The intelligent manufacturing of high‐performance alloys necessitates computational frameworks capable of bridging the complex gap between industrial processing, microstructure, and properties. In this study, an integrated computational framework is presented to optimize the process–structure–property relationships in M42 high‐speed steel (HSS). The framework synergizes furnace computational fluid dynamics (CFD), hot‐working finite‐element (FE) simulations, calculation of phase diagrams–diffusion‐controlled transformations (CALPHAD‐DICTRA) modeling, and machine learning (ML) techniques to predict microstructure evolution throughout the entire hot‐working process. Experimental validation demonstrates a temperature‐prediction accuracy exceeding 85%. Furthermore, ML‐driven optimization forging schedule reduced the number of forging passes from 22 to 19 while eliminating cracking risks, resulting in an 8.3% increase in compressive strength (from 4.8 to 5.2 GPa) and a 2.3 HRC enhancement in hardness. Additionally, DICTRA simulations accurately predicted the dissolution kinetics of M 2 C carbides, showing high consistency with experimental observations. This approach establishes a reliable digital thread for HSS manufacturing, significantly enhancing carbide homogeneity and mechanical performance while providing a systematic pathway for the intelligent design of advanced alloys.
Yin et al. (Tue,) studied this question.