BPCPPDYSOE21 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 3: Poster
SOE 3.2: Poster
Montag, 22. März 2021, 17:30–19:30, SOEp
Persistence length of ride-sharing bus trajectories — •Steffen Mühle and Helge Heuer — Max Planck Institute for Dynamics and Self-Organization, Göttingen, Germany
On-demand ride-sharing services have the potential to drastically decrease urban traffic, mobility costs, carbon emissions and the need for owning a private car. While the benefits of a well-coordinated bus fleet capable of serving live incoming transport requests are compelling, predicting the spatio-temporal dynamics even of single buses is far from trivial. Typically, a bus’ trajectory does not originate in isolation but emerges from its interplay with incoming requests, the street network, other buses and fleet-wide policies.
Given the latter, namely the maximally allowed detour an accepted request may entail, δmax, we treat bus trajectories as random walks and inspect them from the perspective of polymer theory. To this end, we generate random walks purely geometrically, and also run full-scale ride-sharing simulations using MatSim. In both cases, we observe that for long times a bus’ trajectory becomes diffusive, which allows us to assign a persistence length to them.
This creates a quantitative link between the (tunable) parameter δmax and the (observed) typical length scale on which a bus changes its direction, enabling us to predict e.g. how much time a bus spends in a certain district or how far it travels over the course of one day.