| Paper FR-EA-T39.2
Arjomandi, Larry M (Austroads), Radman, Ehsan (Navrood)
Next-Generation Real-World Stress Prediction for Australian Bridges Via Graph Attention-Based Models
Scheduled for presentation during the Regular Session "S39b-Data-Driven Optimization in Intelligent Transportation Systems" (FR-EA-T39), Friday, November 21, 2025,
13:50−14:10, Coolangata 3
2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia
This information is tentative and subject to change. Compiled on October 18, 2025
|
|
| Keywords AI, Machine Learning Techniques for Traffic Demand Forecasting, AI, Machine Learning for Real-time Traffic Flow Prediction and Management, AI, Machine Learning and Predictive Analytics for Traffic Incident Detection and Management
Abstract
With rising traffic and heavier electric vehicles, Australian smart transport systems need advanced bridge stress analysis for resilience. Traditional methods lack speed and scalability. We propose a graph attention network with a variational autoencoder and piecewise neural networks, extracting latent features from Queensland bridge data to predict fatigue and yield stresses. Using 2022 to 2024 traffic data from IoT and weight sensors on heavy vehicles, the model adapts to load variations, achieving fatigue/yield errors of 3.9/7.6 MPa (Concrete), 3.7/10.3 MPa (Steel), and 2.7/8.3 MPa (other materials). Its modular, piecewise design using Mixture-of-Experts architecture outperforms traditional methods, enabling real-time monitoring, stress risk identification, and preventive maintenance.
|
|
This site is protected by copyright and trademark laws under US and International
law.