Optimal Designs for Mixture-Process Experiments with Control and Noise Variables

Abstract

Mixture experiments that involve process variables are widely encountered in industry. When some of the process variables are noise variables, interest focuses on finding levels for the mixture components and the controllable process variables that result in a product that has a desirable mean response and that has small variability in the response transmitted from the noise variables. This is a variation of the robust parameter design problem. Choosing an appropriate experimental design for this type of problem is addressed in the paper. We show how designs that have small prediction variance for the mean and the slope variance can be obtained. We also show how designs that are robust to the level of interaction between control and noise variables can be constructed.

• Mixture-Process Experiment
• Response Surface Methodology
• Robust Parameter Design

Additional information

Notes on contributors

Peter J. Chung

Dr. Chung is a Senior Specialist at Merrill Lynch & Co. His email address is [email protected].

Heidi B. Goldfarb

Dr. Goldfarb is a Research Fellow Statistician in the Research and Development Department. She is a senior member of ASQ. Her email address is [email protected].

Douglas C. Montgomery

Dr. Montgomery is Regents' Professor of Industrial Engineering and Statistics and ASU Foundation Professor of Engineering in the Department of Industrial Engineering. He is a Fellow of ASQ. His email address is [email protected].