Free-floating bike sharing systems are an emerging new generation of bike rentals, that eliminates the needfor specific stations and allows to leave a bicycle (almost) everywhere in the network. Although free-floating bikes allow much greater spontaneity and flexibility for the user, they need additional operationalchallenges especially in facing the bike relocation process. Then, we suggest a methodology able togenerate spatio-temporal clusters of the usage patterns of the available bikes in every zone of the city,forecast the bicycles use trend (by means of Non-linear Autoregressive Neural Networks) for each cluster,and consequently enhance and simplify the relocation process in the network.
Spatio-temporal Clustering and Forecasting method for Free-Floating Bike Sharing Systems / Caggiani, Leonardo; Ottomanelli, Michele; Camporeale, Rosalia; Binetti, Mario (ADVANCES IN INTELLIGENT SYSTEMS AND COMPUTING). - In: Advances in Systems Science : Proceedings of the International Conference on Systems Science 2016 (ICSS 2016) / [a cura di] J. Świątek, J. M. Tomczak. - STAMPA. - Cham : Springer, 2017. - ISBN 978-3-319-48943-8. - pp. 244-254 [10.1007/978-3-319-48944-5_23]
Spatio-temporal Clustering and Forecasting method for Free-Floating Bike Sharing Systems
Leonardo CaggianiMembro del Collaboration Group
;Michele OttomanelliMembro del Collaboration Group
;Rosalia CamporealeMembro del Collaboration Group
;Mario BinettiMembro del Collaboration Group
2017-01-01
Abstract
Free-floating bike sharing systems are an emerging new generation of bike rentals, that eliminates the needfor specific stations and allows to leave a bicycle (almost) everywhere in the network. Although free-floating bikes allow much greater spontaneity and flexibility for the user, they need additional operationalchallenges especially in facing the bike relocation process. Then, we suggest a methodology able togenerate spatio-temporal clusters of the usage patterns of the available bikes in every zone of the city,forecast the bicycles use trend (by means of Non-linear Autoregressive Neural Networks) for each cluster,and consequently enhance and simplify the relocation process in the network.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
