道路网单向交通优化设计模型  

Optimization design model of one-way traffic for road network

作　　者：杨忠振[1] 邬珊华[1] 罗红红[2] 焦新龙[3] 

机构地区：[1]大连海事大学交通运输管理学院,辽宁大连116026 [2]中国科学技术信息研究所,北京100038 [3]宁波工程学院交通与物流学院,浙江宁波315211

出　　处：《交通运输工程学报》2012年第5期72-78,共7页Journal of Traffic and Transportation Engineering

基　　金：国家自然科学基金项目(51078049);宁波市自然科学基金项目(2012A610153)

摘　　要：为了提高道路网的通行能力和服务水平,提出了单向交通组织的双层规划模型。上层模型的目标函数为最小总走行时间、最短绕行距离和最少交叉口冲突点数量,用于求解单向交通的优化方案;下层模型为用户均衡交通量分配模型,用于计算上层模型给出的道路网中的路段交通流特征。提出路段组合、可行解判断等方法减少决策变量和解空间,设计多目标遗传算法进行求解,并以威海市数据进行实例分析。分析结果表明:在实施单向交通组织后,机动车流量与道路容量的平均比值由0.516 1下降至0.487 1,平均值的方差从0.418 9下降至0.379 1,机动车的平均出行时间由27.1min下降至25.4min,单行引起的平均绕行时间仅有0.61min。可见,提出的模型可应用于中等城市道路网的单向交通方案设计。In order to improve the capacity and service level of road network, a bi-level programming model for one-way traffic was put out. The objective functions of upper-level model were the minimum total trip time, the shortest bypassing distance and the least conflict point number, and the upper-level model was used to solve the optimized scheme of one-way traffic. The lower-level model was user equilibrium traffic assignment model, and was used to calculate the traffic flow characteristic of section in road network from the upper-le,）el model. The methods such as road section combination and feasible solution judgment were put out to decrease the decision variables and the solution space, the multi-objective genetic algorithm was designed to find the solutions, and example analysis was carried out by the data in Weihai City. Analysis result indicates that after the implementation of one-way traffic, the average ratio of vehicle flow to road capacity decreases from 0. 516 1 to 0. 487 1, and the variance of average ratio reduces from 0. 418 9 to 0. 379 1. The average trip time of vehicle decreases from 27.1 min to 25.4 min, while the average circumambulated time caused by one way traffic is only 0.61 min. It is clear that the proposed model can be used in the scheme design of one-way traffic in a medium-sized city. 10 figs, 17 refs.

关 键 词：交通规划 单向交通 双层规划模型 用户均衡分配 多目标遗传算法 路段组合 可行解 

分 类 号：U491.13[交通运输工程—交通运输规划与管理]