, category k, in period t [∀t; ∀u; ∀k]

Bt

Planned budged of the salaries of the academic staff for the period t [∀t]

U Pkt , L Pkt

Preferable proportion of staff that belongs to the category k in the period t

\alphaut

Excess of teaching hours to have, at least, the unit u in the period t [∀t]

\lambdakt

Penalty associated to the discrepancy between the preferable and the planned composition of academic staff in the category k, in the period t [∀t]

\mu t

Penalty associated to the maximum discrepancy between the preferable and the planned composition of the academic staff, in the period t [∀t]

\thetak

Factor weight in additional expenditures associated to personnel promotions to category k

rukt_min

Proportion of workers in unit u that can promote to the category k, in period t [∀t], without incurring in additional expenditures

rukt_max

Maximum proportion of workers in unit u that can promote to the category k, in period t [∀t], incurring in additional expenditures

\Deltark

Maximum change of the promotional ratio of category k between two consecutive years\ f orallk|k+ = {∅}

118

R. de la Torre and M. Mateo

Table 2 Variables Variable

Description

wukt ∈ Z+

Number of workers for unit u, category k, and period t [∀t; ∀u; ∀k]

Aut ∈

R+

Quklt ∈ Z+ + w+ ukt ∈ Z + w− ukt ∈ Z

rukt − + δ+ kt , δkt ∈ R

δt ∈ R+

Capacity assigned to part-time lecturers in the unit u in period t [∀t; ∀u] Number of workers who access to the category l from the category k, in the unit u, in the period t [∀t; ∀u; ∀k] Number of workers who are hired from the labor market for the unit u and category k, in the period t [∀t; ∀u; ∀k] Number of fired workers (excluding the previously forecasted) in the unit u and the category k, in the period t [∀t; ∀u; ∀k] Proportion of workers in unit u that can promote, as maximum, to the category k, in period t [∀t] Positive and negative discrepancies, respectively, between the preferable and the planned composition of the staff in the category k in the period t [∀t; ∀k] Maximum discrepancy, between the preferable and the planned composition of  − the academic staff in all categories in period t (i.e., δt = maxk δ+ kt , δkt ) [∀t]

wukt = wukt−1 − L ukt +



Q uskt −

sεk−



+ − Q uklt + wukt − wukt ∀u, t; ∀kε(K C ∪ K P)

lεk+

+ wukt = wukt +



(3) Q uskt ∀u, t; ∀kεK T

(3’)

sεk−

Q uskt ≤ rukt · wuk,t−1 ∀u, t; ∀sεK |s+ = {∅}; ∀kεs+

(4)

rukt_min ≤ rukt ≤ rukt_max [∀u, k, t]

(5)

  rukt − ruk,t−1 ≤ rk [∀u, k, t]

(6)

ruk,t−1 − rukt ≤ rk ∀u, k, t

(7)

wukt ≥

L Pkt ·



− − δkt

(8)

+ ∀u, t; ∀kεK + δkt

(9)

wukt

∀k

wukt ≤ U Pkt ·





wukt

∀k + − + δkt ∀t; ∀kεK δt ≥ δkt

(10)

An MILP Model for Evaluating the Impact of Strategic Decisions …

   ∀u

119

 (ckt · wukt ) + vt · Aut

≤ Bt ∀t

(11)

∀k

+ − − + , wukt , rukt , δkt , δkt , δt ≥ 0 ∀u, t; ∀kεK wukt , Aut , Q uklt , wukt

(12)

Equation (1) presents the objective function. The aim is to minimize the costs associated the workers per each category k, unit u and time to: (i) the salaries of

    − wukt ); t (ckt · wukt ) + vt · Aut ; (ii) penalties for firing staff (c f · ∀u,t

∀k

∀u,k∈K ,t

(iii) discrepancies preferable and the planned composition in academic  between the    + − staff ( + δkt λkt · δkt + μt ·δt +ω·); and (iv) costs associated to workers’ ∀t ∀k∈K ,t    promotion\mathop θk ·ckt · rukt − rukt_min . Constraints (2) determine the min∀u,k,t

imum available capacity (teaching hours) considering the service level. Constraints (3) and (3’) balance the number of the staff members per each category, unit, and period. The number of workers to be promoted is bounded by constraints (4). In constraints (5), rukt is bounded both in terms of the maximum and minimum values it can adopt and, in constraints (6) and (7), it is bounded in terms of admissible increments or decrements over time. Constraints regarding the preferable composition of the academic staff are included in Eqs. (8) and (9). Constraints (10) calculate the maximum discrepancies within all categories and periods. Constraints (11) limit the labor cost of the academic staff within the budget per period. Constraints (12) impose that the variables are non-negative. As now rukt is a variable, in constraints (4) two variables are multiplied. To formulate a linear model, it is necessary to linearize them. Therefore, these constraints are replaced by six new sets. The promotional ratio rukt will have a discretized value vri (i = 1,…,NR), in which vr1 = rukt_min and vr N = rukt_max . Similarly, the promotional ratio wukt receives a discretized value vw j (j = 1,…,UW), in which vw1 = 0 and vwU W = U W − 1. This implies that previously the respective numbers NR and UW for possible values are defined. This also implies new variables: the binary variables yriukt that equal 1 in the case rukt = vri ; the binary variables yriukt that equal 1 in the case wukt = vw j ; and the binary variables yr wi juskt that equal 1 in the case previous variables yriukt = 1 and yw juk,t−1 = 1. About the new equations, the product between variables rukt and wukt in Eq. (4) is replaced with the product of the binary variable yr wi juskt and the parameters vri and vw j , thus yielding a linear function. The value for variable yr wi juskt is computed from variables yriukt and yw jus,t−1 . Complementary, for each unit u, category k and period t there is one and only one pair of indices i and j for which yriukt and yw just , respectively, equal to 1. Finally, the value of rukt and wukt is determined in (4e) and (4f), respectively, to be used in the rest of the model.

120

R. de la Torre and M. Mateo

5 Conclusions and Further Research This paper presents a model to determine the required economic resources for workers’ promotion according to a preferable staff composition. As a parameter in previous models becomes a variable, linearization is necessary. In future works, several computational scenarios should be evaluated for different initial and preferable workforce structures or different temporal trends in budget and demand.

References 1. Agasisti T, Arnaboldi M, Azzone G (2008) Strategic management accounting in universities: the Italian experience. High Educ 55(1):1–15 2. Ahn H-S, Righter R, Shanthikumar JG (2005) Staffing decisions for heterogeneous workers with turnover. Math Methods Oper Res 62(3):499–514 3. Corominas A, Lusa A, Olivella J (2012) A detailed workforce planning model including nonlinear dependence of capacity on the size of the staff and cash management. Eur J Oper Res 216(2):445–458 4. de la Torre R, Lusa, A, Mateo M (2014) Methodology for the strategic capacity planning in universities. Managing Complex 5. de la Torre R, Lusa A, Mateo M (2016) A MILP model for the long term academic staff size and composition planning in public universities. Omega 63:1–11 6. Ernst AT, Jiang H, Krishnamoorthy M, Sier D (2004) Staff scheduling and rostering: a review of applications, methods and models. Eur J Oper Res 153:3–27 7. Heimerl C, Kolisch R (2010) Scheduling and staffing multiple projects with a multi-skilled workforce. OR Spectrum 32:343–368 8. Kim S, Nembhard D (2010) Cross-trained staffing level with heterogeneous learning/forgetting. IEEE Trans Eng Manag 57(4):560–574 9. Llinàs-Audet X, Girotto M, Solé F (2010) University strategic management and the efficacy of the managerial tools: the case of the Spanish universities. Revista de Educación 355:33–54 10. Machuca JAD, González-Zamora MM, Aguilar-Escobar VG (2006) Service operations management research. J Operat Manag 25:585–603 11. Mckelvet M, Holmen M (2009) Learning to compete in European universities: from social institution to knowledge business. Edward Elgar Publishing Limited, Cheltenham, UK 12. Roth AV, Menor LJ (2003) Insights into service operations management: a research agenda. Prod Oper Manag 12(2);ABI/INFORM Complete, 145 13. Shattock M (2003) Managing successful universities. Society for Research in Higher Education & Open University Press, London, UK 14. Song H, Huang H-C (2008) A successive convex approximation method for multistage workforce capacity planning problem with turnover. Eur J Oper Res 188:29–48 15. Ugboro I, Obeng K, Spam O (2011) Strategic planning as an effective tool of strategic management in public sector organizations: evidence from public transit organizations. Adm Soc 43(1):87–123

An Approach to Explore Historical Construction Accident Data Using Data Mining Techniques María Martínez Rojas, Antonio Trillo Cabello, Mª del Carmen Pardo Ferreira, and Juan Carlos Rubio Romero

Abstract Construction worksites are characterized by their dynamic and complex nature, making that work safety awareness a major concern during the project life cycle. In this regard, the analysis of historical data might be useful to identify the most frequent relationship between the variables of accidents in order to help safety practitioners in the task of prioritizing preventive actions. In this work, we propose an approach that will allow to explore unknown relations, expressed as association rules, among diverse variables from a database of construction accidents’ data. These association rules may be useful for efficient safety prevention and control. Keywords Construction accident · Safety · Management · Data mining · Association rule · Causation patterns

1 Introduction Construction is accepted as the most hazardous industries [33, 30, 32]. Due to the high number of accidents registered annually by the construction sector and the serious consequences this has for workers, companies, and society, occupational safety and health (OSH) is a very important issue for all the groups involved [12]. Reducing the number of accidents implies increasing the knowledge about the causes of the accident. Accident investigation is a post-accident technique that aims M. Martínez Rojas (B) · M. C. Pardo Ferreira · J. C. Rubio Romero Dpto. de Economía y Administración de Empresas, Escuela de Ingenierías Industriales, Universidad de Málaga, C/Dr, Ortiz Ramos S/N, 29071 Málaga, Spain e-mail: [email protected] M. C. Pardo Ferreira e-mail: [email protected] J. C. Rubio Romero e-mail: [email protected] A. Trillo Cabello Universidad de Málaga, C/Dr, Ortiz Ramos S/N, 29071 Málaga, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_15

121

122

M. Martínez Rojas et al.

to obtain from this preventive failure an accurate and objective information that allows to improve the identification of existing risks and to control them adequately and adequately [5, 19, 28]. The cost of accidents at work is not negligible. Its impact produces important economic losses, without forgetting the ethical and legal considerations [14]. In this regard, it is necessary to examine the facts, circumstances, and causes of accidents in order to establish policies to reduce the number and severity of accidents that occur at a construction worksite [17]. Because no single factor provides a complete explanation of the high incidence rate in the construction industry, it would be interesting to study the combinatorial effect of multiple factors [8]. To do this, it might be beneficial to explore historical accidents from data provided by work accident reports so as to investigate its causes and preventively to learn from these situations [27]. Nowadays, new technologies are transforming all stages of the engineering and construction process, even safety management [23]. Some examples of such technologies are building information modeling [20, 25], simulation and visualization [16], GIS [3], and databases [24]. Nowadays, the ability to manage large amounts of data is a fundamental issue in a knowledge-based society. In the same way, the ability to extract knowledge from large datasets is increasingly important for organizations. For this purpose, techniques that have been widely applied in other domains, as is the case of data mining enable, are used to explore relations in large amounts of data [32]. Therefore, the main contribution of this work is to propose an approach to explore unknown relations expressed as association rules among diverse variables from a database of construction accident data that occurred in Spain building construction sites. These association rules may be useful for efficient safety prevention, control, and education. The paper is structured as follows. Section 2 outlines some related works both from the perspective of relevant factors identified by authors in the literature and proposals that apply these techniques. Section 3 details the proposed methodology and, finally, Sect. 4 presents the conclusions and guidelines for future research.

2 Related Works This section presents previous works from two different perspectives: Firstly, from the perspective of recognized factors by other authors that are related to the causes of the accident; and secondly, from the perspective of proposals that have applied data mining techniques for the management of construction safety.

An Approach to Explore Historical Construction Accident Data …

123

2.1 Variables The information system that provides the data for this study is the database of work accidents managed by the Spanish Ministry of Employment and Social Safety [26]. The notification of accidents in Spain respects the classification schemes for coding that are harmonized at the European level, following the ESAW methodology [13]. This system provides very relevant information to evaluate and develop policies in the field of health and safety at work both at state level and at territorial and sectoral level. In the literature, we can find a significant number of studies that have analyzed the characteristics and sequences of detailed accident compilations, evaluating various factors which contribute to the high rate of accidents in the construction sector. The following are some examples of relevant variables identified by the authors. A study of accidents in the construction sector in Spain between 2003 and 2008 [2] showed that the severity of accidents is related to variables such as age or size of the company. Similarly, an analysis of accidents at construction sites and their severity [6] found that variables such as age, type of contract, or time of accident influenced the severity of the accident. Otherwise, conditions in the workplace have a very high prevalence in the most frequent causes of accidents in construction, especially in incidents involving falls from height [9]. In addition, an investigation concerning accidents which occurred in the construction industry in the United States between 2002 and 2011 [10] revealed that the workers’ characteristics, the conditions of the workplace, and the work environment are influential factors in the cause of the accident. An analysis of people injured in work accidents in the construction industry [18] concluded that people working on fixed-term contracts suffer accidents more frequently. In a study on the occurrence of accidents in public and private projects in the construction industry [8], factors and rules of occurrence were identified and their results showed that the combination of these factors can be used as an important reference to prevent such accidents from occurring. Otherwise, the results found in an investigation to identify the most frequent causes of accidents in the construction sector [7] indicated that there are significant associations between the type of cause identified and the various stages of the construction process. Initially, we considered the complete set of ESAW variables set by the Spanish official information system for notification of work accidents. Finally, we will select a smaller number of factors using criteria such as the preliminary results of previous projects on this subject and our previous experience.

124

M. Martínez Rojas et al.

2.2 Data Mining Nowadays, tools are needed to allow identification of valid and useful patterns due to the fact of increasing use of databases to store information [22]. For this purpose, data mining techniques present a significant potential to address the problem of transforming implicit knowledge in data into explicit knowledge [21]. These techniques are varied according to the objective to be achieved, as, for example: to discover groups of data (e.g., cluster analysis) or to discover dependencies among data (e.g., association rule mining). In the literature, several proposals can be found applying data mining techniques to address diverse issues of the construction domain. For example, construction productivity and schedule analysis [31], document identification [1], and project performance prediction [11]. Regarding safety management, Rivas et al. [29] evaluate diverse data mining techniques (such as Bayesian networks, decision rules, classification trees, logistic regression, and support vector machines) with the aim of reducing workplace accident rates. Similarly, Fan and Li [15] present an approach to reduce construction accidents by means of retrieving similar cases from an electronic case repository of construction accidents. Tixier et al. [32] develop an approach to extract valuable and actionable information from injury reports in an unsupervised manner. Finally, Shin et al. [30] explore the correlations among diverse attributes of construction accidents, which were not discovered using existing descriptive statistics or multidimensional data analysis. These recent proposals highlight the wide acceptance of these techniques to support the decision-making process of diverse important task in the construction domain, concretely, the safety management.

3 Methodology This section presents the proposed methodology. To do this, firstly, we detail the preprocessing of data and, secondly, we provide more information about the selected data mining technique: the association rule mining. Before applying data mining techniques, it is necessary to obtain the set of data regarding occupational accidents in the construction sector from the period 2003–2015 from the annual digital database on accidents of the Spanish Ministry of Employment and Social Safety. This database contains the accidents of all sectors and 58 diverse variables, so the first stage is devoted to filter the necessary data and variables according to the analysis presented in Sect. 2.1. Figure 1 illustrates the different stages of this methodology with the aim of extract the needed data from the mentioned database. Table 1 details the 17 selected variables associated with the construction accidents where they are described and coded in order to facilitate the association rule process.

An Approach to Explore Historical Construction Accident Data …

125

Fig. 1 Stages of the proposed methodology

Table 1 Selected variables Worker’s personal data

Sex Age Seniority in the job Type of contract

Workplace’s data

Code of main economic activity Main company or subcontracted company Temporary employment agency The form of prevention organization adopted by the company

Data of accident scene and workplace

Code of main economic activity of workplace

Data of accident

Weekday of accident Time of day of accident Work time. How many hours he or she has worked in the time of accident? Usual job. Did the usual job to the person who has suffered the accident? Workplace risk assessment. Is there risk assessment about workplace? Kind of work. What kind of work she is done by the injured person or what work process involved? Deviation. What abnormal event away from the usual work process triggered the accident? Make contact. How has the injured person who has suffered the accident?

3.1 Association Rule Among the various data mining techniques, “association rule mining” is selected in this work because it is a well-known technique to discover cause-and-effect patterns of variables.

126

M. Martínez Rojas et al.

Initially, this technique was proposed for market basket analysis [4]. As a result, this technique provides a measure considering the degree of association between the variables, and it clusters the data using the variables that are highly involved in the association [30]. Additionally, three metrics are usually used to measure the obtained association rules: (i) the first metric involves support, which is the probability that two events may occur simultaneously. (ii) The second metric, confidence, is the posterior confidence or the conditional probability of event B given that event A has already occurred. Finally, (iii) the third metric, lift, is the value where actual confidence is divided by the acquired confidence obtained by assuming that the occurrence of two events (e.g., A, B) is statistically independent. These metrics might provide unknown relations among the considered variables that will enable to design preventive measures in the work of greatest risk.

4 Conclusions In this paper, we propose an approach to explore unknown relations among diverse variables from a large database by means of association rule mining technique. In this regard, we analyze previous work regarding accident variables and data mining techniques. The application of these techniques may be very useful for decisionmaking in order to develop a set of actions that can be developed to prevent the occurrence of future accidents. Concerning further work, we consider to implement the proposed methodology for the dataset of construction accidents data that occurred in Spain building construction sites. Acknowledgements This work has been partially supported by the Spanish Ministry of Economic, Industry and Competitiveness for financing project BIA2016-79270-P and the postdoctoral program (FJCI-2015-24093). It is also supported by the Ministry of Education, Culture and Sports of the Government of Spain for the predoctoral contracts “Formación del Profesorado Universitario” (FPU 2016/03298).

References 1. Antony Chettupuzha AJ, Haas CT (2015) Algorithm for determining the criticality of documents within a construction information system. J Comput Civ Eng 30(3):04015039 2. Arquillos AL, Romero JCR, Gibb A (2012) Analysis of construction accidents in Spain, 2003– 2008. J Saf Res 43(5–6):381–388 3. Bansal V (2011) Application of geographic information systems in construction safety planning. Int J Project Manage 29(1):66–77 4. Berry MJ, Linoff G (1997) Data mining techniques: for marketing, sales, and customer support. Wiley 5. Boraiko C, Beardsley T, Wright E (2008) Accident investigations one element of an effective safety culture. Prof Saf 53(09)

An Approach to Explore Historical Construction Accident Data …

127

6. Camino López MA, Ritzel DO, Fontaneda I, González Alcantara OJ (2008) Construction industry accidents in spain. J Saf Res 39(5):497–507 7. Carrillo-Castrillo JA, Trillo-Cabello AF, Rubio-Romero JC (2017) Construction accidents: identification of the main associations between causes, mechanisms and stages of the construction process. Int J Occup Saf Ergon 23(2):240–250 8. Cheng CW, Leu SS, Cheng YM, Wu TC, Lin CC (2012) Applying data mining techniques to explore factors contributing to occupational injuries in Taiwan’s construction industry. Accid Anal Prev 48:214–222. https://doi.org/10.1016/j.aap.2011.04.014 9. Chi CF, Chang TC, Ting HI (2005) Accident patterns and prevention measures for fatal occupational falls in the construction industry. Appl Ergon 36(4):391–400 10. Chi S, Han S, Kim DY (2013) The relationship between unsafe working conditions and workers: Behaviour and impact of working conditions on injury severity in U.S. construction industry. J Constr Eng Manag 139:826–838 11. Chi S, Suk SJ, Kang Y, Mulva SP (2012) Development of a data mining-based analysis framework for multi-attribute construction project information. Adv Eng Inform 26(3):574–581 12. Comunicación de la comisión al parlamento europeo, al consejo, al comité económico y social europeo y al comité de las regiones (2017) Trabajo más seguro y saludable para todos - Modernización de la legislación y las políticas de la UE de salud y seguridad en el trabajo. Bruselas, 10.1.2017 13. Eurostat (2001) European statistics on accidents at work. Methodology- 2001 edition. https:// doi.org/10.2785/40882 14. EU-OSHA (2017) An international comparison of the cost of work-related accidents and illnesses. Publications Office of the European Union, Luxembourg. https://osha.europa.eu/en/ tools-and-publications/publications/international-comparison-cost-work-related-accidentsand/view 15. Fan H, Li H (2013) Retrieving similar cases for alternative dispute resolution in construction accidents using text mining techniques. Autom Constr 34(1):85–91 16. Guo H, Li H, Li V (2013) VP-based safety management in large-scale construction projects: a conceptual framework. Autom Constr 34(1):16–24 17. Harms-Ringdahl L (2004) Relationships between accident investigations, risk analysis, and safety management. J Hazard Mater 111(1–3):13–19 18. Hoła B, Szóstak M (2017) An occupational profile of people injured in accidents at work in the polish construction Industry. Procedia Eng 208:43–51 19. Khanzode VV, Maiti J, Ray P (2012) Occupational injury and accident research: a comprehensive review. Saf Sci 50(5):1355–1367 20. Kim H, Lee H-S, Park M, Chung B, Hwang S (2013) Information retrieval framework for hazard identification in construction. J Comput Civ Eng 04014052. https://doi.org/10.1061/ (asce)cp.1943-5487.0000340 21. Liew MP, Rosenblatt J (2003) Using data mining techniques for improving building life cycle 22. Martínez-Rojas M, Marín N, Molina C, Vila MA (2016) An intelligent system for cost data handling in construction projects. In: 2016 IEEE international conference on fuzzy systems (FUZZ-IEEE), pp 359–366 23. Martínez-Rojas M, Marín N, Vila MA (2015) The role of information technologies to address data handling in construction project management. J Comput Civ Eng 30(4):04015064 24. Martínez-Rojas M, Pardo Ferreira MC, López-Arquillos A, Rubio-Romero JC (2017) A preliminary approach for accident analysis in construction industry using the multidimensional model. In: International symposium on occupational safety and hygiene, Guimaraes, Portugal 25. Martínez-Aires MD, López-Alonso M, Martínez-Rojas M (2018) Building information modeling and safety management: a systematic review. Saf Sci 101:11–18 26. Ministerio de Empleo y Seguridad Social (2017) http://www.empleo.gob.es/index.htm 27. Papadopoulos G, Georgiadou P, Papazoglou C, Michaliou K (2010) Occupational and public health and safety in a changing work environment: an integrated approach for risk assessment and prevention. Saf Sci 48(8):943–949

128

M. Martínez Rojas et al.

28. Pillay M (2015) Accident causation, prevention and safety management: a review of the stateof-the-art. Procedia Manuf 3:1838–1845 29. Rivas T, Paz M, Martín J, Matías J, García J, Taboada J (2011) Explaining and predicting workplace accidents using data-mining techniques. Reliab Eng Syst Saf 96(7):739–747 30. Shin DP, Park YJ, Seo J, Lee DE (2017) Association rules mined from construction accident data. KSCE J Civ Eng 1–13 31. Soibelman L, Hyunjoo K (2000) Generating construction knowledge with knowledge discovery in databases. Comput Civ Build Eng 2(3):914–921 32. Tixier AJP, Hallowell MR, Rajagopalan B, Bowman D (2017) Construction safety clash detection: identifying safety incompatibilities among fundamental attributes using data mining. Autom Constr 74:39–54 33. Xia N, Wanga X, Griffin MA, Chunlin W, Liu B (2017) Do we see how they perceive risk? An integrated analysis of risk perception and its effect on workplace safety behavior. Accid Anal Prev 106:234–242

A Non-parametric Enhancement of the Fill Rate Estimation Eugenia Babiloni, Ester Guijarro, and Juan Ramon Trapero

Abstract One of the main objectives of an inventory system consists of guarantee a target service level. However, classical approaches are based on assumptions that introduce significant errors. This fact can lead to design inventory policies that do not guarantee the achievement of the target service level. This paper suggests an expression to compute the fill rate for the continuous base-stock policy based on the classic derivation that neglects undershoots at the reorder point. Subsequently, the error incurred for neglecting the undershoots is calculated and corrected by means of the state-dependent parameter approach expressed in a state space framework. The results show that the empirical approach reduces the classic fill rate bias. Simulation is employed to corroborate these results. Keywords Inventory · Continuous review · Fill rate · State-dependent parameter · Lost sales

1 Introduction and Literature Review Inventory systems are classified into continuous and periodic review policies. If the status of the inventory is known at any moment, then the inventory is managed using a continuous policy. In this case, once the inventory position (i.e., on-hand stock + onorder stock − backorders) drops to the reorder point, s or ROP, a new replenishment order is launched and received L units of time later. This order may be constant, Q, as is the case of a traditional (s, Q) policy, or variable until the base-stock level, S, E. Babiloni · E. Guijarro (B) Dpto. de Organización de Empresas, Universitat Politècnica de València, Camino de Vera S/N, 46022 Valencia, Spain e-mail: [email protected] E. Babiloni e-mail: [email protected] J. R. Trapero Dpt. of Business Administration, University of Castilla-La Mancha, 13071 Ciudad Real, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_16

129

130

E. Babiloni et al.

is reached, and thereafter managed by means of a (s, S) policy. This paper focuses on the latter, which is normally used to manage strategic items. The difficulty of the policies that involve reorder point is precisely the fact that when a replenishment order is launched, the inventory position may not be exactly at the reorder point, but a certain amount below it. This amount is called undershoot at ROP. Despite there are several inventory models that consider explicitly the presence of undershoots (see, for example, [3, 11, 15, 16]), the probability of having undershoots at ROP has not been introduced either in the common derivation of continuous policies [13] or in the classic approach to compute the fraction of demand that is immediately satisfied from the stock available on the shelf, i.e., the fill rate (β) (see, for example, [1, 9–10, 12]). However, neglecting undershoots may lead to higher inventory costs and lower service levels than desired for. The objective of this paper is twofold. First, we suggest an expression to compute the fill rate for the continuous base-stock policy based on the common derivation that neglects undershoots at ROP, named β Classic in the rest. After that, we assess the performance of β Classic against simulation to characterize errors that come from neglecting undershoots. Finally, an empirical approach is employed to minimize the differences between β Classic and the simulated fill rate. This empirical approach is based on the state-dependent parameter (SDP) estimation procedure [9] and the use of the information available (as the reorder point) and statistical distribution parameters. Note that this approach utilizes a new perspective on inventory control typically dominated by operational research theoretical developments [13] or control theory [4]. This novel viewpoint intends to use all the available information to improve the design of inventory policies aligned with the development of big data platforms [5]. The remainder of the paper is organized as follows. Section 2 describes the inventory system and introduces the notation and general assumptions of this paper. Section 3 dedicates to the classic fill rate estimation in a lost sales context that is evaluated using simulation. Section 4 dedicates to correct the bias that presents the classic approach applying the state-dependent parameter (SDP) estimation procedure. Finally, Sect. 5 presents the main conclusions and further research of this work.

2 Inventory System, Assumptions, and Notation This paper considers a single-item single-echelon inventory system with stochastic demand which is modeled by any discrete distribution. The stock is controlled according to a continuous base-stock (s, S) policy for the lost sales case (i.e., unfilled demand is lost). In this system, when the inventory position is at or below the reorder point s, a sufficient amount is ordered to raise the inventory position up to the basestock level S. The replenishment order is received L unit of times after being launched. The notations in the rest of the paper are: s

reorder point, ROP (units),

A Non-parametric Enhancement of the Fill Rate Estimation

S L z0 Dτ DL f t (·) F t (·)

131

base-stock level (units), lead time for the replenishment order (time), on-hand stock at the beginning of the cycle and at order delivery (units), accumulated demand from the beginning of the cycle to the ROP (units), accumulated demand during the lead time (units), probability mass function of demand at t, cumulative distribution function of demand during t unit of times.

General assumptions of this paper are: (i) time is organized in a numerable and infinite succession of equally spaced instants; (ii) L is known and constant; (iii) there is never more than one order outstanding which implies that s < S-s; (iv) the replenishment order is added to the inventory at the end of the instant in which it is received; (v) demand process is discrete, stationary, and i.i.d; and (vi) excess demand is lost.

3 On the Impact of Neglecting Undershoots at ROP in the Estimation of the Fill Rate for the Continuous Base-Stock Policy According to the fill rate definition, this service level can be computed through the complementary of the quotient between the expected unfilled demand per replenishment cycle (also known as expected shortage) and the total expected demand, i.e., βClassic = 1 −

E(unfilled demand) E(total demand)

(1)

The common approach to categorize continuous inventory policies is based on neglecting undershoots at ROP. For the continuous base-stock policy, it has two important implications: on the one hand, it implies that the order quantity is fixed and equal to S – s. On the other hand, when neglecting undershoots, the system is only out of stock during the lead time. Taking into account both implications, the ∞ expected unfilled demand is straightforwardly computed as: i=s+1 (i − s) · f L (i). To compute the expected total demand per replenishment cycle, we consider it as the result of adding up the accumulated demand from the beginning of the cycle (i.e., at the moment the order is delivered) until the ROP is reached, Dτ , and the accumulated demand during the lead time, DL . To compute Dτ , we need to know the stock balance at the beginning of the cycle. Since this paper assumes lost sales, the on-hand stock at this moment is z 0 = S − s + E[s − D L ]+ . Therefore, the demand that is required to reach exactly the reorder point is Dτ = z 0 −s = S−2s+E[s − D L ]+ . Consequently: ∞

βClassic = 1 −

S − 2s +

(i − s) · f L (i) ∞ j=0 (s − j) · f L ( j) + k=0 k · f L (k)

s

i=s+1

(2)

132 Table 1 Set of data

E. Babiloni et al. Lead time L = 3 Order quantity S = 7; 8; 9; 10; 11; 12; 13 Reorder point s = 2; 3; 4 Demand variability (Poisson distributed) λ = 0.1; 0.5; 1; 1.5; 2; 2.5

In order to assess the impact of neglecting undershoots; we carry out a simulation with 20,000 consecutive periods and compute β Sim . This simulation uses the data from Table 1 which encompasses 90 different cases with pure poisson demands. Thirty replications are applied to each case using the averages of the values as β Sim . Figure 2 shows the comparison between β Classic and β Sim . As it can be observed, β Classic systematically underestimates β Sim . This behavior was to be expected if we analyze the impact that neglecting undershoots have. For example, if the inventory position and the on-hand stock are 1 unit above the ROP at a given moment t and an order for 2 units is received at t + 1, a replenishment order is launched when the inventory position, and the on-hand stock are equal to exactly s-1 units. Therefore, the on-hand stock remaining on shelf is 1 unit less than what is expected. Thus, β Classic is higher than the real one and therefore stockouts are larger than expected.

4 Correction of the Fill Rate Bias Based on the SDP Approach Considering Fig. 1, we can observe that β Classic overestimates β Sim , however, such an overestimation is not constant, that is, when the target fill rate falls between 0.6 and 0.9, the difference between both approaches is more plausible. Therefore, that Fig. 1 β Classic versus β Sim for the cases from Table 1

1,00

0,90

β Classic

0,80

0,70

0,60

0,50 0,50

0,60

0,70

βSim

0,80

0,90

1,00

A Non-parametric Enhancement of the Fill Rate Estimation

133

plot suggests the presence of non-linearities that depends on the target fill rate. To deal with these potential non-linearities, the SDP approach can be a feasible solution [9]. An initial model that relates β Classic and β Sim is as follows: βSDP = α0 + α1 (F R) · βClassic + α2 · μ + α3 · s + 

(3)

in which α1 (F R) is the state-dependent parameter. In other words, the parameter α1 depends on the state FR, that is the target fill rate. However, the relationship between α1 and FR is unknown. The SDP approach allows us to explore by a nonparametric curve the pattern of such non-linearities [9]. Note that α0 , α2 and α3 are assumed constant and  is a Gaussian error term with zero mean and variance σ 2 . An initial possibility to allow α1 to vary with FR would be to describe the stochastic properties of α1 . To do that, a state vector with two dimensions can be employed, where the stochastic properties are governed by a generalized random walk [6]. Note that the SDP follows a stochastic structure equivalent to the well-known Timevarying parameters (TVP) [9], typically found in a time series context. In fact, the main difference of SDP lies in that the data is sorted with respect to the state instead of the time. Since we are interested in looking for the variations of α1 with respect to the target fill rate instead of time, the SDP framework is the most adequate tool. The SDP method incorporates smoothing recursive algorithms, particularly, the fixed interval smoothing (FIS) to provide state-dependent parameter estimates. It should be noted that the nature of the SDP estimate is non-parametric, i.e., it is a curve, which can be non-linear, of the state-dependent parameter α1 (F R) against the state FR. Among the different TVP types, the α1 stochasticity follows an integrated random walk which is defined as        α1 (k + 1) 11 0 α1 (k) = + (4) α1∗ (k + 1) α1∗ (k) 01 η∗ (k) where α1 and α1∗ are associated with the level and trend of the SDP; the variations of α1 and α1∗ are considered in the model by the random Gaussian noise η∗ (k) with mean zero and variance σα2 . Note that k = 1, …, N is the sample index organized with respect to the FR size and it is not the time index. The full model expressed as a state space (SS) system is comprised of the observation equation in (3) and the state equations in (4). One of the main advantages of using the SS formulation is that wellknown recursive algorithms such as the Kalman filter (KF) in Kalman [7] and (FIS) in Bryson and Ho [2] can be employed. However, all the system matrices should be determined beforehand to use such algorithms. In addition, the noise variances of the observation equation (σ 2 ) and the state equations (σα2 ) are the unknown parameters. Note that such unknown variances can be also called hyper-parameters to avoid any confusion with the states in (4). To reduce the number of unknown parameters, the variance σα2 can be normalized by the innovations variance (σ 2 ) obtaining the noise variance ratio (NVR = σα2 /σ 2 ). Typical optimization routines as maximum likelihood in the time domain can be used to determine the NVR. For the interested reader, more examples using the SDP in different disciplines in conjunction with a full description

134

E. Babiloni et al.

Fig. 2 a SDP α1 (FR) sorted with respect to FR. b The proposed SDP (β SDP =fitSDP) on the hold-out sample (last 30 observations)

of the methodology can be found in Young et al. [9]. Furthermore, it is important to note that SDP algorithms are already implemented within the CAPTAIN toolbox [14] to be used with MATLAB/Simulink™ software. Figure 2a depicts the estimation of α1 (F R) against FR. According to this figure, there exists a variation of α1 with respect to FR. Note that the parameter value decreases for values of SL between 0.6 and 0.9. The hold-in sample utilized to estimate the SDP consisted of 70 observations out of 90. Figure 2b shows the achieved fill rate produced by the exact approach (β Sim ), the classic approach (β Classic ) and the proposed SDP (β SDP =fitSDP) on the hold-out sample (last 30 observations). The results show how the bias has been substantially reduced.

5 Conclusion and Further Research This paper explores the classical approach to compute the fill rate when inventory is managed by the continuous base-stock policy. Such an approach, named classic in this paper, ignores the possible presence of undershoots at the reorder point which leads to a bias estimation of the fill rate as Fig. 1 shows. In this work, we have addressed the problem that neglecting undershoots introduces via a non-parametric approach based on SDP combined with a state space framework instead of deriving a complicated analytical expression that includes undershoots. Several simulations showed that the non-parametric SDP method decreases the classic fill rate bias with respect to a target fill rate as shown in Fig. 2b. This paper intends to be an initial work to develop data-driven methods capable of designing stock control policies. The idea is to develop a set of methods that

A Non-parametric Enhancement of the Fill Rate Estimation

135

allows using the potential benefits of big data applications to enhance the design of inventory policies. Further research should apply the methodology showed here to real demand data to make the results more solid. Additionally, probability forecasts of undershoots statistical distribution function should also be provided to enhance the reduction of fill rate bias and to bridge the traditional gap between forecasting and stock control. Acknowledgements This work was supported by the European Regional Development Fund and Spanish Government (MINECO/FEDER, UE) under the project with reference DPI2015-64133-R and by the Generalitat Valenciana under the project with reference GV/2017/032.

References 1. Axsäter S (2006) A simple procedure for determining order quantities under a fill rate constraint and normally distributed lead-time demand. Eur J Oper Res 174(1):480–491 2. Bryson A, Ho Y (1969) Applied optimal control, optimization, estimation and control. Blaisdell, New York 3. Cohen MA, Kleindorfer PR, Lee HL (1988) Service constrained (s, S) inventory systems with priority demand classes and lost sales. Manage Sci 34(4):482–499 4. Dejonckheere J, Disney SM, Lambrecht MR, Towill DR (2003) Measuring and avoiding the bullwhip effect: a control theoretic approach. Eur J Oper Res 147:567–590 5. Gandomi A, Haider M (2015) Beyond the hype: big data concepts, methods, and analytics. Int J Inf Manage 35(2):137–144 6. Jakeman AJ, Young PC (1984) Recursive filtering and the inversion of ill-posed causal problems. Util Math 35:351–376 7. Kalman RE (1960) A new approach to linear filtering and prediction problems. ASME Trans J Basic Eng 83(D):95–108 8. Larsen C, Thorstenson A (2008) A comparison between the order and the volume fill rate for a base-stock inventory control system under a compound renewal demand process. J Oper Res Soc 59(6):798–804 9. Larsen C, Thorstenson A (2014) The order and volume fill rates in inventory control systems. Int J Prod Econ 147:13–19 10. Platt DE, Robinson LW, Freund RB (1997) Tractable (Q, R) heuristic models for constrained service levels. Manage Sci 43(7):951–965 11. Schneider H (1978) Methods for determining the re-order point of an (s, S) ordering policy when a service level is specified. J Oper Res Soc 29(12):1181–193 12. Schneider H (1981) Effect of service-levels on order-points or order-levels in inventory models. Int J Prod Res 19(6):615–631 13. Silver EA, Pyke DF, Peterson R (1998) Inventory management and production planning and scheduling. Wiley, USA 14. Taylor CJ, Pedregal DJ, Young PC, Tych W (2007) Environmental time series analysis and forecasting with the captain toolbox. Environ Model Softw 22(6):797–814 15. Tijms HC, Groenevelt H (1984) Simple approximations for the reorder point in periodic and continuous review (s, S) inventory systems with service level constraints. Eur J Oper Res 17(2):175–190 16. Vincent P (1985) Exact fill rates for items with erratic demand patterns. INFOR Inf Syst Oper Res 23(2):171–181

FAHP Applications for Manufacturing Environments: A Contemporary Review and Classification Victor Anaya Fons, Raúl Rodríguez Rodríguez, and Angel Ortiz

Abstract This paper presents a review of recent academic papers (since January 2017 until February 2018) on the FAHP applications on manufacturing environments. The main aim of the research was to identify whether FAHP was still of interest to the academic community, as a previous phase to decide its application on further research that the authors aim to undertake. The main results point out that FAHP applications on manufacturing environments: (1) are still of interest at the light of both the number of publications and the quality of the journals; (2) FAHP is usually not used in isolation but complemented with other tools, mainly TOPSIS; (3) the most common applications are for evaluating and/or selecting suppliers, resources, sustainability, or risk factors; (4) There is a lack of applications on core operational improvement processes. Keywords MADM · FAHP · Review · Manufacturing

1 Introduction The multi-attribute decision-making (MADM) is a kind of multi-criteria decisionmaking (MCDM) method that has been very popular in the last two decades with a large base of applications in industrial areas as manufacturing, agriculture, logistics, construction, healthcare, banking, education, energy or government, among others. MADM comprises several techniques such as simple additive method (SAW) [8], elimination et choix traduisant la realité (ELECTRE) [31], preference ranking organisation method for enrichment evaluations (PROMETHEE) [4], technique for order V. Anaya Fons (B) · R. Rodríguez Rodríguez · A. Ortiz CIGIP, Universitat Politècnica de València, Camino de Vera, s/n Ed. 8B - 2ª Planta, L Ciudad Politécnica de la Innovación, 46022 Valencia, Spain e-mail: [email protected] R. Rodríguez Rodríguez e-mail: [email protected] A. Ortiz e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_17

137

138

V. Anaya Fons et al.

preference by similarity to ideal solution (TOPSIS) [11], and analytical hierarchy process (AHP) [24]. AHP and its fuzzy counterpart FAHP the techniques most broadly applied in the industry. There are papers that review the application of FAHP in the industry being [15] the most contemporary article, with an analysis considering articles until June 2016. This paper upgrades those results analyzing how FAHP applications have evolved in the last 2 years (2017 and 2018) and concludes on the current interest of FAHP as a MADM technique. Section 2 presents FAHP technique as a fuzzy extension to the original AHP. Section 3 introduces the review methodology carried out. Section 4 presents the review results obtained. Finally, Sect. 5 concludes with the conclusions and the future research actions.

2 FAHP The analytic hierarchy process (AHP) is a technique invented by Saaty in 1980 [24] to support complex decision-making processes where different alternatives to achieve a unique goal should be systematically analyzed. AHP technique considers the human judgment when evaluating the different alternatives and provides mechanism to formalize and validate their input with concepts such as the consistency index (CI). Subsequently, van Laarhoven [16] applied the fuzzy logic theory making AHP more flexible for users to provide inputs. Inputs are based on looser judgments that are fuzzed and crisp in pre and post-phases added to Saaty’s AHP method. The FAHP Process Is Composed of the Following Steps: (a) (b) (c) (d)

Identifying objectives. Identifying alternatives. Criteria identification for alternatives comparation and weight calculation. Analysis of the options according to criteria identified through a pair-wise comparation of each pair of alternatives. (e) Ranking and decision-making.

3 Research Methodology This paper considers some existing classification methodologies such as those at [15, 29] when defining the classification dimensions considered on the review. These dimensions are: • Application Area: In our review, we only consider papers on the manufacturing application area.

FAHP Applications for Manufacturing Environments …

139

• Theme: the generic purpose for applying FAHP, which could be selection and/or evaluation. • Complementary tools: other MADM techniques applied. • Year: year of publication. • Journal: name of the scientific journal where the paper was published. In addition to the previous dimensions, the article’s information that was synthesized was: • Title: title of the article. • Authors: list of authors publishing the article. • Specific objective: rationale of applying FAHP to a specific manufacturing industry. The review process was targeted in ScienceDirect library for the years 2017 and 2018. A range of journals has been considered and only manufacturing applications have been contemplated. The total amount of papers reviewed after filtering the corpus with the previous criteria is 25 articles, which makes sense due to the methodology approach and the time span (14 months). An aside before continuing introducing the review is on the theme dimension category. Themes proposed in this article are based on those in [15], but naming could be confusing without a deeper explanation. • Selection: the article’s goal is to pick one alternative over the others for satisfying an objective. Computed values for each alternative represent the level of competent of the alternative regarding the intended objective. • Evaluation: the article’s purpose is the evaluation of alternatives to obtain a score used for ranking them, with the goal to select one or more alternatives to satisfy an objective.

4 Review Results The review carried out has considered the articles reported in Table 1 indicating the specific objective of the FAHP application and other techniques applied along. Additionally, Fig. 1 shows the yearly distribution of papers according to their specific thematic, showing a lighter preference in the application of FAHP to selection problems, reaffirmed the same trends that those stated in [15].

140

V. Anaya Fons et al.

Table 1 FAHP recent references on the topic of manufacturing Nº

Authors

Specific objective

Other tool(s) used

M1

Fallahpour et al. [9]

Sustainable supplier selection

FTOPSIS, Mann-Whitney U-test

M2

López and Ishizaka [18]

Selection of location criteria of supplier

Fuzzy Cognitive Maps

M3

Bhat and Kumar [5]

Evaluation of supply chain complexity drivers

DEMATEL, ANP, FTOPSIS and FVIKOR

M4

Sarkar et al. [26]

Supplier selection

DEMATEL, ANP, FTOPSIS and FVIKOR

M5

Akbas and Bilgen [1]

Selection of the ideal gas fuel

TOPSIS, FQFD

M6

Parkouhi and Ghadikolaei [22]

Supplier selection by resilience approach

FANP, VIKOR

M7

Zimmer et al. [33]

Assessment of global supply chains risks

Leontief’s Input Output model

M8

Kopacz et al. [14]

Assessment of sustainable development of hard coal mining industry

Monte Carlo Simulation, bootstrap, copila, frequency distributions

M9

Jian et al. [13]

Evaluation of product maintainability

TOPSIS

M10

Srisawat et al. [28]

Evaluating the spatial efficiency of transport logistics

GIS

M11

Seitia et al. [27]

Assessment and selection of proper maintenance strategy in risky situations

FAD

M12

Jayawickrama et al. [12]

Evaluation of manufacturing plant sustainability

M13

Loh et al. [17]

Evaluation of port-centric supply chain disruption threats

M14

Rodríguez et al. [23]

Evaluation of suitable sites for allocation of bioenergy plants

GIS

M15

Hsu et al. [10]

Selection of important elements in sustainability indicators

QFD, FDM, TOPSIS

M16

Zavadskas et al. [32]

Selection of facilities management strategy

TOPSIS

M17

Sa˘ ¸ gban¸sua and Balo [25]

Selection of the best turbines (continued)

FAHP Applications for Manufacturing Environments …

141

Table 1 (continued) Nº

Authors

Specific objective

Other tool(s) used

M18

Moayyedian et al. [20]

Evaluation of the best alternatives for the moldability index

TOPSIS, Taguchi

M19

Modak et al. [21]

Evaluation of the strategic decision of outsourcing

BSC

M20

Chen et al. [7]

Safety assessment of natural gas purification plant

M21

Breaz et al. [6]

Selecting industrial robots

M22

Al Garni and Awasthi [3]

Selection of power plant site

GIS

M23

Wan et al. [30]

Supplier selection

ANP, ELECTRE II

M24

Akhundzadeh and Shirazi [2]

Technology selection and evaluation

M25

Mahjouri et al. [19]

Assessment of sustainability indicators for wastewater treatment systems

Fuzzy Delphi

Fig. 1 Yearly distribution of papers according to theme

5 Conclusions and Further Research The main conclusions deduced from the previous contemporary review are: • FAHP is still a popular and used technique applied on industrial manufacturing environments and applications and with an interest regarding research purposes. • Though it is feasible to apply FAHP as a unique tool, when selecting evaluating alternatives is far more common to complement it with another tools, mainly TOPSIS.

