Robust design is a well-known quality improvement method that focuses on building quality into the design of products and services. Yet, most well established robust design models only consider a single performance measure and their prioritization schemes do not always address the inherent goal of robust design. This paper aims to propose a new robust design method for multiple quality characteristics where the goal is to first reduce the variability of the system under investigation and then attempt to locate the mean at the desired target value. The paper investigates the use of a response surface approach and a sequential optimization strategy to create a flexible and structured method for modeling multiresponse problems in the context of robust design. Nonlinear programming is used as an optimization tool. The proposed methodology is demonstrated through a numerical example. The results obtained from this example are compared to that of the traditional robust design method. For comparison purposes, the traditional robust design optimization models are reformulated within the nonlinear programming framework developed here. The proposed methodology provides enhanced optimal robust design solutions consistently. This paper is perhaps the first study on the prioritized response robust design with the consideration of multiple quality characteristics. The findings and key observations of this paper will be of significant value to the quality and reliability engineering/management community.
|Number of pages||16|
|Journal||International Journal of Quality and Reliability Management|
|Publication status||Published - 2009|
- design for quality
- quality improvement
- design engineering