分散搜索算法研究进展  

Advances in Scatter Search

作　　者：王晓晴[1] 唐加福[1] 韩毅[1] 

机构地区：[1]东北大学流程工业综合自动化教育部重点实验室,沈阳110004

出　　处：《系统仿真学报》2009年第11期3155-3160,共6页Journal of System Simulation

基　　金：国家自然科学基金(70601004;70625001;70721001);教育部科技研究重点项目(104064);教育部新世纪优秀人才支持计划(NCET-04-280)

摘　　要：分散搜索是进化计算领域一种新兴有效的计算技术,近几年受到学术界和工程界的广泛关注。分散搜索采用基于种群的全局搜索策略,较少地利用搜索过程的随机性,注重于采用一系列系统性方法来构建新解,提高搜索的集中性和多样性。阐述和剖析了分散搜索的基本原理和常用流程。在此基础上,对算法框架中的参考集更新方法、子集合并方法和内存策略等重要机制进行了比较系统地深入分析。重点探讨了分散搜索在多目标优化、连续优化以及混合优化等复杂环境下的研究。论述了分散搜索在物流与供应链、生产管理和图像处理等领域的典型应用情况并展望了分散搜索的发展前景。Scatter search （SS） is a novel and effective computing method, which receives increasing attention from both academic and industry fields in recent years. Scatter search adapts a population-based global search strategy, and makes only a limited use of randomization. The intensification and diversification of search can be significantly improved by constructing solutions systematical!y. The fundamental principles and framework of the scatter search were described. Particularly, reference set update method, subset combination method and memory strategy in its framework were discussed in detail. The researches on scatter search including multi-objective optimization, continuous optimization and hybrid optimization were extensively reviewed. The applications of scatter search in logistics and supply chain management, production management, image processing and so on were discussed. Moreover, future research directions of scatter search were stated.

关 键 词：分散搜索 参考集更新 子集合并 路径重连 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]