Abstract: We construct a model of traffic circle aiming at determining the traffic circle capacity and maximizing the capacity under the condition of safe traffic circulation. The capacity of the circle, with the threshold as its critical point, determines the selection of the control method in terms of whether to position traffic lights: if traffic lights are not needed, the stop and yield signs are designed to regulate the traffic of the circle; if traffic lights are needed, different measures must be adopted according to the period of time when the one-day traffic flow of the circle is bigger or smaller than its capacity (i.e. during the rush and non-rush hours). During the non-rush hours, only the gleaming of the yellow lights is necessary; while during the rush hours, time allocation is designed.

In the model, the motorcade analysis method is adopted to obtain the decisive factors determining the capacity of the circle (i.e. the capacity of the intersection), and further to get the capacity of the circle. When designing traffic lights, the traffic flows during different periods of time are considered to decide time allocation; when designing time allocation, the frequently used Webster Formula is adopted. In the present paper, the time allocation design is optimized by further considering the influence of the average delay time, average stops and capacity so that a more accurate time allocation is obtained, that is the optimized effective green-light time at one interval.

Based on the model and the controlling method adopted in the present paper, an empirical study of the circling intersection is carried out in Yiwu, and the traffic capacity of the circle obtained: , by which the traffic control method is determined to be traffic lights positioning. When designing time allocation, the Webster Formula is utilized to obtain an effective green-light time, and compare this time with the optimized one, our model proves to be effective.

 

Keywords: motorcade analysis method; traffic capacity; time allocation design; Webster formula