A hybrid FEM–machine learning framework for multi-objective optimization of turning parameters in Hastelloy X

Machining of Hastelloy X, a nickel-based superalloy widely used in aerospace and high-temperature applications, is critically important due to its superior mechanical strength and oxidation resistance, which also make it difficult to machine. This study proposes an integrated framework for optimizing cutting parameters in the turning of Hastelloy X by combining the finite element method (FEM) with artificial intelligence (AI) techniques. Extreme Gradient Boosting (XGBoost) was employed to predict key machining responses, including resultant force (Fr), cutting power (Pc), and material removal rate (MRR). Multi-objective optimization was performed using the Nondominated Sorting Genetic Algorithm-III (NSGA-III), generating a set of Pareto-optimal solutions that balance machining performance criteria. The Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) was then applied to rank these alternatives and determine the most suitable parameter combination. FEM simulations, calibrated with experimental data from the literature, demonstrated high predictive accuracy. Validation results showed low error rates of 2.75% for Fr, 3.48% for Pc, and 0.66% for MRR, confirming the reliability of the proposed approach. The findings indicate that the integrated FEM–AI framework effectively enhances machining efficiency, reduces energy consumption, and supports sustainable manufacturing practices.