The Laser Powder Bed Fusion (PBF-LB) technique for nitinol (NiTi) offers a viable manufacturing strategy for producing intricate, high-performance components for biomedical, energy, and aerospace applications. However, the alloy's strong sensitivity to thermal history and composition, coupled with the complexity of the PBF-LB process, introduces challenges such as porosity, cracking, oxidation, residual stress, and surface roughness that directly influence functional behaviour including superelasticity and the shape memory effect. This review synthesizes current modelling efforts aimed at predicting key process–structure–property relationships in PBF-LB NiTi, including melt pool geometry and stability, thermal history, microstructural evolution, defect formation, residual stress development, and functional outcomes such as phase transformation temperatures. Both physics-based and data-driven modelling approaches are examined. Thermal and melt pool simulations, multi-physics finite element models, and microstructural prediction frameworks provide mechanistic insight into process–material interactions, while machine learning, surrogate, and hybrid models enable rapid parameter optimization and defect prediction. A central NiTi-specific insight emerging from the literature is that accurate prediction of local thermal history and composition evolution is critical for controlling phase transformation behaviour and thus functional performance. Physics-informed machine learning frameworks are highlighted as promising strategies to combine mechanistic fidelity with computational efficiency, supporting more reliable process optimization and functional property control in PBF-LB NiTi.
Rangaswamy et al. (Thu,) studied this question.