Machine Learning for Additive Manufacturing and Hot Isostatic Pressing based on Quality Characteristics

Additive manufacturing (AM) using laser powder bed fusion (PBF-LB/M) has many advantages such as design freedom and sustainability. However, AM also suffers from internal defects that limit its functionality. Furthermore, post-treatment methods like hot isostatic pressing (HIP) promise to enhance material properties. This work aims to develop a machine learning model using artificial neural networks, to predict PBF-LB/M or HIP process parameters based on quality characteristics.