基于Logistic回归的高速公路交通事故后果的影响因素分析

Analysis of Influencing Factors of Expressway Accident Consequences Based on Binomial Logistic Regression Analysis Model

文中基于某双向四车道高速公路连续9年的交通事故数据,将交通事故分为严重事故、一般事故2种类型,利用二项Logistic回归分析模型,选取了涉事车型、事发天气、事发时间、特殊路段类型、事故类型和事发月份共六类自变量进行分析,筛选其中的对高速公路的交通事故后果有明显影响的因素.研究结果表明:涉事车型、事发时间、特殊路段类型、事故类型和事发月份均对交通事故的严重程度有显著性影响,并从车型管制、加强路面巡逻、隧道及长陡坡路段采取主动安全措施、增强被动防护以提高路侧设施的保护能力、宣传教育等方面,提出了减缓交通事故严重程度的综合措施建议.

Based on the traffic accident data of a two-way four-lane expressway for nine consecutive years,the traffic accidents were divided into two types:serious accidents and general accidents.By using binomial logistic regression analysis model,six independent variables were selected for analysis,including vehicle type,weather,time,special road section type,accident type and incident month,and the factors that have obvious influence on the traffic accident consequences of the expressway were screened out.The results show that the vehicle type involved,the time of the accident,the type of special road section,the type of accident and the month of the accident all have significant influence on the severity of traffic accidents,but the weather factor has no obvious influence.From the aspects of vehicle type control,strengthening road patrol,taking active safety measures in tunnels and long steep slope sections,strengthening passive protection to improve the protection ability of roadside facilities,publicity and education,etc.,this paper puts forward comprehensive measures and suggestions to slow down the severity of traffic accidents.