SummaryWith the increase of the operation mileage of the urban rail transit and the extension of lines, the networked operation has become an inevitable trend of rail transit operation. Once an emergency takes place, it will cause delays of the large‐scale operation or even more serious mass safety incidents. First, the topology structure of urban rail transit network is analyzed. According to the static characteristics of the network clustering coefficient, degree, etc, a network model of directional empowerment network is constructed that can be used for analysis of the delay propagation scope. Second, the characteristics of the mass passenger flow distribution of rail transit in emergency is investigated and deeply explored in this paper. Based on this, the model of the delay spread and dissipation is constructed according to the emergency handling time and dissipation time of the passenger congestion. Finally, based on the cellular automaton model, the evolution rule for the delay spreading of the station state is established. The simulation results show that the established model has good reliability and accuracy, which can be adopted to predict the duration and spread of the delay in the emergency. This provides a basis for adjusting the transport organization plan and evacuating passenger flow in emergency.