Evolutionary algorithms with user’s preferences for solving hybrid interval multi-objective optimization problems

doi:10.1007/s10489-015-0658-x

CORC > 兰州理工大学 > 兰州理工大学

	Evolutionary algorithms with user’s preferences for solving hybrid interval multi-objective optimization problems
	Gong, Dunwei 2,3; Liu, Yiping 2; Ji, Xinfang 1; Sun, Jing 4
刊名	Applied Intelligence
	2015-10-22
卷号	43 期号:3 页码:676-694
关键词	Multiobjective optimization Quality control Evolutionary optimizations Hybrid objective Interval objectives Interval parameter Multi-objective optimization problem Optimization algorithms Optimization problems User preference
ISSN号	0924669X
DOI	10.1007/s10489-015-0658-x
英文摘要	Hybrid interval multi-objective optimization problems are common in real-world applications. These problems involve both explicit and implicit objectives, and the values of these objectives are intervals. Few previous methods are suitable for them. An evolutionary algorithm with a large population and a user’s interval preferences was presented to effectively solve the problems in this paper. In the proposed algorithm, a similarity-based strategy was employed to estimate the interval values of implicit objectives of evolutionary individuals that the user had not evaluated in order to alleviate user fatigue; the user’s preferences to different objectives were expressed precisely as intervals by solving an auxiliary optimization problem; a sorting scheme based on the user’s preferences was proposed to guide the population evolving toward the user’s preferred regions. We applied the proposed method to an interior layout problem, which is a typical optimization problem with both interval parameters in the explicit objective and interval value of the implicit objective. The proposed algorithm was compared with four other optimization algorithms on the interior layout problem. Experimental results validated its effectiveness and superiority over the compared algorithms in terms of solution quality and the number of user’s evaluations. © 2015, Springer Science+Business Media New York.
资助项目	Natural Science Foundation of Jiangsu Province[BK2012566]
WOS研究方向	Computer Science
语种	英语
出版者	Kluwer Academic Publishers
WOS记录号	WOS:000361391200013
状态	已发表
内容类型	期刊论文
源URL	[http://ir.lut.edu.cn/handle/2XXMBERH/113555]
专题	兰州理工大学
通讯作者	Liu, Yiping
作者单位	1.China Univ Min & Technol, Yinchuan Coll, Dept Mech Power & Informat Engn, Yinchuan, Peoples R China 2.China Univ Min & Technol, Sch Informat & Elect Engn, Xuzhou, Peoples R China 3.LanZhou Univ Technol, Sch Elect Engn & Informat Engn, Lanzhou, Peoples R China 4.Huai Hai Inst Technol, Sch Sci, Lianyungang, Peoples R China
推荐引用方式 GB/T 7714	Gong, Dunwei,Liu, Yiping,Ji, Xinfang,et al. Evolutionary algorithms with user’s preferences for solving hybrid interval multi-objective optimization problems[J]. Applied Intelligence,2015,43(3):676-694.
APA	Gong, Dunwei,Liu, Yiping,Ji, Xinfang,&Sun, Jing.(2015).Evolutionary algorithms with user’s preferences for solving hybrid interval multi-objective optimization problems.Applied Intelligence,43(3),676-694.
MLA	Gong, Dunwei,et al."Evolutionary algorithms with user’s preferences for solving hybrid interval multi-objective optimization problems".Applied Intelligence 43.3(2015):676-694.