Response Surface Methodology: Process and Product Optimization Using Designed Experiments / Edition 3

Response Surface Methodology: Process and Product Optimization Using Designed Experiments / Edition 3

by Raymond H. Myers, Douglas C. Montgomery, Christine M. Anderson-Cook
     
 

ISBN-10: 0470174463

ISBN-13: 9780470174463

Pub. Date: 12/31/2008

Publisher: Wiley

Praise for the Second Edition:

"This book [is for] anyone who would like a good, solid understanding of response surface methodology. The book is easy to read, easy to understand, and very applicable. The examples are excellent and facilitate learning of the concepts and methods."
Journal of Quality

…  See more details below

Overview

Praise for the Second Edition:

"This book [is for] anyone who would like a good, solid understanding of response surface methodology. The book is easy to read, easy to understand, and very applicable. The examples are excellent and facilitate learning of the concepts and methods."
Journal of Quality Technology

Complete with updates that capture the important advances in the field of experimental design, Response Surface Methodology, Third Edition successfully provides a basic foundation for understanding and implementing response surface methodology (RSM) in modern applications. The book continues to outline the essential statistical experimental design fundamentals, regression modeling techniques, and elementary optimization methods that are needed to fit a response surface model from experimental data. With its wealth of new examples and use of the most up-to-date software packages, this book serves as a complete and modern introduction to RSM and its uses across scientific and industrial research.

This new edition maintains its accessible approach to RSM, with coverage of classical and modern response surface designs. Numerous new developments in RSM are also treated in full, including optimal designs for RSM, robust design, methods for design evaluation, and experiments with restrictions on randomization as well as the expanded integration of these concepts into computer software. Additional features of the Third Edition include:

  • Inclusion of split-plot designs in discussion of two-level factorial designs, two-level fractional factorial designs, steepest ascent, and second-order models

  • A new section on the Hoke design for second-order response surfaces

  • New material on experiments with computer models

  • Updated optimization techniques useful in RSM, including multiple responses

  • Thorough treatment of presented examples and experiments using JMP 7, Design-Expert Version 7, and SAS software packages

  • Revised and new exercises at the end of each chapter

  • An extensive references section, directing the reader to the most current RSM research

Assuming only a fundamental background in statistical models and matrix algebra, Response Surface Methodology, Third Edition is an ideal book for statistics, engineering, and physical sciences courses at the upper-undergraduate and graduate levels. It is also a valuable reference for applied statisticians and practicing engineers.

Read More

Product Details

ISBN-13:
9780470174463
Publisher:
Wiley
Publication date:
12/31/2008
Series:
Wiley Series in Probability and Statistics Series, #705
Edition description:
New Edition
Pages:
704
Product dimensions:
7.20(w) x 10.00(h) x 1.50(d)

