Improved Real-coded Genetic Algorithm for Planar Maximal Covering Location Problem

Facility location is a strategic optimization problem in organizational decision making. Maximal coverage problems are a class of widely used facility location problems. In this paper, PMCLP problem is considered, and a real-coded genetic algorithm is introduced to attain near-optimal answers for PMCLP in short time. Initial solutions of the problem are obtained by a randomized greedy heuristic. These answers are then evolved by innovative demand-based crossover operator and a virtual-force mutation operator in successive iterations. To evaluate effectiveness and efficiency of the proposed method and its genetic operators, several medium and large-scale problems are solved in our experiments. Results are then compared with an exact algorithm. The experimental results show that the proposed real-coded genetic algorithm provides near-optimal answers to large-scale problems in tolerable time.

Keywords:

facility location problems , PMCLP , genetic algorithm