OPTIMAL PARAMETER DESIGN BY REGRESSION TECHNIQUE AND GREY RELATIONAL ANALYSIS

 Abstract—This research proposes an approach for optimizing multiple responses in the Taguchi method using regression models and grey relational analysis. In this approach, each response is transformed into signal-to-noise (S/N) ratio. The S/N ratios are then utilized to model each response with process factors and complete the responses for all factor level combinations. The grey relational analysis is then used to combine the quality response at each experiment into a single grey grade. Typically, the larger grey grade indicates better performance. Thus, the factor level with the largest level grade is selected as the optimal level for that factor. Three case studies in manufacturing applications on the Taguchi method are utilized for illustration of the proposed approach. It is concluded that the proposed approach is efficient for finding global optimal factor levels. Moreover, this approach can be used with incomplete data. Finally, the regression models can be used to determine the process factors that significantly affect quality response. between process factors (independent factors) with the quality response (dependent factor). An efficient mathematical technique for underplaying the relationship between the quality response and process factors is the multiple regression models (7). Grey relational analysis has been reported efficient in transforming multiple quality responses into a single grade. Several researches (8-9) have used the grey grade for deciding the optimal factor levels. In this context, this paper proposes an approach for optimizing multiple quality responses in the Taguchi method using regression models and grey relational analysis; where the former will be used to complete the response values for all factor level combinations, whereas the latter will be used to determine optimal factor levels. This research is organized into the following sequence. Section two outlines the proposed approach steps. Section three provides illustrative case studies. Finally, the conclusions are made in section four.