ABSTRACT

Circular intersections have been used in transportation systems since the 1900s. Three types of circular intersections have been used in the United States: traffic circles, rotaries and roundabouts. While the use of traffic circles and rotaries in recent decades was found to have resulted in high crash rates, safety issues have been mitigated for roundabouts through the use of improved geometric designs.

Nevertheless, all three types of circular intersections face capacity problems during periods of high traffic volume, resulting in long queues and delays. Signal metering was introduced to reduce long queues and delays on the dominant approaches to circular intersections by stopping the flow of traffic from other approaches. This methodology was found to ease congestion for circular intersections with historically high traffic volumes. However, most signal metering at those intersections employ fixed signal timing, in which the metering rate is not responsive to changes in traffic condition. This study investigates the performance of an adaptive metering system for circular intersections. The system was implemented on a real traffic circle having high and unbalanced volumes. The model was calibrated, and a case study was simulated for peak-hour traffic conditions. Using the PTV VISSIM application programming interface, the algorithm was tested and the performance of the system was compared to the current intersection operation. The results showed that adaptive metering can significantly reduce delays and queues at a traffic circle. This preliminary study can be a useful reference for the development of priority- controlled circular intersections.