A multiagent evolutionary algorithm for constraint satisfaction problems
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Hi-index | 0.00 |
From the viewpoint of decision making process, it brings inconveniences for decision makers to select one (few) proper solution(s). Thus we propose preference oriented two-layered multiagent evolutionary algorithm (TL-MAEA) to meet customers' needs. The algorithm has a structure of two layers: in the top layer, preference relations among multiple objectives are calculated through interactions with the decision maker; while in the bottom layer, MAEA is employed to obtain the optimal solution corresponding to the preference relations. In the experimental, 12 benchmark problems are used to test the algorithm. The results show that the proposed algorithm is effective.