AN EXAMINATION OF PEDESTRIAN CROSSWALK DESIGN AT ROUNDABOUTS

This study uses a rational approach to examine three elements of pedestrian design at roundabouts – crosswalk location, crosswalk angle, and exit design – and attempts to recommend how they could be implemented to achieve the best possible balance of pedestrian safety and convenience. This was done using the following objectives: • Maximize visibility (of approaching vehicles to pedestrians, and of pedestrians to drivers). • Minimize impacts of vehicle queues (on pedestrians and on other drivers). • Minimize walking distances. • Minimize pedestrian exposure in the crosswalk. • Minimize vehicle speeds through the crosswalk. • Maximize pedestrian refuge width. The evaluation shows that the only clear preference in terms of pedestrian design at a roundabout is that a crosswalk be located one passenger car length from the roundabout on the exit side as opposed to two. This was because it was shown to minimize walking distances, minimize vehicle speeds through the crosswalk, and maximize pedestrian refuge width in the splitter island. There was determined to be no overall preference between straight and angled crosswalks, and no overall difference between a curved and a more tangential exit design. The evaluation was undertaken for an example two-lane roundabout. It is possible that different results may be obtained with different geometric designs. This could include exits much more curved than for the example roundabout, as well as single-lane roundabouts, three-lane roundabouts, flared entries and tapered exits, and crosswalk locations of three car lengths from the roundabout. It is also possible that different results may be obtained with a different evaluation procedure that assigns weights to the study objectives based on their level of pedestrian safety or convenience. Regardless of the specifics of the evaluation results, it can be seen that the objectives associated with pedestrian crosswalks at roundabouts compete with each other. Weighting the objectives may be appropriate to account for site context. A more human factors approach may allow a better understanding of the driver and pedestrian tasks associated with using a roundabout, and therefore a better means of evaluating various pedestrian design elements.