Incorporating Neighborhood Reduction for the Solution of the Planar p-Median Problem

Two efficient neighbourhood reduction schemes are proposed for the solution of the p-Median problem on the plane. Their integration into a local search significantly reduces the run time with an insignificant deterioration in the quality of the solution. For completeness this fast local search is also embedded into one of the most powerful meta-heuristic algorithms recently developed for this continuous location problem. Excellent results for instances with up to 1060 demand points with various values of p are reported. Eight new best known solutions for ten instances of a large problem with 3,038 demand points and up to 500 facilities are also found