北京建筑大学学报2025，Vol.41Issue(2)：62-71,10.DOI:10.19740/j.2096-9872.2025.02.08
基于随机参数Logit模型的夜间行人-机动车事故严重程度致因分析
Analysis of the Causes of Severity in Nighttime Pedestrian-Motor Vehicle Accidents Based on Random Parameter Logit Model
摘要
Abstract
The frequency of nighttime traffic accidents is high and easily leads to serious consequences,especially posing a threat to the safety of pedestrians as a vulnerable group in traffic.Therefore,it is necessary to conduct in-depth research on the factors influencing nighttime pedestrian-vehicle accidents.Considering the heterogeneity of accident data,a random parameter Logit model is used to establish nighttime pedestrian-vehicle accident data from 2017 to 2022.The severity of the accidents is divided into three categories:no/possible injury,minor injury(non-disabling injury)and severe injury(death/disabling injury).A total of 30 influencing factors are selected as independent variables from pedestrian characteristics,driver characteristics,collision characteristics,road characteristics,time,and environmental characteristics,aiming to explore the factors influencing the severity of nighttime pedestrian-vehicle accidents and their heterogeneity.The results indicate that driver alcohol consumption,speed limit 32～40 km/h,presence of street lighting and driver injury are four random parameter variables displaying unobserved heterogeneity in influencing the severity of nighttime pedestrian-vehicle accidents.Variables such as pedestrian injury,traveling direction(straight),urban development level and road grade(local road)are significantly associated with the occurrence of minor and severe injuries in accidents.关键词
交通安全/夜间事故/事故严重程度分析/随机参数Logit模型/异质性Key words
traffic safety/nighttime crash/crash injury severity analysis/random parameters Logit model/heterogeneity分类
土木建筑引用本文复制引用
唐玉洁,焦朋朋,王健宇,陈世杰..基于随机参数Logit模型的夜间行人-机动车事故严重程度致因分析[J].北京建筑大学学报,2025,41(2):62-71,10.基金项目
北京市社会科学基金项目(21GLA010) （21GLA010）
北京建筑大学研究生创新项目(PG2024054). （PG2024054）
