摘要
为了探究城市快速路上可变信息板VMS对驾驶员路径选择行为的影响,首先对驾驶员进行了RP和SP调查。然后基于得到的调查数据,采用Logistic作为连接函数,利用SAS进行广义线性混合模型的拟合,也同时进行了经典广义线性模型的回归。结果表明:对比于经典线性回归模型,广义线性混合模型更为有效地体现了同一受访者重复观测值之间的相关性;同时对驾驶员自身属性进行了分析,对VMS的设计给出建议。
In order to explore the influence of Variable Message Sign （VMS） on the driver＇s route selection behavior onurban expressway, the RP and SP investigation were conducted on the selected driver. Based on the survey data obtained,the Generalized Linear Mixed Model （GLMM ） was fitted withLogistic as the connection function by SAS. Meanwhile, thedata was also fitted by Generalized Linear Model （GLM） . Theresults show that the GLMM -based route choice model ismore effective on the correlations among the repeated observations from the same respondent. Furthermore, the driver＇s ownattributed were analyzed, and the advice of VMS design wasgiven.
作者
程邦录
干宏程
CHENG Banglu;GAN Hongcheng(Management School , University of Shanghai for Science and Technology, Shanghai 200093, Chin)
出处
《物流科技》
2018年第6期83-86,96,共5页
Logistics Sci-Tech
关键词
广义线性混合模型
快速路
选择行为
可变信息板
generalized linear mixed model
expressway
selection behavior
variable message sign
