交通运输系统工程与信息 ›› 2011, Vol. 11 ›› Issue (2): 100-105.

• 系统工程理论与方法 • 上一篇    下一篇

基于网络对偶均衡的有边约束的交通流分配模型

何胜学*,何建佳,徐福缘   

  1. 上海理工大学 管理学院,上海 200093
  • 收稿日期:2010-11-03 修回日期:2011-01-11 出版日期:2011-04-28 发布日期:2011-05-10
  • 通讯作者: 何胜学
  • 作者简介:何胜学(1976-),男,陕西三原人,讲师.
  • 基金资助:

    国家自然科学基金 (70672110);上海市(第三期)重点学科(S30504);上海市研究生创新基金项目(JWCXSL1001).

Side Constrained Traffic Assignment Model Based on Network Dual Equilibrium

HE Sheng-xue, HE Jian-jia, XU Fu-yuan   

  1. School of Management, University of Shanghai for Science and Technology, Shanghai 200093,China
  • Received:2010-11-03 Revised:2011-01-11 Online:2011-04-28 Published:2011-05-10
  • Contact: HE Sheng-xue

摘要: 利用网络对偶均衡理论,依据“局部近视”用户均衡原则建立了具有一般边约束的网络交通流分配模型. 将交通网络中的流量与行程时间看作一对对偶的变量. 从网络的基本组成元素入手,首先考虑网络节点的流量守恒条件与节点距起点最小行程时间对偶关系,然后考虑路段流量与“局部近视”用户路段行程时间约束条件的对偶关系,最后通过整合上述对偶关系,并增加一般边约束建立了新的交通流分配模型. 分析了模型求解过程中如何体现“优先出牌”与“在途调整弹性”两个择路行为假设. 利用模型求解结果中分起讫点对的路段流量唯一的特点,给出了确定有效路径集的搜索算法. 用算例验证了模型及算法的有效性,并对具有一般边约束的流量分配模型的计算结果从拥挤收费和路段排队延误角度进行了解释.

关键词: 系统工程, 交通流分配, 网络均衡, 边约束, 有效路径

Abstract: One side constrained traffic assignment model is presented based on local nearsighted user equilibrium principle by taking advantage of network dual equilibrium theorem. The traffic flow and travel time are regarded as a pair of dual variables. The analysis is started from the basic elements of network. At first, the dual relationships between the conservation conditions of node’s flow and the shortest travel time from origin to the node are considered. Then the dual relationships between link flow and travel time constrains on link with nearsighted users are also analyzed. At last, through combining these dual relationships and adding generalized side constraints, new traffic assignment model is built. How to embody two behavioral assumptions (priority of playing cards and elasticity of route adjusting on route) in the proposed algorithm to the model is analyzed. A searching method of effective path set is given by using the model’s solution results of unique link flow with respect to origin and destination pair. Several numeral examples are given to prove the validity and efficiency of the model and related algorithm. The results for the side constrained model are explained by congestion pricing and queuing delay on links.

Key words: system engineering, traffic assignment, network equilibrium, side constraints, effective path

中图分类号: