The Maximal Covering Location Problem (MCLP) seeks to locate facilities in order to maximize the serviced population, considering a given distance or standard service time. Various extensions of this model have been proposed to enhance its applicability, e.g., probabilistic models for maximum location-allocation coverage with waiting time or queue length constraints for congested systems, taking into account one or more servers per service center. In this paper we present two procedures for solving a probabilistic model, which considers one server per center, using Lagrangian relaxation and the Constructive Genetic Algorithm. Extensive tests of these approaches are presented and their results compared.
location problems; maximal coverage; lagrangian relaxation; constructive genetic algorithm