Detailed User Demand Modeling

Background: Modeling user demand in Bike Sharing Systems (BSSs) using net demand (returns minus withdrawals) is common for operational planning tasks like bike relocation.

Question / Future Work: Developing more detailed representations of user demand, such as modeling the demand for a trip from a specific bike station to another (Origin-Destination pair), or capturing the actual origin and destination of a user’s trip, could lead to a more accurate estimation of the link between request forecasting and decision-making for bike sharing systems.

Why It Matters: OD-level forecasting provides richer spatial information than single-station net demand, which could lead to significantly more optimized and localized relocation strategies.

Evidence: More detailed representations, such as, demand for a trip from a specific bike station to another, or from the actual origin to destination of the user, would allow for a more accurate estimation of the relationship between request forecasting and decision-making for bike sharing systems.