Bicycle Improvement Design Problem using real bike trajectories

Read full text Summarise with VIRAT
Authors: Félix Repusseau, Tifenn Rault, Emmanuel Néron
Cited by: 0
References: 42

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

The development of low-carbon mobility such as cycling is a crucial issue today, especially in urban areas where the extensive use of cars causes difficulties such as pollution, traffic congestion, and overcrowded parking. Existing literature highlights that a crucial factor in the adoption of bicycle commuting is the availability and density of bike networks ( Schoner and Levinson, 2014 , Buehler and Dill, 2016 ). In this context, computational methods can serve as decision support systems to determine how bicycle network should be expanded. Recent years have seen two key technological developments that can be integrated into these systems: (i) collaborative open data platforms such as OpenStreetMap that offer accurate information about cycling networks, and (ii) dedicated navigation apps for cyclists, such as Geovelo [from La Compagnie des Mobilités (Geovelo, 2009)], that record and analyze cycling journeys. In this study, we propose using a set of GPS tracks of the routes taken by cyclists to recommend budget-constrained improvements to cycling networks. Our objective is to determine where bicycle facilities should be located to improve overall user satisfaction. To address this problem we introduce a mathematical model and heuristics. In our approach, we show that individual GPS trajectories can play a crucial role in effectively improving the cycling network. First, the use of origin–destination pairs corresponding to actual trajectories is useful for determining where the network needs to be improved; second, actual bicycle trajectories are used to determine each cyclist’s trade-offs between distance and danger. These individual trade-offs are considered in our optimization model and result evaluation, allowing us to account for the heterogeneity of cyclists. The use of a representative set of trajectories is therefore essential. Experiments using real datasets are presented to evaluate the quality of the proposed methods.

Fields of Study & Subjects

Urban Transport and Accessibility Transportation Planning and Optimization Vehicle Routing Optimization Methods

Publication details

Journal / source
Computers & Operations Research
Volume
192
Pages
107474–107474
Publication date
22 March 2026
Language
EN