Design and Analysis of Simulation Experiments

by Kleijnen, Jack P. C.

• ISBN13: 9780387718125
• ISBN10: 0387718125
• Format: Hardcover
• Copyright: 2007-10-19
• Publisher: Springer Verlag

Summary

DASE: Design and Analysis of Simulation Experiments focuses on statistical methods for discrete-event simulation (such as queuing and inventory simulations). In addition, the book discusses DASE for deterministic simulation (such as engineering and physics simulations). The text presents both classic and modern statistical designs. Classic designs (e.g., fractional factorials) assume only a few factors with a few values per factor. The resulting input/output data of the simulation experiment are analyzed through low-order polynomials, which are linear regression (meta) models. Modern designs allow many more factors, possible with many values per factor. These designs include group screening (e.g., Sequential Bifurcation, SB) and space filling designs (e.g., Latin Hypercube Sampling, LHS). The data resulting from these modern designs may be analyzed through low-order polynomials for group screening and various metamodel types (e.g., Kriging) for LHS. DASE: Design and Analysis of Simulation Experiments is an authoritative textbook and reference work for researchers, graduate students, and technical practitioners in simulation. Basic knowledge of simulation and mathematical statistics are expected; however, the book does summarize these basics, for the readers' convenience. In addition, the book provides relatively simple solutions for (a) selecting problems to simulate, (b) how to analyze the resulting data from simulation, and (c) computationally challenging simulation problems.

Author Biography

Jack Kleijnen is well known internationally for being a leading researcher in simulation for more than 30 years. He is the author of highly cited books in the area of statistical techniques in simulation that were published between 1974 and 1992. He is an excellent writer and researcher, and hence, ideally suited to write this important book for the field.  On 25 February 2008 Her Majesty Beatrix, Queen of the Netherlands, appointed Jack Kleijnen a Knight in the Order of the Netherlands Lion. 

Table of Contents

Introduction	p. 1
What is simulation?	p. 1
What is DASE?	p. 7
DASE symbols and terms	p. 10
Solutions for exercises	p. 12
Low-order polynomial regression metamodels and their designs: basics	p. 15
Introduction	p. 16
Linear regression analysis: basics	p. 19
Linear regression analysis: first-order polynomials	p. 27
First-order polynomial with a single factor	p. 27
First-order polynomial with several factors	p. 28
Designs for first-order polynomials: resolution-III	p. 36
2[superscript k-p] designs of resolution-III	p. 36
Plackett-Burman designs of resolution-III	p. 39
Regression analysis: factor interactions	p. 40
Designs allowing two-factor interactions: resolution-IV	p. 42
Designs for two-factor interactions: resolution-V	p. 46
Regression analysis: second-order polynomials	p. 49
Designs for second-degree polynomials: Central Composite Designs (CCDs)	p. 50
Optimal designs and other designs	p. 51
Validation of metamodels	p. 54
Coefficients of determination and correlation coefficients	p. 54
Cross-validation	p. 57
More simulation applications	p. 63
Conclusions	p. 66
Appendix: coding of nominal factors	p. 66
Solutions for exercises	p. 69
Classic assumptions revisited	p. 73
Introduction	p. 73
Multivariate simulation output	p. 74
Designs for multivariate simulation output	p. 77
Nonnormal simulation output	p. 78
Realistic normality assumption?	p. 78
Testing the normality assumption	p. 79
Transformations of simulation I/O data, jackknifing, and bootstrapping	p. 80
Heterogeneous simulation output variances	p. 87
Realistic constant variance assumption?	p. 87
Testing for constant variances	p. 88
Variance stabilizing transformations	p. 89
LS estimators in case of heterogeneous variances	p. 89
Designs in case of heterogeneous variances	p. 92
Common random numbers (CRN)	p. 93
Realistic CRN assumption?	p. 94
Alternative analysis methods	p. 94
Designs in case of CRN	p. 96
Nonvalid low-order polynomial metamodel	p. 97
Testing the validity of the metamodel	p. 97
Transformations of independent and dependent regression variables	p. 98
Adding high-order terms to a low-order polynomial metamodel	p. 98
Nonlinear metamodels	p. 99
Conclusions	p. 99
Solutions for exercises	p. 100
Simulation optimization	p. 101
Introduction	p. 101
RSM: classic variant	p. 105
Generalized RSM: multiple outputs and constraints	p. 110
Testing an estimated optimum: KKT conditions	p. 116
Risk analysis	p. 123
Latin Hypercube Sampling (LHS)	p. 126
Robust optimization: Taguchian approach	p. 130
Case study: Ericsson's supply chain	p. 135
Conclusions	p. 137
Solutions for exercises	p. 138
Kriging metamodels	p. 139
Introduction	p. 139
Kriging basics	p. 140
Kriging: new results	p. 147
Designs for Kriging	p. 149
Predictor variance in random simulation	p. 151
Predictor variance in deterministic simulation	p. 152
Related designs	p. 154
Conclusions	p. 155
Solutions for exercises	p. 156
Screening designs	p. 157
Introduction	p. 157
Sequential Bifurcation	p. 160
Outline of simplest SB	p. 160
Mathematical details of simplest SB	p. 165
Case study: Ericsson's supply chain	p. 167
SB with two-factor interactions	p. 169
Conclusions	p. 171
Solutions for exercises	p. 172
Epilogue	p. 173
References	p. 175
Index	p. 211
Table of Contents provided by Ingram. All Rights Reserved.