A reasonable location planning of emergency facility is an important work in the emergency management of urban rail transit. Based on the complex network theory, we propose a reliable emergency facility location model that aiming to improve the coverage level for the nodes with high vulnerability. This is mainly due to that the nodes with the different vulnerabilities in the network have different requirements for coverage level. We research on the node vulnerability of urban rail transit using complex network theory, and identifies the key nodes in the network before location planning. The vulnerable nodes in the network will be given higher weight to improve their coverage level in the location model, so that the equilibrium and efficiency of emergency are enhanced. To efficiently solve the model, we develop a heuristic based on Lagrangian relaxation. Finally, the model and heuristic are applied to the emergency facility location in urban rail transit (URT) network in Nanjing, China. The proposed method is effective and robust, and contributes greatly to the performance of emergency facility location planning in emergency management of URT.