142

V. Anaya Fons et al.

• The most common applications of FAHP (either isolated or in combination with other tools) are: Supplier selection, raw material/resources selection, plant location, and evaluation of sustainability and risk factors on companies. • Existing applications for evaluation purposes are mainly on both management and support process, whereas there is a lack of applications on core operational improvement processes, such as manufacturing processes or optimal resource allocation

References 1. Akba¸s H, Bilgen B (2017) An integrated fuzzy QFD and TOPSIS methodology for choosing the ideal gas fuel at WWTPs. Energy 125:484–497 2. Akhundzadeh M, Shirazi B (2017) Technology selection and evaluation in Iran’s pulp and paper industry using 2-filterd fuzzy decision making method. J Clean Prod Part 4 142:3028–3043 3. Al Garni H, Awasthi A (2017) Solar PV power plant site selection using a GIS-AHP based approach with application in Saudi Arabia. Appl Energy 206:1225–1240 4. Behzadian M, Kazemzadeh RB, Albadvi A, Aghdasi M (2010) PROMETHEE: a comprehensive literature review on methodologies and applications. Eur J Oper Res 200(1):198–215 5. Bhat SA, Kumar A (2018) An integrated fuzzy approach for prioritizing supply chain complexity drivers of an Indian mining equipment manufacturer by Kavilal EG, Venkatesan SP, Kumar KDH (2017) Resour Policy 51:204–218. Suggested modification. Resources Policy. ISSN 0301-4207 6. Breaz RE, Bologa O, Racz SG (2017) Selecting industrial robots for milling applications using AHP. Proc Comput Sci 122:346–353 7. Chen K, Khan F, Jing X (2018) Safety assessment of natural gas purification plant. Process Saf Environ Prot 113:459–466 8. Churchman CW, Ackoff RL (2016) An approximate measure of value. J Oper Res Soc Am 2(1):1954 9. Fallahpour A, Olugu E, Musa S, Wong K, Noori S (2017) A decision support model for sustainable supplier selection in sustainable supply chain management. Comput Ind Eng 105:391–410 10. Hsu C, Chang A, Luo W (2017) Identifying key performance factors for sustainability development of SMEs—integrating QFD and fuzzy MADM methods. J Clean Prod 161:629–645 11. Hwang CL (1981) Multiple attribute decision making methods and applications: a state-of-theart survey. Springer, New York 12. Jayawickrama HMMM, Kulatunga AK, Mathavan S (2017) Fuzzy AHP based plant sustainability evaluation method. Procedia Manuf 8:571–578 13. Jian X, Cai S, Chen Q (2017) A study on the evaluation of product maintainability based on the life cycle theory. J Clean Prod 141:481–491 14. Kopacz M, Kryzia D, Kryzia K (2017) Assessment of sustainable development of hard coal mining industry in Poland with use of bootstrap sampling and copula-based Monte Carlo simulation. J Clean Prod 159:359–373 15. Kubler S, Robert J, Derigent W, Voisin A, Traon Y (2016) A state-of the-art survey & testbed of fuzzy AHP (FAHP) applications. Expert Syst Appl ISSN 65:398–422 ISSN 0957-4174 16. van Laarhoven (1983) A fuzzy extensión of Saaty’s priority theory. Fuzzy Sets Syst 11(1):199– 227 17. Loh HS, Zhou Q, Thai VV, Wong YK, Yuen KF (2017) Fuzzy comprehensive evaluation of port-centric supply chain disruption threats. Ocean Coast Manag 148:53–62

FAHP Applications for Manufacturing Environments …

143

18. López C, Ishizaka A (2017) A hybrid FCM-AHP approach to predict impacts of offshore outsourcing location decisions on supply chain resilience. J Bus Res ISSN 0148-2963 19. Mahjouri M, Ishak MB, Torabian A, Manaf LA, Halimoon N (2017) The application of a hybrid model for identifying and ranking indicators for assessing the sustainability of wastewater treatment systems. Sustain Prod Consum 10:21–37 20. Moayyedian M, Abhary K, Marian R (2018) Optimization of injection molding process based on fuzzy quality evaluation and Taguchi experimental design. CIRP J Manuf Sci Technol 21. Modak M, Pathak K, Ghosh KK (2017) Performance evaluation of outsourcing decision using a BSC and Fuzzy AHP approach: a case of the Indian coal mining organization. Resour Policy 52:181–191 22. Parkouhi S, Ghadikolaei AS (2017) A resilience approach for supplier selection: using fuzzy analytic network process and grey VIKOR techniques. J Clean Prod 161:431–451 23. Rodríguez R, Gauthier-Maradei P, Escalante H (2017) Fuzzy spatial decision tool to rank suitable sites for allocation of bioenergy plants based on crop residue. Biomass Bioenerg 100:17–30 24. Saaty TL (1980) The analytic hierarchy process. McGraw-Hill, New York 25. Sa˘ ¸ gban¸sua L, Balo F (2017) Multi-criteria decision making for 1.5 MW wind turbine selection. Procedia Computer Science 111:413–419 26. Sarkar S, Pratihar D, Sarkar B (2018) An integrated fuzzy multiple criteria supplier selection approach and its application in a welding company. J Manuf Syst 46:163–178 27. Seiti H, Hafezalkotob A, Fattahi R (2017) Extending a pessimistic–optimistic fuzzy information axiom based approach considering acceptable risk: application in the selection of maintenance strategy. Appl Soft Comput 28. Srisawat P, Kronprasert N, Arunotayanun K (2017) Development of decision support system for evaluating spatial efficiency of regional transport logistics. Transp Res Procedia 25:4832–4851 29. Vaidya O, Kumer S (2006) Analytic hierarchy process: an overview of applications. Eur J Oper Res 169(1):1–29 30. Wan S, Xu G, Dong J (2017) Supplier selection using ANP and ELECTRE II in interval 2-tuple linguistic environment. Inf Sci 385–386:19–38 31. Yu X, Zhang S, Liao X, Qi X (2018) ELECTRE methods in prioritized MCDM environment. Inf Sci 424:301–316 32. Zavadskas EK, Turskis Z, Vilutien˙e T, Lepkova N (2017) Integrated group fuzzy multi-criteria model: case of facilities management strategy selection. Expert Syst Appl 82:317–331 33. Zimmer K, Fröhling M, Breun P, Schultmann F (2017) Assessing social risks of global supply chains: a quantitative analytical approach and its application to supplier selection in the German automotive industry. J Clean Prod 149:96–109

A MILP Approach to Maximize Productivity in Mixed-Model Assembly Lines Joaquín Bautista and Rocío Alfaro-Pozo

Abstract Two mathematical models are proposed in order to obtain assembly line configurations that guarantee a maximum productivity while the ergonomic risk of workstations is controlled and the available space for the line is respected. Both models, with different optimization approaches, are assessed through a daily demand plan linked with the Nissan powertrain plant in Barcelona. Results show the influence of limitations on the efficient assignment of operations to workstations of the line. The effectiveness of models for establishing the number of workstations and the cycle time of the assembly line that maximizes productivity is demonstrated given a demand plan. Keywords Assembly line · Balancing problem · Productivity · Ergonomic risk

1 Preliminaries Progress in technology, the ever-increasing requirements of consumers and an ever more globalized market have been factors influential in the creation of mass customization. This competitive and changeable environment requires companies to modify their activities in order to offer customers those they want, when they want, and how they want. Obviously, this need for satisfying consumers’ needs goes hand-in-hand with the control of production costs. Accordingly, mass customization has become one of the leading strategies used in production industries to make them more competitive. However, this type of manufacturing strategy supposes production systems that must be able to adjust to the different options of products, as well as the changes in production mixes, without incurring high costs. J. Bautista (B) IOC-ETSEIB, Universitat Politècnica de Catalunya, 08028 Barcelona, Spain e-mail: [email protected] R. Alfaro-Pozo EAE Business School, 08015 Barcelona, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_18

145

146

J. Bautista and R. Alfaro-Pozo

Within manufacturing systems based on mixed-model assembly lines (MMALs), the configuration of the assembly line is a crucial stage on the design of lines highly efficient and cost-effective. MMALs must be able to endure the manufacturing of a wide range of product options and frequent variations of production mixes. Accordingly, an efficient and robust design is essential for addressing the desirable flexibility and maintaining the advantages from mass production. The assembly line balancing problems (ALBPs) are focused on this issue. Given the set of tasks needed to manufacture the products, the challenge is to assign these tasks to the set of workstations that made up the line in accordance with one or more optimization criteria. Obviously, this assignment is subject to a set of restrictions linked with different attributes, such as temporal restrictions from tasks or workstations, spatial restrictions, or ergonomic conditions. Indeed, ALBPs have been categorized in literature regarding the type of restrictions and the optimization criteria. Thus, Baybars [1] differentiated between the simple assembly line balancing problems (SALBPs) [9], and the generalized assembly line balancing problems (GALBPs), [2], among others). Scholl and Becker [9] classified later the SALBP family according the optimization criteria, giving rise four problem types. The SALBP-F that determines the feasibility of a line configuration given a number of stations, m, and a cycle time, c. The SALBP-1 that aims at minimizing the number of workstations, K, given the cycle time, c. The SALBP-2 that minimizes the cycle time, c (i.e., maximizes the production rate), given a number of workstations, K. And the SALBP-E that focus on maximizing the line productivity by means of minimizing simultaneously c and m. Despite all research about ALBPs, the SALBP-E literature is not too much extensive due to non-linearity and difficult resolution of the problem [7, 10]. For that reason, we propose to address the SALBP-E from not only a productivity perspective but also considering the comfort of operators that work in the line. Specifically, first we propose an extension of SALBP-E based on the time and space assembly line balancing problems (TSALBP), which were proposed by Bautista and Pereira [6]: the T S AL B P − m × c. In addition to the spatial limitations for workstations, that are already considered in the TSALBP family, we also consider in this work the maximum ergonomic risk allowed per workstation, similarly to Bautista et al. [4] and [3]. Secondly, we propose an alternative model whose objective is minimizing discrepancies between the workload times of workstations. This model, which also considers the spatial and ergonomic limitations from the first model, allows us to assess the quality of assembly line configurations in terms of (1) line’s efficiency and (2) workloads’ balance between workstations. Finally, both models are solved through mixed integer linear programming (MILP).

