Learning traffic flows: Graph Neural Networks for Metamodelling Traffic Assignment

TL;DR

This paper introduces a graph neural network-based metamodel for the Traffic Assignment Problem, significantly reducing computational costs and enabling real-time large-scale transportation scenario analysis.

Contribution

It presents a novel message-passing neural network approach that mimics traditional traffic simulation algorithms for more accurate equilibrium flow prediction.

Findings

Model outperforms conventional deep learning methods.

Demonstrates robustness on out-of-distribution data.

Reduces computational time for large-scale networks.

Abstract

The Traffic Assignment Problem is a fundamental, yet computationally expensive, task in transportation modeling, especially for large-scale networks. Traditional methods require iterative simulations to reach equilibrium, making real-time or large-scale scenario analysis challenging. In this paper, we propose a learning-based approach using Message-Passing Neural Networks as a metamodel to approximate the equilibrium flow of the Stochastic User Equilibrium assignment. Our model is designed to mimic the algorithmic structure used in conventional traffic simulators allowing it to better capture the underlying process rather than just the data. We benchmark it against other conventional deep learning techniques and evaluate the model's robustness by testing its ability to predict traffic flows on input data outside the domain on which it was trained. This approach offers a promising solution for accelerating out-of-distribution scenario assessments, reducing computational costs in large-scale transportation planning, and enabling real-time decision-making.