• 论文 •    

应用混合蚁群算法求解模糊作业车间调度问题

宋晓宇, 朱云龙, 尹朝万, 李富明,   

  1. 1.中国科学院 沈阳自动化研究所,辽宁沈阳110016;2.中国科学院 研究生院,北京100039
  • 出版日期:2007-01-15 发布日期:2007-01-25

Hybrid ant colony algorithm for fuzzy Job Shop scheduling

SONG Xiao-yu, ZHU Yun-long,YIN Chao-wan,LI Fu-ming   

  1. 1.Shenyang Inst. of Automation, Chinese Academy of Sciences, Shenyang110016, China; 2.Graduate Sch. of Chinese Academy of Sciences, Beijing100039, China
  • Online:2007-01-15 Published:2007-01-25

摘要: 为解决蚁群算法求解时间过长和易陷入局部最优的问题,提出了一种求解模糊作业车间调度问题的混合算法,该算法将蚁群算法用于全局搜索。为了提高搜索效率,根据作业车间调度问题解的特征,提出一种基于关键工序的邻域搜索方法,并使用此邻域搜索方法的禁忌搜索算法嵌入蚁群算法。利用禁忌搜索算法较强的局部搜索能力,提高了蚁群算法的优化能力,改善了作业车间调度问题解的质量。实验结果验证了该混合搜索算法的有效性,其优化效果优于并行遗传算法和禁忌搜索算法。

关键词: 蚁群算法, 禁忌搜索, 混合算法, 模糊加工时间

Abstract: Due to the low efficiency of ant colony algorithms and easily plunged into local optimal, a hybrid ant colony algorithm for fuzzy Job Shop scheduling was proposed. In this method, the ant colony algorithm was served as a global search algorithm. According to the characteristics of Job Shop scheduling, a new TS algorithm based on the critical operation was presented, which was used as the local search algorithm in the hybrid ant colony algorithm. Because of the stronger local search ability of TS algorithm, this hybrid algorithm could improve the solutions′ quality. Experimental results verified the effectiveness of the proposed hybrid algorithm by comparing with TS algorithm and parallel genetic algorithms.

Key words: ant colony algorithm, taboo search, hybrid algorithm, fuzzy processing time

中图分类号: