Research on Freeway Traffic Anomalies Based on Association Analysis

In order to study the characteristics and influencing factors of freeway traffic anomalies, this paper employs association analysis to investigate abnormal traffic incidents on G4218 in Sichuan. The study reveals strong associations between traffic anomalies on bridges and rainy weather, long downhill slopes, and curves. Traffic anomalies in tunnels often occur on downhill sections of long tunnels on cloudy days. Furthermore, anomalies on long uphill and downhill slopes are significantly associated with high and low speeds, respectively. Additionally, through sequence pattern mining, the study categorizes traffic anomalies in the research section into three classes based on duration and impact range. It is found that the most severe class of traffic anomalies shows a strong correlation with risky driving behavior. These findings could contribute to the early detection of traffic anomalies and the implementation of corresponding measures, reducing losses caused by traffic abnormalities and promoting road safety in other regions.