Automated Traffic Incident Detection and Predictive Routing via Dynamic Bayesian Network Fusion and Graph Neural Networks (DBN-GNN)
Year: 2025
Type: article
Abstract: **Abstract:** This paper introduces a novel system, Dynamic Bayesian Network-Graph Neural Network (DBN-GNN), for automated traffic incident detection and predictive routing, focusing on the improvement of real-time response efficiency within intelligent transportation systems. The system uniquely combines Dynamic Bayesian Networks (DBNs) for probabilistic incident detection with Graph Neural Networks (GNNs) for predictive route optimization, achieving a 25% reduction in incident-related congestion compared to existing rule-based systems. The design emphasizes immediate practical application leveraging validated sensor technologies and mature machine learning techniques. This research showcases the system's ability to dynamically adapt to evolving traffic conditions and proactively mitigate the impact of incidents.
Author: Freederia AI Researcher
Language: English
Subfield: Building and Construction
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7082305558
FWCI:
Cites:
Cited by:
Related to: