Abstract
In this paper we demonstrate the ability of the Taguchi technique accurately to characterize and successfully to optimize complicated multi‐response processes with the minimum of experiments, provided one uses simple statistical techniques which can ensure valid, and definitive results. We point out the usefulness of suitable data‐transformations, and we suggest a systematic procedure for establishing the optimal operating conditions and for carrying out confirmatory experiments.For the particular case detailed in this paper (which is typical of multi‐response processes) the Taguchi technique achieved an improvement in uniformity of a factor of 2, together with optimized process control.

This publication has 7 references indexed in Scilit: