Benyoucef / Grabot Artificial Intelligence Techniques for Networked Manufacturing Enterprises Management

1. Auflage 2010
ISBN: 978-1-84996-119-6
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 508 Seiten

Reihe: Springer Series in Advanced Manufacturing

Artificial Intelligence Techniques for Networked Manufacturing Enterprises Management addresses prominent concepts and applications of AI technologies in the management of networked manufacturing enterprises. The aim of this book is to align latest practices, innovation and case studies with academic frameworks and theories, where AI techniques are used efficiently for networked manufacturing enterprises. More specifically, it includes the latest research results and projects at different levels addressing quick-response system, theoretical performance analysis, performance and capability demonstration. The role of emerging AI technologies in the modelling, evaluation and optimisation of networked enterprises' activities at different decision levels is also covered. Artificial Intelligence Techniques for Networked Manufacturing Enterprises Management is a valuable guide for postgraduates and researchers in industrial engineering, computer science, automation and operations research.

Dr. Lyes Benyoucef received his PhD in Operations Research at the National Polytechnic Institute of Grenoble, France, in 2000 and his HDR (Research Director Thesis) degree from the University of Metz, France, in 2008. He is a senior researcher (CR1-HDR) at INRIA (the French National Institute for Research in Computer Science and Control). His main research interests include modelling and performance evaluation; and the simulation and optimization of supply chains and e-sourcing technologies. Prof. Bernard Grabot teaches production management, industrial organization and ERP systems at the National Engineering School of Tarbes, France. He is a member of IFAC working groups on knowledge-based enterprise and editor-in-chief of the international journal, Engineering Applications of Artificial Intelligence. His main research interests concern the implementation of ERP systems, supply chain management and knowledge engineering.

Autoren/Hrsg.

