Routing Trains with Consideration of Congestion-Induced Link and Node Delay

Abstract

North American heavy haul railroads are experiencing growth in traffic demand and increases in capacity constraints. An appropriate service-design process is thus crucial to allow for more effective network capacity planning and efficient rail operations. Existing approaches to freight train routing usually ignore the congestion-induced delay and fail to consider the dynamics in link and node delay in response to traffic volume. Furthermore, different train types can have substantially different operating characteristics, including maximum speed, power-to-ton ratio, and dispatching priority. This heterogeneity causes conflicts between trains that can increase delays and reduce capacity. Therefore, traffic volume and heterogeneity should both be incorporated into the decision process for service design. In this research, the authors proposed to add an additional traffic routing process in service design and developed a novel optimization framework to route trains by considering link and node delay, directional traffic, heterogeneity, and various delay costs according to train types. To improve the solution efficiency, a solution algorithm using Lagrangian relaxation (LR) techniques is also presented. Numerical results show that the optimal strategy has the potential to substantially reduce operational costs. Using this tool can help design better traffic routing, more efficient capacity utilization, and reliable services.