Simulation Methods for Reliability and Availability of Complex Systems (eBook)

Javier Faulin is associate professor of statistics and operations research at the Public University of Navarra, Navarra, Spain. He is also a mentor and professor of operations research for the Spanish Open University (UNED), Madrid, Spain.Dr Faulin has been a visiting professor at the University of Sabana, Bogotá, Colombia, and the University of Montreal, Montreal, Canada, and a visiting scholar at the University of Cincinnati, Ohio, USA, and University College of Business, Dublin, Ireland. He also spent three years at the University of Surrey, Guildford, UK, as associate lecturer in decision making in business, and in 2007 and 2008 was a professeur invité at the Université de Rennes, Rennes, France.Angel A. Juan is an associate professor of simulation and data analysis at the Open University of Catalonia. He is also a lecturer of statistics at the Technical University of Catalonia.Dr Juan has also been a teacher of mathematics and statistics at Elian's Boston School, Boston, USA; an assistant professor of mathematics at the University of Alicante, Alicante, Spain; a teacher of mathematics and computer science for the Catalan government's Department of Education; assistant professor of applied statistics, simulation of computer networks and mathematics at the Open University of Catalonia, Barcelona, Spain; and a teacher of programming languages, computer networks and database management systems for the Catalan government's Department of Education.Sebastián Martorell is associate professor in nuclear engineering and director of the Department of Chemical and Nuclear Engineering at the Universidad Politécnica de Valencia, from which he also received his MSc and PhD.Dr Martorell is vice-chairman of the European Safety and Reliability Association (ESRA). He is also a member of the editorial board of the European Journal of Industrial Engineering.Jose Emmanuel Ramirez-Marquez is assistant professor at the Stevens Institute of Technology, Hoboken, USA. He is also director of the Quality Control and Reliability Engineering Division Board for the Institute of Industrial Engineers.Prior to receiving his PhD in 2004, Dr Ramirez-Marquez was an area officer for the Secretaria de Hacienda y Credito Publico Servicio de Administracion Tributaria, Ciudad de Mexico, Mexico, and a graduate assistant at Rutgers, New Brunswick, USA.

Foreword 6
Preface 8
Contents 12
Part I: Fundamentals of Simulation in Reliability and Availability Issues 19
Chapter 1: Reliability Estimation by Advanced Monte Carlo Simulation 20
1.1 Introduction 21
1.2 Simulation Methods Implemented in this Study 23
1.2.1 The Subset Simulation Method 23
1.2.2 The Line Sampling Method 27
1.3 Simulation Methods Considered for Comparison 30
1.3.1 The Importance Sampling Method 31
1.3.2 The Dimensionality Reduction Method 32
1.3.3 The Orthogonal Axis Method 33
1.4 Application 1: the Cracked-plate Model 34
1.4.1 The Mechanical Model 35
1.4.2 The Structural Reliability Model 35
1.4.3 Case Studies 36
1.4.4 Results 36
1.5 Application 2: Thermal-fatigue Crack Growth Model 40
1.5.1 The Mechanical Model 41
1.5.2 The Structural Reliability Model 42
1.5.3 Case Studies 43
1.5.4 Results 43
1.6 Summary and Critical Discussion of the Techniques 46
Appendix 1. Markov Chain Monte Carlo Simulation 51
Appendix 2. The Line Sampling Algorithm 52
References 55
Chapter 2: Dynamic Fault Tree Analysis: Simulation Approach 57
2.1 Fault Tree Analysis: Static Versus Dynamic 57
2.2 Dynamic Fault Tree Gates 58
2.3 Effect of Static Gate Representation in Place of Dynamic Gates 61
2.4 Solving Dynamic Fault Trees 62
2.5 Modular Solution for Dynamic Fault Trees 62
2.6 Numerical Method 64
2.6.1 PAND Gate 64
2.6.2 SEQ Gate 65
2.6.3 SPARE Gate 65
2.7 Monte Carlo Simulation Approach for Solving Dynamic Fault Trees 66
2.7.1 PAND Gate 67
2.7.2 SPARE Gate 68
2.7.3 FDEP Gate 69
2.7.4 SEQ Gate 69
2.8 Example 1: Simplified Electrical (AC) Power Supply System of Typical Nuclear Power Plant 71
2.8.1 Solution with Analytical Approach 72
2.8.2 Solution with Monte Carlo Simulation 73
2.9 Example 2: Reactor Regulation System of a Nuclear Power Plant 76
2.9.1 Dynamic Fault Tree Modeling 77
2.10 Summary 77
References 79
Chapter 3: Analysis and Improvements of Path-based Methods for Monte Carlo Reliability Evaluation of Static Models 81
3.1 Introduction 82
3.2 Standard Monte Carlo Reliability Evaluation 84
3.3 A Path-based Approach 85
3.4 Robustness Analysis of the Algorithm 87
3.5 Improvement 90
3.6 Acceleration by Randomized Quasi-Monte Carlo 92
3.6.1 Quasi-Monte Carlo Methods 93
3.6.2 Randomized Quasi-Monte Carlo Methods 94
3.6.3 Application to Our Static Reliability Problem 95
3.6.4 Numerical Results 97
3.7 Conclusions 99
References 99
Chapter 4: Variate Generation in Reliability 101
4.1 Generating Random Lifetimes 101
4.1.1 Density-based Methods 103
4.1.2 Hazard-based Methods 105
4.2 Generating Stochastic Processes 107
4.2.1 Counting Processes 107
4.2.2 Poisson Processes 108
4.2.3 Renewal Processes 109
4.2.4 Alternating Renewal Processes 110
4.2.5 Nonhomogeneous Poisson Processes 110
4.2.6 Markov Models 111
4.2.7 Other Variants 111
4.2.8 Random Process Generation 112
4.3 Survival Models Involving Covariates 115
4.3.1 Accelerated Life Model 116
4.3.2 Proportional Hazards Model 116
4.3.3 Random Lifetime Generation 116
4.4 Conclusions and Further Reading 118
References 118
Part II: Simulation Applications in Reliability 120
Chapter 5: Simulation-based Methods for Studying Reliability and Preventive Maintenance of Public Infrastructure 121
5.1 Introduction 121
5.2 The Power of Simulation 122
5.3 Case Studies 123
5.3.1 Emergency Response 124
5.3.2 Preventive Maintenance of Bridges 128
5.4 Conclusions 133
References 134
Chapter 6: Reliability Models for Data Integration Systems 136
6.1 Introduction 136
6.2 Data Quality Concepts 139
6.2.1 Freshness and Accuracy Definitions 139
6.2.2 Data Integration System 140
6.2.3 Data Integration Systems Quality Evaluation 142
6.3 Reliability Models for Quality Management in Data Integration Systems 144
6.3.1 Single State Quality Evaluation in Data Integration Systems 145
6.3.2 Reliability-based Quality Behavior Models 146
6.4 Monte Carlo Simulation for Evaluating Data Integration Systems Reliability 151
6.5 Conclusions 155
References 156
Chapter 7: Power Distribution System Reliability Evaluation Using Both Analytical Reliability Network Equivalent Technique and Time-sequential Simulation Approach 158
7.1 Introduction 158
7.2 Basic Distribution System Reliability Indices 160
7.2.1 Basic Load Point Indices 160
7.2.2 Basic System Indices 161
7.3 Analytical Reliability Network Equivalent Technique 162
7.3.1 Definition of a General Feeder 163
7.3.2 Basic Formulas for a General Feeder 163
7.3.3 Network Reliability Equivalent 166
7.3.4 Evaluation Procedure 167
7.3.5 Example 168
7.4 Time-sequential Simulation Technique 171
7.4.1 Element Models and Parameters 171
7.4.2 Probability Distributions of the Element Parameters 172
7.4.3 Exponential Distribution 173
7.4.4 Generation of Random Numbers 174
7.4.5 Determination of Failed Load Point 174
