Multi-Objective Optimization of AISI P20 Mold Steel Machining in Dry Conditions Using Machine Learning-TOPSIS Approach

In the present research, AISI P20 mold steel was processed using the milling process. The machining parameters considered in the present work were speed, depth of cut (DoC), and feed (F). The experiments were designed according to an L-27 orthogonal array; therefore, a total of 27 experiments were conducted with different settings of machining parameters. The response parameters investigated in the present work were material removal rate (MRR), surface roughness (Ra, Rt, and Rz), power consumption (PC), and temperature (Temp). The machine learning (ML) approach was implemented for the prediction of response parameters, and the corresponding error percentage was investigated between experimental values and predicted values (using the ML approach). The technique for order of preference by similarity to ideal solution (TOPSIS) approach was used to normalize all response parameters and convert them into a single performance index (Pi). An analysis of variance (ANOVA) was conducted using the design of experiments, and the optimized setting of machining parameters was investigated. The optimized settings suggested by the integrated ML-TOPSIS approach were as follows: speed, 150 m/min; DoC, 1 mm; F, 0.06 mm/tooth. The confirmation results using these parameters suggested a close agreement and confirmed the suitability of the proposed approach in the parametric evaluation of a milling machine while processing P20 mold steel. It was found that the maximum percentage error between the predicted and experimental values using the proposed approach was 3.43%.