• 论文 •    

粒子群优化算法求解多模式项目再调度问题

程序,吴澄   

  1. 清华大学 自动化系,北京100084
  • 出版日期:2009-01-15 发布日期:2009-01-25

Solving multi-mode project rescheduling problem with PSO algorithm

CHENG Xu, WU Cheng   

  1. Department of Automation, Tsinghua University, Beijing 100084, China
  • Online:2009-01-15 Published:2009-01-25

摘要: 针对有资源约束的项目调度问题中,项目执行环境发生变更时的项目再调度问题,分析了项目执行环境变更的情况和项目再调度变更费用的构成,建立了以再调度变更费用为优化目标的多模式项目再调度模型并应用基于粒子群优化算法进行求解。在该算法中,粒子分为优先级粒子和模式粒子,实现对活动优先顺序和模式的优化。部分初始优先级粒子位置矢量通过启发式规则产生,以提高解的质量。通过仿真计算分析了关键参数的选择,并与其他启发式算法进行了比较。仿真结果表明,该算法能有效地求解多模式项目的再调度问题,在实际工程中有一定应用价值。

关键词: 项目管理, 粒子群优化算法, 多模式, 项目调度, 再调度, 启发式算法

Abstract: To deal with project scheduling problem with resource constraints and with changes in project execution circumstances, the circumstance changes and rescheduling cost constitution were analyzed. A multi-mode project rescheduling model with the optimization object of minimizing rescheduling cost was constructed,and an algorithm based on Particle Swarm Optimization (PSO) was used to solve. In this algorithm, particles were divided into priority particles and mode particles to optimize the activity priority and mode. Partial position vectors of initial priority particles were generated by heuristic rules to improve the quality of solution. Selection of key parameters were analyzed by simulations. Comparison of performances of PSO-based algorithm with other heuristic algorithms were conducted. Results of simulations showed that this algorithm could effectively solve multi-mode project rescheduling problems.

Key words: project management, particle swarm optimization algorithm, multi-mode, project scheduling, rescheduling, heuristic algorithms

中图分类号: