| Abstract Scope |
Machine learning and uncertainty-aware materials design are enabling a new paradigm for resilient manufacturing in resource-constrained environments. This talk will describe recent advances that combine AI, additive manufacturing, and adaptive materials engineering to accelerate qualification of novel materials systems derived from uncertain or variable feedstocks. The presentation will highlight how physics-informed machine learning can predict composition–processing–property relationships for additively manufactured components produced from recycled and mixed-source alloys, enabling rapid production of mission-critical hardware in austere environments. Complementary efforts in integrated computational frameworks for accelerated qualification and process control will also be discussed. Together, these examples demonstrate how data-driven materials engineering can reduce experimental burden, quantify uncertainty, and improve robustness in advanced manufacturing workflows, enabling faster deployment of next-generation structural materials. |