The analysis and determination of the failure rate on distribution system components is the basis of the system reliability assessment. According to the fact that different failure causes coexist and the statistic period is short and failure data is incomplete, this paper proposes a new approach for the evaluation of failure rate on distribution system components. Firstly, the mixed model of random and aging failure is established based on the analysis of component failure causes. Then, according to the collected component failure data, the maximum likelihood estimation method is directly applied to estimate the unknown parameters of the mixed model in the case of a longer statistic failure period, and the EM algorithm is applied to the type-Ⅰ doubly censored failure data for the small and highly censored samples. As a result, the mixture failure rate function of the component is obtained. Finally, the failure rates of 10 kV aerial lines in the distribution system are calculated, and the availability and practicality of the proposed method is verified through the comparison of the results among different statistical periods and different methods.