摘要

There exist uncertainties in the prediction of electronic device reliability based on PoF (physics of failure) model. Based on the combination of Kolmogorov-Smirnov test (KS-test) and Monte-Carlo simulation method this paper presents a modified method for reliability prediction of electronic devices considering limited test failure data. The process capability index (Cpk) is used to quantitatively characterize the main factors of model uncertainties. Firstly the degree of fitting between the theoretical probability distribution of electronic device failures based on PoF by using the Monte-Carlo simulation method and the practical probability distribution of electronic device failures based on test or field failure data is tested by using K-S lest method. Secondly the corrected coefficient of the model is optimized. Finally, a solder thermal fatigue life assessment model and some test failure data are used to verify the proposed method in the illustrative example. The prediction results calculated by modified model are consistent with lest results.