This PhD project aims to develop a novel modelling framework of mechanical component lifecycle for better RUL prediction in terms of uncertainty estimation and reduction. Remaining useful life (RUL) prediction is a process using prediction methods to forecast the future performance of components or systems and obtain the time left before them loses its operation ability. Knowing the RUL of a system is essential for maintenance decision making and contingency mitigation. Its ultimate aim is to increase availability, reliability and safety, and reduce maintenance costs. Apart from maintenance, the estimation of RUL has also an important role in the management of product reuse and recycle which has strategic impacts on energy consumption, raw material use, pollution and landfill.