A Parallel Ant Colony Optimization for the Maximum-Weight Clique Problem

In this paper, we propose a parallel ant colony optimization based metaheuristic for solving the maximum-weight clique problem, which is a variation of the maximum clique problem. The advised parallel computing model is based on concept of cooperation among multiple ant colonies system. The cooperation system consists of a message center and a number of ant colonies. Each ant colony attempts to explore the solution space by using its own search strategy. The message center first collects the solution information from different ant colonies, and then it shares the current best solution with them. The performance of the proposed method was evaluated on a set of the standard benchmark instances from literature. The obtained results were compared to those reached by the Cplex solver and the best solutions reported in the literature. From the experimental results, one can observe that encouraging results have been obtained.