The current state of metallurgy is changing towards anticipatory and information-focused engineering. Historically, the development of alloy and optimization of its processes were based on trial and error experimentation with the assistance of thermodynamics and kinetics. With the advent of high-performance computing, physics-based numerical techniques, including Density Functional Theory (DFT), molecular dynamics (MD), Monte Carlo (MC) simulations, phase-field modeling, crystal plasticity and finite element methods (FEM) could now be applied. The models enable to quantitatively describe phase stability, microstructure development, deformation processes, and defects.

Эшету М. (науч. рук. Новиков А.С.) Application of machine learning and computer modeling in metallurgy // Сборник тезисов докладов конгресса молодых ученых. Электронное издание. – СПб: Университет ИТМО, [2026]. URL: https://kmu.itmo.ru/digests/article/16053