Learning heuristics for transit network design and improvement with deep reinforcement learning
Document Type
Journal Article
Publication Date
2025
Subject Area
place - north america, place - urban, planning - network design, planning - route design
Keywords
Public transit, urban planning, neural networks, reinforcement learning, optimisation
Abstract
Planning a network of public transit routes is a challenging optimisation problem. Metaheuristic algorithms search through the space of possible transit networks by applying heuristics that randomly alter routes in a network. Existing algorithms almost exclusively use heuristics that modify the network in purely random ways. In this work, we explore whether we can obtain better transit networks using more intelligent heuristics, that modify networks according to a learned preference function instead of at random. We use reinforcement learning to train graph neural nets to act as heuristics. These neural heuristics yield improved results on benchmark synthetic cities with 70 nodes or more, and achieve new state-of-the-art results on the challenging Mumford benchmark. They also improve upon a simulation of the real transit network in the city of Laval, Canada, achieving cost savings of up to 19% over the city's existing transit network.
Rights
Permission to publish the abstract has been given by Taylor&Francis, copyright remains with them.
Recommended Citation
Holliday, A., El-Geneidy, A., & Dudek, G. (2025). Learning heuristics for transit network design and improvement with deep reinforcement learning. Transportmetrica B: Transport Dynamics, 13(1), 2561863.
