CASE WESTERN RESERVE UNIVERSITY

STATISTICS COLLOQUIUM

RESPONSE SURFACES AND TAGUCHI METHODS

Dennis K.J. Lin

Penn State University

Friday, April 4,1997

3:30 pm - refreshments

4:00 pm - talk

Room 327, Yost Hall

Abstract

Response surface methodology (RSM) is designed to find optimal settings for a set of input variables which optimize the response. It has been proven to be a useful tool in modern quality engineering. Some recent advances in RSM will be introduced.

The mean response is typically the major concern in RSM. As a result, the homogeneous variance assumption is required. When such an assumption is violated, it is wise to deal with both mean and variance responses --an important contribution due to the leading quality consultant in Japan, Genichi Taguchi. Some of his ideas will be re-examined from the RSM perspective. Recent results on the dual response approach as well as designing product- array experiments will be discussed.