Enhancing drilling performance in 3D printed PLA implants application of PIV and ML models

Fused Deposition Modeling (FDM) is a common additive manufacturing technique known for its ability to quickly produce complex geometries and features according to customer requirements in a short timeframe. Parts produced by FDM technique with (Polylactic Acid) PLA can be used extensively as it is biocompatible in nature. The present investigation involves drilling the fabricated circular discs to assess the material removal rate (MRR) and hole profile. The inputs considered are spindle speed (SS), feed rate (fr), and drill diameter (dia), in relation to the output characteristics such as MRR, surface roughness (Ra and Rz), circularity, and cylindricity. This work employs the Proximity Indexed Value (PIV) tool to predict the near-optimal value for improving hole quality and enhancing the drilling process. In addition, Artificial neural network (ANN), Support vector machine (SVM), and Random Forest (RF) models were employed to predict the optimized results by PIV method. The tests show that the RF model, which has a R² value of 99.16%, does a better job of describing the desired outcome than the ANN model (89.45%) and the SVM model (89.08%). This also proves that the machine learning (ML) models offer better prediction to the optimization of drilling parameters with reference to the quality of the workpiece machined.