基于VMS信息差异的驾驶人路径选择行为  

Driver’s route choice behavior based on variable message sign (VMS) information differences

作　　者：兆雪[1] 赵文静 马壮林[3] 张荣辉 ZHAO Xue;ZHAO Wen-jing;MA Zhuang-lin;ZHANG Rong-hui(School of Science,Xi'an Shiyou University,Xi'an 710065,Shaanxi,China;School of Traffic and Transportation Engineering,Central South University,Changsha 410075,Hunan,China;College of Transportation Engineering,Chang'an University,Xi'an 710064,Shaanxi,China)

机构地区：[1]西安石油大学理学院,陕西西安710065 [2]中南大学交通运输工程学院,湖南长沙410075 [3]长安大学运输工程学院,陕西西安710064

出　　处：《长安大学学报（自然科学版）》2020年第3期91-99,共9页Journal of Chang’an University(Natural Science Edition)

基　　金：教育部人文社会科学研究项目(18YJCZH130);陕西省自然科学基础研究计划项目(2017JM5084);中央高校基本科研业务费专项资金项目(300102228202,300102229666)。

摘　　要：为了探究可变信息标志(variable message sign, VMS)条件下驾驶人路径选择行为的影响因素,以西安市私家车和出租车驾驶人为调查对象,采用正交试验法设计调查问卷,采用行为(revealed preference,RP)调查和意愿(stated preference,SP)调查结合的方法,获取特定情景下驾驶人的路径选择行为;采用二项Logit模型构建驾驶人路径选择行为决策模型,并分别用Jackknife和Bootstrap抽样技术修正所构建的模型,对比分析上述3种模型的精度。在基于Bootstrap技术的二项Logit模型的基础上,分别讨论不同的VMS发布形式、拥堵原因、拥堵时间和更新频率对驾驶人路径选择行为的影响。研究结果表明:Jackknife技术不能提高Logit模型精度,而Bootstrap技术可以显著地提高Logit模型的精度;文字式和图文混合式的VMS发布形式对驾驶人路径选择行为有显著的正影响,且影响程度相差不大,而图形式的VMS发布形式对驾驶人路径选择行为没有显著的影响;拥堵原因对驾驶人改变路径的影响强度从大到小依次为交通事故、车流量大、道路施工;随着拥堵时间的增加,受过高等教育的驾驶人选择替代路径的概率逐渐增大;驾驶人更愿意接收到更新频率为30、60 s的建议绕行信息。研究结论有助于交通管理部门设计更可接受、易理解的VMS信息,从而为改善交通状况和提供个性化的出行信息服务奠定理论基础。In order to explore the factors affecting drivers’ route choice behavior under variable message sign(VMS) information environment, the orthogonal experimental was applied to design the questionnaire, and the data about private cars and taxi drivers’ route choice behavior in a specific situation was surveyed by a combination of revealed preference(RP) and stated preference(SP) survey method. A binary Logit(BL) model was used to establish the drivers’ route choice behavior decision model. Two BL models of drivers’ route choice behavior based on Jackknife and Bootstrap technology were built respectively, and the fitting accuracy of above three models were compared and analyzed. On the basis of the BL model based on Bootstrap technology, the impact of display form, congestion cause, congestion time, and update frequency in VMS on drivers’ route choice behavior were classified and discussed. The results show that Jackknife technology could not improve the accuracy of BL model, while Bootstrap technology could significantly improve the accuracy of BL model. The display format in VMS of text and text-graph have a significantly positive and approximate effect on drivers’ route choice behavior, while the effect of the graphic-aided information is not significant. In terms of influential degree of congestion cause, the information is ranked in the following order, traffic accident, heave traffic and road construction. With the increasing of congestion time, the probability of higher education drivers choosing alternative routes gradually increases. Drivers are more likely to receive the suggested alternative route information with the update frequency for 30 and 60 seconds. The conclusion of this study could assist traffic authorities to design the most acceptable and understandable VMS information, and lay a theory foundation for improving traffic situation and providing personal travel information services.

关 键 词：交通工程 VMS信息 路径选择 二项Logit模型 Jackknife和Bootstrap技术 

分 类 号：U491[交通运输工程—交通运输规划与管理]