Statistical Quality Control, 7th Edition

14Process Optimization with Designed Experiments

CHAPTER OUTLINE

14.1   RESPONSE SURFACE METHODS AND DESIGNS

14.1.1    The Method of Steepest Ascent

14.1.2    Analysis of a Second-Order Response Surface

14.2   PROCESS ROBUSTNESS STUDIES

14.2.1    Background

14.2.2    The Response Surface Approach to Process Robustness Studies

14.3 EVOLUTIONARY OPERATION

Supplemental Material for Chapter 14

S14.1   Response Surface Designs

S14.2   More about Robust Design and Process Robustness Studies

The supplemental material is on the textbook Website www.wiley.com/college/montgomery.

CHAPTER OVERVIEW AND LEARNING OBJECTIVES

In Chapter 13, we focused on factorial and fractional factorial designs. These designs are very useful for factor screening—that is, identifying the most important factors that affect the performance of a process. Sometimes this is called process characterization. Once the appropriate subset of process variables is identified, the next step is usually process optimization, or finding the set of operating conditions for the process variables that result in the best process performance. This chapter gives a brief account of how designed experiments can be used in process optimization.

We discuss and illustrate response surface methodology, an approach to optimization developed in the early 1950s and initially applied in the chemical and process industries. This is probably the most widely used and successful optimization technique based on designed experiments. Then we discuss ...