A Multi-objective Optimization Algorithm Based on Preference Three-Way Decomposition
Most of refining processes were optimized using single objective approach, but practically such complex processes must be optimized with several objectives. Inspired by the theory of three-way decisions, a multi-objective optimization algorithm based on preference three-way decomposition is proposed in this paper. First, according to the preferences of the DM, the analytic hierarchy process (AHP) is used to sort objectives. Then, based on the idea of three-way decisions, these objectives are divided into three sub-parts as the primary objective set, the secondary objective set and the general objective set. Besides, a multi-group parallel optimization algorithm is presented to solve each sub-optimization problem. Finally, based on Non-dominated set of the three sub-problems, a set of external preservation sets are formed so as to get the optimal set that the DM is interested in. Experimental results show that the proposed method can reduce the workload of the DM and obtain more accurately converge to the optimal frontiers of the optimization problems.
KeywordsMulti-objective optimization Decision making Preference information Three-way decisions Decomposition
This work was supported in part by the National Natural Science Foundation of China under Grant Nos. 61751312, 61533020 and 61379114.
- 1.Deb, K., Thiele, L., Laumanns, M., Zitzler, E.: Scalable test problems for evolutionary multiobjective optimization. In: Abraham, A., Jain, L., Goldberg, R. (eds.) Evolutionary Multiobjective Optimization. Advanced Information and Knowledge Processing. Springer, London (2005). https://doi.org/10.1007/1-84628-137-7_6CrossRefMATHGoogle Scholar
- 3.Deb, K., Sundar, J.: Reference point based multi-objective optimization using evolutionary algorithms. In: Conference on Genetic and Evolutionary Computation, pp. 635–642. ACM (2006)Google Scholar
- 6.Narzisi, G.: Multi-objective optimization: a quick introduction. New York University lectures (2008)Google Scholar
- 7.Purshouse, R.C., Fleming, P.J.: Evolutionary many-objective optimisation: an exploratory analysis. Evol. Comput. 3, 2066–2073 (2003). CECGoogle Scholar
- 10.Zeman, J.: Ontological and gnoseological aspects of contradiction and their importance in analysis of the development of scientific knowledge. In: Handbook of Professional Ethics for Psychologists (1984)Google Scholar