Particle Swarm Optimization algorithm for the continuous p-median location problems

A Particle Swarm Optimization (PSO) algorithm is proposed for the solution of large-scale continuous p-median location problems in plane. A PSO algorithm previously developed for the solution of the continuous p-center has been extended to solve the p-median location problem. The p-median location problem seeks to locate p facilities in plane so that the sum of Euclidean distances from p medians to a set of n demand points is minimized. The p-median problem is proven to be NP-hard. In this paper, we introduce a new PSO algorithm for large-scale p-median location problems. Computational results show that the proposed PSO algorithm is able to find good feasible solutions for instances from TSP library with demand points up to 9,976 and medians (p) up to 75. Our proposed PSO algorithm outperformed one of the most recent aggregation methods, which has been developed to solve large-scale p-median.