Algoritma K-Nearest Neighbor untuk Memperoleh Nilai Prediksi dari Kekasaran Permukaan pada Operasi Pembubutan Baja Aisi 304 Berbantuan Teknik Mql

Abstract

This research aims to apply the K-Nearest Neighbors (K-NN) algorithm to predict surface roughness in the machining operation of AISI 304 steel with the Minimal Quantity Lubrication (MQL) technique. Surface roughness is a critical parameter in the machining industry that affects product quality. In the experiment, surface roughness data were obtained from a research paper conducted by Dubey et al. (2022), considering operational parameters such as cutting speed, depth of cut, feed rate, and nanoparticle concentration. The K-NN model is used to predict surface roughness based on specific configurations of operational parameters. The research results indicate that the K-NN model can provide accurate predictions with a significant coefficient of determination (R2) and low error rates (MSE and MAPE). This study provides a foundation for improving efficiency and product quality in the machining industry with the assistance of machine learning technology, especially the K-NN algorithm.