Towards a Sustainable City Policy for Managing Shared Dockless E-Scooters

Based on the anonymized dataset of one million shared scooter rides collected by the Tel Aviv municipality, we investigate the behavior of the shared e-scooter users in Tel Aviv and the resulting dynamics of the scooters' spatio-temporal patterns. The users' choice of the shared e-scooters as a transportation mode follows fast and frugal heuristics: the user who activated the shared scooter's application continues with the ride when the closest available scooter is close enough, at up to 50-100 meters from her. Riders strongly prefer bike paths for riding and essentially deviate from the shortest path between the origin and destination and choose the routes with the higher percentage of bike paths. The longer the route, the higher the fraction of the bike paths in the rider's route and the difference between the shortest and chosen paths. We also demonstrate that the operators fail to match the supply to the demand and the scooters are oversupplied in the central part of the city, where the demand is high, and undersupplied in the rest of the city, where the demand is essentially lower. Based on the analysis we propose a new policy of spatial adjustment between the demand and supply.