EVALUATING THE PERFORMANCE OF A ROUNDABOUT

The current NAASRA method of evaluating the performance of roundabouts is based on studies which provided only a general indication of their performance. These studies did not provide an indication of the effect of the entry geometry. This report outlines an extensive field study of driver behavior at roundabouts and quantifies this behavior to enable design equations to be developed. It is recommended that a roundabout be analyzed as a number of connected T-interssections, that the entering drivers give way to all circulating drivers, that the drivers in different entry lanes behave differently, and that different entry lanes have a different capacity. The entering drivers' behavior was quantified in terms of the conventional follow-on times and critical acceptance gap terms. The entry lane with the greatest flow was termed the 'dominant stream' and its follow-on time was found to be a function of the entry geometry and the circulating flow. The follow-on times for other entry lanes were a function of the dominant stream follow-on time and entry lane flows. The critical acceptance gap was a function of the follow-on times and the entry geometry. The level of platooning in the major stream greatly affects the performance of a roundabout. A technique for accounting for this platooning is also given in this report.