Bikelution: Federated Gradient-Boosting for Scalable Shared Micro-Mobility Demand Forecasting

TL;DR

Bikelution introduces a federated learning approach using gradient-boosted trees for accurate, privacy-preserving bike demand forecasting in shared micro-mobility systems, outperforming existing methods.

Contribution

The paper presents a novel federated learning framework for demand forecasting that maintains privacy and achieves comparable accuracy to centralized models.

Findings

Bikelution performs similarly to centralized ML models in demand prediction.

It outperforms current state-of-the-art demand forecasting methods.

The approach demonstrates the feasibility of privacy-preserving micro-mobility analytics.

Abstract

The rapid growth of dockless bike-sharing systems has generated massive spatio-temporal datasets useful for fleet allocation, congestion reduction, and sustainable mobility. Bike demand, however, depends on several external factors, making traditional time-series models insufficient. Centralized Machine Learning (CML) yields high-accuracy forecasts but raises privacy and bandwidth issues when data are distributed across edge devices. To overcome these limitations, we propose Bikelution, an efficient Federated Learning (FL) solution based on gradient-boosted trees that preserves privacy while delivering accurate mid-term demand forecasts up to six hours ahead. Experiments on three real-world BSS datasets show that Bikelution is comparable to its CML-based variant and outperforms the current state-of-the-art. The results highlight the feasibility of privacy-aware demand forecasting and outline the trade-offs between FL and CML approaches.