Lifetime data modelling and reliability analysis based on modified weibull extension distribution and Bayesian approach

Satellite reliability has long been a critical issue in space industry. In-orbit lifetime data modelling and the corresponding reliability analysis are indispensable aspects for satellite reliability. However, limited by available in-orbit lifetime data, these two aspects have not been well studied. Aiming to mitigate this limitation, we conduct a coherent research on satellite reliability analysis based on an in-orbit database of 1584 satellites. A nonparametric reliability analysis is firstly implemented on the data set to identify the characteristics of failure rate. Then, the modified Weibull extension distribution is used to model the satellite reliability. Point estimation and interval estimation are carried out by utilizing a Bayesian method. Posterior distributions of model parameters are obtained and reserved for further reliability analysis of newly launched satellites. The results presented in this paper can provide helpful information to the space industry for design improvement and test planning of new satellites.