Development of an experiment-based robust design paradigm for multiple quality characteristics using physical programming

Abstract

The well-known quality improvement methodology, robust design, is a powerful and cost-effective technique for building quality into the design of products and processes. Although several approaches to robust design have been proposed in the literature, little attention has been given to the development of a flexible robust design model. Specifically, flexibility is needed in order to consider multiple quality characteristics simultaneously, just as customers do when judging products, and to capture design preferences with a reasonable degree of accuracy. Physical programming, a relatively new optimization technique, is an effective tool that can be used to transform design preferences into specific weighted objectives. In this paper, we extend the basic concept of physical programming to robust design by establishing the links of experimental design and response surface methodology to address designers' preferences in a multiresponse robust design paradigm. A numerical example is used to show the proposed procedure and the results obtained are validated through a sensitivity study.

Development of a sequential optimization procedure for robust design and tolerance design within a bi-objective paradigm

“…Chen et al (1999), Lewis et al (2006), Govindaluri and Cho (2007), Hong and Cho (2007) and Kovach et al Cho (2008) studied other multidisciplinary RD models. A weighted sum approach based on the weight w can be formulated as follows:…”

Multiple response optimization: a global criterion‐based method

“…Examples of loss function-based approaches for multiple dual response optimization are presented in References [36][37][38][39][40]. Other popular alternative approaches are the fuzzy, compromise programming, nonlinear goal programming and physical programming approaches presented in References [41][42][43][44], respectively. The proposed method can also be used for optimizing dual and The dual and multiple dual optimization problems require replicates at each design run.…”

Review of Multiresponse Process Optimisation Methods

“…RSM has been used in diverse applications for solving multiresponse optimisation problems, such as optimisation of numerical control machine (Berni and Gonnelli 2006), optimisation of various machining processes (Aggarwal and Singh 2005), modelling and analysis of laser drilling processes (Kuar et al 2006;Ghoreishi et al 2006), optimisation of laser shock peening process to improve performances of micro-electro-mechanical system (MEMS) (Zhu et al 2012), optimisation of laser welding of stainless steels (Khan et al 2012), predictive modelling and optimisation of Nd:YAG laser micro-turning of ceramics (Kibria et al 2013), optimisation of wire electric discharge machining (WEDM) in processing high strength low-alloy steel (HSLA) (Sharma et al 2013), optimisation of WEDM in processing a pure titanium , modelling of plasma spray coating process (Datta et al 2013), optimisation of selective laser sintering process used to produce PA12/MWCNT nanocomposite (Paggi et al 2013), and others (Tsui et al 2004;Kovach et al 2008;Timothy et al 2004;Robinson et al 2004).…”

“…Wu (2002) presented an approach for multiresponse optimisation based on a quality loss function, multiple regressions and mathematical programming. Kovach et al (2008) proposed the method based on physical programming and RSM methodology to solve multiresponse optimisation problems. However, it includes shortcomings of both approaches.…”