Sustainable Machining of Nimonic 80A: Machine Learning-Driven Optimization of Energy Consumption Under Environmentally Conscious Cooling Strategies

Abstract Machining advanced materials like Nimonic 80A in a way that is both environmentally friendly and energy efficient is becoming more important around the world. However, achieving an acceptable level of machining performance without harming the environment remains challenging. The present study is focused on resolving this issue. Machine Learning (ML) was employed to predict and optimize cutting power and specific cutting energy (SCE). Four novel sustainable cooling and lubrication (C/L) environments were studied which are dry, minimum quantity lubrication (MQL), cryogenic CO 2 , and hybrid CO 2 + MQL configuration. The main contribution of the study is the systematic comparison of ML techniques such as support vector machine (SVM), random forest (RF), and multi-layer perceptron (MLP) for getting the prediction and optimization of the cutting power and SCE among different C/L environments. Among all C/L environments, the hybrid (CO 2 + MQL) environment produced the best results, significantly reducing power consumption and SCE. Under the best conditions, the hybrid method used about 47% less power and about 80% less SCE than dry machining. ML analysis showed the superior predictive capability of the MLP model, which achieved the lowest error rates and highest coefficients (R = 0.9996 for power; R = 0.9873 for SCE). These results demonstrate the strong potential of combining hybrid cooling with intelligent ML prediction for sustainable and energy-efficient machining of Nimonic 80A.