Abstract: In order to solve the problem that the accuracy of current traffic conflict data collection method is not high and the outcome severity of traffic conflict is not considered, the typical highway workzone (subgrade stage) was selected as the research object, high-precision regional vehicle operation videos were captured by high-altitude unmanned aerial vehicle, and continuous vehicle microdata were obtained through video and conflict recognition program. Based on the theory and process of vehicle collision, the outcome severity of traffic conflict model was derived and the index of traffic conflict severity rates was used to assess traffic safety. The results show that traffic volume and large truck rate have a significant impact on the traffic conflict severity, that when the traffic volume is less than 700 pcu/h, the conflict severity is low, that when the traffic volume exceeds 900 pcu/h, the conflict severity increases rapidly, that the higher the large truck rate is, the higher the conflict severity will be, and that the traffic safety in highway workzone can be improved by diverting traffic volume and large trucks through traffic organizations.

Key words: traffic conflict, highway workzone, outcome severity of traffic conflict, vehicle collision theory, unmanned aerial vehicle