Traffic conflicts in the lane-switching sections at highway reconstruction zones.

Introduction: There are designated sections for lane-shifting in several highway reconstruction and expansion zones. Similar to the bottleneck sections of highways, these sections are characterized by poor pavement surface conditions, disorderly traffic flow, and high safety risk. This study examined the continuous track data of 1,297 vehicles collected using an area tracking radar.

Method: The data from the lane shifting sections were analyzed in contrast with the regular section data. Further, the single-vehicle attributes, traffic flow factors, and the respective road characteristics in the lane-shifting sections were also taken into account. In addition, the Bayesian network model was established to analyze the uncertain interaction between the various other influencing factors. The K-Fold cross validation method was used to evaluate the model.

Results: The results showed that the model has a high reliability. The analysis of the model revealed that the significant influencing factors in decreasing order of their influence on the traffic conflict are: the curve radius, cumulative turning angle per unit length, standard deviation of the single-vehicle speed, vehicle type, average speed, and the standard deviation of the traffic flow speed. The probability of traffic conflicts is estimated to be 44.05% when large vehicles pass through the lane- shifting section while it is 30.85% for small vehicles. The probabilities of traffic conflict are 19.95%, 34.88%, and 54.79% when the turning angles per unit length are 0.20 °/m, 0.37 °/m, and 0.63 °/m, respectively.

Practical Applications: The results support the view that the highway authorities help reduce traffic risks on lane change sections by diverting large vehicles, implementing speed limits on road sections, and increasing the turning angle per unit length of vehicles.