Scheduling of Coupled Train Platoons for Metro Networks: A Passenger Demand-Oriented Approach

Abstract

In recent years the concept of virtual coupling, where multiple train units are virtually coupled into a platoon with very short following distances, has received considerable attention in the railway transportation field. This study introduces this concept into the train scheduling problem to improve line capacity and reduce congestion in urban metro networks. With consideration of the time-dependent passenger demand, train (platoon) loading capacity, and limited rolling stock resources, specifically, a mixed integer linear programming model is developed to simultaneously generate the platoon (de)coupling strategies, orders of trains, and their arrival/departure times at each station in the metro network. Several model improvement strategies, for example, model linearization and determination of big- $M$ values, are proposed to enhance the computational efficiency of the model. Finally, numerical experiments based on historical passenger data from the Beijing metro network are implemented to verify the effectiveness of the approach. The results demonstrate that the introduction of train platoons of different sizes can evidently reduce station congestion, while the percentage improvement greatly depends on the distribution of passenger demand.