Passenger Delay in a Rapid Transit Station

An analytical queueing model for predicting passenger delay in a busy rapid transit station is developed and tested. The model accounts for the “lumpy” arrival pattern of passengers exiting trains. Passenger delay is divided into two components: “standard” delay and “interference” delay. Standard delay is the delay incurred when all of the passengers on a train are served before the next train arrives. Interference delay is the added delay when the next train arrives before all of the passengers are served. When a station serves two equally busy tracks, the standard delay per passenger is proportional to the headway and the utilization (arrival rate divided by service rate). The interference delay per passenger is proportional to the headway and the square of the utilization. The size of the headway reflects the “lumpiness” of the arrival pattern. When the headway is small, passengers arrive at a nearly constant rate and delay per passenger is small. When the headway is large, passengers arrive in larger lumps and delay per passenger is large.