Deep Learning for Short-Term Prediction of Available Bikes on Bike-Sharing Stations

Bike-sharing is adopted as a valid alternative to traditional public transports since they are eco-friendly, prevent traffic congestions, reduce the probability of social contacts which happens on most of the public means. However, some problems may occur such as the irregular distribution of bikes on the stations/racks/areas, and the difficulty of knowing in advance the rack 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. Thus, providing predictions can be useful to improve the quality of service, especially in those cases in which the bike racks are used for e-bikes, which need to be recharged. This paper compares the state-of-the-art techniques to predict the number of available bikes and free bike-slots in bike-sharing stations (i.e., bike racks). To this end, a set of features and predictive models were compared to identify the best models and predictors for short-term predictions of 15, 30, 45, and 60 minutes. The study demonstrated that deep learning and in particular Bidirectional Long Short-Term Memory networks (Bi-LSTM) offer a robust approach for the implementation of reliable and fast predictions of available bikes, even with a limited amount of historical data. The paper also reported an analysis of feature relevance based on SHAP that demonstrated the validity of the model for different behaviours of the clusters. The solution and its validation were derived by using data collected in bike-stations in the cities of Siena and Pisa (Italy), in the context of Sii-Mobility National Research Project on Mobility and Transport and Snap4City Smart City IoT infrastructure.