A public road experiment at the '4-arm' junction A2/A28 (Wincheap roundabout, Canterbury) has been carried out to compare the capacities of two alternative offside priority roundabout designs. As expected, higher capacities were obtained with the design having wider entries combined with a reduced island size. Closing one arm of the roundabout had only a small (and not statistically significant) effect on the full capacity. Possible effects on the individual entry capacities of the number of entries queueing and of the flow leaving the roundabout by the previous exit were investigated; neither effect was found to be significant. Entry capacity relationships obtained from the data by linear regression were used to test a roundabout capacity balancing computer program by comparing the predicted operating capacities with those observed. The results showed close agreement. (Author/publisher)
Abstract