Bike-sharing systems have been growing rapidly worldwide over the past decade. In times of high traffic, pollution concerns, and decrease of physical activity, bike-sharing perfectly meets the needs of the present. Crucial to the success of such systems is the proper location of stations to attract users and to have permanent availability of stations for renting and returning bicycles. Nonetheless, user demand for bicycles and bike parking racks is unbalanced during the day, which may induce overloading of some stations and depletion of others. This study investigated a bike-sharing programme in Belfast which regularly encounters the problem of overcrowding and empty stations. A rebalancing approach is introduced based on a tree based predictive model for bike and rack demand. The novelty of this methodology is its use of station location features, such as proximity to the parks or train stations for demand prediction. This study also proposes an approach to select new sites to expand the system. The approach is applied to the Belfast Bike system where two new locations are successfully proposed.
Optimizing the Belfast Bike Sharing Scheme
Lect. Notes in Networks, Syst.
Proceedings of SAI Intelligent Systems Conference ; 2021 ; Amsterdam September 02, 2021 - September 03, 2021
03.08.2021
14 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Springer Verlag | 2023
|Springer Verlag | 2020
|DataCite | 2024
|