Process Optimization: A Statistical Approach

ISBN-13: 9781441943965 / Angielski / Miękka / 2010 / 462 str.

Enrique del Castillo

Opis

PROCESS OPTIMIZATION: A Statistical Approach is a textbook for a course in experimental optimization techniques for industrial production processes and other "noisy" systems where the main emphasis is process optimization. The book can also be used as a reference text by Industrial, Quality and Process Engineers and Applied Statisticians working in industry, in particular, in semiconductor/electronics manufacturing and in biotech manufacturing industries. The major features of PROCESS OPTIMIZATION: A Statistical Approach are:

• It provides a complete exposition of mainstream experimental design techniques, including designs for first and second order models, response surface and optimal designs;
• Discusses mainstream response surface method in detail, including unconstrained and constrained (i.e., ridge analysis and dual and multiple response) approaches;
• Includes an extensive discussion of Robust Parameter Design (RPD) problems, including experimental design issues such as Split Plot designs and recent optimization approaches used for RPD;
• Presents a detailed treatment of Bayesian Optimization approaches based on experimental data (including an introduction to Bayesian inference), including single and multiple response optimization and model robust optimization;
• Provides an in-depth presentation of the statistical issues that arise in optimization problems, including confidence regions on the optimal settings of a process, stopping rules in experimental optimization and more;
• Contains a discussion on robust optimization methods as used in mathematical programming and their application in response surface optimization;
• Offers software programs written in MATLAB and MAPLE to implement Bayesian and frequentist process optimization methods;
• Provides an introduction to the optimization of computer and simulation experiments including and introduction to stochastic approximation and stochastic perturbation stochastic approximation (SPSA) methods;
• Includes an introduction to Kriging methods and experimental design for computer experiments;
• Provides extensive appendices on Linear Regression, ANOVA, and Optimization Results.

Spis treści

Preliminaries.- An Overview of Empirical Process Optimization.- Elements of Response Surface Methods.- Optimization Of First Order Models.- Experimental Designs For First Order Models.- Analysis and Optimization of Second Order Models.- Experimental Designs for Second Order Models.- Statistical Inference in Process Optimization.- Statistical Inference in First Order RSM Optimization.- Statistical Inference in Second Order RSM Optimization.- Bias Vs. Variance.- Robust Parameter Design and Robust Optimization.- Robust Parameter Design.- Robust Optimization.- Bayesian Approaches in Process Optimization.- to Bayesian Inference.- Bayesian Methods for Process Optimization.- to Optimization of Simulation and Computer Models.- Simulation Optimization.- Kriging and Computer Experiments.- Appendices.- Basics of Linear Regression.- Analysis of Variance.- Matrix Algebra and Optimization Results.- Some Probability Results Used in Bayesian Inference.