Optimisation of Multiple Quality Characteristics Based on Percentage Reduction of Taguchi's Quality Loss

Abstract

The traditional Taguchi method was focused on ensuring good performance at the parameter design stage with one quality characteristic, but most products and processes have multiple quality characteristics. The optimal parameter design minimises the total quality loss for multiple quality characteristics. Several studies have presented approaches addressing multiple quality characteristics. Most of these papers were concerned with maximising the parameter combination of SN ratios. This paper presents an approach for multiple quality characteristics based on the percentage reduction of quality loss. A multiple regression is employed to fit the function of percentage reduction of quality loss and to optimise the parameter design with multiple quality characteristics using mathematical programming. The results reveal the advantages of this approach are that the optimal parameter design is the same as the traditional Taguchi method for the single quality characteristic; the optimal design maximises the amount of reduction of total quality loss for multiple quality characteristics.