Benyoucef, Lyes

Grabot, Bernard

Inhaltsverzeichnis

1;Series Editor’s Foreword;6
2;Preface;7
3;Acknowledgments;12
4;Contents;13
5;1. Intelligent Manufacturing Systems;21
5.1;1.1 Introduction;21
5.2;1.2 Traditional Manufacturing Systems;23
5.3;1.3 Changes in Manufacturing Systems: a Historical Perspective;26
5.4;1.4 Artificial Intelligence and Intelligent Manufacturing Systems;33
5.4.1;1.4.1 Technologies of Artificial Intelligence;33
5.4.2;1.4.2 Intelligent Manufacturing Systems;42
5.5;1.5 Properties of Intelligent Manufacturing Systems;45
5.6;1.6 Architecture of Intelligent Manufacturing Systems;47
5.7;1.7 Holonic Manufacturing Systems;53
5.8;1.8 Applications of Intelligent Manufacturing Systems;57
5.9;1.9 Conclusions;59
5.10;References;59
6;2. Agent-based System for Knowledge Acquisition and Management Within a Networked Enterprise;62
6.1;2.1 Agent-based and Related Systems;62
6.1.1;2.1.1 Origins of Agent Research;63
6.1.2;2.1.2 Definition of an Agent;65
6.1.3;2.1.3 Agent Architectures;66
6.1.4;2.1.4 Agent Types and Applications;69
6.1.5;2.1.5 Machine Learning for Generation of Knowledge Bases;72
6.2;2.2 Product Fault Knowledge Acquisition and Management;76
6.2.1;2.2.1 Automating Knowledge Base Management;76
6.2.2;2.2.2 Analysis of Knowledge Base Management Process;77
6.3;2.3 Agent System for Knowledge Acquisition and Management;82
6.3.1;2.3.1 User Agent;84
6.3.2;2.3.2 Server Agent;89
6.3.3;2.3.3 Testing;98
6.4;2.4 Conclusions;100
6.5;References;100
7;3. Multi-agent Simulation-based Decision Support System and Application in Networked Manufacturing Enterprises;106
7.1;3.1 Introduction;106
7.2;3.2 Literature Review;107
7.2.1;3.2.1 Simulation Methods;108
7.2.2;3.2.2 Multi-agent Simulation;108
7.2.3;3.2.3 Tools and Applications;109
7.3;3.3. Problem Description and Approach;110
7.3.1;3.3.1 Platform Architecture;111
7.3.2;3.3.2 Multi-agent Supply Chain Simulation Model;112
7.4;3.4 Modelling and Analysis;113
7.4.1;3.4.1 Baseline Simulation Model;114
7.4.2;3.4.2 Scenario Analysis;114
7.4.3;3.4.3 Inventory Control Policy;115
7.4.4;3.4.4 Forecast Accuracy;116
7.4.5;3.4.5 Build-to-plan vs. Build-to-order;118
7.4.6;3.4.6 Procurement Policy;119
7.4.7;3.4.7 Truck Utilization;120
7.5;3.5 Conclusions and Perspectives;122
7.6;References;123
8;4. A Collaborative Decision-making Approach for Supply Chain Based on a Multi-agent System;125
8.1;4.1 Introduction;126
8.2;4.2 Distributed Simulation and Supply Chain Management;128
8.2.1;4.2.1 Decision-making and Multi-agent Approaches;129
8.2.2;4.2.2 Multi-agent Systems Simulation;129
8.3;4.3 The Supply Chain Modeling;130
8.3.1;4.3.1 Supply Chain Modeling Methodology;131
8.3.2;4.3.2 The Safety Inventory Case Study;132
8.4;4.4 The Multi-agent Architecture;134
8.4.1;4.4.1 The Modeling Principle;134
8.4.2;4.4.2 Architecture;135
8.4.3;4.4.3 Negotiation Protocols;138
8.5;4.5 Industrial Case Study;140
8.6;4.6 Conclusion and Perspectives;143
8.7;References;144
9;5. Web-service-based e-Collaborative Framework to Provide Production Control with Effective Outsourcing;146
9.1;5.1 Introduction;147
9.2;5.2 Literature Review;148
9.3;5.3 Design of Web-services-based e-Collaborative Framework;150
9.3.1;5.3.1 The Proposed Framework;150
9.3.2;5.3.2 Design of System Components of Model;151
9.3.3;5.3.3 The Mathematical Model;156
9.3.4;5.3.4 Overall Objective Function;159
9.4;5.4 Protocols of Functional Agents;160
9.4.1;5.4.1 Machine Agent Protocol;160
9.4.2;5.4.2 Production Controlling Agent Protocol;161
9.4.3;5.4.3 Task Management Agent Protocol;162
9.4.4;5.4.4 Hierarchy of Steps in Web-service-based e-collaborative System;162
9.5;5.5 Results and Discussion;162
9.5.1;5.5.1 Sample-sort Simulated Annealing Algorithm;165
9.5.2;5.5.2 Experimental Analysis;165
9.6;5.6 Conclusions;174
9.7;References;175
10;6. Isoarchic and Multi-criteria Control of Supply Chain Network;177
10.1;6.1 Introduction;177
10.2;6.2 Supply Chain Management Limits;179
10.3;6.3 Control of a Dynamic Logistic Network: Isoarchic and Multi-criteria Control;180
10.3.1;6.3.1 Description of the Interacting Entities;181
10.3.2;6.3.2 Definition of Self-organized Control;182
10.3.3;6.3.3 Support Structure: Autonomous Control Entity;187
10.4;6.4 Distributed Simulation of a Dynamic Logistic Network;190
10.4.1;6.4.1 High-level Architecture Components;190
10.4.2;6.4.2 Integration of the DEVS-ACE Models in High-level Architecture Environment;191
10.5;6.5 Experiments via Simulation;191
10.6;6.6 Conclusion and Future Work;194
10.7;References;195
11;7. Supply Chain Management Under Uncertainties: Lot-sizing and Scheduling Rules;197
11.1;7.1 Introduction;197
11.2;7.2 Short Presentation of Lot-sizing and Scheduling Problems;200
