Long-Term Predictions of Bike-Sharing Stations? Bikes Availability

Bike-sharing systems are present in many cities as a valid alternative to fuel-based public transports since they are eco-friendly, prevent traffic congestions, reduce the probability of social contacts. On the other hand, bike-sharing present some problems such as the irregular distribution of bikes on the stations/racks/areas (still very used for e-bikes) and for the final users the difficulty of knowing in advance their status with a certain degree of confidence, whether there will be available bikes at a specific bike-station at a certain time of the day, or a free slot for leaving the rented bike. Therefore, providing predictions can be useful for improving the quality of e-bike services. This paper presents a technique to predict the number of available bikes and free bike slots in bike-sharing stations (the best solution for e-bikes). To this end, a set of features and predictive models have been developed and compared to identify the best prediction model for long-term predictions (24 hours in advance). The solution and its validation have been performed by using data collected in bike stations in the cities of Siena and Pisa, in the context of Sii-Mobility National Research Project on Mobility and Transport and Snap4City Smart City IoT infrastructure. The Random Forest (RF) and Gradient Boosting Machine (GBM) offer a robust approach for the implementation of reliable and fast predictions of available bikes in terms of flexibility and robustness to critical cases, producing longterms predictions in critical conditions (i.e., when there are only few remaining available bikes on the