Kordon Applying Computational Intelligence

How to Create Value
1. Auflage 2009
ISBN: 978-3-540-69913-2
Verlag: Springer Berlin Heidelberg
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 470 Seiten

In theory, there is no difference between theory and practice. But, in practice, there is. Jan L. A. van de Snepscheut The ?ow of academic ideas in the area of computational intelligence has penetrated industry with tremendous speed and persistence. Thousands of applications have proved the practical potential of fuzzy logic, neural networks, evolutionary com- tation, swarm intelligence, and intelligent agents even before their theoretical foundation is completely understood. And the popularity is rising. Some software vendors have pronounced the new machine learning gold rush to 'Transfer Data into Gold'. New buzzwords like 'data mining', 'genetic algorithms', and 'swarm optimization' have enriched the top executives' vocabulary to make them look more 'visionary' for the 21st century. The phrase 'fuzzy math' became political jargon after being used by US President George W. Bush in one of the election debates in the campaign in 2000. Even process operators are discussing the perf- mance of neural networks with the same passion as the performance of the Dallas Cowboys. However, for most of the engineers and scientists introducing computational intelligence technologies into practice, looking at the growing number of new approaches, and understanding their theoretical principles and potential for value creation becomes a more and more dif?cult task.

Arthur K. Kordon is a Data Mining and Modeling Leader in the Data Mining and Modeling Capability of The Dow Chemical Company. He is an internationally recognized expert in applying emerging technologies in industry, and has given talks and chaired panels on the topic at the major computational intelligence conferences such as WCCI and GECCO. He has successfully introduced several novel technologies for improved manufacturing and new product design in the chemical industry, and his research interests include application issues of computational intelligence, robust empirical modeling, intelligent process monitoring and control, and data mining.

Autoren/Hrsg.

Kordon, Arthur

Inhaltsverzeichnis

1;FM;2
1.1;Outline placeholder;1
1.1.1;Motivation;7
1.1.2;Purpose of the Book;8
1.1.3;Who Is This Book for?;10
1.1.4;How This Book Is Structured;12
1.1.5;What This Book Is NOT About;13
1.1.6;Features of the Book;14
2;p1;21
2.1;Part I: Computational Intelligence in a Nutshell;21
3;144233_1_En_1_Chapter_OnlinePDF;22
3.1;Chapter 1: Artificial vs. Computational Intelligence;22
3.1.1;Artificial Intelligence: The Pioneer;23
3.1.1.1;Practical Definition of Applied Artificial Intelligence;23
3.1.1.2;Key Practical Artificial Intelligence Approaches;24
3.1.1.2.1;ExpertExpert Systems (RuleRule-Based and Frame-Based);25
3.1.1.2.1.1;RuleRule-Based ExpertExpert Systems;25
3.1.1.2.1.2;Frame-Based ExpertExpert Systems;26
3.1.1.2.2;Inference Mechanisms;26
3.1.1.2.3;Backward Chaining;27
3.1.1.2.4;Forward Chaining;27
3.1.1.2.5;Case-Based ReasoningReasoning;27
3.1.1.2.6;Knowledge Management;28
3.1.1.3;Applied Artificial Intelligence Success Stories;30
3.1.1.3.1;Integrated SoftwareSoftware InfrastructureInfrastructure for Applied AIApplied AI;30
3.1.1.3.2;Technical Advantages of Applied AIApplied AI;30
3.1.1.3.3;Domain Expertise Is Captured;31
3.1.1.3.4;Knowledge Is Presented in Natural Language;31
