Public transport grid analysis using multi-scale backbones

Details

Supervisors

Delvenne, Jean-Charles

Degree label

Master [120] : ingénieur civil en science des données, à finalité spécialisée

Abstract

By modeling public transportation systems and experimenting with a few scenarios coupled with graph theory methods, this study aims to discuss their approaches while also examining infrastructural, operational guidelines, and real implementation challenges. The realm of graph theory faces no limitations, as nodes and edges can be freely created. However, a transportation grid must respect urban spaces and topological constraints such as leveling off and lakes, for example. Still, analytical tools can help us identify operational opportunities, map expansion, and optimize efforts to bridge delivery and demand. In this study, we will present different rankings to highlight the relevance discrepancy between infrastructure, service demand, and potential stress (according to toy metrics). We will also discuss other possible metrics to help identify service quality gaps between areas, such as a mobility index based on the number of stations, vehicle speed, and average speed. In other words, it sheds light on the fact that some neighborhoods are better served by the transport system, raising important moral questions about profitability and social responsibility through a common service. Additionally, we extrapolate on how machine learning can be used to bring efficiency without affecting the equality of our common right to come and go.