An intelligent approach for multi-response optimisation of WEDM parameters

Abstract

Usually, multi-criteria decision making methods are embedded with design of experiment (DOE) approach for handling the multi-response optimisation problems. However, uncertainties, impreciseness and arbitrary human judgement for weight assignment to criteria and alternatives result in inferior solutions. To overcome this limitation, an intelligent approach based on neuro-fuzzy system is proposed for converting multi-responses into single equivalent response. To illustrate the superiority of the proposed approach, a complex case study of taper cutting operation using wire electrical discharge machining (WEDM) process is considered. The effect of process parameters on equivalent response has been studied in detail and the relationship between the input parameters and responses are established by means of a nonlinear regression analysis resulting in a valid mathematical model. Finally, optimal parameter setting is obtained by recently proposed meta-heuristics like bat algorithm.