A new method for product field reliability assessment based on accelerated life test

Abstract

Accelerated life testing (ALT) is one of the major methods of rapidly evaluating product field reliability and life. The topic of how to improve assessment precision using ALT and field observation data has attracted research interest recently. A new method of field reliability assessment is proposed in this paper. Firstly, in view of data fusion, a more accurate model is established by integrating failure information from both the field and ALT data. Secondly, posterior parameter estimation, and field reliability and life are rapidly obtained through Bayesian analysis based on MCMC, which can effectively reduce the computational complexity of the model. Finally, a comprehensive set of experiments is conducted and the results show that reliability and life assessment based on this method is more precise.