中国机械工程 ›› 2010, Vol. 21 ›› Issue (10): 1173-1178.

• 机械基础工程 • 上一篇    下一篇

小生境蚁群优化及其在JSSP中的应用研究

甘屹;李胜;张志伟
  

  1. 上海理工大学,上海,200093
  • 出版日期:2010-05-25 发布日期:2010-06-02
  • 基金资助:
    国家自然科学基金资助项目(0505030);上海市教育委员会“曙光”计划资助项目(07SG51);上海市教育委员会重点学科建设项目(J50503)
    National Natural Science Foundation of China(No. 0505030)

Study on Job Shop Scheduling Problems Based on Microhabitat Ant Colony Optimization

Gan Yi;Li Sheng;Zhang Zhiwei
  

  1. University of Shanghai for Science and Technology, Shanghai, 200093
  • Online:2010-05-25 Published:2010-06-02
  • Supported by:
    National Natural Science Foundation of China(No. 0505030)

摘要:

借鉴蚁群的并行、多样化寻优活动,提出蚁群基本调度规则。为了改进优化性能,提出小生境蚁群优化策略,从信息素分布的时变性、蚂蚁更新信息素策略和信息交流突变性方面改进了基本蚁群算法,提出将小生境蚁群优化调度规则(MACO SR)用于求解车间调度问题的方法。并在MACO SR的启发函数、更新路径等环节中加入蚂蚁等待时间要素。通过求解目标函数为最小化最大加工完成时间的车间调度问题,并与基本蚁群算法、蚁群基本调度规则进行比较,证明了小生境蚁群优化调度规则能获得相当好的优化结果,具有较好的寻优性能。

关键词:

Abstract:

Simulating the parallel and multiform optimized actions of ant colony, basic scheduling rules of ants (BSRA) were built up. To improve the optimized performance, microhabitat ant colony optimization stratage (MACOS) was built up to improve basic ant colony optimization (ACO) by pheromone time-varying distribution, pheromone updating tactic and information exchanging mutation. Ant colony optimization scheduling rules (MACO SR) were built up based on MACOS to solve JSSP. And waiting time was integrated into the heuristic function and path updating of MACO SR. MACO SR can achieve more satisfactory results than basic ACO and BSRA for JSSP with the objects function of minimum make-span. And MACO SR also shows significant optimization performance.

Key words:

中图分类号: