Reliability analysis based on the Wiener process integrated with historical degradation data

For high-reliability and long-life electronic devices, reliability analyses based on degradation data with small sample sizes are an outstanding question. Recently, increasing attention has been paid to the abundant historical degradation data. Exploiting the information in these historical data efficiently and integrating them to benefit the current reliability analysis is an important issue. This study proposes a new integrating method for the historical and current degradation data based on the Wiener process by considering the consistency of failure mechanisms between the different batches of degradation data. Simulation studies show that the new method has superior reliability estimation, and is robust to the assumption of consistent failure mechanisms. Finally, the proposed method is used to analyze a real data set consisting of the metal-oxide-semiconductor field-effect transistor (MOSFET) degradation data.