About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Artificial Intelligence Applications in Integrated Computational Materials Engineering (AI-ICME)
|
| Presentation Title |
Leveraging Large Language Models for Inverse Design of Processing Parameters in Materials Engineering |
| Author(s) |
Jing Luo, Yuxuan Xiao, Khalid A. El-Awady, Jaafar El-Awady |
| On-Site Speaker (Planned) |
Jing Luo |
| Abstract Scope |
Designing processing routes to achieve targeted mechanical properties remains a core challenge in materials engineering. Traditional approaches rely heavily on expert intuition and trial-and-error experimentation, as the relationships between processing conditions and material performance are often buried in unstructured literature. In this work, we introduce a large language model (LLM)-driven agent designed to propose candidate processing parameters given a known material and a desired yield strength. By fine-tuning the Gemma3 12B model and incorporating retrieval-augmented generation (RAG) techniques on a curated literature text, we enable the system to extract, reason over, and recombine processing–property knowledge. The resulting agent can generate plausible processing strategies across diverse alloy systems, providing a data-driven foundation for experimental planning. This approach empowers materials discovery through the intelligent design of experiments, offering a scalable method for navigating complex processing-property landscapes. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, ICME, Mechanical Properties |