Robust parameter design of dynamic multi‐response system: a new integrated method

Abstract

Purpose

The purpose of this paper is to propose a new method of robust parameter design for dynamic multi‐response system. The objectives are to resolve the correlations among multiple responses and the uncertainty of system with incomplete information.

Design/methodology/approach

First, desirability function is used to measure dynamic system sensitivity and system variation, and principal component analyses on the two indices are conducted. Second, the grey relational grade (GRD) between principal component sequences of the two indices and their respective ideal sequences, gained by grey relational analysis, is converted to an integrated GRD (IGRD) index by means of TOPSIS method, and then the optimal level combination of controllable factors is identified based on the IGRD index.

Findings

It was found that the optimal factor level combination obtained by the proposed method is nearest the ideal solution and farthest from the negative ideal solution. The validity and superiority of the proposed method are confirmed through two illustrative examples.

Research limitations/implications

It should be noted that the proposed method fails to consider the interaction effects between controllable factors and noise factors.

Originality/value

The method proposed in the paper effectively integrates several common methods to optimize a dynamic multiple responses system based on Taguchi's robust parameter design. These methods do not involve complicated mathematical theory, and are therefore easy for practitioners to use in engineering practice.

Citation

Wang, J., Ma, Y. and Su, G. (2011), "Robust parameter design of dynamic multi‐response system: a new integrated method", Asian Journal on Quality, Vol. 12 No. 1, pp. 67-79. https://doi.org/10.1108/15982681111140552