Rail electrification network, within the concept of smart grid, integrates various technologies and is operated in an environment where the behavior and failure modes of the system are difficult to model. It has been proven that modern electrical networks are rather complex, involving multi-dependencies between components (also called system variables) and uncertainties about these dependencies. Modeling and quantification of the reliability for a large system, which requires the handling of dependencies and uncertainties is a complex task, especially for the system where high availability is required. System design includes historical experiences and evidence; therefore, system correctly performs its intended functions. However, wrong method or system model for the purpose of reliability analysis can lead to over or underestimation of the system reliability. In this work, Hierarchical Bayesian Networks are applied to model and assess the reliability of a large and complex rail electrification network and the reliability critical items are identified.