Research on influencing factors of pedestrian-vehicle conflict safety based on Ordered Probit model
To investigate and quantitatively analyze the factors influencing the severity of pedestrian-vehicle conflict,the video data of pedestrian-vehicle conflict at three signal intersections was collected.The pedestrian-vehicle trajectories were calibra-ted by using T-Analyst,then the conflict index(PET)was calculated,and the 85%bit cumulative frequency curve method was used to classify the pedestrian-vehicle conflict levels.Ten factors were selected as variables from pedestrain,vehicle,road,and environmental characteristics,then an Ordered Probit model was constructed to establish the significant influencing factors on the severity of pedestrian-vehicle conflict,and the degree of influence of different significant factors was quantified by marginal effect.The results of the study show that the degree of pedestrian-vehicle conflict is significantly affected by fac-tors such as pedestrian red-light running,age,pedestrian traffic volume,sidewalk occupancy,sidewalk starting and ending points,vehicle speed changes,and traffic flow,and compared to their respective reference quantities,the probability of serious conflict caused by pedestrian running red light,vehicle acceleration through the conflict point,pedestrian starting point,and elderly increases by 8.2%,5.9%,5.1%and 4.4%,respectively.Compared to low flow traffic flow,the high flow of vehicles and pedestrian traffic increases the probability of serious conflict by 3.7%and 2.5%,respectively,but the sidewalk occupan-cy make the probability of serious conflict decrease by 6.8%.The research results can provide a theoretical basis for the de-sign and implementation of pedestrian crossing traffic safety facilities at signal intersections.
signal intersectionpedestrian safetyseverity of pedestrian-vehicle conflictOrdered Probit modelmarginal effect
NETL
NSTL
万方数据
