基于混合Logit模型的轨道交通安全统计生命价值

doi:10.16265/j.cnki.issn1003-3033.2020.04.029

中国安全科学学报 ›› 2020, Vol. 30 ›› Issue (4): 186-191.doi: 10.16265/j.cnki.issn1003-3033.2020.04.029

基于混合Logit模型的轨道交通安全统计生命价值

刘文歌副教授, 杨晶讲师

大连交通大学经济管理学院,辽宁大连 116028

收稿日期:2020-01-11 修回日期:2020-03-19 出版日期:2020-04-28 发布日期:2021-01-27
作者简介:刘文歌(1981—),女,辽宁大连人,博士,副教授,主要从事交通安全与交通经济方面的研究。E-mail:18737653@qq.com。
基金资助:
教育部人文社会科学研究项目(16YJC630075);国家自然科学基金资助(51608088);辽宁省社会科学规划基金资助(L15BGL003);辽宁省教育厅科学研究项目(JDW2016002)。

Value of statistical life in rail traffic safety based on mixed Logit model

LIU Wen'ge, YANG Jing

School of Economics and Management, Dalian Jiaotong University, Dalian, Liaoning 116028, China

Received:2020-01-11 Revised:2020-03-19 Online:2020-04-28 Published:2021-01-27

摘要/Abstract

摘要： 为了提供轨道交通安全项目经济评价的基础指标,开展轨道交通安全统计生命价值(VOSL)测算研究。基于参数服从正态分布和对数正态分布的混合Logit(ML)模型,构建轨道交通安全VOSL测算模型;以“大连至沈阳出行路径选择”为研究场景设计出行路径选择调查问卷,实施交通意向调查并获得调查数据;利用Monte Carlo仿真算法进行模型标定和模型比较分析,最终获得轨道交通安全VOSL测算值及其分布函数。结果表明:参数服从对数正态分布的ML-2模型,其拟合优度为0.213 5,命中率为82.49%,模型精确性较高;基于ML-2模型的轨道交通安全VOSL服从参数为(0.615 5,0.514 8²)的对数正态分布,其数学期望为616.9万元,且在取值为410万元时达到最大概率。

关键词: 轨道交通安全, 统计生命价值(VOSL), 混合Logit(ML)模型, Monte Carlo仿真算法, 意愿选择法

Abstract: In order to provide basic indicators for economic evaluation of rail traffic safety projects, estimation studies on VOSL in rail traffic safety were carried out. Firstly, VOSL calculation model was constructed based on ML model with normal distribution and lognormal distribution. Secondly, route-choice questionnaires were designed with hypothetical scenario of "Dalian to Shenyang route choice", and then survey was conducted and data were obtained. Finally, the model was calibrated by Monte Carlo simulation and comparatively analyzed before VOSL and its distribution function were gained. The results show that ML-2 model with lognormal distribution has high accuracy, whose goodness of fit is 0.213 5 and hit ratio 82.49%. VOSL based on ML-2 model obeys lognormal distribution with parameters (0.615 5,0.514 8²), who has a mathematical expectation of 6.169 million RMB and reaches the maximum probability at 4.1 million RMB.

Key words: rail traffic safety, value of a statistical life(VOSL), mixed Logit(ML) model, Monte Carlo simulation, stated choice method

中图分类号:

X915.1

刘文歌, 杨晶. 基于混合Logit模型的轨道交通安全统计生命价值[J]. 中国安全科学学报, 2020, 30(4): 186-191.

LIU Wen'ge, YANG Jing. Value of statistical life in rail traffic safety based on mixed Logit model[J]. China Safety Science Journal, 2020, 30(4): 186-191.

参考文献

[1] 沧浪之歌. 解读致命真相:德国史上最严重高铁事故揭秘[EB/OL]. [2018-12-26]. http://www.360doc.com/content/18/1126/20/31326041_797415186.shtml.
[2] 2008年1月23日山东动车组D59次列车撞倒铁道施工人员18死9伤[EB/OL]. [2015-01-22]. https://www.bbaqw.com/js/82.htm.
[3] LIU Feng. “7.23”温州动车事故六周年祭—还原当年动车事故真相—为什么要埋掉车头[EB/OL]. [2017-07-24]. http://www.chinairn.com/hyzx/20170724/144451773.shtml.
[4] SCHELLING T C. The life you save may be your own [C]. Problems in Public Expenditure Analysis, 1968: 127-162.
[5] 刘文歌.道路交通安全统计生命价值评价研究:基于四边界二分式条件价值法[J]. 中国安全科学学,2019, 29(5): 131-137
LIU Wen'ge. Evaluation study on VOSL in road traffic safety based on quadruple-bound dichotomous CVM[J]. China Safety Science Journal, 2019, 29(5): 131-137.
[6] RIZZI L I, ORTÜZAR J. Stated preference in the valuation of interurban road safety[J]. Accident Analysis and Prevention, 2003, 35(1): 9-22.
[7] IRAGÜEN P, ORTÜZAR J. Willingness-to-pay for reducing fatal accident risk in urban areas: an internet-based web page stated preference survey[J]. Accident Analysis and Prevention, 2004, 36(1): 513-524.
[8] RHEINBERGER C M. A mixed logit approach to study preferences for safety on alpine roads[J]. Environmental and Resource Economics, 2011, 49(1): 121-146.
[9] 刘文歌, 赵胜川. 基于Logit模型的道路交通统计生命价值[J]. 交通运输系统工程与信息, 2013,13(1): 137-141.
LIU Wen'ge, ZHAO Shengchuan. The value of statistical life in road traffic based on logit model[J]. Journal of Transportation Systems Engineering and Information Technology, 2013,13(1): 137-141.
[10] 刘文歌, 赵胜川. 基于Mixed Logit模型的道路交通统计生命价值评价研究[J]. 统计信息论坛, 2013, 28(2): 109-112.
LIU Wen'ge, ZHAO Shengchuan. The evaluation study on the value of a statistical life in road traffic based on mixed logit model[J]. Statistics & Information Forum, 2013, 28(2): 109-112.
[11] ZHAO Yang, PAN Liu, XIN Xu. Estimation of social value of statistical life using willingness-to-pay method in Nanjing, China[J]. Accident Analysis and Prevention, 2016,95: 308-316.
[12] MANN T. Rail safety and the value of a life[N]. Wall Street Journal, 2013-06-18(1).
[13] COVEY J, ROBINSON A, JONES-LEE M, et al. Responsibility, scale and the valuation of rail safety[J]. Journal of Risk Uncertain, 2010, 40: 85-108.
[14] JONES-LEE M, SPACKMAN M. The development of road and rail transport safety valuation in the United Kingdom[J]. Research in Transportation Economics, 2013, 43: 23-40.
[15] 杨利强, 黄亮, 张宁, 等. 基于Mixed Logit模型的城市轨道交通出行时间价值估计[J]. 城市轨道交通研究, 2015(5): 29-32,37.
YANG Liqiang, HUANG Liang, ZHANG Ning, et al. Calculation of travel time value for urban rail transit based on mixed logit model [J]. Urban Mass Transit, 2015(5): 29-32,37.
[16] 晶荣宏祥.辽宁2017年各地工资排行榜出炉[EB/OL]. [ 2017-02-22]. http://www.360doc.com/content/17/0222/20/33542116_631215151.shtml.
[17] 国家铁路局. 2017铁路安全情况公告[EB/OL]. [ 2018-03-30]. http://www.nra.gov.cn/xwzx/xwdt/xwlb/201803/t20180330_54494.shtml.
[18] MATTHIAS Gsul, 胡予红, 周旋,等. 道路交通运输安全发展报告(2017)[J]. 中国应急管理,2018(2): 48-58.
[19] 宗会明, 何舟, 杨庆媛, 等. 基于SP方法的重庆居民轨道交通出行时间价值及影响因素分析[J]. 现代城市研究, 2014(2): 115-120.
ZONG Huiming, HE Zhou, YANG Qingyuan, et al. The value of travel time of the rail transportation based on SP method: a case study of Chongqing city[J]. Modern Urban Research, 2014(2): 115-120.
[20] 赵胜川, 王喜文, 姚荣涵, 等. 基于Mixed Logit模型的私家车通勤出行时间价值[J]. 吉林大学学报, 2010, 40(2): 406-411.
ZHAO Shengchuan, WANG Xiwen, YAO Ronghan, et al. Value of driver commuting travel time based on mixed logit model[J]. Journal of Jilin University 2010, 40(2): 406-411.
[21] 周洋. 深圳大学居民出行时间价值标定及其在交通分布预测中的运用:以东莞市为例[D]. 深圳:深圳大学,2016.
ZHOU Yang. The value of time calibration and its application in traffic distribution prediction: taking Dongguan as an example[D]. Shenzhen: Shenzhen University, 2016.

基于混合Logit模型的轨道交通安全统计生命价值

Value of statistical life in rail traffic safety based on mixed Logit model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

本文评价

[1]	刘文歌. 道路交通安全统计生命价值评价研究:基于四边界二分式条件价值法[J]. 中国安全科学学报, 2019, 29(5): 131-137.
[2]	刘文歌，赵胜川. 道路交通安全统计生命价值评价研究——基于单边界和双边界二分式条件价值法[J]. 中国安全科学学报, 2013, 23(11): 138-.