Space-Time Prediction of Bike Share Demand

As bike sharing became more and more popular in Philadelphia, people tend to use Indego bikes quite often, which has created new challenges for the company to distribute and organize the bikes. Occasionally, people may see no bikes at bike stations during peak hours, and there can also be excess bikes at stations that people barely use. Therefore, “re-balancing”, getting bikes to stations that are anticipated to have demand but lack bikes, is important for city management.

​

In this project, I used predictive modeling to solve this problem. I will predict the demand for the next two weeks in Philadelphia by using the time-space predictive model. Users can see when demand for bikes might drive stations to run out of bikes, and then move excess bikes from elsewhere. As a result, program managers for a bike-share system could reasonably anticipate demand and allocate bikes ahead of time in a efficient way.

​

Please click the button below to see the final deliverable.