7.4.6 Consideration of Overlapping Times 176
7.4.7 Reliability Indices and Their Distributions 176
7.4.8 Simulation Procedure 177
7.4.9 Stopping Rules 178
7.4.10 Example 178
7.4.11 Load Point and System Indices 178
7.4.12 Probability Distributions of the Load Point Indices 179
7.5 Summary 183
References 184
Chapter 8: Application of Reliability, Availability, and Maintainability Simulation to Process Industries: a Case Study 186
8.1 Introduction 186
8.2 Reliability, Availability, and Maintainability Analysis 187
8.3 Reliability Engineering in the Process Industry 187
8.4 Applicability of RAM Analysis to the Process Industry 188
8.5 Features of the Present Work 189
8.5.1 Software Used 190
8.6 Case Study 190
8.6.1 Natural-gas Processing Plant Reliability Block Diagram Modeling 191
8.6.2 Failure and Repair Data 197
8.6.3 Phase Diagram and Variable Throughput 198
8.6.4 Hidden and Degraded Failures Modeling 199
8.6.5 Maintenance Modeling 200
8.6.6 Crews and Spares Resources 203
8.6.7 Results 204
8.6.8 Bad Actors Identification 205
8.6.9 Cost Analysis 206
8.6.10 Sensitivity Analysis 207
8.7 Conclusion 208
References 209
Chapter 9: Potential Applications of Discrete-event Simulation and Fuzzy Rule-based Systems to Structural Reliability and Availability 211
9.1 Introduction 212
9.2 Basic Concepts on Structural Reliability 212
9.3 Component-level Versus Structural-level Reliability 213
9.4 Contribution of Probabilistic-based Approaches 214
9.5 Analytical Versus Simulation-based Approaches 214
9.6 Use of Simulation in Structural Reliability 215
9.7 Our Approach to the Structural Reliability Problem 216
9.8 Numerical Example 1: Structural Reliability 218
9.9 Numerical Example 2: Structural Availability 221
9.10 Future Work: Adding Fuzzy Rule-based Systems 223
9.11 Conclusions 224
References 225
Part III: Simulation Applications in Availability and Maintenance 227
Chapter 10: Maintenance Manpower Modeling: A Tool for Human Systems Integration Practitioners to Estimate Manpower, Personnel, and Training Requirements 228
10.1 Introduction 228
10.2 IMPRINT – an Human Systems Integration and MANPRINT Tool 229
10.3 Understanding the Maintenance Module 230
10.3.1 System Data 231
10.3.2 Scenario Data 233
10.4 Maintenance Modeling Architecture 234
10.4.1 The Static Model – the Brain Behind It All 235
10.4.2 A Simple Example – Putting It All Together 238
10.5 Results 239
10.6 Additional Powerful Features 240
10.6.1 System Data Importing Capabilities 240
10.6.2 Performance Moderator Effects on Repair Times 240
10.6.3 Visualization 241
10.7 Summary 241
References 242
Chapter 11: Application of Monte Carlo Simulation for the Estimation of Production Availability in Offshore Installations 244
11.1 Introduction 244
11.1.1 Offshore Installations 244
11.1.2 Reliability Engineering Features of Offshore Installations 245
11.1.3 Production Availability for Offshore Installations 246
11.2 Availability Estimation by Monte Carlo Simulation 247
11.3 A Pilot Case Study: Production Availability Estimation 252
11.3.1 System Functional Description 253
11.3.2 Component Failures and Repair Rates 254
11.3.3 Production Reconfiguration 255
11.3.4 Maintenance Strategies 255
11.3.5 Operational Information 258
11.3.6 Monte Carlo Simulation Model 258
11.4 Commercial Tools 261
11.5 Conclusions 262
References 263
Chapter 12: Simulation of Maintained Multicomponent Systems for Dependability Assessment 264
12.1 Maintenance Modeling for Availability Assessment 264
12.2 A Generic Approach to Model Complex Maintained Systems 266
12.3 Use of Petri Nets for Maintained System Modeling 268
12.3.1 Petri Nets Basics 268
12.3.2 Component Modeling 269
12.3.3 System Modeling 273
12.4 Model Simulation and Dependability Performance Assessment 275
12.5 Performance Assessment of a Turbo-lubricating System 276
12.5.1 Presentation of the Case Study 276
12.5.2 Assessment of the Maintained System Unavailability 279
12.5.3 Other Dependability Analysis 280
12.6 Conclusion 281
References 282
Chapter 13: Availability Estimation via Simulation for Optical Wireless Communication 284
13.1 Introduction 284
13.2 Availability 285
13.3 Availability Estimation 286
13.3.1 Fog Models 286
13.3.2 Rain Model 288
13.3.3 Snow Model 289
13.3.4 Link Budget Consideration 289
13.3.5 Measurement Setup and Availability Estimation via Simulation for Fog Events 290
13.3.6 Measurement Setup and Availability Estimation via Simulation for Rain Events 297
13.3.7 Availability Estimation via Simulation for Snow Events 299
13.3.8 Availability Estimation of Hybrid Networks: an Attempt to Improve Availability 301
13.3.9 Simulation Effects on Analysis 303
13.4 Conclusion 305
References 305
About the Editors 307
About the Contributors 309
Index 320

