In the context of increasing road traffic, congestion inevitably leads to significant economic, environmental, and social costs. The causes of congestion at controlled intersections can vary widely, ranging from the presence of infrastructure and growing urban development to obstacles on the road surface and adverse weather conditions. One of the key challenges in addressing congestion is the difficulty of obtaining accurate data on traffic flow and road occupancy. This paper introduces a new term, “congestion level” of an intersection, which implies the creation of a specific “profile” for each intersection. Based on this approach, models are developed to predict the increase in congestion length under various scenarios, including changes in the number of traffic lanes, as well as models for estimating the duration of congestion dissipation. The article presents a methodology for calculating and classifying levels of congestion complexity, which allows for more effective analysis and management of congestion. The results of this research will be useful for making informed management decisions regarding traffic regulation in urban transportation networks, thereby contributing to an improvement in the overall congestion situation and enhancing the quality of life for city residents.