A MILP Approach to Maximize Productivity …

147

2 Maximizing Productivity by Considering Physical Ergonomic Risks and Spatial Restrictions Workers of assembly lines are exposed to numerous risk factors that can cause them musculoskeletal disorders [8]. While it is truth that companies are aware of the importance of labor risk prevention, they usually perform the risk analysis once the line is in operation. However, considering ergonomic risk in the design phase can reduce cost and time. Based on the above and following previous work [3–5], next we present a mathematical model whose objective is to obtain assembly line configurations with the maximum productivity. In addition, the assembly line configurations are subject to a maximum ergonomic risk and maximum linear area per workstation. The parameters and variables of the model are the following:

Parameters J

Set of elemental tasks ( j = 1 . . . ..|J |)

K

Set of workstations (k = 1 . . . ..|K |)



Set of ergonomic risk factors (φ = 1, . . . , ||). This work deals only with the physical factor, therefore || = 1

tj

Processing time of an elemental task j ∈ J (at normal work pace)

aj

Area or space (linear) required by the task j ∈ J

χφ, j

Ergonomic risk category associated to the task j ∈ J regarding the risk factor φ ∈ . In this work χ j (|| = 1)

Rφ, j

Ergonomic risk (ergo-seconds) associated to the task j ∈ J regarding the risk factor φ ∈  : Rφ, j = t j · χφ, j . In this work R j = t j · χ j (|| = 1)

Pj

Set of tasks that precede task j ∈ J

m min

Minimum number of workstations of the assembly line

m max

Maximum number of workstations of the assembly line

Amax k

Maximum area available at workstation k ∈ K . In this work: Amax = Amax (∀k ∈ K ) k

max Rφ, k

Maximum ergonomic risk allowed at workstation k ∈ K according to the ergonomic risk factor φ ∈ . In this work: Rkmax (|| = 1)

Variables c

Cycle time (seconds). Time allowed to workstations to work on a product

m

Number of workstations required by the assembly line, (m = |K |)

x j,k

Binary variable equal to 1 whether task j ∈ J is assigned to workstation k ∈ K

And the mathematical model, T S AL B P − m × c_erg, is: Min z = m × c

(1)

148

J. Bautista and R. Alfaro-Pozo

m ≥ m min m max

x j,k = 1 ( j = 1, . . . , |J |)

k=1

m− c− Amax − k

|J | j=1

m max

j=1

(4)

t j · x j,k ≥ 0 (k = 1, . . . , m max )

(5)

a j · x j, k ≥ 0 (k = 1, . . . , m max ) ∧ (φ = 1, . . . ||)

max Rφ, k −

|J |

(3)

k · x j,k ≥ 0 ( j = 1, . . . , |J |)

k=1

|J |

(2)

(6)

Rφ, j · x j, k ≥ 0 (k = 1, . . . , m max )

(7)

  k · x j,k − xi,k ≥ 0 (i ∈ P j , j = 1, . . . |J |)

(8)

x j,k ∈ {0, 1} ( j = 1, . . . , |J |) ∧ (k = 1, . . . , m max )

(9)

m max k=1

j=1

In the model, the Eq. (1) represents the maximization of productivity by multiplying the number of workstations per the cycle time of the line. Constraints (2) and (3) guarantee a minimum number of workstations of the line and the assignment of each task to only one workstation. Constraints (4) and (5) determine de number of workstations and the cycle time of the line. Constraints (6) and (7) limit the area and the ergonomic risk per workstation. Finally, constraints (8) and (9) set the precedence rules of operations and set the assignment variables as binary. It should be noted that the model is not linear. Then, to solve it through MILP we adapt the objective function and the constraint (2) as follows: min(max c)

(10)

m = m max

(11)

Equations (3)–(9). Accordingly, we will solve the simplified model for different values for the number of workstations.

3 Balancing of Workload Time of Assembly Line Stations From models proposed in Bautista et al. [5], next we present a mathematical model whose objective is minimizing differences between the processing times of the set of tasks assigned to workstations. Thereby, the workload time assigned to each

A MILP Approach to Maximize Productivity …

149

workstation will be more or less balanced between the set of workstations of the line. Similarly to the T S AL B P − m × c_erg model, this one considers the spatial and ergonomic attributes. Thus, we guarantee assembly line configurations feasible in terms of the length of the plant and secure in terms of physical ergonomic risks. The additional parameters and variables for the model are the following:

Additional parameters K

Set of workstations (k = 1 . . . ..|K |). In this model |K | = m max = m

c

Cycle time. Time allowed to workstation to work on a product

T med

Average processing time (at normal activity) of each workstation while manufacturing a product unit. That is, T med =

1 |K |

|J | 

tj.

j=1

Additional variables



 j ∈ J : x j, k = 1 .

Sk

Workload of workstation k ∈ K : Sk =

T (Sk )

Processing time required (at normal activity) by the workload Sk : T (Sk ) =

δk+ (T )

Over-time (at normal work pace) required at workstation k ∈ K with respect to the  + average value. That is: δk+ (T ) = T (Sk ) − T med

δk− (T )

Defect of processing time required by the workstation k ∈ K (at normal activity) with  + respect to the average value. That is, δk− (T ) = T med − T (Sk )

 j ∈ Sk

tj

And the mathematical model, T S AL B P −  R (T )_erg, is the following: Min  R (T ) =

|K |   + δk (T ) + δk− (T )

(12)

k=1

Subject to: (3)–(9). |J |

x j,k ≥ 1 (k = 1, . . . , |K |)

(13)

t j · x j, k − δk+ (T ) + δk− (T ) = T med (k = 1, . . . , |K |)

(14)

j=1

|J | j=1

δk+ (T ), δk− (T ) ≥ 0 (k = 1, . . . , |K |)

(15)

In the model, Eq. (12) involves the minimization of discrepancies between workload times of workstations and the average cycle time given the number of stations. Constraints (13) guarantee that no workstation is empty, and constraints (14) and (15) determine difference between workload time associated to each workstation and the average workload time.

150

J. Bautista and R. Alfaro-Pozo

4 Computational Experience In order to assess the performance of models, we use a demand plan with 270 engines, which is part of the Nissan-9Eng case. The total demand is equally distributed between nine types of engines: three types of engines for Sport Utility Vehicles ( p1 , . . . , p3 ), two types for vans ( p4 , p5 ), and four types for medium tonnage trucks ( p6 , . . . , p9 ); each one with its temporal, spatial, and ergonomic attributes. Both models are solved through the CPLEX (v11.0) solver on a Mac Pro computer with an Intel Xeon, 3.0 GHz CPU and 2 GB RAM memory under the Windows XP operating system and allowing a maximum CPU time of 7200 s. This has involved the following considerations: • The T S AL B P − m × c_erg is solved as a mono-objective model. Specifically, given a fixed number of workstations for the assembly line, the objective is to minimize the cycle time that corresponds with the maximum workload time from the set of workstations of the assembly line. Obviously, this simplification of the model supposes a sweep of the number of workstations of the assembly line. Indeed, the model is solved for |K | = {19, 20, 21, 22, 23, 24}. • Considering only the physical factor, both models are subjected to a maximum ergonomic risk per workstation equal to 360 ergo-seconds (Rkmax = 360e − s). • Two values for the maximum spatial area per workstation are considered for both models, Amax = {4, 5} meters. k Table 1 shows the results given by both models in regard with the cycle time of the line. No model obtains line configurations with less than 21 workstations and four meters per workstation. On the other hand, as it was to be expected, the T S AL B P − m × c_erg model obtains better results in regard with the maximum workload of workstations in almost all cases. The T S AL B P −  R (T )_erg model only equals the results from the first model in cases with 24 and 25 workstations and 5 M per workstations. By contrast, considering the rectangular distance between workloads assigned to workstations and the ideal cycle time,1 the T S AL B P −  R (T ) − erg is clearly Table 1 Cycle time of assembly line configurations given by both the simplified T S AL B P − m × c_erg model with Amax = 4m (M1) or Amax = 5m (M2) and the T S AL B P −  R (T )_erg k k model with Amax = 4m (M3) or Amax = 5m (M4) k k |K | = 19

|K | = 20

|K | = 21 230

160

150

130

130

200

165

152

146

135

130

125

275

175

170

145

135

163

150

140

130

125

M1 M2 M3 M4 1 The

250

175

|K | = 22

|K | = 23

|K | = 24

|K | = 25

ideal cycle time is equal to the average of processing times of operations in regard with the number of workstations of the assembly line.

A MILP Approach to Maximize Productivity …

151

Table 2 Overall discrepancy of workload time of workstations from line configurations given by both the simplified T S AL B P − m × c_erg model with Amax = 4m (M1) or Amax = 5m (M2) k k and the T S AL B P −  R (T )_erg model with Amax = 4m (M3) or Amax = 5m (M4) k k |K | = 19

|K | = 20

M1 M2

531.37

242.00

M3 M4

351.58

184.00

|K | = 21

|K | = 22

|K | = 23

|K | = 24

|K | = 25

603.62

436.36

392.00

134.17

223.60

178.57

214.55

110.00

138.50

145.20

511.90

343.64

200.00

107.67

84.40

148.57

110.00

70.00

53.33

28.80

superior to the T S AL B P − m × c_erg model in all cases (minimum improvement of 17.9%— R (T )T S AL B P− R (T )_erg = 511.90 versus  R (T )T S AL B P−m×c_erg = 603.62 for |K | = 21, Amax = 5). As it can be seen in Table 2, time dispersion decreases when increasing the number of workstations. Indeed, when the line has 25 workstations and five meters per station, the workload times are around the ideal cycle time with only a maximum distance de 5.4 s.

4.1 Productivity Assessment We assess the quality of solutions in terms of productivity. We analyze the product m × c in order to determine the efficiency of the line through Eq. 16. η(m, c) =

1 |J | · tj j=1 mc

(16)

As we can see in Fig. 1, the T S AL B P − m × c_erg model reaches the best efficiency for an assembly line configuration with 23 workstations, with a maximum linear area of five meters per station and with a maximum ergonomic risk of 360 ergoseconds. This configuration supposes a cycle time of 135 s, and therefore, involves a minor/moderate risk (risk category below 2.7 for all workstations, Bautista et al. [5].

