About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Tribology: Advances in Friction, Wear and Lubrication of Interfaces
|
| Presentation Title |
Optimizing Alloy Design using Machine Learning |
| Author(s) |
David Montes De Oca Zapiain, Nathan Brown, Tomas Babuska, Michael Chandross, Scott Bobbit, John Curry |
| On-Site Speaker (Planned) |
David Montes De Oca Zapiain |
| Abstract Scope |
Designing alloys that exhibit desirable tribological properties requires an efficient exploration of a vast design space and a successful integration of results from multiple data-sources. Despite their accuracy and high-throughput nature, existing methods are ill-equipped to efficiently explore the space to identify factors that foment the desired properties, because they are not capable to extract usable knowledge from previously obtained results. Therefore, new data-driven protocols are desperately needed. In this work, we leverage machine learning to establish accurate and computationally efficient linkages between the desired properties of the alloys and their corresponding inputs. Additionally, once these linkages have been established, we adequately build an effective data-driven protocol to efficiently explore the design space by guiding where the subsequent measurements and results should be obtained. Therefore, this work presents a powerful tool for screening tribological properties. SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525. SAND No. SAND2025-07941A |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Machine Learning, Thin Films and Interfaces, Mechanical Properties |