Methodology for Detecting Anomalous Traffic Flows in the Network Based on the Hurst Parameter

This paper investigates the use of the Hurst index for traffic flow analysis based on data from intelligent transportation system (ITS) detectors. The values of Hurst index (H) for different time intervals during the day ranged from 0,1295 to 0,4178, indicating the presence of antipersistence and trendiness in certain time periods. These results confirm that the Hurst index can be applied for predictive analytics and adjusting traffic light modes to prevent traffic congestion. A comparison of the results of different data analysis methods and data from different regions has revealed significant differences, which emphasizes the importance of considering the specifics of transportation systems in the analysis. The Hurst index estimation methodology has proven to be an effective tool for identifying anomalies in traffic flows and optimizing traffic management, especially when compared to alternative methods.