11.2.1;7.2.1 Basic Models and Extensions;203
11.2.2;7.2.2 Lot-sizing and Scheduling Under Uncertainties;207
11.2.3;7.2.3 Optimization Techniques;212
11.3;7.3 A Case Study;213
11.3.1;7.3.1 Description of the Case Study;213
11.3.2;7.3.2 Representation of Uncertainties;216
11.3.3;7.3.3 Objective Function;218
11.3.4;7.3.4 Decomposition for Optimization;219
11.3.5;7.3.5 Numerical Example;226
11.3.6;7.3.6 Numerical Results;230
11.4;7.4 Conclusions;232
11.5;References;233
12;8. Meta-heuristics for Real-time Routing Selection in Flexible Manufacturing Systems;237
12.1;8.1 Introduction;238
12.2;8.2 Literature Review;239
12.3;8.3 Job-shop;241
12.3.1;8.3.1 Job-shop Problem;241
12.3.2;8.3.2 Simulation of a Flexible Manufacturing System Environment;242
12.4;8.4 Dissimilarity Maximization Method and Modified DMM Rules;244
12.4.1;8.4.1 Dissimilarity Maximization Method for Real-time Routing Selection;244
12.4.2;8.4.2 Modified Dissimilarity Maximization Method for Real-time Routing Selection;245
12.5;8.5 Meta-heuristics for Job-shop Routing;246
12.5.1;8.5.1 Ant Colony Optimization;246
12.5.2;8.5.2 Simulated Annealing;247
12.5.3;8.5.3 Particle Swarm Optimization;248
12.5.4;8.5.4 Genetic Algorithms;250
12.5.5;8.5.5 Taboo Search;251
12.5.6;8.5.6 Electromagnetism-like Method;252
12.6;8.6 Performance Evaluation of Routing Selection Methods;253
12.6.1;8.6.1 System Simulation Without Presence of Breakdown;253
12.6.2;8.6.2 System Simulation With Presence of Breakdown;257
12.7;8.7 Conclusions;262
12.8;References;263
13;9. Meta-heuristic Approaches for Multi-objective Simulation-based Optimization in Supply Chain Inventory Management;265
13.1;9.1 Introduction;265
13.2;9.2 Literature Review;267
13.3;9.3 Problem Formulation;269
13.4;9.4 Implementation of Selected Evolutionary Algorithms;272
13.4.1;9.4.1 Non-dominated Sorting Genetic Algorithm II;272
13.4.2;9.4.2 Strength Pareto Evolutionary Algorithm II;274
13.4.3;9.4.3 Strength Pareto Evolutionary Algorithm IIb;275
13.4.4;9.4.4 Multi-objective Particle Swarm Optimization;276
13.5;9.5 Computational Experiments and Analysis;278
13.5.1;9.5.1 Evaluation Criteria;278
13.5.2;9.5.2 Parameters Used;279
13.5.3;9.5.3 Experimental Results;279
13.6;9.6 Conclusion;283
13.7;References;284
14;10. Diverse Risk/Cost Balancing Strategies for Flexible Tool Management in a Supply Network;286
14.1;10.1 Introduction;286
14.2;10.2 Multi-agent Tool Management System;287
14.3;10.3 Flexible Tool Management Strategies;290
14.3.1;10.3.1 Current Inventory-Level Decision Making (INVADAPT);292
14.3.2;10.3.2 Fixed-Horizon Inventory-Level Decision Making (I-FUTURE);294
14.3.3;10.3.3 Variable-Horizon Inventory-Level Decision Making (INVADAPT_NS);295
14.4;10.4 Tool Inventory Management Simulations;296
14.4.1;10.4.1 Simulation Through INVADAPT;302
14.4.2;10.4.2 Simulation Through I-FUTURE;309
14.4.3;10.4.3 Simulation Through INVADAPT_NS;316
14.5;10.5 Test Case Applications;324
14.6;10.6 Conclusions and Future Work;327
14.7;References;327
15;11. Intelligent Integrated Maintenance Policies for Manufacturing Systems;329
15.1;11.1 Introduction;329
15.2;11.2 Optimal Maintenance Policy Considering the Influence of the Production Plan on the Deterioration of the Manufacturing System;332
15.2.1;11.2.1 Problem Statement;332
15.2.2;11.2.2 Problem Formulation;333
15.2.3;11.2.3 Influence of Manufacturing System Deterioration on the Optimal Production Plan;338
15.2.4;11.2.4 Optimization of the Maintenance Policy;340
15.3;11.3 Intelligent Periodic Preventive Maintenance Policy in Finite Horizon with an Adaptive Failure Law;344
15.3.1;11.3.1 Problem Description;344
15.3.2;11.3.2 Model Formulation and Intelligent Determination of the Optimal Solution;346
15.3.3;11.3.3 Numerical Example;349
15.4;11.4 Conclusions;352
15.5;References;353
16;12. Enhancing the Effectiveness of Multi-pass Scheduling Through Optimization via Simulation;354
16.1;12.1 Introduction;355
16.2;12.2 Multi-pass Scheduling Using Nested Partitions and Optimal Computing Budget Allocation;359
16.2.1;12.2.1 The Proposed Multi-pass Scheduling Framework;359
16.2.2;12.2.2 The Outer Loop: Nested Partitions;359
16.2.3;12.2.3 The Inner Loop: Optimal Computing Budget Allocation;361
16.3;12.3 Implementation of Nested Partitions and Optimal Computing Budget Allocation;362
16.3.1;12.3.1 Nested Partitions: Partitioning Strategy;362
16.3.2;12.3.2 Nested Partitions: Sampling Strategy;364
16.3.3;12.3.3 Nested Partitions: Backtracking Strategy;364
16.3.4;12.3.4 Optimal Computing Budget Allocation: Ranking and Selection Strategy;365
16.3.5;12.3.5 Performance of Nested Partitions and Optimal Computing Budget Allocation;366
