Abstract:

The robust parameter design problem for dynamic multi-response with the skewness characteristics and the great influence on the optimization and robustness of dynamic responses from the scale and skewness are discussed. A new approach based on the multivariate skew normal distribution and response surface methodology is proposed to resolve the problem about the robust parameter design and optimization for dynamic multi-response. Firstly, the responses in different levels of the signal factor are given and their distribution types are determined based on the non-parametric test methods. Secondly, the joint models for the location, scale and skewness for each response are defined based on the response surface methodology, and then the total expected loss function based on the multivariate skew normal distribution is proposed.Finally, a specific example is used to illustrate the effectiveness of the proposed approach when the hybrid genetic algorithm is used to optimize the total expected loss function. The result shows that the proposed method can produce more effective solutions in terms of the robust parameter design for dynamic multi-response with the skew normal distribution characteristics.