Bike-Sharing Docking Stations Identification Using Clustering Methods in Lisbon City

Abstract

Urban clean mobility has enormous impacts on environmental, economic and social levels, promoting important eco-friendly means of sustainable transportation. Soft mobility (specially bike-sharing services) plays a crucial role in these initiatives since it provides an alternative for hydrocarbon fuel vehicles inside the cities. However, choosing the best location to install soft mobility docks can be a difficult task since many variables should be considered (e.g. proximity to bike paths, points of interest, transportation access hubs, schools, etc.). On the other hand, mobile data from personal cellphones can provide critical information regarding demographic rate, traffic trajectories, origin/destination points, etc., which can aid in the installation of soft mobility platforms. This paper presents a decision support system to study both existent and new bike-sharing docking stations, using mobile data and clustering techniques for three Lisbon council parishes: Beato, Marvila and Parque das Nações.