计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (28): 236-238.DOI: 10.3778/j.issn.1002-8331.2008.28.077

• 工程与应用 • 上一篇    下一篇

基于遗传算法的动态MC2运输问题求解

薛 锋,罗 建   

  1. 西南交通大学 交通运输学院,成都 610031
  • 收稿日期:2007-11-13 修回日期:2008-03-12 出版日期:2008-10-01 发布日期:2008-10-01
  • 通讯作者: 薛 锋

Solution of dynamic MC2 transportation problem based on genetic algorithm

XUE Feng,LUO Jian   

  1. College of Traffic and Transportation,Southwest Jiaotong University,Chengdu 610031,China
  • Received:2007-11-13 Revised:2008-03-12 Online:2008-10-01 Published:2008-10-01
  • Contact: XUE Feng

摘要: 动态MC2运输问题是描述多阶段供求波动的运输问题,其模型框架可以应用到很多领域。目前对动态MC2运输模型的求解主要采用传统的单纯形法,针对该问题的特殊性采用具有全局搜索能力的遗传算法进行求解。通过三维数组编码,设计有效的交叉、变异算子和适应度函数,克服了单纯形法求解该问题出现的并行性差、求解整数规划困难的不足。用Matlab7.0编程对算法进行检验,结果表明经过特殊设计的遗传算法能够很好地解决动态MC2运输问题。

关键词: 多目标, 多约束, 动态运输问题, 遗传算法

Abstract: Dynamic MC2(Multiple Criteria and Multiple Constraints) transportation problem describes the supply and demand fluctuates of many stages,and its model frame can be applied to a lot of fields.Dynamic MC2 transportation model is solved mainly by the traditional simplex method at present,genetic algorithm can solve it using global searching according to its particularity.The improving genetic algorithm can overcome the deficiencies of poor parallelism and difficult solving integer program by simplex method with three-dimensional group code,and then effective crossover and mutation operator,fitness function are designed.Examining the algorithm with Matlab7.0 programming,the result shows that dynamic MC2 transportation problem can be solved by hybrid genetic algorithm.

Key words: multiple criteria, multiple constraints, dynamic transportation problem, genetic algorithm