基于虚拟对比控制组的交通拥堵费安全影响研究  

Safety Effects of Congestion Charge Based onVirtual Synthetic Control Group

作　　者：李科君 李豪杰 朱曼曼 LI Kejun;LI Haojie;ZHU Manman(Urban CLP Big Data Research Institute Co.,Ltd.,Guiyang 550000,Guizhou,China;Transportation College,Southeast University,Nanjing 211189,Jiangsu,China;Jiangsu Key Laboratory of Urban ITS,Transportation College,Southeast University,Nanjing 211189,Jiangsu,China;Jiangsu Province Collaborative Innovation Center ofModern Urban Traffic Technologies,Southeast University,Nanjing 211189,Jiangsu,China)

机构地区：[1]中电科大数据研究院有限公司,贵州贵阳550000 [2]东南大学交通学院,江苏南京211189 [3]东南大学交通学院城市智能交通江苏省重点实验室,江苏南京211189 [4]现代城市交通技术江苏高校协同创新中心,江苏南京211189

出　　处：《重庆交通大学学报（自然科学版）》2020年第9期17-24,共8页Journal of Chongqing Jiaotong University(Natural Science)

摘　　要：以伦敦为研究案例,采用虚拟对比控制组的因果效应方法,估计伦敦中心区实施拥堵收费对交通事故伤亡人数的影响。通过对比伦敦实际发生的伤亡人数和虚拟对比控制组估计的伤亡人数,分析发现交通拥堵收费区的交通事故总伤亡人数减少了4.29%,其中轻伤和重伤数量各自减少了5.05%和12.12%。采用安慰剂试验对模型进行验证,结果显示伦敦交通拥堵费的评估效应达到95%的置信水平。研究结果表明:在交通安全分析研究中,当没有合适的对比控制组的情况下,虚拟对比控制方法可作为一个有效的研究方法。Taking London as a case study,the causal effect method of virtual synthetic control group was used to estimate the impact of congestion pricing on traffic accident casualties in central London.By comparing the actual number of casualties in London and the number of casualties estimated by the virtual synthetic control group,the total number of casualties in traffic congestion toll area decreased by 4.29%,in which the number of minor injuries and serious injuries decreased by 5.05%and 12.12%respectively.The proposed model was validated by placebo test.The results show that the evaluation effect of congestion charge in London reaches 95%confidence level.The results show that the virtual synthetic control method can be used as an effective research method when there is no suitable comparative control group in traffic safety analysis.

关 键 词：交通运输工程 伦敦交通拥堵费 虚拟对比控制组 因果效应 交通政策 安慰剂试验 

分 类 号：U121[交通运输工程]