基于随机参数有序Logit的山区高速公路追尾事故严重程度分析

Using Random Logistic Ordered Logit to Analyze Severity of Rear-End Collision Accidents on Mountain Highways

【目的】基于2012年到2017年汉宁高速河池至棋盘关段的1039条追尾事故数据,对山区高速公路追尾事故的主要影响因素进行分析。【方法】根据事故发生的时间,按白天和夜晚对数据进行分类,分别建立随机参数Logit模型。通过计算各显著变量的边际效用值,来定量分析各影响因素对追尾事故的影响程度。【结果】试验结果表明:白天和夜晚的事故严重程度、影响因素及影响程度均存在明显差异。【结论】性别、道路纵坡因素仅在夜晚模型中显著,节假日因素仅在白天模型中显著。部分变量虽在夜晚和白天模型中都显著,但边际效应却明显不同。

[Purposes]Based on the data of 1039 rear-end accidents from Hechi to Qipanguan section of Hanning Expressway from 2012 to 2017,the main influencing factors of rear-end accidents on mountainous expressways were analyzed.[Methods]According to the time of the accident,the data were classified by day and night,and the random parameter Logit model was established respectively.By calculating the marginal utility value of each significant variable,the influence degree of each influencing factor on the rear-end accident was quantitatively analyzed.[Findings]The results showed that there were significant differences in the severity,influencing factors and influence degree of accidents between daytime and nighttime.[Conclusions]Gender and road longitudinal slope factors are only significant in the night model,and holiday factors are only significant in the daytime model.Although some variables are significant in the daytime and nighttime models,the marginal effects are significantly different.