A Stochastic Model for the Bike-sharing
Conférence : Communications avec actes dans un congrès international
Bike Sharing Systems (BSS) offer a sustainable and
flexible solution to urban mobility, but their rapid growth as a
viable and popular transportation alternative has exposed major
challenges. Due to asymmetric user flows throughout the day
they suffer from chronic imbalances in bike distribution, badly
impacting both the system reliability and user satisfaction. In
this paper, we address this issue with an optimized repositioning
strategy that lowers environmental impact while improving
operational efficiency and user satisfaction. After a data analysis
phase, we present in detail a stochastic bike-sharing repositioning
model that explicitly incorporates uncertainty on bike-demand
prediction at each station, thereby improving the robustness of
decision-making and the overall reliability of the system. To find
the global optimum we then decompose the stochastic model
into two manageable sub-problems and iteratively solve them
to optimality with exact methods. We finally validates both our
mathematical model and our algorithm by solving a synthetic
instance, showing the potential of our method.