A Novel Surface Roughness Estimation and Optimization Model for Turning Process Using RSM-JAYA Method

This study proposes a novel model to estimate and optimize the surface roughness factor (SRF) of duplex steel along with other control variables required in the turning process during machining. Response surface methodology (RSM) has been utilized to correlate the control variables and generate the objective function for estimating the surface roughness factor. Further, the evolutionary JAYA algorithm has been implemented to optimize the developed objective function of the SRF. The simulation results analysis reveals that the proposed technique reduces the average prediction error from 14.63 to 5.028% compared with the prevailing techniques. The prediction coefficient of correlation (R2) for the roughness factor was found to be 0.9971, which means the model can predict new observations accurately. Also, the adjusted R2 has been recorded to be 0.9943, which indicates the model behaves well when adjustments are made to variables or new variables are added or eliminated. Performance comparison has been performed with other existing models to establish the supremacy of the proposed one.