Hybrid Algorithm for Winner Determination Problem in Combinatorial Auctions

Genetic algorithm and ant colony algorithm are the methods for solving winner determination problem in combinatorial auctions. Because of their mutual compensability: ant colony algorithm can make up for the shortage of feedback information and the slowness of solving solution in genetic algorithm, while genetic algorithm is able to enhance the speed of ant colony algorithm in solving solution and avoid its liability to local optimal solution in some measure, this paper proposes an improved hybrid ant genetic algorithm is adopted to solve the problem. The algorithm improves the local pheromone, accelerates the convergence and enhances the efficiency in solving the phenomenon of pre-maturity and stagnation in process of convergence. The experiment results show that it is much better in both the speed of convergence and the quality of the solution compared with genetic algorithm and ant colony algorithm.