Prediction of cutting performance in slot milling process of AISI 316 considering energy efficiency using experimental and machine learning methods

Purpose The aim of the study is to optimize the cutting parameters (cutting tool diameter, cutting speed and feed) to minimize energy consumption and surface roughness in the slot milling process of AISI 316 stainless steel on CNC milling machine. Design/methodology/approach Growing environmental concerns and cost reduction efforts around the world have made energy efficiency in manufacturing processes a priority goal. Improving energy efficiency in the machining sector is one of the biggest challenges in this area, and slot milling is a critical manufacturing process that directly affects energy consumption. Cutting power, cutting force and surface roughness values were measured during the experimental process. In addition, energy performance of the process was evaluated by calculating specific energy consumption (SEC) and specific cutting energy consumption (SCEC). Experimental data were modeled using machine learning methods of regression analysis and artificial neural networks (ANN). Findings As a result, the lowest SEC and SCEC values, that is the highest energy efficiency, were obtained at 12 mm tool diameter, 75 m/min cutting speed and 0.25 mm/tooth feed. In addition, the optimum cutting parameters for different machining scenarios (roughing and finishing) were determined taking into account the purposes of the machining process (max. or min of energy efficiency, machining time, surface quality, etc.). The optimum cutting parameters for general purpose slot milling and acceptable machining purposes were found to be 12 mm tool diameter, 150 m/min cutting speed and 0.15 mm/tooth feed. Originality/value This study emphasizes the critical importance of energy efficiency and the correct selection of machining parameters for sustainable manufacturing practices. Highlights Slot milling cutting performance of AISI 316 Measurement of cutting power, cutting force and surface roughness Prediction with Regression and ANN methods