The assessment of reliability in the distribution system is a crucial undertaking that plays a vital role in ensuring uninterrupted and high-quality power delivery, we propose a reliability assessment approach for the distribution system utilizing a dynamic Bayesian network. Firstly, we model the components of the distribution system and their relationships. By collecting historical data and expert knowledge, a dynamic Bayesian network model is constructed, which can describe the state evolution and fault propagation process of distribution system and take into account the interaction between different components and the conditional probability changing with time. Subsequently, we employ the dynamic Bayesian network framework to assess the reliability. By incorporating reliability indicators and operational constraints, we have the means to conduct a quantitative analysis on the dependability of the distribution system. Eventually, we verify the effectiveness of the proposed method through a case study of an actual distribution system. The findings indicate that the reliability evaluation method based on dynamic Bayesian network can accurately assess the dependability of the distribution and provide important decision support for system operation and maintenance