Investigating Novel Approaches to Overcome Data Deficiency in Prognostics and Health Management

Recently, the prognostics and health management (PHM) is becoming more important in many industries since it enhances reliability and efficiency during the operation. Among the several issues in the current PHM research, it may be of no doubt that the data deficiency of field data may be one of the most challenges, making the PHM applications difficult in the real industry. Performance of prognostics requires large amount of run-to-fail data, which is however not easy due to the high cost or inability to run until failure under real operation. This is overcome by exploiting two resources in terms of data: 1) accelerated life data at laboratory 2) real data with suspension at field. In this study, a new algorithm is developed that combines these two data sources for better prognostics.