中国安全科学学报 ›› 2019, Vol. 29 ›› Issue (4): 31-36.doi: 10.16265/j.cnki.issn1003-3033.2019.04.006

• 安全系统学 • 上一篇    下一篇

山区二级公路交通事故致因的时间演化机制

戢晓峰1 教授, 李德林1, 杨文臣2   

  1. 1 昆明理工大学 交通工程学院,云南 昆明 650504;
    2 云南省交通规划设计研究院 陆地交通气象灾害防治技术国家工程实验室,云南 昆明 650031
  • 收稿日期:2019-01-24 修回日期:2019-03-19 出版日期:2019-04-28 发布日期:2020-11-04
  • 作者简介:戢晓峰 (1982—),男,湖北随州人,博士,教授,博士生导师,主要从事交通运输安全、交通行为与交通网络等方面的研究。E-mail:yiluxinshi@sina.com。
  • 基金资助:
    国家重点研发计划项目(2017YFC0803906);陆地交通气象灾害防治技术国家工程实验室开放研究基金资助(NEL-2019-05)。

Temporal evolution analysis on causes of traffic accidents occurring on secondary highways in mountain areas

JI Xiaofeng1, LI Delin1, YANG Wenchen2   

  1. 1 Faculty of Traffic Engineering, Kunming University of Science and Technology, Kunming Yunnan 650504, China;
    2 National Engineering Laboratory for Surface Transportation Weather Impacts Prevention, Yunnan Institute of Transportation Planning and Design, Kunming Yunnan 650200, China
  • Received:2019-01-24 Revised:2019-03-19 Online:2019-04-28 Published:2020-11-04

摘要: 为获取二级公路事故致因的时间演化机制,根据云南省元(谋)双(柏)二级公路2012—2017年事故数据,以交通事故严重程度为因变量,以时间、交通运行环境及交通事故参与者等为自变量,基于熵权-逼近理想解排序(TOPSIS)法量化交通事故严重程度;采用相关性分析法分析自变量与因变量间的相关程度;依托Logistic模型建立各年交通事故严重程度分析模型,并从拟合优度检验和预测准度检验等2方面检验模型的信度与效度。结果表明:事故发生天气、时间段、事故形态、是否涉及无防护车辆以及是否涉及大型车辆等与事故严重程度相关性较高,且形态为碰撞行人、涉及无防护车辆或大型车辆的事故严重程度较高。

关键词: 交通安全, 二级公路, 交通事故严重程度, Logistic模型, 熵权-逼近理想解排序(TOPSIS)法

Abstract: In order to study the temporal evolution of traffic accidents causes, the entropy-TOPSIS is adopted to quantify the severity of traffic accidents based on the accident data of Yuan-Shuang secondary highway from 2012 to 2017 and using the severity of accidents as dependent variable and the time, traffic environment and participants of accidents as independent variables. Then the correlation analysis was made to study the degree of correlation between dependent variables and independent variables. Finally, a severity analysis model of traffic accidents on each year was established based on Logistic model and its reliability and validity were verified by taking goodness of fit test and prediction accuracy test. The results show that the severity of accidents is highly correlated with factors like weather, time period, accident type, involvement of unprotected vehicles or not and involvement of oversize vehicles or not, and the accident would be severer when an unprotected or oversize vehicle hits pedestrians.

Key words: traffic safety, secondary road, traffic accident severity, Logistic model, entropy-technique for order preference by similarity to ideal solution (TOPSIS)

中图分类号: