A Roundabout Capacity Survey Method Using Gap Acceptance Cycles and Vehicle Trajectory Data

This detailed research report describes a new gap acceptance capacity survey method using vehicle trajectory data and gap acceptance cycles based on a traffic signal analogy approach. Developed using data at a single-lane roundabout, the method also applies to multi-lane roundabouts, opposed turns at sign-controlled and signalised intersections, and various merge situations. By observing the departure of queued entry vehicles during saturated periods, the method defines reliable cycles and simultaneously measures follow-up headway, critical gap, and capacity cycle-by-cycle. Observed follow-up headway and critical gap values were used to assess estimates of average blocked time, unblocked time, gap acceptance cycle time and capacity using a number of arrival headway distribution models for the circulating stream. The results confirmed the validity of the survey method, supporting the novel signal analogy concept (used for developing SIDRA models) with observations of real-world data. Assessment of bunching in circulating flows showed that models estimating lower bunching performed better. The assessments of gap acceptance capacity models were not conclusive as they were limited by the small size of the dataset available from this survey. Future research is recommended by implementing the new survey method for assessing the models to estimate gap acceptance capacity and related parameters at different traffic facilities.