EUVIMEDEUVIMED entdecken
Crossref

Machine Learning Approaches for Heat Treatment in Thermal Manufacturing Processes: A Review

Adib Bazgir, Yuwen Zhang

ASME Journal of Heat and Mass Transfer · 01.01.2026 · 148 (1)

DOI: 10.1115/1.4069270 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ASME Journal of Heat and Mass Transfer, online erschienen am 02.09.2025. Verfasst von Adib Bazgir und 1 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 156 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This review focuses on the paradigm shift from machine learning (ML) to heat treatment, a subset of manufacturing thermal processing. The thermal processing encompasses a range of techniques to tailor the microstructures of materials. Still, this review focuses on ML and the challenges associated with optimizing heat treatment processes such as annealing, quenching, tempering, and precipitation hardening. Controlling heat treatment processes to achieve the desired mechanical, electrical, and functional properties can often be inefficient and costly. In this review, numerous ML approaches are detailed, including supervised learning algorithms such as support vector machines and random forests, neural networks (including convolutional neural networks and physics-informed neural networks), and ML interatomic Potentials that can improve predictive modeling of properties exhibited by materials (e.g., hardness, strength, and fatigue life) and microstructural features related to heat treatment. It also reviews advancements in automated process optimization driven by ML to determine ideal heat treatment parameters and the alloy composition, and the automated characterization of microstructures. Although considerable advancements exist, there remain barriers to overcome, including data availability, model explainability, and implementation in industry. This review presents a vision for the future in which physics-informed and interpretable ML will replace human decision-making, integrated with digital twins and advanced artificial intelligence, enabling less human oversight, more efficient, and smarter heat treatment processes.

Zitieren

APA

Bazgir, A., & Zhang, Y. (2026). Machine Learning Approaches for Heat Treatment in Thermal Manufacturing Processes: A Review. ASME Journal of Heat and Mass Transfer, 148(1). https://doi.org/10.1115/1.4069270

Vancouver

Bazgir A, Zhang Y. Machine Learning Approaches for Heat Treatment in Thermal Manufacturing Processes: A Review. ASME Journal of Heat and Mass Transfer. 2026;148(1). doi:10.1115/1.4069270

BibTeX
@article{Bazgir2026,
  author = {Bazgir, Adib and Zhang, Yuwen},
  title = {Machine Learning Approaches for Heat Treatment in Thermal Manufacturing Processes: A Review},
  journal = {ASME Journal of Heat and Mass Transfer},
  year = {2026},
  volume = {148},
  number = {1},
  doi = {10.1115/1.4069270}
}
Kostenlos starten →