Experimental investigations on the machining performance of the WEDM process of NiTi SMA by comparison of Taguchi and Box-Behnken design approaches

Abstract

The present study aims to investigate the wire electrical discharge machining (WEDM) performance of NiTi shape memory alloy (SMA) by using two widely accepted approaches of design of experiment: Taguchi and Box-Behnken design (BBD) of response surface methodology (RSM). The key important input variables of pulse-on time (T_on), pulse-off time (T_off), and discharge current (I_p) were selected along with material removal rate (MRR), and surface roughness (SR) as response measures. Taguchi's L9 array and BBD with 15 runs were employed. Analysis of variance (ANOVA) and coefficient of determination (R²) validated the adequacy of the developed empirical relations. The teaching–learning-based optimization (TLBO) algorithm was further applied to identify optimal machining conditions. For the Taguchi design, optimal parameters were T_on = 40 μs, T_off = 18 μs, and I_p = 4 A, yielding MRR = 1.5769 g/min and SR = 3.57 μm. For the BBD-RSM design, the optimal parameters were T_on = 40 μs, T_off = 17 μs, and I_p = 5 A, resulting in MRR = 1.7321 g/min and SR = 3.38 μm. The optimal results for both designs have shown robust performance. However, the BBD-RSM design approach has outperformed the Taguchi design approach in both MRR and SR under the given conditions. The BBD-RSM design outperformed the Taguchi design, with enhancements of 9.84 % in MRR and 5.62 % in SR. This comparative analysis highlights the robustness of both approaches while demonstrating the superior optimization capability of the BBD-RSM design for machining NiTi SMA.