Critical subway stations identification for passenger flow control by applying network controllability

Passenger flow control on a limited number of stations is a common way to maintain the service of a subway system at a desired level. Critical subway stations identification is the key for controlling a complex subway system. Based on network controllability analysis, this paper proposes a passenger flow network (PFN) for a subway system to capture the state dependencies among stations, and develops the analysis procedure for solving the station selection problem. Practical analysis is conducted using data from the Shanghai subway system. The results show that the critical station identifications are dependent on the time period for control. The stations connected with few stations, or with low passenger volume may also be critical for subway system performance, especially stations with low inflow. This study contributes a quantitative method for subway system managers to develop effective passenger flow control strategies.