Applying Particle Swarm Optimization for the Absolute p-center Problem

Locating facilities on anywhere on a network is known as the absolute p-center location problem and it is proven to be NP-hard problem. Most of the recent approaches solve largescale vertex p-center location problem in which facilities can be located only on the nodes of the network. However, rarely algorithms are developed to solve large-scale absolute p-center problem. Particle swarm optimization (PSO) is a metaheuristic approach, which recently proved to be a successful approach in solving complex continuous optimization problems. In this paper we present a PSO algorithm for the absolute p-center problem to minimize the maximum distance from each customer to his/her nearest facility. We have tested our proposed algorithm on a set of 12 problems from “Beasley OR Library” and compared the results of vertex location problem to those of the absolute location problem. The numerical experiments show that PSO algorithm can solve optimally large-scale location problems with networks up to 16,200 edges. Keywordsparticle swarm optimization;location problem;absolute p-center.