Exploratory Simulation of Pedestrian Crossings at Roundabouts

There is presently much debate over the issue of pedestrian access to highway facilities, particularly for those who are blind or function with low vision. This issue is now the focus of current United States Access Board draft recommendations, which propose that signalization is necessary to ensure safe access for blind pedestrians at roundabouts. The objective of this research was to capture the gap acceptance behavior for both sighted and blind pedestrians near roundabouts (revealed from a parallel research effort) and integrate that behavior into a microscopic simulation model of pedestrian and vehicular traffic operations at a roundabout. This process enables the quantification of the mutual impact of pedestrian crossing behavior on vehicle operation, and of motorized traffic impact on pedestrian access (or delay). In addition, it allows the conduct of experiments targeted at evaluating various pedestrian treatments at or in the vicinity of the roundabout. The simulation results indicated that pedestrian delay increases in a nonlinear fashion as vehicle volume increased. In addition, while there was a small difference in sighted pedestrian delays at the entry and exit legs, that difference was more pronounced for blind pedestrians who experienced higher delays on the exit side. Finally, an analysis of pedestrian crossing treatments indicated that placing a pedestrian-actuated, signalized crossing upstream/ downstream of the roundabout results in delays to blind pedestrians that are comparable to those experienced by sighted pedestrians that cross at the unsignalized splitter island. The location of the crossing was assessed on the basis of queue spillback probabilities onto the roundabout exit leg.