Solving the Capacitated Clustering Problem Using Simulated Annealing and Genetic Algorithms

The Capacitated Clustering Problem (CCP) is a classical location problem with various applications in data mining. In the capacitated clustering problem, a set of n entities is to be partitioned into p disjoint clusters, such that the total dissimilarity within each cluster is minimized subject to constraints on maximum cluster capacity. Dissimilarity of a cluster is the sum of the dissimilarities between each entity that belongs to the cluster and the median associated with the cluster. In this paper two solution methods proposed for the problem. First method is a simulation annealing algorithm which uses different neighborhood structures randomly. The second method is a genetic algorithm approach which strengthened by a heuristic local search method. Computational results of test samples from literature demonstrate the robustness and efficiency of the proposed solution methods. This confirms that the proposed algorithm provides high quality solutions in reasonable time.