Integer programming and ant colony optimization for planning intermodal freight transportation operations

In this paper we deal with the operational planning of transportation operations in an intermodal network. The objective is to satisfy a given transportation demand by using road vehicles and trains, in order to minimize the total transportation cost and meeting a set of operational constraints. We propose a linear integer programming model and an Ant Colony Optimization metaheuristic approach in a pure and a hybrid version. We present and compare the results obtained testing all the approaches on a benchmark set made of randomly generated problem instances. The tests show the appreciable behavior of the IP model, that however requires a considerable amount of time, and the ability of the hybrid ACO to generate high quality solutions for all the benchmark instances with a quite reduced computational effort.