Modeling and Optimization Using Taguchi Based Grey Relation and Genetic Algorithm: A Comparative Case Study

摘要

This paper discusses the use of Taguchi based grey relation and Genetic Algorithm (GA) for optimizing process parameters of a case study related to laser transmission welding process of thermoplastics. The earlier researcher applied grey relational analysis to convert multiple qualitycharacteristics to a single performance grey relational grade and predicted the optimal results. In this work, Response Surface Methodology (RSM) is employed to develop the mathematical model for predicting the responses. A separate second degree regression model with best fit is developedfor the weld strength and weld width consisting three important welding parameters namely laser power, welding speed and focal position. The fit summary for weld strength and weld width suggests the quadratic relationship where the additional terms are significant and there is a convenientagreement between predicted and actual values. For optimization of the developed models, GA is applied for single and multi-objective optimization by applying prior weightages to the responses. The obtained results are compared with the earlier results and it is found that the results obtainedin present work shows considerable improvement over the individual and combined grey relational grade. The results of optimization show that the optimal parameters can be predicted effectively with less computation and are superior to grey relational analysis so as to improve multiple qualitycharacteristics.