基于级差效益的驾驶员个性化路径选择  

Driver’s Personalized Route Choice Based on Differential Benefit

作　　者：张惠玲 陈华伟[1] 敖谷昌 ZHANG Hui-ling;CHEN Hua-wei;AO Gu-chang(Chongqing Key Lab of Traffic System&Safety in Mountain Cities,Chongqing 400074,China;School of Traffic&Transportation,Chongqing Jiaotong University,Chongqing 400074,China)

机构地区：[1]山地城市交通系统与安全重庆市重点实验室,重庆400074 [2]重庆交通大学交通运输学院,重庆400074

出　　处：《公路交通科技》2019年第7期114-120,共7页Journal of Highway and Transportation Research and Development

基　　金：国家自然科学基金项目(51508061);重庆市教委科学技术研究项目(KJQN201800727);山地城市交通系统与安全重庆市重点实验室开放基金项目(2018TSSMC03)

摘　　要：在公路成网的条件下,出行路径选择成为智能交通系统的研究重点。考虑到路径选择的多属性和不确定性,提出一种基于级差效益的、体现驾驶员个性化出行需求的路径选择模型,该模型通过计算可行路径的综合级差效益来量化公路综合属性值,可以克服公路综合属性值在量化过程中的强主观性,同时,根据驾驶员的出行需求为其规划出行路径,能够避免单一优化目标下最优路径的车辆过度聚集所造成路网交通流失衡。模型首先参考不确定型多属性决策模型,构建公路级差效益的综合评价指标体系,获取各个评价指标的级差效益定量化模型,以此得到规范化决策矩阵;然后针对驾驶员需求的主观性和不确定性,利用模糊分析法确定驾驶员个性化路径选择的指标权重,求得可行路径的综合级差效益;最后根据可行路径的综合级差效益,对可行路径方案进行排序,从而选出最优路径。以重庆市黔江区公路网为例,基于级差效益的驾驶员个性化路径选择模型为两个具有不同出行需求的驾驶员规划了不同的路径,为关心成本的驾驶员1规划了路径2,为关心安全的驾驶员2规划了路径1,克服公路属性值在量化过程中的强主观性,同时,避免单一优化目标下最优路径的车辆过度聚集,可为公路信息服务和驾驶员出行选择提供依据。Under the condition of highway network, travel route choice has become a research emphasis of ITS. Considering the multi-attribute and uncertainty of route choice, a route choice model which based on differential benefit and reflects the driver’s personalized travel demand is proposed. By calculating the comprehensive differential benefit of feasible routes to quantify the comprehensive attribute value of highway, the model can overcome the strong subjectivity of the comprehensive attribute value of highway in the quantification process. At the same time, planning the travel route according to the driver’s travel demand, the model can avoid the traffic flow imbalance in road network caused by the excessive aggregation of vehicles in the optimal route with single optimization objective. First, referring to the uncertain multiple-attribute decision-making model, the model constructs the comprehensive evaluation index system of highway differential benefit, and obtains the quantitative model of differential benefit of each evaluation index, so as to get the normalized decision matrix. Next, according to the driver’s subjectivity and uncertainty, the model uses the fuzzy analysis to determine the index weight of driver’s personalized route choice, and solves the comprehensive differential benefit of feasible routes. Last, according to the comprehensive differential benefit of the feasible routes, the model sorts the feasible schemes to select the optimal route. Taking the highway network of Qianjiang District in Chongqing for example, different routes for 2 drivers with different travel demands are planned by the driver’s personalized route choice model based on differential benefits, in which route 2 is planned for driver 1 concerned with cost, route 1 is planned for driver 2 concerned with safety, overcoming the strong subjectivity of highway attribute values in the quantification process, and avoiding the excessive aggregation of vehicles in the optimal route with single optimization objective, which can

关 键 词：智能交通 驾驶员路径选择 级差效益 多属性决策 个性化 

分 类 号：U491.25[交通运输工程—交通运输规划与管理] U116.2[交通运输工程—道路与铁道工程]