基于PCA与偏好强度可调的DEA方法的多响应稳健参数设计

Optimizing Multi-Response Problem in Robust Design Using Principal Components Analysis and Dea with Adjustable Preference Intensity

稳健参数设计旨在选择可控因子的水平组合来降低产品或过程对噪声因子的敏感性,能够有效地提高和改善产品的质量.针对多响应参数优化设计中响应间存在相关性的问题,提出了基于主成分分析(Principal Component Analysis,PCA)与偏好强度可调的数据包络分析(Data Envelopment Analysis,DEA)的稳健参数设计;并结合实例验证了方法的有效性.研究结果表明,结合PCA与偏好强度可调的DEA方法能够有效地实现相关多响应的参数优化设计.

Robust parameter design is an efficient approach for obtaining the best combi- nation of control factors with the lowest societal cost solution. However, it can only be used to optimize single response problems and not multi-response. In practice, customers are con- cerned about multiple quality characteristics, which are usually correlated. So the Principal Component Analysis （PCA） is found to optimize multi-response problems in robust parame- ter design. Although, a set of original response can be transformed into a set of uncorrelated components by PCA, it has some shortcomings. Thus, in this study, we propose an effective procedure for multiple responses by using Data Envelopment Analysis （DEA） with adjustable preference intensity, which can efficiently overcome the shortcoming of Data Envelopment Analysis （DEA） when normalizing the inputs and outputs differently, and PCA. By combin- ing DEA and PCA, the proposed procedure optimizes multiple responses simultaneously, and overcomes the limitatioas of PCA. An example is given to demonstrate the effectiveness of proposed approach.