16.4;12.4 Experimental Design and Analysis;367
16.4.1;12.4.1 Experimental Assumptions;367
16.4.2;12.4.2 Experimental Design;369
16.4.3;12.4.3 Experimental Results for the Probability of Correct Selection;370
16.5;12.5 Conclusions;376
16.6;References;377
17;13. Intelligent Techniques for Safety Stock Optimization in Networked Manufacturing Systems;379
17.1;13.1 Introduction;379
17.2;13.2 Multi-echelon Inventory Control in Networked Manufacturing Systems;381
17.3;13.3 Literature Review;386
17.4;13.4 System Safety Stock Optimization in n-echelon Distribution Systems;389
17.4.1;13.4.1 Introduction of a Two-echelon Distribution System;390
17.4.2;13.4.2 Characterization of the Back-order Service Time in the Central Warehouse;392
17.4.3;13.4.3 Characterization of the Actual Retailer Replenishment Lead Time;394
17.4.4;13.4.4 System Safety Stock Optimization in a Two-echelon Distribution System;396
17.5;13.5 System Safety Stock Optimization in n-echelon Assembly Systems;398
17.5.1;13.5.1 Introduction of a Two-echelon Assembly System;398
17.5.2;13.5.2 Characterization of the Back-order Service Time for a Subset of Components;401
17.5.3;13.5.3 Characterization of the Incoming Service Time to the Assembly;402
17.5.4;13.5.4 Characterization of the Actual Assembly Lead Time;403
17.5.5;13.5.5 System Safety Stock Optimization in a Two-echelon Assembly System;404
17.5.6;13.5.6 Distribution System as Special Case of the Assembly System;406
17.6;13.6 System Safety Stock Optimization in Generic Networks;407
17.6.1;13.6.1 Introduction of a Spanning Tree System;407
17.6.2;13.6.2 Characterization of Actual Replenishment Lead Times in a Spanning Tree System;414
17.6.3;13.6.3 System Safety Stock Optimization for a Spanning Tree System;427
17.6.4;13.6.4 Extension to Generic Systems;428
17.7;13.7 Conclusions;430
17.8;References;431
18;14. Real-world Service Interaction with Enterprise Systems in Dynamic Manufacturing Environments;434
18.1;14.1 Motivation;435
18.2;14.2 Real-world Awareness;437
18.2.1;14.2.1 Device Integration Protocols;437
18.2.2;14.2.2 Device-to-Business Coupling;439
18.2.3;14.2.3 Integrating Heterogeneous Devices;440
18.3;14.3 Enterprise Integration;441
18.4;14.4 Integrating Manufacturing Equipment with the SOCRADES Integration Architecture;443
18.5;14.5 Towards Dynamic Adaptation;446
18.5.1;14.5.1 Simulation;448
18.5.2;14.5.2 Self-healing Mechanisms;449
18.5.3;14.5.3 Self-optimizing Mechanisms;450
18.6;14.6 Concept Validation in Prototypes;451
18.6.1;14.6.1 Machine Monitoring, Dynamic Decision and OrderAdaptation;452
18.6.2;14.6.2 The Future Shop Floor: Mashup of Heterogeneous Service-oriented-architecture Devices and Services;455
18.6.3;14.6.3 Dynamic Supply Chain Management Adaptation;457
18.6.4;14.6.4 Taming Protocol Heterogeneity for Enterprise Services;460
18.6.5;14.6.5 Energy Monitoring and Control via Representational State Transfer;462
18.7;14.7 Discussion and Future Directions;465
18.8;14.8 Conclusions and Future Work;466
18.9;References;466
19;15. Factory of the Future: A Service-oriented System of Modular, Dynamic Reconfigurable and Collaborative Systems;469
19.1;15.1 Introduction;470
19.2;15.2 The Emergence of Cooperating Objects;471
19.3;15.3 The Cross-layer Service-oriented-architecture-driven Shop Floor;473
19.4;15.4 Dynamic Reconfiguration of a Service-oriented-architecture-based Collaborative Shop Floor;475
19.4.1;15.4.1 Methodology;476
19.4.2;15.4.2 Example;478
19.5;15.5 Analysis Behind the Engineering Methods and Tools;479
19.5.1;15.5.1 Applying Functional Analysis to Validate Service Composition Paths in High-level Petri-net-based Orchestration Models;479
19.5.2;15.5.2 Example;482
19.6;15.6 A Service-oriented-architecture-based Collaborative Production Management and Control System Engineering Application;484
19.7;15.7 Conclusions and Future Work;489
19.8;References;490
20;16. A Service-oriented Shop Floor to Support Collaboration in Manufacturing Networks;492
20.1;16.1 Introduction;492
20.2;16.2 Agility in Manufacturing;494
20.3;16.3 Collaborative Networks;496
20.4;16.4 Service-oriented Architecture as a Method to Support Agility and Collaboration;497
20.5;16.5 Architecture;502
20.5.1;16.5.1 The Role of Componentization;502
20.5.2;16.5.2 Service Exposure, Composition and Aggregation;504
20.5.3;16.5.3 The Role of Orchestration in Control, Monitoring and Diagnosis;505
20.6;16.6 An Implementation;507
20.6.1;16.6.1 Manufacturing Device Service Implementation;507
20.6.2;16.6.2 Coalition Leader Service Implementation;507
20.6.3;16.6.3 Application Example: a Collaborative Pick-and-place Operation;508
20.7;16.7 Conclusions;510
20.8;References;511
21;Index;513

