Exploring Urban Traffic: Uncovering Sectional Anomalies through an Optimization Framework

Sectional anomaly detection in urban traffic is challenging due to weather or driving behavior uncertainties. Identifying these anomalies is crucial for maintaining traffic safety and effective administration. This paper proposes a framework for optimization to uncover anomalous patterns in traffic flow time series. The proposed methodology involves the analysis of potential anomalies through the utilization of clustering techniques. An optimization framework is employed to identify particular sections of traffic data that deviate significantly from standard patterns by assuming that smaller clusters are likely to be abnormal. The results were validated through a comparison with established methodologies. The experimental results indicate that the conventional optimization methods can effectively estimate the sectional irregularities with high accuracy. The paper also examines the consequences of detecting anomalies in urban traffic management.