On the nested p-center problem with min-max regret objective

Description

In managerial decision-making, consistency over a time horizon is very important, although it is ignored by many facility location problems. One way to achieve consistency is by forcing a multi-period facility location solution to fulfill the nesting property. In the recently introduced nested p-center problem, which is an extension of the classical (vertex) p-center problem, we are given a finite time horizon and at each time period, we are allowed to open a given number of facilities. The open facilities must fulfill the nesting property, i.e., at each time period they need to be supersets of the set of open facilities of the previous time period. In our work, we consider the nested p-center problem with minimizing the maximal relative regret as objective function, where the regret is defined with respect to the optimal solution for each time period. We present mixed integer linear programming formulations for the problem. Based on these formulations, we develop branch-and-cut solution algorithms, which include various speed-up ingredients such as preprocessing and cut-lifting. We present a computational study to assess the performance of our solution algorithms.

Period	22 May 2024
Event title	International Symposium on Combinatorial Optimization
Event type	Conference
Location	SpainShow on map