Urban intermodal terminals: The entropy maximising facility location problem

•Entropy maximization approach to competitive facility location problems.•Equivalent MILP shown to be a limiting case of the entropy maximizing problem.•Proof of existence and uniqueness of a solution.•Algorithms for solving the entropy maximizing facility location problem.•Model re-formulation into nested logit model for forecasting and policy testing with numerical examples to illustrate key features of the model.•.

An important problem confronting port cities is where and how to accommodate port growth. Larger ships combined with increased container throughput require more yard space and generate more traffic, straining the urban fabric in the vicinity of the port. A promising solution to this problem is the development of urban intermodal container terminals (IMTs) that interface with both road and rail (or possibly inland waterway) networks. This raises two linked choices; where to locate the intermodal terminals and what will be their likely usage by multiple shippers, each having a choice of whether or not to use the IMT as part of an intermodal transport chain. The use of an IMT by a shipper indicates the shipper's choice of intermodal transport, which comprises a combined use of a high capacity mode (rail or barge between the port and the IMT) and trucks (between the IMT and the cargo origin or destination). The overall problem therefore comprises a mode choice problem embedded within a facility location problem. This paper employs the method of entropy maximisation to combine a logit mode choice model with a facility location model, leading to a non-linear mixed integer programming model. The principal features of the entropy maximising facility location model are illustrated by small and large numerical examples.