3.1.1.3.5;RuleRule Structure Is Uniform;31
3.1.1.3.6;Interpretive Capability;31
3.1.1.3.7;Separation of Knowledge from Inference Mechanisms;32
3.1.1.3.8;Application Areas of AI;32
3.1.1.3.9;Advisory SystemsAdvisory systems;32
3.1.1.3.10;Decision-MakingDecision-making;33
3.1.1.3.11;PlanningPlanning;33
3.1.1.3.12;Selection;33
3.1.1.3.13;Diagnostics;33
3.1.1.3.14;Preserving Knowledge;34
3.1.1.3.15;Examples of Successful AI Real-World Applications;34
3.1.1.4;Applied Artificial Intelligence Issues;36
3.1.1.4.1;Technical Issues of Applied AI;36
3.1.1.4.1.1;Knowledge ConsistencyKnowledge consistency;36
3.1.1.4.1.2;Scale-upScale-up;36
3.1.1.4.1.3;Static Nature, No LearningLearning;36
3.1.1.4.1.4;Subjective Nature of Representing Intelligence;37
3.1.1.4.2;InfrastructureInfrastructure Issues of Applied AI;37
3.1.1.4.2.1;Limited Computer Capabilities in the Early 1980s;37
3.1.1.4.2.2;High Total Cost of Ownership;37
3.1.1.4.3;People Issues of Applied AI;38
3.1.1.4.3.1;Knowledge Extraction;38
3.1.1.4.3.2;Incompetence DistributionIncompetence distribution;38
3.1.1.4.3.3;Legal Issues;39
3.1.1.5;Artificial Intelligence Application Lessons;39
3.1.1.5.1;Application AI Lesson 1: Do not create unrealistic expectationsUnrealistic expectations;39
3.1.1.5.2;Application AI Lesson 2: Do not push new technologies by campaigns;39
3.1.1.5.3;Application AI Lesson 3: Do not underestimate maintenance and support;40
3.1.1.5.4;Application AI Lesson 4: Clarify and demonstrate value as soon as possible;40
3.1.1.5.5;Application AI Lesson 5: Develop a strategy for sustainable application growth;40
3.1.1.5.6;Application AI Lesson 6: Link the success of the application with incentives to all stakeholders;41
3.1.2;Computational Intelligence: The Successor;41
3.1.2.1;Practical Definition of Applied Computational Intelligence;42
3.1.2.2;Key Computational Intelligence Approaches;43
3.1.2.2.1;Fuzzy SystemsFuzzy Systems;44
3.1.2.2.2;Neural NetworkNeural networks;44
3.1.2.2.3;Support Vector Machines;44
3.1.2.2.4;Evolutionary ComputationEvolutionary computation;45
3.1.2.2.5;Swarm Intelligence;45
3.1.2.2.6;Intelligent Agents;45
3.1.3;Key Differences Between AI and CI;46
3.1.3.1;Key Technical Differences;46
3.1.3.1.1;Key Difference #1 - On the main source of representing intelligence;46
3.1.3.1.2;Key Difference #2 - On the mechanisms of processing intelligence;46
3.1.3.1.3;Key Difference #3 - On the interactions with the environment;47
3.1.3.2;The Ultimate Difference;47
3.1.3.3;An Integrated View;48
3.1.4;Summary;48
3.1.5;The Bottom Line;49
3.1.6;Suggested Reading;49
4;144233_1_En_2_Chapter_OnlinePDF;50
4.1;Chapter 2: A Roadmap Through the Computational Intelligence Maze;50
4.1.1;Strengths and Weaknesses of CI Approaches;50
4.1.1.1;Strengths and Weaknesses of Fuzzy Systems;51
4.1.1.2;Strengths and Weaknesses of Neural Networks;53
4.1.1.3;Strengths and Weaknesses of Support Vector Machines;54
4.1.1.4;Strengths and Weaknesses of Evolutionary Computation;56
4.1.1.5;Strengths and Weaknesses of Swarm IntelligenceSwarm Intelligence;58
4.1.1.6;Strengths and Weaknesses of Intelligent Agents;59
4.1.2;Key Scientific Principles of Computational Intelligence;61
4.1.2.1;Bio-inspired Computing;61
4.1.2.2;LearningLearning Systems;63
4.1.2.3;Computer Science;64
4.1.3;Key Application Areas of Computational Intelligence;64
4.1.3.1;InventionInvention of New Products;64
4.1.3.2;Systems Design;66
4.1.3.3;ManufacturingManufacturing;66
4.1.3.4;Supply Chain;66
4.1.3.5;Market Analysis;67
4.1.3.6;Financial Modeling;67
