Fine-Grained Urban Traffic Forecasting on Metropolis-Scale Road Networks

2025-10-02

AI Summary

Existing traffic forecasting benchmarks suffer from missing road topology, sparse road attributes, and insufficient scale—particularly hindering fine-grained prediction on dense, structurally complex urban road networks. Method: We introduce a large-scale real-world urban road network dataset covering nearly 100,000 road segments, enriched with comprehensive road attributes and high-frequency traffic flow/speed measurements; and propose a lightweight graph neural network (GNN) architecture that explicitly encodes road network topology and multidimensional road features—eliminating the need for dedicated temporal modules. Contribution/Results: Our dataset exceeds mainstream benchmarks in scale by over one order of magnitude. The proposed model achieves superior prediction accuracy and significantly improved inference efficiency compared to state-of-the-art spatiotemporal models, while offering enhanced scalability. Empirical evaluation demonstrates consistent performance gains across diverse urban scenarios. This work establishes a new benchmark and paradigm for city-scale traffic forecasting.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Graph-based Machine Learning
• Computer Vision: Large Vision Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Security and Privacy: Large-scale security measurements

Abstract

Traffic forecasting on road networks is a complex task of significant practical importance that has recently attracted considerable attention from the machine learning community, with spatiotemporal graph neural networks (GNNs) becoming the most popular approach. The proper evaluation of traffic forecasting methods requires realistic datasets, but current publicly available benchmarks have significant drawbacks, including the absence of information about road connectivity for road graph construction, limited information about road properties, and a relatively small number of road segments that falls short of real-world applications. Further, current datasets mostly contain information about intercity highways with sparsely located sensors, while city road networks arguably present a more challenging forecasting task due to much denser roads and more complex urban traffic patterns. In this work, we provide a more complete, realistic, and challenging benchmark for traffic forecasting by releasing datasets representing the road networks of two major cities, with the largest containing almost 100,000 road segments (more than a 10-fold increase relative to existing datasets). Our datasets contain rich road features and provide fine-grained data about both traffic volume and traffic speed, allowing for building more holistic traffic forecasting systems. We show that most current implementations of neural spatiotemporal models for traffic forecasting have problems scaling to datasets of our size. To overcome this issue, we propose an alternative approach to neural traffic forecasting that uses a GNN without a dedicated module for temporal sequence processing, thus achieving much better scalability, while also demonstrating stronger forecasting performance. We hope our datasets and modeling insights will serve as a valuable resource for research in traffic forecasting.

Problem

• Addressing scalability issues in large-scale urban traffic forecasting
• Providing realistic city road network datasets with rich features
• Overcoming limitations of sparse highway-focused traffic datasets

Innovation

• Uses GNN without temporal module for scalability
• Introduces metropolis-scale dataset with 100k roads
• Provides fine-grained traffic volume and speed data