Robust Design of Static and Dynamic Systems

Taguchi's robust design is an important method to reduce the effect of noise factors in the ideal function of product or process and unit-to-unit variation in order to improve the quality of products or processes and lead to higher customer satisfaction and operational performance. Many publications have revealed approaches to the multiple quality characteristics problem. Most did not consider the measurement units and quality loss coefficients varied with quality characteristics, or experimental data involving the static and dynamic quality characteristics. The objective of robust design is to minimize the total quality loss. The purpose of this paper is to resolve the relationship between the quality loss function and SN ratio. According to definitions of the quality loss function and SN ratio, the average SN ratio for a factor level can be transformed into a PQL (proportion of quality loss for a factor level) value, which will convert the multiple quality characteristics into a single problem. The presented model of robust design is based on minimizing the total quality loss. From the proposed method and optimization steps, the result can address complex parameter design varying with the property and objective of experimental data for promoting the product quality.