"Chapter 10 Maintenance Manpower Modeling: A Tool for Human Systems Integration Practitioners to Estimate Manpower, Personnel, and Training Requirements (p. 217-218)

Mala Gosakan and Susan Murray

Abstract This chapter discusses the maintenance manpower modeling capability in the Improved Performance Research Integration Tool (IMPRINT) that supports the Army’s unit of action. IMPRINT has been developed by the US Army Research Laboratory (ARL) Human Research and Engineering Directorate (HRED) in order to support the Army’s need to consider soldiers’ capabilities during the early phases of the weapon system acquisition process. The purpose of IMPRINT modeling is to consider soldiers’ performance as one element of the total system readiness equation. IMPRINT has been available since the mid 1990s, but the newest version includes significant advances.

10.1 Introduction

Even as the far-reaching implications of the next generation of weapons and information systems are being constantly redefined, one piece which has been and will continue to be central to the process is human involvement. The impacts of human performance on system performance are significant.

Human systems integration (HSI) is primarily a concept to focus on the human element in the system design process [18]. The ability thus to include and consider human involvement early in the process of system development cycle will only ease mobilization, readiness, and sustainability of the newly developed system. The Department of Defense therefore has placed increased emphasis on applying HSI concepts to evaluate and improve the performance of complex systems [16].

The US Army was the first large organization to implement HSI approach and reap the benefits of it by creating the Manpower and Personnel IntegrationManagement and Technical Program (MANPRINT) [24, 25]. As stated in the MANPRINT handbook, MANPRINT is a comprehensive management and technical program that focuses on the integration of human considerations (i.e., capabilities and limitations) into the system acquisition process.

The goal of MANPRINT is to enhance soldier-system design, reduce life-cycle ownership costs, and optimize total system performance. To facilitate this, MANPRINT is divided into the following seven domains: manpower, personnel capabilities, training, human factors engineering, system safety, health hazards, and soldier survivability. Themanpower domain focuses on the number of people required and available to operate, maintain, sustain, and provide training for systems.

The domain of personnel addresses the cognitive and physical characteristics and capabilities required to be able to train for, operate, maintain, and sustain materiel and information systems. The training domain is defined as the instruction, education, on-the-job, or selfdevelopment training required providing all personnel and units with essential job skills, and knowledge to effectively operate, deploy/employ, maintain, and support the system."