Abstract:Metropolitan traffic networks are becoming increasingly complex due to the growing population and diverse range of travel modes. However, the limited installation of continuous count stations leads to partially observable networks, posing a significant challenge for effective highway planning and design practices at various scales. Travel demand models have been developed and calibrated using sparse traffic counts at the metropolitan level. Nevertheless, these models are cumbersome to recalibrate and rerun whenever network changes occur. To overcome this challenge, we propose a flexible learning-based approach that extracts embedded knowledge from large-scale activity-based travel demand models to estimate Annual Average Daily Traffic (AADT). The approach offers two primary advantages: (1) directly learning network flow patterns based on segment attributes and network topology that can be transferred across regions, and (2) enabling efficient and reliable AADT estimation for projects of various scales. Our study explores a wide range of machine learning techniques, including novel graph neural networks that explicitly account for network topology, as well as modern and traditional regression and regression kriging models, which either disregard or implicitly consider network topology. We conducted extensive experiments using the loaded network data from the activity-based travel demand model for the Atlanta metropolitan area. Our findings underscore the importance of network topology in AADT estimation, with the diffusion graph convolutional network model demonstrating the best performance in both transductive and inductive settings. Additionally, modern tree ensemble models such as random forest regressor and CatBoost, despite their ignorance of network topology, show the second-best inductive performance with relatively lightweight structures.

Efficiently Measure the Topologies of Large-Scale Networks under the Guidance of Neural Network Gradients

Network Topology Inference by Exploring Underlying Traffic Behaviors

Network Routing Topology Inference from End-to-End Measurements.

Controlling Directed Networks with Evolving Topologies

Analyzing the importance of network topology in AADT estimation: insights from travel demand models using graph neural networks

Efficient and Dynamic Routing Topology Inference from End-to-end Measurements.

Network Topology Discovery Based on Ip Address Inferring

DeepComNet: Performance evaluation of network topologies using graph-based deep learning

Investigation of a GNN approach to mitigate congestion in a realistic MANET scenario

Learning Load Balancing with GNN in MPTCP-Enabled Heterogeneous Networks

Network Topology Inference Based on Traceroute and Tomography

Multi-source collaboration topology measurement method based on Trace NET and its application

Taking A Free Ride for Routing Topology Inference in Peer-to-Peer Networks

Network Topology Inference Based on Subset Structure Fusion

Bound-Based Network Tomography for Inferring Interesting Path Metrics.

Traffic Analysis of Partially Observable Bearer Network under RouteNet Based GNN Model

Research on the Application of Graph Neural Network in Topology Optimization of Distribution Network

Xnet: Modeling Network Performance with Graph Neural Networks

Routing in Small Satellite Networks: A GNN-based Learning Approach

Characterizing the Influence of Topology on Graph Learning Tasks

NeuralMon - Graph Neural Network for Flow Measurement Allocation.