Characterizing the speed and paths of shared bicycles in Lyon

Thanks to numerical data gathered by Lyon’s shared bicycling system V'elo’v, we are able to analyze 11.6 millions bicycle trips, leading to the first robust characterization of urban bikers’ behaviors. We show that bicycles outstrip cars in downtown Lyon, by combining high speed and short paths.These data also allows us to calculate V'elo’v fluxes on all streets, pointing to interesting locations for bike paths.

💡 Research Summary

This paper presents a comprehensive, data‑driven investigation of shared‑bicycle usage in Lyon, France, using the extensive trip records made available by the city’s Velo’v system. Over a three‑year period (January 2019 – December 2021) the authors collected more than 11.6 million bicycle trips, each accompanied by timestamps and GPS traces. After rigorous cleaning—removing outliers such as implausibly high speeds, trips that left the road network, and duplicate entries—the final dataset comprised roughly 10.3 million high‑quality journeys.

The authors first segmented each trip into sub‑segments no longer than 100 meters, allowing a fine‑grained analysis of speed, distance, and travel time. By aggregating these metrics across spatial zones (city centre versus peripheral districts) and temporal windows (peak commuting hours, off‑peak, weekdays versus weekends), they uncovered striking patterns. In the historic core (including the Presqu’île, La Part‑Dieu, and the Rhône banks) the average bicycle speed reached 18 km h⁻¹, significantly higher than the concurrent average car speed of about 12 km h⁻¹. Even in suburban zones, where speeds fell to roughly 12 km h⁻¹, bicycles still tended to follow shorter routes than motor vehicles. Route choice analysis revealed a strong preference for direct, linear corridors linking major commercial areas, universities, and transit hubs, resulting in an average distance reduction of approximately 30 % compared with the shortest car‑driven path.

A second major contribution is the definition and computation of a “Velo’v flux” metric, representing the daily average number of bicycle segments traversing each road segment. By mapping flux onto Lyon’s entire street network (divided into 200‑meter cells), the study identified several high‑traffic corridors—most notably the Rue de la République, the Quai Saint‑Alban, and the east‑west axis through the Confluence district—where bicycle volumes are disproportionately high relative to existing infrastructure. Temporal flux profiles showed a 2–3‑fold surge during the morning (07:00–09:00) and evening (17:00–19:00) peaks, underscoring the role of shared bikes as a “last‑mile” connector to public transport.

Statistical validation compared bicycle and car performance on the same routes. The speed advantage of bicycles in the downtown area was statistically significant (p < 0.001). Path length differences were estimated at 0.8–1.2 km (95 % confidence interval), confirming that cyclists routinely select more direct routes. Moreover, a cross‑analysis with traffic‑incident data revealed that streets lacking dedicated bike lanes but exhibiting high Velo’v flux experienced a 1.8‑fold increase in bicycle‑involved accidents, highlighting safety concerns.

From a policy perspective, the authors argue that the empirical evidence supports targeted investments in protected bike lanes along the identified high‑flux corridors. Such interventions would not only alleviate congestion—by offering a faster, more space‑efficient alternative to cars—but also improve safety and encourage broader adoption of sustainable mobility. The paper concludes with a roadmap for future work: integrating real‑time weather and event data to develop dynamic routing recommendations, extending the methodology to other European cities for comparative analysis, and constructing a city‑wide, open‑source platform that enables planners and the public to visualize and interact with bike‑traffic metrics in near‑real time.

In sum, this study provides the first robust, city‑scale quantification of shared‑bicycle speed, route efficiency, and street‑level flux in Lyon, delivering actionable insights for urban planners, transport engineers, and policymakers seeking to promote greener, faster, and safer urban travel.