This paper studies the timing and site selection of the standby trains launch in the case of large passenger flow in congestion metro stations, which focuses on the evacuation of large passenger subway stations for passengers. By exploiting problem-specific knowledge, we present an optimization-based approach that deals with the issue of timing and site of launching standby trains by developing a multi-objective optimization model in this paper. We first set up a criterion for the definition which explains what large passenger station is. Once the passenger flow status of a metro station meets the criterion, we find the time to launch standby trains. Second, we develop a multi-objective optimization model aiming to minimize waiting times of passengers in the congestion station. We take different subway stations with the conditions of launching standby train as input variables and use train tracking interval as constraints of the multi-objective optimization model. Third, using the Particle Swarm Optimization with Penalty Function to solve the model, we get the departure time of standby trains and the follow-up trains so that passengers can leave the congestion station with minimum waiting times. Finally, we apply the developed model to Guangzhou Metro Line 8 to verify the effectiveness of the proposed method. The conclusion shows that reasonable selection of the time and site of the standby train can minimize the waiting time of passengers at large passenger flow stations.

CICTP