| Abstract Scope |
Computational materials and its various sub-disciplines, such as material-related machine learning, have a wide variety of uses in AM. Applications of computational material methods range from full-physics modeling of process with predictions of the resulting material structure and properties, to much simpler tasks that aid in witness specimen evaluations in production. The use of computational material-related models in certification is a relatively new endeavor for the discipline, which by good measure is rooted in conservatism. The chosen path to implementation of such capabilities into certification activities is important. An incremental approach that focuses first on small, well-validated contributions to the certification process should be the priority now. These “small wins” help pave the way for acceptance of increasingly complex computational material contributions to certification rationale. In the meantime, advanced computational material models continue to evolve and shape our understanding of the AM process, providing a foundation for potentially revolutionary improvements. |