Optimisation of Electrical Discharge Machining Processing for AZ91 Magnesium Alloy using Coupled AHP-Taguchi Analyses-GA Method with the Rank Selection Approach

Despite being contemporary, the wire electrical discharge machining (EDM) industry is burdened with complicated and challenging problems. However, the double optimisation method involving Taguchi analyses and genetic algorithms is a powerful tool to help tackle some of these problems. This article evaluates the wire EDM process through a rank-based genetic algorithm coupled with the AHP-Taguchi analyses using the AZ91 magnesium alloy for the first time in the literature. The rank selection method was used at the selection stage of the operations. Six parameters, namely pulse on time, pulse off time, wire feed, wire tension, pulse current and gap voltage, were the process parameters. For all the methods, the total values were computed and compared for the selection, cross-over and mutation operations. It was found that the total values at the selection stage for each of the methods, namely AHP-Taguchi-GA, AHP-Taguchi-Pareto-GA and AHP-Taguchi-ABC-GA methods, were 2750, 4176 and 6306 (best value as Part A), respectively. For all the methods, there was a 25.35% improvement in total value at the cross-over stage compared with the selection stage. The improvement in the total values of the mutation over cross-over and mutation over selection was 53.84% and 92.84%, respectively. These improvement values were for the AHP-Taguchi-GA method but also turned out to be the same for the AHP-Taguchi-Pareto-GA and AHP-Taguchi-ABC methods. The principal advantage of the rank selection method introduced in the present study is to avoid quick convergence. This article is beneficial to the process engineers aimed at improving the wire electrical discharge machining process.