Journal
IEEE TRANSACTIONS ON CYBERNETICS
Volume 49, Issue 1, Pages 287-300Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCYB.2017.2772250
Keywords
Artificial bee colony (ABC) algorithm; decomposition strategy; dynamic resource allocation; many-objective optimization
Categories
Funding
- National Natural Science Foundation of China [61773410, 61673403, 61472143]
- Foundation of Key Laboratory of Machine Intelligence and Advanced Computing of the Ministry of Education [MSC-201602A]
- Scientific Research Special Plan of Guangzhou Science and Technology Programme [201607010045]
- Excellent Graduate Student Innovation Program from the Collaborative Innovation Center of High Performance Computing
Ask authors/readers for more resources
In this paper, a decomposition-based artificial bee colony (ABC) algorithm is proposed to handle many-objective optimization problems (MaOPs). In the proposed algorithm, an MaOP is converted into a number of subproblems which are simultaneously optimized by a modified ABC algorithm. The hybrid of the decomposition-based algorithm and the ABC algorithm can make full use of the advantages of both algorithms. The former, with the help of a set of weight vectors, is able to maintain a good diversity among solutions, while the latter, with a fast convergence speed, is highly effective when solving a scalar optimization problem. Therefore, the convergence and diversity would be well balanced in the new algorithm. Moreover, subproblems in the proposed algorithm are handled unequally, and computational resources are dynamically allocated through specially designed onlooker bees and scout bees. The proposed algorithm is compared with five state-of-the-art many-objective evolutionary algorithms on 13 test problems with up to 50 objectives. It is shown by the experimental results that the proposed algorithm performs better than or comparably to other algorithms in terms of both quality of the final solution set and efficiency of the algorithms. Finally, as shown by the Wilcoxon signed-rank test results, the onlooker bees and scout bees indeed contribute to performance improvements of the algorithm. Given the high quality of solutions and the rapid running speed, the proposed algorithm could be a promising tool when approximating a set of well-converged and properly distributed nondominated solutions for MaOPs.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available