Title: QFD-based optimisation using response surface method

Authors: Young-soon Yang, Beom-seon Jang, Yun-seog Yeun

Addresses: Research Institute of Marien System Engineering, Department of Naval Architecture and Ocean Engineering, Seoul National Univeristy, San 56-1 Shillim-dong, Kwanak-gu, Seoul 151-742, Korea. Department of Mechanical Design Engineering, Daejin University, San 11-1 Sundanri, Pochen, Kyonggi-do, 467-711, Korea

Abstract: This paper suggests a customer-oriented optimisation method, called Quality Function Deployment (QFD) based optimisation method, to determine target levels of multiple design objectives that attain the maximum customer satisfaction level, and also satisfy a Pareto optimality condition. The customer satisfaction level is defined by QFD, and the numerical association is identified through the use of the Response Surface Method (RSM) that constructs an approximate model for Pareto set. This decision is made by solving an optimisation problem of which objective function is the customer satisfaction level and constraint of the non-inferior solution that follows customers| preferences most. This regression model enables to not only easily investigate both the feasibility and the Pareto-optimality for a set of design objective values, but also support trade-off among various perspectives by having a clear understanding of the association. It also allows the exploration of the effects of ambiguity included in QFD-based optimisation. Since there always exists ambiguity that originates from the process of the value judgements in QFD, it is necessary to investigate what effects the ambiguity has on the determination of the target levels. To deal with such ambiguity systematically, this paper suggests an exploratory approach, and introduces orthogonal array and Analysis of Variance (ANOVA) table developed by Taguchi.

Keywords: Quality Function Deployment; optimisation; Pareto set; RSM.

DOI: 10.1504/IJVD.2002.001990

International Journal of Vehicle Design, 2002 Vol.28 No.4, pp.275-293

Published online: 15 Aug 2003 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article