Accident-Driven Congestion Prediction and Simulation: An Explainable Framework Using Advanced Clustering and Bayesian Networks

2025-07-30

AI Summary

Traffic accidents trigger urban congestion and cascading effects—including increased delays, elevated emissions, and heightened safety risks. To address this, we propose an interpretable accident-driven congestion prediction and simulation framework. Methodologically, we integrate AutoML-enhanced Deep Embedded Clustering (DEC) for automated accident pattern identification and data labeling, coupled with a Bayesian Network (BN) to model causal relationships between accidents and congestion—enabling probabilistic and interpretable congestion forecasting. We further validate predictions across diverse scenarios using SUMO-based microscopic traffic simulation. Experimental results demonstrate that AutoML-DEC substantially outperforms conventional clustering methods; the BN achieves 95.6% prediction accuracy; and simulation validation confirms strong alignment between predicted and actual congestion evolution (mean Intersection-over-Union = 0.87). This work advances intelligent traffic management by jointly ensuring high predictive accuracy, model interpretability, and practical deployability—establishing a novel paradigm for data-driven, causally grounded traffic governance.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Causality

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Traffic congestion due to uncertainties, such as accidents, is a significant issue in urban areas, as the ripple effect of accidents causes longer delays, increased emissions, and safety concerns. To address this issue, we propose a robust framework for predicting the impact of accidents on congestion. We implement Automated Machine Learning (AutoML)-enhanced Deep Embedding Clustering (DEC) to assign congestion labels to accident data and predict congestion probability using a Bayesian Network (BN). The Simulation of Urban Mobility (SUMO) simulation is utilized to evaluate the correctness of BN predictions using evidence-based scenarios. Results demonstrate that the AutoML-enhanced DEC has outperformed traditional clustering approaches. The performance of the proposed BN model achieved an overall accuracy of 95.6%, indicating its ability to understand the complex relationship of accidents causing congestion. Validation in SUMO with evidence-based scenarios demonstrated that the BN model's prediction of congestion states closely matches those of SUMO, indicating the high reliability of the proposed BN model in ensuring smooth urban mobility.

Problem

• Predicting traffic congestion caused by accidents using clustering and Bayesian Networks
• Evaluating congestion prediction accuracy with simulation-based evidence scenarios
• Improving urban mobility reliability through explainable accident impact analysis

Innovation

• AutoML-enhanced Deep Embedding Clustering for congestion labeling
• Bayesian Network for predicting accident-induced congestion probability
• SUMO simulation for evidence-based validation of congestion predictions