Genetic algorithm and local search comparison for solving bi-objective p-Median problem

Abstract

This paper presents two algorithms, which are a nondominated sorting genetic algorithm II (NSGA-II) and an indicator-based multi-objective local search (IBMOLS), for solving a bi-objective p-Median problem. The bi-objective p-Median problem is a problem of finding p location points to install facilities from a set of m candidates. This problem considers two objectives: minimizing the sum of the distances from each customer to the nearest facility and minimizing the sum of the costs to install each facility in the selected location points. NSGA-II and IBMOLS are efficient algorithms in the area of multi-objective optimization. Experiments are conducted on generated instances. Hypervolume values of the approximate Pareto fronts are computed and the obtained results from IBMOLS and NSGA-II are compared.