The project investigates how sensor data, intelligent data logging and data mining techniques can be combined to predict the aging state, remaining useful life and safety performance of infrastructures. It examines how prognostics and health management methods can be integrated with probabilistic safety assessment to dynamically update failure probabilities and risk rankings. Further research questions concern how data fusion, trend analysis and model updating can improve prediction accuracy, and how risk-informed predictive maintenance strategies can be derived from continuously updated safety information.