Towards the Development of a Multi-Modal Community-Based AM Database

Metal additive manufacturing (AM) is increasingly being sought after in critical military, aerospace, and biomedical manufacturing applications for its capability to create near-net shaped parts while minimizing time and material cost. In order to fabricate critical parts with orientation dependent properties required by industries, a complete understanding of microstructural heterogeneities (MH) and the ability to control these MH using AM process parameters is needed. These knowledge gaps are a serious impediment to AM part qualification and industry adoption [1]. Additionally, the non-equilibrium process conditions during AM of parts have led to the development of new alloys suitable for AM processes [2]. Therefore, strategies to accelerate AM part qualification is a major challenge that faces the AM community. The reasons for these barriers to qualification of AM parts are 1) the lack of spatial understanding in the hierarchy of defects and MH found in AM parts, 2) the lack of data-driven approaches for discovering new materials/alloys suitable for AM and 3) the lack of standardized high-volume process-structureproperty (PSP) datasets for AM builds.