A defect-based physics-informed machine learning framework for fatigue finite life prediction in additive manufacturing

•A machine Learning-based approach and phenomenological laws are combined to account for several influencing factors and improve fatigue life prediction.•A PINN framework is employed to predict finite fatigue life in materials containing defects.•A newly developed LEFM semi-empirical model is employed to represent the physics of the problem.•Results from an additively manufactured material are taken from the literature to demonstrate the effectiveness of the proposed method.