5 Conclusions Both mathematical models have been proposed and evaluated in order to obtain mixed-model assembly lines configurations with the maximum efficiency. The first one is in line with the problem from literature that is named T S AL B P − m × c. Specifically, the T S AL B P − m × c_erg incorporates not only temporal and spatial attributes but also the ergonomic risk of workstations and its objective is focused on minimizing the product between the number of workstations and the cycle time, that is, equal to maximize the efficiency of the line. The second one is a workload

152

J. Bautista and R. Alfaro-Pozo

Fig. 1 Efficiency of line configurations given by the simplified T S AL B P − m × c_erg model with a spatial limitation (Amax = 4m (M1) or Amax = 5m (M2), and by the T S AL B P −  R (T )_erg k k model with a spatial limitation (Amax = 4m (M3) or Amax = 5m (M4) k k

balance model. The T S AL B P −  R (T )_erg model minimizes distances between the workload time of workstations in regard with the cycle time of the line considering the total processing times of operations and the number of workstations. Results in terms of efficiency show the model T S AL B P − m × c_erg gives line configurations more efficient. However, these configurations present higher dispersions of workload times between the set of workstations. Acknowledgements This research was subsidized by the Ministry of Economy and Competitiveness of the Government of Spain through project OPTHEUS (ref. PGC2018-095080-B-I00), including European Regional Development Funds (ERDF).

References 1. Baybars I (1986) A survey of exact algorithms for the simple assembly line balancing problem. Manag Sci 32(8):909–93. https://doi.org/10.1287/mnsc.32.8.909 2. Battaïa O, Dolgui A (2012) Reduction approaches for a generalized line balancing problem. Comput Oper Res 39(10):2337–2345. https://doi.org/10.1016/j.cor.2011.11.022 3. Bautista J, Alfaro-Pozo R (2017) Minimizing the ergonomic risk dispersion between the workstations of an assembly line. In: Proceedings of the 11th International Conference on Industrial Engineering and Industrial Management. XXI Congreso de Ingeniería de Organización, Valencia. https://www.researchgate.net/publication/318362597 4. Bautista J, Alfaro-Pozo R, Batalla-García C (2016a) Maximizing comfort in assembly lines with temporal, spatial and ergonomic attributes. Int J Comput Intell Syst 9(4):788–799. http:// doi.org/10.1080/18756891.2016.1204125

A MILP Approach to Maximize Productivity …

153

5. Bautista J, Batalla-García C, Alfaro-Pozo R (2016b) Models for assembly line balancing by temporal, spatial and ergonomic risk attributes. Eur. J. Oper. Res. 251:814–829. https://doi.org/ 10.1016/j.ejor.2015.12.042 6. Bautista J, Pereira J (2007) Ant algorithms for a time and space constrained assembly line balancing problem. Eur J Oper Res 177(3):2016–2032. https://doi.org/10.1016/j.ejor.2005. 12.017 7. Esmaeilbeigi R, Naderi B, Charkhgard P (2015) The type E simple assembly line balancing problem: a mixed integer linear programming formulation. Comput Oper Res 64:168–177. https://doi.org/10.1016/j.cor.2015.05.017 8. Otto A, Battaïa O (2017) Reducing physical ergonomic risks at assembly lines by line balancing and job rotation: a survey. Comput Ind Eng 111:467–480. https://doi.org/10.1016/j.cie.2017. 04.011 9. Scholl A, Becker C (2006) State-of-the-art exact and heuristic solution procedures for simple assembly line balancing. Eur J Oper Res 168(3):666–693. https://doi.org/10.1016/j.ejor.2004. 07.022 10. Th Zacharia P, Nearchou AC (2013) A meta-heuristic algorithm for the fuzzy assembly line balancing type-E problem. Comput Oper Res 40(12):3033–3040. https://doi.org/10.1016/j.cor. 2013.07.012

Production Typologies in Production Scheduling: Identification and Management Pilar I. Vidal-Carreras, Julio J. Garcia-Sabater, Angel Ruiz, and Julien Maheut

Abstract Production environments vastly vary in the industrial reality and can also do so with time. To manage production as optimally as possible, having tools that guide the people in charge of production would greatly help to know the best approach to take to perform the task according to the factory’s production typology. These tools should also be able to identify when the production typology has been amended and a new approach is required to optimally manage it. The present work aims to present a structure that allows the production typologies in the company to be classified, and to link them to more suitable management approaches in order to specifically manage production scheduling. Keywords Production typologies · Practical production scheduling · Economic lot scheduling problem

1 Introduction The industrial reality is completely diverse and not always stable with time. Having tools to guide the people in charge of production management about the most appropriate way to work in line with the factory’s production typology would be most helpful. These tools should also be capable of identifying when the production typology has been modified, and a new approach is needed to optimally manage P. I. Vidal-Carreras (B) · J. J. Garcia-Sabater · J. Maheut Grupo ROGLE, Dpto. de Organización de Empresas, Universitat Politècnica de València, Camino de Vera S/N, 46022 Valencia, Spain e-mail: [email protected] J. J. Garcia-Sabater e-mail: [email protected] J. Maheut e-mail: [email protected] A. Ruiz Department of Operations and Decision Systems, Laval University, Quebec, Canada e-mail: [email protected] © Springer Nature Switzerland AG 2020 R. de Castro and G. Giménez (eds.), Advances in Engineering Networks, Lecture Notes in Management and Industrial Engineering, https://doi.org/10.1007/978-3-030-44530-0_19

155

156

P. I. Vidal-Carreras et al.

it. The object of this study is to present the structure that allows the production typologies in a company to be classified and to link them with more suitable management approaches, specifically for managing production scheduling. Although the literature on production scheduling includes many works that present algorithms and optimum solution methods [4], most provide optimum results for given datasets. When a situation goes beyond a circle of reliability as regards the data provided in an article, performance considerably reduces with most of these methods [5], which work well for certain characteristics and, in agreement with this line of work, would be one production typology or several, but not others. The problem with many of these works is that the production typology for which they have been designed has not been defined, which makes its practical application difficult. The present work centers on classifying the production typologies that may appear in a specific production configuration. The employed configuration is based on a machine with one stage capable of producing different products, and some costs and times are incurred when the production lot is changed. This is a common configuration in industry because it is often more economic to purchase quite a fast machine capable of producing many products than purchasing many specialized machines [11, 13]. Regarding the production scheduling of this production configuration, as the present work indicates in detail, is clearly framed within the ELSP problem (the economic lot scheduling problem). The literature contains works within the so-called “practical production scheduling” [10] that are related with the production typologies concept contemplated herein, specifically those related with the definition of the production or manufacturing environment. These works identify the relevant factors for characterizing a manufacturing environment. Some have used surveys with heads of production in companies to identify lists of characteristics [9, 15]. The work proposed here considers the applicable characteristics of these studies as they are applied to the ELSP setting, as well as other specific ELSP characteristics. Our work objectives are: • Categorize the production typologies in the production scheduling area for the typical production configuration of the ELSP • Propose the suitable solution approaches for all the production typologies. To this end, our article is arranged as follows: after this introduction in Sect. 1, Sect. 2 briefly describes the production configuration to be considered. Section 3 presents the method to be followed and how it was developed. Finally, Sect. 4 offers conclusions and future research lines.

2 Production Configuration The contemplated production configuration characteristics are the following: • There is only one machine and one stage (single stage, single machine) that produces different products

Production Typologies in Production Scheduling …

157

• Only one product i can be produced on the machine during time t • Some setup times (tsi) and setup costs (csi) are incurred during production, which are known and independent of the sequence • A holding cost exists for product hi, which is known, constant, and proportional to the inventory • The demand di ratio and the production pi ratios are known and constant • Production capacity is limited, but is sufficient to meet total demand • No pending orders are allowed, so total demand must be met during the period in which it is produced As previously mentioned, the production scheduling problem of several products in accordance with this production configuration is known in the literature as the ELSP [2, 11, 16]. This problem is based on scheduling the production of several products that use the capacity of a single machine [3]. The solution for the problem basically lies in defining the lot size and the feasible production schedules that minimize total costs with no pending orders. Although the ELSP is a problem and not just a production configuration, in this section we refer to this configuration as configuration ELSP, rather than a production configuration that is typically framed within the ELSP, throughout this paper for clarity reasons.

3 Method and Development The stages considered to fulfill the objectives are: Sect. 3.1 identify the variables to be considered; Sect. 3.2 identify typologies; and Sect. 3.3 propose tactics for the recommended typologies.

3.1 Identify the Variables to Be Considered To define the different typologies in the ELSP configuration, the variables to consider are those that define the model; that is, number of items, setup time duration, demand quotient and the di/pi production ratio, setup cost and holding cost. In the related literature, two considerations are indicated to choose these variables. As mentioned in the introduction, these factors allow production environments to be identified. Moreover, when works on simulations for the ELSP were analyzed, it was found that these parameters were those that had been modified [6, 8].

158

P. I. Vidal-Carreras et al.

3.2 Identify Typologies To deal with production typologies from the qualitative analysis perspective, we considered combining the variables in the previous section that did not refer to costs, but to the actual structure of the problem. Thus we combined three variables: No. Items, di/pi, and Tsi. These variables were classified into different ranges. The following range was used for the number of different items to be processed on the machine: a few, some, or many items. To define the exactness of the value, we must consider the industry being operated in. By way of example, with an engraving machine, a few items would comprise five different items, some items would lie between five and 10, which would be the mean value, and more than 10 items would be considered many items. The combination of variables di/pi indicates how fast the machine works if the demand of products is taken into account. So when di/pi