Exclusive Content & Downloads from ASQ

Evaluating Mixture-Process Designs with Control and Noise Variables

Summary: [This abstract is based on the authors' abstract.] In some mixture-process experiments, certain process variables are noise variables that cannot be controlled in normal process operation. This constitutes a robust design setting requiring a response model combining mixture, process, and noise variables. Models for the mean response and for the variance of the response for this response model are subsequently developed. Expressions are developed for the scaled and unscaled prediction variances of both models. Competing designs for such situations are evaluated using variance dispersion graphs and fraction of design space plots of the prediction variance values.

Please sign-in or register to download this information. Registration is FREE and gives you access to ASQ's articles, case studies and general information.