Robust uncapacitated hub location

Abstract

In this paper we present robust counterparts for uncapacitated hub location problems in which the level of conservatism can be controlled by means of a budget of uncertainty. We study three particular cases for which the parameters are subject to interval uncertainty: demand, transportation cost, and both simultaneously. We present mixed integer programming formulations for each of these cases and a branch-and-cut algorithm to solve the latter. We present computational results to evaluate the performance of the proposed formulations when solved with a general purpose solver and study the structure of the solutions to each of the robust counterparts. We also compare the performance between solutions obtained from a commensurable stochastic model and those from our robust counterparts in both risk neutral and worst-case settings.