期刊文献+

山区高速公路货车事故严重度致因及随机参数分析 被引量:1

Analysis on Causes and Random Parameters of Truck Accident Severity in Mountainous Expressway
原文传递
导出
摘要 为明晰山区高速公路货车交通事故严重程度的影响因素,并量化交通事故数据异质性对货车交通事故严重程度的差异性,在分析四川省某山区高速公路2013—2018年693起货车交通事故时空特性的基础上,以交通事故严重程度为因变量,从人、车、路、环境、事故5个方面选择14个影响货车事故严重程度的潜在因素,采用随机参数有序Logit模型建立了货车事故严重程度异质性效应模型,并采用平均边际系数分析了显著变量对不同等级货车事故严重程度的影响方向及程度。结果表明:与传统的有序Logit模型相比,随机参数有序Logit模型的拟合优度和模型稳定性较高,且可以识别出货车事故严重程度影响因素的异质性影响效应,更适用于山区高速货车事故致因精细化建模;驾驶员违规操作、路面潮湿、不利天气、货车-大型货车、货车-中小型货车、平曲路段、夜间无照明、冬季、追尾、翻车10个变量显著影响山区高速公路货车事故严重程度。其中,驾驶员违规操作、货车-大型货车事故、追尾碰撞3个变量为随机参数,其估计系数分别服从(0.479,1.344),(1.042,0.545),(-0.489,1.098)正态分布,对货车事故严重程度具有显著的异质性影响效应。研究结果可为制订山区高速公路货车安全管理条例提供一定的理论基础,减少货车事故率及其严重度。 In order to clarify the influencing factors of the severity of truck traffic accidents on expressways in mountainous areas,and quantify the difference of the severity of truck traffic accidents caused by the heterogeneity of traffic accident data,based on the analysis of the spatio-temporal traffic accidents of 693 trucks on a mountain expressway in Sichuan Province from 2013 to 2018,taking the severity of traffic accidents as the dependent variable,selecting 14 potential factors affecting the severity of truck accidents from aspects of human,vehicle,road,environment and accident,the heterogeneity effect model of truck accident severity is established by using random parameter ordered Logit model,and the influence direction and degree of significant variables on the severity of different graded truck accidents are analyzed by using average marginal coefficient.The result shows that(1)Compared with the traditional ordered Logit model,the random parameter ordered Logit model has a higher goodness-of-fit and model stability,and can identify the heterogeneous influence effect of the factors influencing the severity of truck accidents,which is more suitable for the fine modeling of the cause of accidents of the truck accidents on mountainous expressway.(2)Driver's illegal operation,wet road,adverse weather,truck and large truck,truck and small/medium truck,flat and curved road section,no lighting at night,winter,rear-end collision,and overturning significantly affect the severity of truck accidents on expressway.Among them,driver's illegal operation,large truck accident and rear-end collision are random parameters,and their estimated coefficients obey the normal distribution of(0.479,1.344),(1.042,0.545),and(-0.489,1.098)respectively,which has significant heterogeneity effect on the severity of truck accidents.The result could provide a certain theoretical for the formulation of regulations on truck safety management on mountain expressway,and reduce the truck accident rate and its severity.
作者 李硕 马玉坤 韩晖 廖文洲 LI Shuo;MA Yu-kun;HAN Hui;LIAO Wen-zhou(Zhongjiaojian Jijiao Expressway Investment Development Co.,Ltd.,Shijiazhuang Hebei 050000,China;ROIH Traffic Safety Co.,Ltd.,Beijing 100088)
出处 《公路交通科技》 CAS CSCD 北大核心 2023年第4期228-236,共9页 Journal of Highway and Transportation Research and Development
基金 河北省交通运输厅科技项目(TH1202016)。
关键词 交通工程 货车事故严重程度 随机参数有序Logit模型 山区公路 异质性 traffic engineering severity of truck accident random parameter ordered Logit model mountainous highway heterogeneity
  • 相关文献

参考文献5

二级参考文献44

共引文献51

同被引文献14

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部