Abstract

To route and schedule trains over a large complex network can be computationally intensive. One way to reduce complexity could be to "aggregate" suitable sections of a network. In this paper, we present a simulation-based technique to generate delay estimates over track segments as a function of traffic conditions, as well as network topology. We test our technique by comparing the delay estimates obtained for a network in Los Angeles with the delays obtained from the simulation model developed by Lu et al. [Lu, Q., Dessouky, M.M., Leachman, R.C., 2004. Modeling of train movements through complex networks. ACM Transactions on Modeling and Computer Simulation 14, 48-75], which has been shown to be representative of the real-world delay values. Railway dispatchers could route and schedule freight trains over large networks by using our technique to estimate delay across aggregated network sections. (C) 2009 Published by Elsevier Ltd.