4.1.3.7;Modeling Social Behavior;68
4.1.3.8;Health;68
4.1.3.9;LeisureLeisure;68
4.1.4;Summary;69
4.1.5;The Bottom Line;69
4.1.6;Suggested Reading;69
5;144233_1_En_3_Chapter_OnlinePDF;70
5.1;Chapter 3: Let's Get Fuzzy;70
5.1.1;Fuzzy Systems in a Nutshell;70
5.1.1.1;Dealing With AmbiguityAmbiguity;71
5.1.1.2;Fuzzy Sets;73
5.1.1.3;Fuzzy Systems Created By Experts;74
5.1.1.4;Fuzzy Systems Created by DataData;77
5.1.2;Benefits of Fuzzy Systems;79
5.1.3;Fuzzy Systems Issues;81
5.1.4;How to Apply Fuzzy Systems;82
5.1.4.1;When Do We Need Fuzzy Systems?;82
5.1.4.2;Applying ExpertExpert-Based Fuzzy Systems;82
5.1.4.3;Applying DataData-Based Fuzzy Systems;83
5.1.5;Typical Applications of Fuzzy Systems;84
5.1.6;Fuzzy Systems MarketingMarketing;87
5.1.7;Available Resources for Fuzzy Systems;89
5.1.7.1;Key Websites;89
5.1.7.2;Selected SoftwareSoftware;90
5.1.8;Summary;90
5.1.9;The Bottom Line;90
5.1.10;Suggested Reading;91
6;144233_1_En_4_Chapter_OnlinePDF;92
6.1;Chapter 4: Machine LearningLearning: The Ghost in the Learning Machine;92
6.1.1;Neural Networks in a Nutshell;95
6.1.1.1;Biological Neurons and Neural Networks;95
6.1.1.2;Artificial Neurons and Neural Networks;96
6.1.1.3;Back-propagation;99
6.1.1.4;Neural Network Structures;101
6.1.2;Support Vector Machines in a Nutshell;103
6.1.2.1;Statistical LearningLearning Theory;104
6.1.2.2;Structural Risk Minimization;105
6.1.2.3;Support Vector Machines for ClassificationClassification;106
6.1.2.4;Support Vector Machines for RegressionRegression;109
6.1.3;Benefits of Machine LearningLearning;110
6.1.3.1;Comparison Between Neural Networks and SVM;110
6.1.3.1.1;Key Difference #1 - On the method´s basis;111
6.1.3.1.2;Key Difference #2 - On the necessary dataData for model developmentModel;111
6.1.3.1.3;Key Difference #3 - On the optimization typeOptimization;111
6.1.3.1.4;Key Difference #4 - On the generalization capability;112
6.1.3.1.5;Key Difference #5 - On the number of modelModel outputs;112
6.1.3.2;Benefits of Neural Networks;112
6.1.3.3;Benefits of Support Vector Machines;113
6.1.4;Machine LearningLearning Issues;115
6.1.4.1;Neural Networks Issues;115
6.1.4.2;Support Vector Machines Issues;115
6.1.5;How to Apply Machine LearningLearning Systems;116
6.1.5.1;When Do We Need Machine LearningLearning?;116
6.1.5.2;Applying Machine LearningLearning Systems;117
6.1.5.3;Applying Neural Networks: An Example;119
6.1.5.4;Applying Support Vector Machines: An Example;122
6.1.6;Typical Machine LearningLearning Applications;124
6.1.6.1;Typical Applications of Neural Networks;124
6.1.6.2;Typical Applications of Support Vector Machines;127
6.1.7;Machine LearningLearning MarketingMarketing;127
6.1.7.1;Neural Networks MarketingMarketing;127
6.1.7.2;Support Vector Machines MarketingMarketing;129
6.1.8;Available Resources for Machine LearningLearning;130
6.1.8.1;Key Websites;130
6.1.8.2;Key SoftwareSoftware;130
6.1.9;Summary;131
6.1.10;Suggested Reading;132
7;144233_1_En_5_Chapter_OnlinePDF;133
8;144233_1_En_6_Chapter_OnlinePDF;163
9;144233_1_En_7_Chapter_OnlinePDF;193
9.1;Chapter 7: Intelligent Agents: The Computer Intelligence Agency (CIA);193
9.1.1;Intelligent Agents in a Nutshell;194
9.1.1.1;Complex Systems;195
9.1.1.2;Intelligent Agents;197
9.1.1.3;Agent-Based Integrators;199
9.1.1.4;Agent-Based Systems;201
9.1.2;Benefits of Intelligent Agents;204
9.1.2.1;Comparison Between Agents and Objects;204
9.1.2.2;Comparison Between Agents and ExpertExpert Systems;205
9.1.2.3;Benefits of Agent-Based Modeling;205
9.1.3;Intelligent Agents Issues;207
9.1.4;How to Apply Intelligent Agents;208
9.1.4.1;When Do We Need Intelligent Agents?;208
9.1.4.2;Applying Intelligent Agents;209
9.1.5;Typical Applications of Intelligent Agents;211
9.1.6;Intelligent Agents MarketingMarketing;214
9.1.7;Available Resources for Intelligent Agents;217
9.1.7.1;Key Websites;217
9.1.7.2;Selected SoftwareSoftware;217
9.1.8;Summary;217
9.1.9;Suggested Reading;218
10;p2;219
10.1;Part II: Computational Intelligence Creates Value;219
11;144233_1_En_8_Chapter_OnlinePDF;220
11.1;Chapter 8: Why We Need Intelligent Solutions;220
11.1.1;Beat CompetitionCompetition;221
11.1.1.1;Effective Utilization of Emerging Technologies;221
11.1.1.2;Fast Response to Changing Environment;222
11.1.1.3;Effective Operation in the Global Economy;222
11.1.1.4;Flexible Strategy;223
11.1.1.5;Low Cost of Operation;224
11.1.2;Accelerate Innovations;224
11.1.2.1;Business Impact Analysis of InnovationInnovation;225
11.1.2.2;Automatic Novelty GenerationGeneration;225
11.1.2.3;Rapid ExplorationExploration of New Ideas;226
11.1.2.4;Fast CommercializationCommercialization in Practice;226
11.1.3;Produce Efficiently;227
11.1.3.1;Accurate Production PlanningPlanning;227
11.1.3.2;Enhanced ObservabilityObservability of Processes;228
11.1.3.3;Broad Product and Process OptimizationOptimization;228
11.1.3.4;Advanced Process Control;229
11.1.3.5;Improved Operating Discipline;229
11.1.4;Distribute Effectively;229
11.1.4.1;Estimate Demand;230
11.1.4.2;Handle Global Market ComplexityComplexity;230
11.1.4.3;Real-Time Operation;231
11.1.4.4;Optimal SchedulingScheduling;231
11.1.5;Impress Customers;232
11.1.5.1;Analyze Customers;232
11.1.5.2;Deliver Simple Solutions;233
11.1.5.3;Create a Visionary Image;233
11.1.5.4;Broaden Customer Base;234
11.1.6;Enhance CreativityCreativity;234
11.1.6.1;Reduce Routine Operations Related to Intelligence;235
11.1.6.2;Magnify Imagination;235
11.1.6.3;Add Intellectual Sensors;236
11.1.6.4;Increase Cognitive Productivity;236
11.1.7;Attract Investors;237
11.1.7.1;Intelligence and Growth;237
11.1.7.2;High-Tech Magnetism;238
11.1.7.3;Technology Credibility;238
11.1.7.4;Technology SustainabilitySustainability;239
11.1.8;Improve National Defense;239
11.1.8.1;Intelligent Intelligence;240
11.1.8.2;RobotRobot Soldiers;240
11.1.8.3;Smart Weapons;241
11.1.8.4;Cyber Wars;241
11.1.9;Protect Health;242
11.1.9.1;Medical Diagnosis;242
11.1.9.2;Personal Health Modeling;243
11.1.9.3;Health Monitoring;244
11.1.9.4;Personal Health Advisor;244
11.1.10;Have Fun;245
11.1.10.1;Intelligent GamesGames;245
11.1.10.2;Funny Education;246
11.1.10.3;Smart Toys;246
11.1.10.4;Evolutionary Art;247
11.1.10.5;Virtual HollywoodVirtual Hollywood;247
11.1.11;Summary;248
11.1.12;The Bottom Line;248
11.1.13;Suggested Reading;248
12;144233_1_En_9_Chapter_OnlinePDF;249
12.1;Chapter 9: Competitive Advantages of Computational Intelligence;249
12.1.1;Competitive Advantage of a Research Approach;249
12.1.1.1;Step 1: Clarify Technical Superiority;250
12.1.1.2;Step 2: Demonstrate Low Total Cost of Ownership;251
12.1.1.3;Step 3: Apply in Areas with High Impact;252
12.1.2;Key Competitive Approaches to Computational Intelligence;253
12.1.2.1;Competitor #1: First-Principles Modeling;254
12.1.2.2;Competitor #2: Statistical Modeling;257
12.1.2.3;Competitor #3: HeuristicsHeuristics;259
12.1.2.4;Competitor #4: Classical OptimizationOptimization;261
12.1.3;How Computational Intelligence Beats the CompetitionCompetition;263
