Predicting Citi Bike Demand Evolution Using Dynamic Graphs

TL;DR

This paper explores the use of graph neural networks to predict bike demand in NYC's Citi Bike system, aiming to improve capacity management by accurately forecasting station-specific demand patterns.

Contribution

It introduces a novel application of dynamic graph neural networks for demand prediction in bike sharing systems, addressing the challenge of variable demand.

Findings

Graph neural network model effectively predicts demand patterns.

Improved capacity management potential for bike sharing systems.

Demonstrates the applicability of dynamic graphs in transportation demand forecasting.

Abstract

Bike sharing systems often suffer from poor capacity management as a result of variable demand. These bike sharing systems would benefit from models to predict demand in order to moderate the number of bikes stored at each station. In this paper, we attempt to apply a graph neural network model to predict bike demand in the New York City, Citi Bike dataset.