Learning from user's behaviour of some well-known congested traffic networks

2025/08/20 by Isolda Cardoso, Cardoso, Isolda, Lucas David Venturato +3
Computer Science · Social Sciences · Physics and Astronomy · #Data Management and Algorithms #Human Mobility and Location-Based Analysis #Complex Network Analysis Techniques

paper · pdf · doi:10.48550/arxiv.2508.14804

Abstract

The traffic assignment problem (TAP) aims to predict how traffic flows distribute themselves across a road network, traditionally requiring computationally expensive iterative simulations to reach a user equilibrium (UE) where no driver can unilaterally reduce their travel time. Recent developments in machine learning (ML), particularly Graph Neural Networks (GNNs) and hybrid approaches, aim to solve this faster while maintaining accuracy

Citations

• WardropNet: Traffic Flow Predictions via Equilibrium-Augmented Learning
• End-to-End Heterogeneous Graph Neural Networks for Traffic Assignment
• End-to-end heterogeneous graph neural networks for traffic assignment
• Data-Driven Traffic Assignment: A Novel Approach for Learning Traffic Flow Patterns Using a Graph Convolutional Neural Network
• Machine-Learned Prediction Equilibrium for Dynamic Traffic Assignment

Related

• Learning traffic flows: Graph Neural Networks for Metamodelling Traffic Assignment
• Data-Driven Traffic Assignment: A Novel Approach for Learning Traffic Flow Patterns Using a Graph Convolutional Neural Network
• Multi-Class Traffic Assignment using Multi-View Heterogeneous Graph Attention Networks
• An Approach to Avoid the Unreal High Flows on Congested Links and Investigates the Evolution of Congestion over Network
• End-to-End Heterogeneous Graph Neural Networks for Traffic Assignment