12.1.3.1;Creating ``Objective Intelligence´´;263
12.1.3.2;Dealing with UncertaintyUncertainty;265
12.1.3.3;Dealing with ComplexityComplexity;267
12.1.3.4;Generating Novelty;268
12.1.3.5;Low-Cost Modeling and OptimizationOptimization;270
12.1.4;Summary;271
12.1.5;Suggested Reading;272
13;144233_1_En_10_Chapter_OnlinePDF;273
13.1;Chapter 10: Issues in Applying Computational Intelligence;273
13.1.1;Technology Risks;273
13.1.1.1;The Change Function;274
13.1.1.2;Technocentric Culture;275
13.1.1.3;Increased ComplexityComplexity;276
13.1.1.4;Technology Hype;276
13.1.2;Modeling Fatigue;277
13.1.2.1;The Invasion of First-Principles Models;277
13.1.2.2;Statistical Models Everywhere;278
13.1.2.3;How to Lie with AI;278
13.1.2.4;Anything but ModelModel (ABM) Movement;278
13.1.3;Looks Too Academic;279
13.1.3.1;Difficult to Understand;279
13.1.3.2;Diverse Approaches;280
13.1.3.3;Difficult to Track;280
13.1.3.4;It's Not Yet Ready for IndustryIndustry;281
13.1.4;Perception of High Cost;281
13.1.4.1;Growing R&D Cost;282
13.1.4.2;Expensive InfrastructureInfrastructure;282
13.1.4.3;Expected Training Cost;282
13.1.4.4;Anticipated Maintenance Nightmare;282
13.1.5;Missing InfrastructureInfrastructure;283
13.1.5.1;Specialized Hardware;283
13.1.5.2;Limited SoftwareSoftware;284
13.1.5.3;Unclear Organization Structure;284
13.1.5.4;Work Process Not Defined;285
13.1.6;No MarketingMarketing;285
13.1.6.1;Product Not Clearly Defined;285
13.1.6.2;Unclear Competitive Advantages;286
13.1.6.3;Key Markets Not Identified;286
13.1.6.4;No Advertisement;287
13.1.7;Wrong Expectations;287
13.1.7.1;Magic Bullet;287
13.1.7.2;GIGO 2.0;288
13.1.7.3;SkepticismSkepticism;288
13.1.7.4;ResistanceResistance;289
13.1.8;No Application Methodology;290
13.1.8.1;Method Selection;290
13.1.8.2;Integration Advantages;291
13.1.8.3;Application Sequence;291
13.1.8.4;Few References;291
13.1.9;Summary;291
13.1.10;Suggested Reading;292
14;p3;293
14.1;Part III: Computational Intelligence Application Strategy;293
15;144233_1_En_11_Chapter_OnlinePDF;294
15.1;Chapter 11: Integrate and Conquer;294
15.1.1;11.1The Nasty Reality of Real-World Applications;295
15.1.2;11.2Requirements for Successful Real-World Applications;297
15.1.3;11.3Why Integration Is Critical for Real-World Applications;299
15.1.3.1;11.3.1Benefits of Integration;300
15.1.3.2;11.3.2The Price of Integration;301
15.1.4;11.4Integration Opportunities;302
15.1.4.1;11.4.1Hybrid Intelligent Systems;302
15.1.4.2;11.4.2Integration with First-Principles Models;306
15.1.4.3;11.4.3Integration with Statistical Models;308
15.1.5;11.5Integrated Methodology for Robust Empirical Modeling;309
15.1.5.1;11.5.1Integrated Methodology for Undesigned DataDataThe initial version of the methodology is published in: A. Kordon, G;310
15.1.5.1.1;11.5.1.1Variable Selection;311
15.1.5.1.2;11.5.1.2DataData RecordRecord Selection;313
15.1.5.1.3;11.5.1.3ModelModel GenerationGeneration;313
15.1.5.1.4;11.5.1.5ModelModel Linearization;314
15.1.5.2;11.5.2Integrated Methodology for Designed DataDataThe material in this section was originally published in F. Castillo,;315
15.1.6;11.6Integrated Methodology in Action;316
15.1.7;11.7Summary;324
15.1.8;Suggested Reading;324
16;144233_1_En_12_Chapter_OnlinePDF;325
16.1;Chapter 12: How to Apply Computational Intelligence;325
16.1.1;When Is Computational Intelligence the Right Solution?;325
16.1.2;Obstacles in Applying Computational Intelligence;327
16.1.2.1;Technical Obstacles in Applying CI;327
16.1.2.2;Nontechnical Obstacles in Applying CI;328
16.1.2.3;Checklist ``Are we Ready?´´;330
16.1.3;Methodology for Applying CI in a Business;330
16.1.3.1;Steps for Introducing CI in a Business;332
16.1.3.2;Steps for Applying CI in a Business;333
16.1.3.3;Steps for Leveraging CI in a Business;335
16.1.4;Computational Intelligence Project Management;336
16.1.4.1;Define Project Objectives and Scope;337
16.1.4.2;Define Roles;338
16.1.4.3;Select Computational Intelligence Methods;339
16.1.4.4;Prepare DataData;340
16.1.4.5;Develop ModelModel;342
16.1.4.6;Deploy ModelModel;343
16.1.4.7;ModelModel Maintenance and Support;344
16.1.5;CI for Six Sigma and Design for Six Sigma;345
16.1.5.1;Six SigmaSix Sigma and Design for Six SigmaSigma in IndustryIndustry;346
16.1.5.2;How CI Fits in Design for Six SigmaSix Sigma;351
16.1.6;Summary;354
16.1.7;The Bottom Line;355
16.1.8;Suggested Reading;355
17;144233_1_En_13_Chapter_OnlinePDF;356
17.1;Chapter 13: Computational Intelligence MarketingMarketing;356
17.1.1;Research MarketingMarketing Principles;356
17.1.1.1;Key Elements of MarketingMarketing;357
17.1.1.2;Research MarketingMarketing Strategy;358
17.1.1.2.1;TargetTarget Market IdentificationIdentification;358
17.1.1.2.2;Product Strategy Definition;359
17.1.1.2.3;Promotional Strategy Implementation;360
17.1.2;Techniques - Delivery, Visualization, Humor;361
17.1.2.1;Message Delivery;361
17.1.2.2;Effective VisualizationVisualization;363
17.1.2.2.1;Combining Mind-mapMind-maps and Clip Art;363
17.1.2.2.2;Some VisualizationVisualization Techniques;365
17.1.2.2.3;To PP or Not to PP?;365
17.1.2.3;Humor;367
17.1.2.3.1;Dilbert Cartoons;368
17.1.2.3.2;Useful QuotationsQuotations;368
17.1.2.3.3;Murphy's Laws Related to Computational Intelligence;370
17.1.3;Interactions Between Academia and IndustryIndustry;372
17.1.3.1;Protecting Intellectual Property;372
17.1.3.2;Publishing;374
17.1.3.3;Conference Advertising;375
17.1.3.4;Technology Development;375
17.1.3.5;Interaction with Vendors;376
17.1.4;MarketingMarketing CI to a Technical Audience;376
17.1.4.1;Guidelines for Preparing Technical Presentations for Applied Computational Intelligence;377
17.1.4.2;Key TargetTarget Audience for Technical Presentations;379
17.1.4.2.1;Visionary Guru;380
17.1.4.2.2;Open Mind Guru;380
17.1.4.2.3;Technical King Guru;381
17.1.4.2.4;Political Scientist Guru;381
17.1.4.2.5;Retiring Scientist Guru;382
17.1.4.2.6;1D Mind Guru;382
17.1.5;MarketingMarketing to a Nontechnical Audience;382
17.1.5.1;Guidelines for Preparing Nontechnical Presentations for Applied Computational Intelligence;382
17.1.5.2;Key TargetTarget Audience for Nontechnical Presentations;384
17.1.6;Summary;385
17.1.7;Suggested Reading;386
18;144233_1_En_14_Chapter_OnlinePDF;387
18.1;Chapter 14: Industrial Applications of Computational Intelligence;387
18.1.1;Applications in ManufacturingManufacturing;387
18.1.1.1;Robust Inferential Sensors;388
18.1.1.1.1;Robust Inferential Sensor for Alarm Detection;389
18.1.1.1.2;Robust Inferential Sensor for Product Transition Monitoring;390
18.1.1.1.3;Robust Inferential Sensor for BiomassBiomass Estimation;391
18.1.1.2;Automated Operating Discipline;393
18.1.1.2.1;Knowledge Acquisition from the Experts;394
18.1.1.2.2;Organization of the Knowledge Base;394
18.1.1.2.3;Implementation of Prototype for One Process Unit;396
18.1.1.2.4;Scaling up to the Full System for All Process Units;396
18.1.1.2.5;Operators' Involvement;396
18.1.1.2.6;Value Evaluation;396
18.1.1.3;Empirical EmulatorsEmulators for On-line OptimizationOptimization;397
