Bikeability assessment using machine learning and user survey

Read full text Summarise with VIRAT
Authors: Tristan Lemoalle, Bahman Madadi, Angelo Furno
Cited by: 0
References: 22

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Urban areas face multiple challenges stemming from population growth and the limited capacity of transportation systems. Sustainable transport modes, such as cycling, are crucial for addressing congestion, pollution, and accessibility. Bike infrastructure improves overall accessibility but accessibility also depends on actual and perceived safety and comfort. To measure these nuanced aspects, a broader indicator is needed. Bikeability aims to bridge this gap. It can be approached in two different but complementary ways. Either quantitatively through objective measurements or simulations (e.g., type of infrastructure, slope, speed and traffic) or qualitatively. This paper proposes a hybrid, reproducible framework that combines quantitative and qualitative aspects. Using five indicators based on open data (type of infrastructure, slope, speed of motorized traffic, number of lanes and environment), a user survey and machine learning, we propose a bikeability assessment model. Results from our decision tree model show that physical separation is the primary factor defining bikeability, splitting the remainder of the model into a safety-focused branch and a comfort-focused branch. Cyclists prioritize aesthetics and comfort only after basic infrastructure separation guarantees safety. Validated through a case study of Lyon, France, the results show a high correlation between top-rated segments and the most frequented biking spots in the city. This framework provides urban planners with a scalable, low-cost diagnostic tool to prioritize infrastructure investments without relying on expensive sensing campaigns.

Fields of Study & Subjects

Urban Transport and Accessibility Traffic and Road Safety Transportation Planning and Optimization

Publication details

Journal / source
HAL
Publication date
16 March 2026
Language
EN
Document type
Preprint