Table of Contents

Prefacexiii
1.Introduction1
1.1Response Surface Methodology1
1.2Product Design and Formulation (Mixture Problems)14
1.3Robust Design and Process Robustness Studies15
1.4Useful References on RSM16
2Building Empirical Models17
2.1Linear Regression Models17
2.2Estimation of the Parameters in Linear Regression Models18
2.3Properties of the Least Squares Estimators and Estimation of [sigma superscript 2]25
2.4Hypothesis Testing in Multiple Regression29
2.5Confidence Intervals in Multiple Regression37
2.6Prediction of New Response Observations41
2.7Model Adequacy Checking43
2.8Fitting a Second-Order Model56
2.9Qualitative Regressor Variables65
2.10Transformation of the Response Variable68
Exercises74
3Two-Level Factorial Designs85
3.1Introduction85
3.2The 2[superscript 2] Design85
3.3The 2[superscript 3] Design100
3.4The General 2[superscript k] Design111
3.5A Single Replicate of the 2[superscript k] Design112
3.6The Addition of Center Points to the 2[superscript k] Design128
3.7Blocking in the 2[superscript k] Factorial Design134
Exercise143
4Two-Level Fractional Factorial Designs155
4.1Introduction155
4.2The One-Half Fraction of the 2[superscript k] Design156
4.3The One-Quarter Fraction of the 2[superscript k] Design170
4.4The General 2[superscript k-p] Fractional Factorial Design178
4.5Resolution III Designs183
4.6Resolution IV and V Designs192
4.7Summary194
Exercises195
5Process Improvement with Steepest Ascent203
5.1Determining the Path of Steepest Ascent205
5.2Consideration of Interaction and Curvature213
5.3Effect of Scale (Choosing Range of Factors)218
5.4Confidence Region for Direction of Steepest Ascent220
5.5Steepest Ascent Subject to a Linear Constraint224
Exercises229
6The Analysis of Second-Order Response Surfaces235
6.1Second-Order Response Surface235
6.2Second-Order Approximating Function235
6.3A Formal Analytical Approach to the Second-Order Model241
6.4Ridge Analysis of the Response Surface254
6.5Sampling Properties of Response Surface Results262
6.6Multiple Response Optimization273
6.7Further Comments Concerning Response Surface Analysis286
Exercises287
7Experimental Designs for Fitting Response Surfaces--I303
7.1Desirable Properties of Response Surface Designs303
7.2Operability Region, Region of Interest, and Model Inadequacy304
7.3Design of Experiments for First-Order Models307
7.4Designs for Fitting Second-Order Models321
Exercises366
8Experimental Designs for Fitting Response Surfaces--II377
8.1Designs That Require a Relatively Small Run Size378
8.2General Criteria for Constructing, Evaluating, and Comparing Experimental Designs390
8.3Computer-Generated Designs in RSM413
8.4Some Final Comments Concerning Design Optimality and Computer-Generated Design428
Exercises429
9.Advanced Response Surface Topics--I437
9.1Effects of Model Bias on the Fitted Model and Design437
9.2A Design Criterion Involving Bias and Variance441
9.3RSM in the Presence of Qualitative Variables456
9.4Errors in Control of Design Levels478
9.5Experiments with Computer Models481
9.6Minimum Bias Estimation of Response Surface Models485
9.7Neural Networks489
Exercises492
10.Advanced Response Surface Topics--II500
10.1RSM for Nonnormal Responses--Generalized Linear Models501
10.2Restrictions in Randomization in RSM521
Exercises532
11Robust Parameter Design and Process Robustness Studies536
11.1Introduction536
11.2What Is Parameter Design?536
11.3The Taguchi Approach539
11.4The Response Surface Approach552
11.5Experimental Designs for RPD and Process Robustness Studies586
11.6Dispersion Effects in Highly Fractionated Designs591
Exercises605
12Experiments with Mixtures614
12.1Introduction614
12.2Simplex Designs and Canonical Mixture Polynomials618
12.3Response Trace Plots638
12.4Reparameterizing Canonical Mixture Models to Contain a Constant Term ([beta subscript 0])639
Exercises643
13Other Mixture Design and Analysis Techniques652
13.1Constraints on the Component Proportions652
13.2Mixture Experiments Using Ratios of Components685
13.3Process Variables in Mixture Experiments690
13.4Screening Mixture Components701
Exercises704
14Continuous Process Improvement with Evolutionary Operation715
14.1Introduction715
14.2An Example of EVOP716
14.3EVOP Using Software721
14.4Simplex EVOP725
14.5Some Practical Advice About Using EVOP727
Exercises728
References731
Appendix 1.Variable Selection and Model Building in Regression742
Appendix 2.Multicollinearity and Biased Estimation in Regression759
Appendix 3.Robust Regression770
Appendix 4.Some Mathematical Insights into Ridge Analysis778
Appendix 5.Moment Matrix of a Rotatable Design779
Appendix 6.Rotatability of a Second-Order Equiradial Design784
Appendix 7.Relationship Between D-Optimality and the Volume of a Joint Confidence Ellipsoid on [beta]787
Appendix 8.Relationship Between the Maximum Prediction Variance in a Region and the Number of Parameters789
Appendix 9.The Development of Equation (8.21)790
Appendix 10.Determination of Data Augmentation Result (Choice of x[subscript r + 1] for the Sequential Development of a D-Optimal Design)791
Index793

Read More

Customer Reviews

Average Review:

Write a Review

and post it to your social network

     

Most Helpful Customer Reviews

See all customer reviews >