18.1.1.3.1;Motivation for Developing Empirical EmulatorsEmulators;397
18.1.1.4;Empirical EmulatorsEmulators Structures;397
18.1.1.5;A Case Study: an Empirical Emulator for OptimizationOptimization of an Industrial Chemical ProcessA. Kordon, A. Kalos, and B.;399
18.1.1.5.1;Problem Definition;399
18.1.1.5.2;DataData Preparation;399
18.1.1.5.3;Empirical Emulator Based on Analytic Neural Networks;400
18.1.1.5.4;Empirical EmulatorsEmulators Based on Symbolic RegressionRegression;402
18.1.2;Applications in New Product Development;402
18.1.2.1;Accelerated Fundamental Model BuildingThe material in this section was originally published in: A. Kordon, H. Pham, C. Bosnyak;403
18.1.2.1.1;Potential of GP-Generated Symbolic RegressionRegression in Fundamental Model Building;404
18.1.2.2;Symbolic Regression in Fundamental Modeling of Structure-Properties;406
18.1.2.2.1;Case Study Description;406
18.1.2.2.2;Fundamental ModelModel Building Approach;407
18.1.2.2.3;Symbolic RegressionRegression Approach;407
18.1.2.3;Fast Robust Empirical ModelModel Building;408
18.1.2.4;Symbolic RegressionRegression Models of Blown Film Process Effects;411
18.1.2.4.1;Modeling Scope;411
18.1.2.4.2;Symbolic RegressionRegression ModelModel for DART Impact;411
18.1.2.4.3;Blown Film Process Effects ModelModel Implementation;412
18.1.3;Unsuccessful Computational Intelligence Applications;413
18.1.3.1;Application with Significant Cultural Change;414
18.1.3.2;Applications with Low-Quality DataData;414
18.1.4;Acknowledgements;415
18.1.5;Summary;415
18.1.6;The Bottom Line;415
18.1.7;Suggested Reading;415
19;p4;417
19.1;Part IV: The Future of Computational Intelligence;417
20;144233_1_En_15_Chapter_OnlinePDF;418
20.1;Chapter 15: Future Directions of Applied Computational Intelligence;418
20.1.1;Supply-Demand-Driven Applied Research;418
20.1.1.1;Limitations of Supply-Driven Research;419
20.1.1.2;What Is Supply-Demand Research?;421
20.1.1.3;Advantages of Supply-Demand Research;422
20.1.1.4;Mechanisms of Supply-Demand Research;422
20.1.2;Next-GenerationGeneration Applied Computational Intelligence;424
20.1.2.1;Computing with Words;424
20.1.2.1.1;Basic Principles of Computing with Words;425
20.1.2.1.2;Potential Application Areas for Computing with Words;427
20.1.2.2;Evolving Intelligent Systems;427
20.1.2.2.1;Basic Principles of Evolving Intelligent Systems;428
20.1.2.2.2;Potential Application Areas for Evolving Intelligent Systems;428
20.1.2.3;Co-evolving Systems;430
20.1.2.3.1;Basic Principles of Co-evolving Systems;431
20.1.2.3.2;Potential Application Areas for Co-evolving Systems;432
20.1.2.4;Artificial Immune Systems;433
20.1.2.4.1;Basic Principles of Artificial Immune Systems;434
20.1.2.4.2;Potential Application Areas for Artificial Immune Systems;435
20.1.3;Projected Industrial Needs;436
20.1.3.1;Predictive MarketingMarketing;436
20.1.3.2;Accelerated New Products Diffusion;437
20.1.3.3;High-Throughput InnovationInnovation;438
20.1.3.4;ManufacturingManufacturing at Economic OptimumOptimum;438
20.1.3.5;Predictive Optimal Supply-Chain;439
20.1.3.6;Intelligent Security;439
20.1.3.7;Reduced Virtual Bureaucracy;440
20.1.3.8;Emerging SimplicitySimplicity;440
20.1.3.9;Handling the Curse of Decentralization;441
20.1.4;SustainabilitySustainability of Applied Computational Intelligence;442
20.1.4.1;Potential Roadblocks;443
20.1.4.2;The Fun of Computational Intelligence;444
20.1.5;Summary;444
20.1.6;Suggested Reading;445
21;BM;446
22;Kordon_Index_o;458
22.1;: Index;458