A Sensor-Data-Driven Proactive Accident Detection and Traffic Prediction Method Based on Lane-Level Grid Partitioning and a Three-Dimensional Markov Model

Abstract

Highlights What are the main findings? The proposed sensor-driven method achieves lane-level accident detection and traffic prediction with high accuracy by fusing historical and real-time data within a three-dimensional Markov model. The proactive detection mechanism substantially shortens detection latency, reducing secondary accident risks and congestion propagation. What are the implications of the main findings? The framework offers a practical sensor-enabled solution for real-time traffic surveillance and intelligent transportation management in big-data environments. The method is also applicable to diagnosing congestion and other traffic anomalies, supporting comprehensive roadway incident monitoring and proactive control strategies. A sensor-data-driven proactive accident detection and traffic prediction method based on lane-level grid partitioning and a three-dimensional Markov model. Abstract The timely detection of road traffic accidents is essential for intelligent transportation systems. Leveraging multi-source sensor data including GPS, loop detectors, and vehicular sensors, this study proposes a proactive accident diagnostic method within a big-data framework. We introduce a lane-level traffic state representation that discretizes each lane into rectangular grids, enabling precise evaluation of local traffic conditions. To capture the spatio-temporal propagation of traffic disturbances, a three-dimensional Markov model is adopted, which accounts for both upstream–downstream traffic spread and temporal evolution, as well as historical features, to predict post-accident traffic dynamics. Experimental results demonstrate that the proposed method achieves high-accuracy lane-level accident detection and improves traffic prediction performance through the effective fusion of historical sensor records with real-time streaming data. The proactive detection mechanism efficiently reduces accident identification time, thereby mitigating potential secondary impacts. Additionally, the method proves effective in diagnosing other traffic anomalies, such as congestion, and for continuous monitoring of roadway incidents. These findings provide a practical sensor-enabled solution for accident detection and traffic flow prediction, offering a robust basis for real-time traffic management under intelligent network and big-data environments.