Smart Decisions in Complex Systems

Pierre MASSOTTE, Pr. HDr.Ing., has long worked for IBM in Quality then Advanced Technologies (AoT), then as scientific director in EMEA Manufacturing, to improve European Manufacturing plants and Development Laboratories competitivity. Lately, he joined "Ecole des Mines d'Alès" as Deputy Director within the Nîmes EMA Laboratory. His research and development topics are related to complexity, self-organization, and issues on business competitiveness and sustainability in global companies. He is the co-author of several books in production systems management. He is now involved, as senior consultant, in various 'inclusive society' projects.Patrick CORSI , Dr. Ing. is an international consultant specialized in breakthrough design innovation processes. After an engineering and managerial career in industry with IBM Corp., IBM France, SYSECA/THOMSON-CSF and a successful start-up in artificial intelligence in Paris, he acted as a Project Officer within the European Commission in Brussels, specializing in R&D projects in advanced AI technologies. He is an ex-Associate Professor, a serial business books and eBooks author and a professional speaker, and an Associate Practitioner with Mines ParisTech in a large number of application domains.

• Preface xiiiAcknowledgments xviiList of Acronyms xixIntroduction xxvPart 1 1Chapter 1 The Foundations of Complexity 31.1 Complexities and simplexities: paradigms and perspectives 31.1.1 Positioning the problem 41.1.2 Reminders, basics and neologisms 51.1.3 What are the analytical steps in a complex system? 161.1.4 Organization and management principles in complex systems 311.1.5 Action and decision processes in self-organized systems 351.1.6 Notions of centralization and decentralization 361.2 What is the prerequisite for the handling of a complex system? 431.3 Applications: industrial complex systems 451.3.1 Distributed workshop management system 451.3.2 Analysis and diagnosis of a complex system 471.3.3 Some recommendations and comments to conclude 481.4 Time to conclude 501.4.1 Summary 501.4.2 Lessons and perspectives 51Part 2 53Chapter 2 Evidencing Field Complexity 552.1 Introduction 552.2 Qualitative study of deterministic chaos in a dynamic simple system 582.2.1 Description of a few simple cases 582.2.2 Initial conditions related to the emergence of chaos 592.2.3 Modeling and mathematical analysis of chaos 622.2.4 Application at the level of a simple cell 632.3 Test for the presence of deterministic chaos in a simple dynamic system 682.3.1 Characterization of the systems studied 692.3.2 A general question: is there deterministic chaos? 702.4 Properties of chaos in complex systems 772.4.1 Study of an elementary cell 772.4.2 Complex cellular systems 812.5 Effects of fractal chaos in “Complexity” theory 832.5.1 Organized complexity 832.5.2 Innovative complexity 842.5.3 Random complexity 852.5.4 Principles of implementation 872.6 Self-organization: relations and the role of chaos 872.6.1 Introduction 872.6.2 How to combine self-organization and chaos 882.6.3 Critical self-organized systems 892.6.4 Networked systems and co-operative systems 902.6.5 The three states of a dynamic complex system 932.6.6 Towards a typology of behavioral complexity 942.7 Applications: introduction of new concepts in systems 952.7.1 Questions on the management of complex industrial systems 952.7.2 Implementation of the concepts of chaos and self-organization 962.8 Conclusions 98Chapter 3 The New “Complex” Operational Context 1013.1 The five phases of economy – how everything accelerates at the same time 1013.2 The expected impact on just about everything 105Chapter 4 Taking Up Complexity 1094.1 Taking into account complex models 1094.1.1 A brief overview of the approach called “complexity” 1094.1.2 Another (bio-inspired) vision of the world: universality 1124.1.3 How to address complexity in this universal world? 1154.1.4 The usefulness of this book 1164.2 Economy and management of risks 1174.2.1 Important challenges to raise 1174.2.2 Adapted vocabulary that it is useful to adopt 1184.2.3 What do we mean by dynamic pricing? 119Part 3 121Chapter 5 Tackling Complexity with a Methodology 1235.1 Any methodology must first enrich the systemic interrelationships 1235.1.1 The innovation economy: the dynamic management of innovation 1245.1.2 A basic mechanism of efficient innovation 1255.1.3 The benefits of such a shift mechanism 1265.2 Towards a transdisciplinary co-economy 126Chapter 6 Management and Control of Complex Systems 1296.1 Introduction 1296.2 Complex systems: the alternatives 1326.2.1 Notions of sociability in agent communities 1326.2.2 The evolutionary principles of complex systems 1346.3 Control principles of production systems 1356.3.1 Introduction 1356.3.2 Control: by scheduling or by configuration? 1366.3.3 The tools used in monitoring and control 1406.4 PABADIS: an example of decentralized control 1416.4.1 Introduction 1416.4.2 Context and objectives of the PABADIS project 1426.4.3 Conceptual overview of PABADIS 1426.4.4 Principle of adopted convergence: the inverse solution 1446.4.5 Implementation 1456.5 Generalization of the concepts and mechanisms 1466.5.1 Introduction 1466.5.2 Allocation of resources: the agents in complex production systems 1476.5.3 Allocation of resources: the negotiation protocols 1476.5.4 Optimization of the resource allocation process 1486.6 A basic mechanism of control – the auction 1506.6.1 Introduction 1506.6.2 The mechanism of the auction 1516.6.3 Comparative review of the types of auctions 1536.6.4 Findings on the interest of the auction mechanism 1556.7 The control of self-organized systems 1566.7.1 Introduction 1566.7.2 The types and mechanisms of self-organization 1576.7.3 Towards a dynamic integrated model: Cellular Automata (CA) 1606.7.4 Self-organization: forms and configurations obtained 1656.7.5 Conclusion and implementation of the ACCA concept, a major model 167Chapter 7 Platforms for Taking up Complexity 1697.1 The VFDCS: a platform for implementation 1697.1.1 Controlling the phenomena of self-organization 1717.1.2 Methodology for implementation and the validation of concepts 1727.2 The application of VFDCS: the auction market 1747.2.1 The concept of the “Container” in the auction market 1767.2.2 Feedbacks and results 1767.2.3 Discussion 1787.3 The application of VFDCS: the virtual supply chain 1797.3.1 Introduction 1797.3.2 Architecture of the virtual supply chain 1817.3.3 Results and comments 1847.3.4 Conclusion 1857.3.5 Enhancement of the multi-agent platform 1867.4 General method for the control of systems 1867.4.1 Introduction 1867.4.2 Reminders and definitions 1877.4.3 Analytical approach to consistency 1887.4.4 Methods for the analysis and monitoring of performances 1897.4.5 Critical analysis of the convergence of configurations 1927.5 Conclusions and prospects 1947.5.1 Synthesis 1947.5.2 Discussion 1957.5.3 Comparison of approaches, tools and applications 1977.5.4 Results 199Part 4 201Introduction to Part 4 203Chapter 8 Applying Intrinsic Complexity: The Uberization of the Economy 2078.1 Preamble 2078.2 The context: new opportunities and new consumption needs 2078.3 The domains that are studied in this chapter 2088.4 Concepts, definitions and remainders 2098.4.1 Uberization 2098.4.2 Digitalization of the economy 2108.4.3 Collaborative consumption (CC) 2118.4.4 Model generalization: the sharing economy 2118.4.5 Participatory financing 2118.5 The business model and key elements 2138.5.1 Practicing networks 2138.5.2 Positive and negative impacts of network applications 2148.5.3 The problem of producer–consumers and consumer–producers 2158.5.4 Underlying mechanisms: some differences with the usual economic systems 2168.5.5 A form of social hypocrisy? 2178.5.6 Generalization: the management rules for P2P 2198.6 The problem of property and resource allocation 2208.6.1 The growing role of platforms 2208.6.2 The prisoner’s dilemma 2238.6.3 Games theory: an introduction 2248.6.4 Nonlinear models in game theory 2248.7 The uberization approach in context 2268.7.1 Simplexification 2278.7.2 Increasing complexity: the influence of cognitive approaches 2278.8 Generalization: the complexity of allocation problems 2308.9 Conclusion 234Chapter 9 Computer-assisted Production Management 2359.1 Introduction and reminders 2359.2 Intercommunication networks 2369.2.1 Notions of complexity in networks 2369.2.2 A few concepts of parallelism 2379.2.3 Elements of parallelism and associated architectures 2379.2.4 Transposition into industrial or social applications 2399.3 Communication network topologies 2409.3.1 Some characteristics of different network topologies 2419.3.2 Construction of a hypercube 2429.3.3 Notions of symmetry: cutting a hypercube 2439.3.4 The shortest path between two processors 2449.4 A few important properties 2449.5 Analysis of new concepts and methods in manufacturing sciences: instabilities, responsiveness and flexibility 2469.5.1 General approach: planning and scheduling 2479.5.2 Illustration in management systems 2479.5.3 Problems and remarks 2509.5.4 Improvements in planning and scheduling 2519.5.5 Improvements in configuration/reconfiguration 2529.5.6 Global improvements through simulation 2539.5.7 Inverse modeling and simulation 2549.6 New concepts for managing complex systems 2569.6.1 Traditional approach 2579.6.2 Recent improvements in the management of systems 2609.7 The change of conduct 2649.8 Improvements in manufacturing: process balancing 2669.9 Conclusion: main action principles in complex environments 267Chapter 10 Complexity and Cognitive Robotics 27110.1 Introduction 27110.2 The new industrial revolution 27210.3 The factory of the future: trend or revolution? 27210.4 Inputs for the factory of the future and their impact on the industry’s professions 27510.5 Conditions for success 27610.6 The data sciences 27710.6.1 Introduction to the characteristics of “Big Data” 27710.6.2 The problem of Big Data 27710.6.3 A new profession: the data scientist 27910.6.4 Some ask, how will this be possible? 27910.6.5 The field of large numbers 28010.7 A few technologies in data sciences 28110.7.1 The steps of reasoning based on the experience of the inductive approach and on the verification of hypotheses 28110.7.2 The “Lasso” method 28110.7.3 Kernel regression methods 28210.7.4 The random forests 28310.7.5 Neural networks 28410.7.6 Comments on clustering and graph partitioning issues 28610.7.7 Cognitive informatics – cognitivism 28610.8 Mechanisms of conventional cognitive engineering 28810.9 The new mechanisms of engineering 28910.9.1 Transduction 28910.9.2 Reasoning by constructed analogies 29010.10 The study of links and relationships in large databases 29010.10.1 Comment 29110.11 Application of cognitive robotics: the Watson platform 29110.11.1 Applications 29210.12 The impossibilities and unpredictabilities of complexity 29310.13 Current strategies of digitalization 29510.13.1 Reference examples and discussion 29610.13.2 Gnosis 29810.13.3 “Data is Centric” 29910.14 Conclusion: a maximum risk economy 300Bibliography 303Index 327