基于MNL模型的车车碰撞事故严重程度影响因素辨识方法

A Method for identifying Influencing factors of Severity of Vehicle Crash Based on MNL model

为了研究两车之间事故严重程度的影响因素及其机理,探索在不同严重程度的交通事故中影响因素存在的差异性,采用多项logistic回归(MNL)模型对影响因素进行了分析。基于道路交通事故数据的特征形成了涵盖人员因素、车辆因素、时间因素、环境因素的影响因素集合。构建了适用于分析车车碰撞事故的模型,分析了事故严重程度的各类影响因素,并对不同严重程度的事故进行分类讨论,分析导致其存在差异性的原因。结果表明,在降低车辆间事故的安全风险时,需要重点考虑过失方车辆类型、过失方年龄、碰撞类型、受害方车辆类型、城市分区5项因素。对仅财产损失事故和有受伤但无死亡事故,需关注不同的影响因素组合存在的差异性,其中对仅财产损失事故,正面碰撞、受害方车辆类型为小客车、主城区、近郊区是可能的显著影响因素;对有受伤但无死亡事故,过失方年龄为19~45岁、侧面碰撞是可能的显著影响因素。持续提高对驾驶员的教育警示,使用新技术监督驾驶员的行车状态,提升车辆防御侧面碰撞的能力是可能降低车车碰撞事故严重程度的手段,从而提升城市道路交通的安全水平。

In order to study the influencing factors and mechanism of the severity of accidents between 2 vehicles,and explore the differences of influencing factors in traffic accidents with different severities,the influencing factors are analyzed by using multinomial logistic regression(MNL)model.A set of influencing factors including human factor,vehicle factor,time factor and environmental factor is formed based on the characteristics of road traffic accident data.A model suitable for analyzing vehicle collision accidents is constructed.Various influencing factors of accident severity are analyzed,the accidents with different severities are classified and discussed,and the reasons for the differences are analyzed.The result shows that(1)For reducing the risk of a vehicle-to-vehicle accidents,it is necessary to focus on 5 factors,namely,faulty party’s vehicle type,age of faulty party,collision type,victim’s vehicle type,and urban district.(2)For property damage accidents only and accidents with injuries but no fatalities,it is necessary to pay attention to the differences in different combinations of influencing factors.Among them,for property damage accidents only,frontal collision,victim’s vehicle type being small passenger car,main urban district,and suburban area are possible significant factors.For accidents with injuries but no fatalities,the age of faulty party is 19 to 45 years old,and side collisions are possible significant factors.(3)Continuously improving the education and warnings for drivers,using new technologies to monitor the driving status of drivers,and improving the ability of vehicles to defend against side collisions are possible means to reduce the severity of vehicle-to-vehicle collision accidents and improve the safety level of urban road traffic.