Leseproben

"Chapter 10 Diverse Risk/Cost Balancing Strategies for Flexible Tool Management in a Supply Network (p. 271-272)

D. D’Addona and R. Teti

Abstract This work is a part of a wider-scope research concerned with the development and implementation of a multi-agent tool management system (MATMS) for automatic tool procurement. The design, functioning, and performance of diverse flexible tool management strategies integrated in the MATMS is illustrated here. The MATMS operates in the frame of a negotiation based multiple-supplier network where a turbine blade producer (customer) requires from external tool manufacturers (suppliers) the performance of dressing operations on worn-out cubic boron nitride grinding wheels for nickel base alloy turbine blade fabrication. The diverse FTMS paradigms, configured as domain-specific problem-solving functions operating within the MATMS intelligent agent holding the responsibility for optimum tool inventory sizing and control, have been tested by tool inventory management simulations and comparison with real industrial cases. Keywords Tool management, inventory control, multi-agent systems, supply networks

10.1 Introduction

In recent times, novel software architecture to manage supply networks at the tactical and operational levels has emerged. The supply network is viewed as a system made of a set of intelligent (software) agents, each responsible for one or more activities in the supply network and each interacting with other agents in planning and executing their responsibilities (Fox et al., 2000).

The adoption of agent-based or multi-agent technology is founded on the three main system domain characteristics (Yuan et al., 2001): data, control, expertise or resources are inherently distributed; the system is naturally regarded as a society of autonomous cooperating components; the system contains legacy components that must interact with other, possibly new software components. Supply network management by its very nature has all the above domain characteristics (Sycara, 1998).

A supply network consists of suppliers, factories, warehouses, etc., working together to fabricate products and deliver them to customers. Parties involved in the supply network have their own resources, capabilities, tasks, and objectives. They cooperate with each other autonomously to serve common goals but also have their own interests.

A supply network is dynamic and involves the constant flows of information and materials across multiple functional areas both within and between network members. Multi-agent technology therefore appears to be particularly suitable to support collaboration in supply network management."