Drilling performance prediction of HDPE/Washingtonia fiber biocomposite using RSM, ANN, and GA optimization

Polymeric biocomposites are emerging as a convenient alternative for traditional materials used in light-weight structural applications. Many studies on the machining of biocomposites were published in recent years. However, their number remains limited compared to the number of studies on the machining of glass and carbon fiber–reinforced composites. Therefore, drilling of Washingtonia filifera (WF) fiber-reinforced high-density polyethylene (HDPE) biocomposites was studied in this research by varying various factors, such as drill diameter (d), feed rate (f), and spindle speed (N) using full factorial design (L₂₇). Response surface methodology (RSM) was applied for the drilling experiment and was used in conjunction with artificial neural network (ANN) in mathematical modeling of the drilling operation parameters. The results showed excellent agreement between experimental data and RSM/ANN predictions. Hence, the developed biocomposite HDPE/WF can be used in the polymer field to improve overall product performances. Furthermore, the drilling parameter optimization results obtained by the genetic algorithm (GA) combined with ANN are almost similar to those of the desirability function (DF) of RSM, especially f = 50 mm/min, N = 806 rev/min, and d = 5 mm.