Moment-based availability prediction for bike-sharing systems

Abstract

We study the problem of predicting the future availability of bikes in a bike station through the moment analysis of a PCTMC model with time-dependent rates. Given a target station for prediction, the moments of the number of available bikes in the station at a future time can be derived by a set of moment equations with an initial setup given by the snapshot of the current state of all stations in the system. A directed contribution graph is constructed, and a contribution propagation method is proposed to prune the PCTMC so that it only contains stations which have significant contribution to the journey flows to the target station. Once the moments have been derived, the underlying probability distribution of the available number of bikes is reconstructed through the maximum entropy approach. We illustrate our approach on Santander Cycles, the bike-sharing system in London. The model is parameterised using historical data from Santander Cycles. Experimental results show that our model outperforms a time-inhomogeneous Markov queueing model with respect to several performance metrics for bike availability prediction.

A rule-based model for Seoul Bike sharing demand prediction using weather data

“…Recently, it is found that the rental bike sharing infrastructure, (for example, cycle paths and ways) affected the rental demand, additionally uncovering huge connections amongst temperature factor and land use and bike sharing action (El-Assi, Mahmoud, & Habib, 2017). Future accessibility of bikes in the stations was predicted by examining the moment of a continuous time Markov-chain population model with time-dependent rates (Feng, Hillston, & Reijsbergen, 2017). Investigation was carried out to study various climate conditions and temporal qualities, in stationlevel and framework level examination attributes.…”

Neural Networks with Improved Extreme Learning Machine for Demand Prediction of Bike-sharing

“…The performance is verified by comparing with other models. Feng et al [11] discussed the Markov chain population model to predict bike demands among different travel stations. Kim [12] studied the influence of weather conditions and time characteristics on demands of bike-sharing.…”