Traffic Conflict Analysis at Freeway Interchange Merging Areas: A Joint Model Based on Tobit and Survival Analysis Model

Freeway interchange merging areas suffer from high crash risk due to the frequent lane change behaviors, which bring disturbance to transverse and longitudinal traffic flow. This paper proposes a joint model based on the tobit and survival analysis model to explore the relationship between vehicles’ merging behavior and their traffic conflicts, and to investigate the influencing factors. The trajectory data from Maqun interchange, Nanjing, is used for analysis. The tobit method is applied to model the merging traffic conflict rate, and the survival analysis method is used to model the merging duration of vehicles. Traditional traffic conflict counting models are developed for comparison. The results show that the joint model outperforms other models and there exists coupling relationship between the merging duration and the conflict rate. The speed of the following vehicle in the target lane and the type of lead vehicle in the target lane have significant effects.