Comprehensive assessment on estimating the thermodynamic and mechanical properties of multicomponent Fe–Cr-based alloys using machine learning techniques

Multicomponent Fe–Cr-based alloys are extensively used in various industrial applications due to their favorable thermodynamic and mechanical properties. However, traditional methods of obtaining these properties are often experimentally intensive and resource-consuming. This study investigates the performance of 11 different machine learning algorithms to predict the mixing enthalpy, Young’s modulus, and the shear-to-bulk modulus ratio of Fe–Cr alloys with additions of Ni, Mo, Al, W, V, and Nb. Moreover, it and performs a sensitivity assessment of the key factors affecting these properties of Fe–Cr alloys. Within the study context, the models applied include a mix of simplified and advanced techniques, such as multiple linear regression, artificial neural networks, random forest, and gradient boosting. Accordingly, the significance of this research lies in enhancing the design process of Fe–Cr alloys by providing accurate and computationally efficient predictions of essential material properties and reporting important insights on factors affecting their performance through a machine learning-based sensitivity analysis.