The article is dedicated to a comprehensive assessment of factors affecting road conditions, including the number of vehicles, their speed, distance traveled, idle time, and waiting time. Special attention is given to analyzing the relationships among these parameters and their impact on the formation of congestion. The study investigates the application of machine learning methods, particularly the k Nearest Neighbors (kNN) model, to evaluate the degree of congestion in real-time. Within this work, an analysis of the effectiveness of various metrics such as precision, recall, and F1-score is conducted to assess the quality of the binary classification model. The research findings have identified key features that influence traffic conditions and established their correlations. The obtained data emphasize the appropriateness of using the k Nearest Neighbors model for analyzing and predicting anomalies in traffic situations. Future research aims to deepen the understanding of the characteristics of congestion and further enhance methods for analyzing traffic flows.