Abstract
Among the engineering design methods currently studied in the engineering design community, researchers often identify robust design as one of the most important fields for the purpose of quality improvement. Although the notion of robust design is clearly important, the use of Taguchi's signal-to-noise ratios and crossed arrays to capture variability has been a subject of controversy. As a natural alternative, a response surface approach to robust design has drawn much attention recently. This article first investigates the robust design models based on response surface approach and then proposes a priority-based robust design model to better reflect an engineer's point of view at the early product design stage. Discussions are made and numerical optimizations are performed for the purpose of comparative studies.
Acknowledgments
We thank Professor James R. Simpson in the Department of Industrial Engineering at Florida State University for his insightful comments, which greatly improved the clarity and presentation of this article. We also thank Mr. A. Ritvirool in the Department of Industrial Engineering at Clemson University for his assistance.