This paper outlines the assumptions made in the gap acceptance techniques for estimating capacity and delay in the naasra design guide for roundabouts. This was based on the best information available at the time of writing, but now research is suggesting modifications to the guide. The research is based on the observed behaviour at a number of roundabouts in the eastern states of australia. Possible alternative approaches for the estimation of capacity and delay are given in this paper. The application of the work to other unsignalised intersections is briefly discussed, as is the design of minis roundabouts. The major conclusions in this paper are that bunching in the major or circulation lane has a significant effect on capacity; that driver behaviour differs in different lanes on the same approach; that the follow on time for the dominant entry lane is related to the size of the roundabout; that drivers in other subdominant streams have a longer follow on time and that the critical gap is proportional to the follow on time. (Author/publisher) For the covering record of the symposium, see IRRD No 808826.
Abstract