An improved origin-based algorithm for solving the combined distribution and assignment problem

被引:15
|
作者
Xu, Meng [1 ]
Chen, Anthony [2 ]
Gao, Ziyou [1 ]
机构
[1] Beijing Jiaotong Univ, Inst Syst Sci, Sch Traff & Transportat, Beijing 100044, Peoples R China
[2] Utah State Univ, Dept Civil & Environm Engn, Logan, UT 84322 USA
基金
美国国家科学基金会; 中国国家自然科学基金;
关键词
combined distribution and assignment problem; Evans' algorithm; origin-based algorithm; trip distribution; traffic assignment;
D O I
10.1016/j.ejor.2007.04.047
中图分类号
C93 [管理学];
学科分类号
12 ; 1201 ; 1202 ; 120202 ;
摘要
This paper is to further study the origin-based (013) algorithm for solving the combined distribution and assignment (CDA) problem, where the trip distribution follows a gravity model and the traffic assignment is a user-equilibrium model. Recently, the OB algorithm has shown to be superior to the Frank-Wolfe (FW) algorithm for the traffic assignment (TA) problem and better than the Evans' algorithm for the CDA problem in both computational time and solution accuracy. In this paper, a modified origin-destination (OD) flow update strategy proposed by Huang and Lam [Transportation Research Part B 26 (4), 1992, 325-337] for CDA with the Evans' algorithm is adopted to improve the OB algorithm for solving the CDA problem. Convergence proof of the improved OB algorithm is provided along with some preliminary computational results to demonstrate the effect of the modified OD flow update strategy embedded in the OB algorithm. (c) 2007 Elsevier B.V. All rights reserved.
引用
收藏
页码:354 / 369
页数:16
相关论文
共 50 条