Models to simulate gas turbine blade high cycle fatigue (HCF) failure are presented. The models are integrated into a prognosis capability to predict HCF resistance after a foreign object damage (FOD) event. The capability considers the uncertainty in the FOD condition, dynamic loading, material microstructure and diagnostic data. HCF is predicted as the statistical distribution of the number of cycles to failure (cycle counting) or the probability that the fatigue strength has been exceeded (fatigue limit). The statistical confidence bound are also predicted. The technical approach provides a prognosis capability for situations where it is not possible to directly sense the blade damage state with a high degree of certainty.