A hybrid MCDM approach for parametric optimization of a micro-EDM process

In modern day manufacturing industries, micro-electrical discharge machining (micro-EDM) has emerged out as an efficient material removal process to produce miniaturized components having varying industrial applications. To explore its fullest machining potential, it is always required to operate the micro-EDM process while setting its various input parameters at their optimal levels. In this paper, four popular multi-criteria decision making (MCDM) techniques, in the form of weighted aggregated sum product assessment, technique for order of preference by similarity to ideal solution, combinative distance-based assessment and complex proportional assessment are separately hybridized with teaching-learning-based optimization (TLBO) algorithm to solve the parametric optimization problems of a micro-EDM process. The polynomial regression (PR) models are considered here as the inputs to these hybrid optimizers. Their optimization performance is subsequently validated against the conventionally adopted weighted sum multi-objective optimization (WSMO) approach at four different weight scenarios. It is revealed that for the micro-EDM process, all the MCDM-PR-TLBO approaches provide better solutions as compared to PR-WSMO-TLBO method for the considered weight scenarios. The best performance of the MCDM-PR-TLBO approaches is achieved when 50% weight is assigned to material removal rate. Moreover, it is also noticed that MCDM-PR-TLBO approaches are less computationally intensive than PR-WSMO-TLBO with approximately 9.61-26.70% saving in computational time.