Machinability investigation of natural fibers reinforced polymer matrix composite under drilling: Leveraging machine learning in bioengineering applications

Machinability investigation of natural fibers reinforced polymer matrix composite under drilling: Leveraging machine learning in bioengineering applications

The growing demand for fiber-reinforced polymer (FRP) in industrial applications has prompted the exploration of natural fiber-based composites as a viable alternative to synthetic fibers. Using jute–rattan fiber-reinforced composite offers the potential for environmentally sustainable waste material decomposition and cost reduction compared to conventional fiber materials. This article focuses on the impact of different machining constraints on surface roughness and delamination during the drilling process of the jute–rattan FRP composite. Inspired by this unexplored research area, this article emphasizes the influence of various machining constraints on surface roughness and delamination in drilling jute–rattan FRP composite. Response surface methodology designs the experiment using drill bit material, spindle speed, and feed rate as input variables to measure surface roughness and delamination factors. The technique of order of preference by similarity to the ideal solution method is used to optimize the machining parameters, and for predicting surface roughness and delamination, two machine learning-based models named random forest (RF) and support vector machine (SVM) are utilized. To evaluate the accuracy of the predicted values, the correlation coefficient (R2), mean absolute percentage error, and mean squared error were used. RF performed better in comparison with SVM, with a higher value of R2 for both testing and training datasets, which is 0.997, 0.981, and 0.985 for surface roughness, entry delamination, and exit delamination, respectively. Hence, this study presents an innovative methodology for predicting surface roughness and delamination through machine learning techniques.