Evaluation of a Grid for the Identification of Traffic Congestion Patterns.

Today, urban growth, increased vehicular traffic and congestion have become a key challenge in cities. As a consequence, negative effects on mobility are generated, such as longer travel times, increased environmental pollution, stress for drivers, and difficulties in urban traffic planning and management. Understanding and analyzing congestion patterns is essential to effectively address this problem and develop more efficient traffic management strategies. Some research has proposed various solutions to address vehicular congestion, such as the use of algorithms for traffic data analysis, the implementation of intelligent traffic management systems, and the optimization of road infrastructure. The proposed methodology uses dynamic clustering techniques and the analysis of historical information to analyze vehicular congestion patterns, implementing the DyClee algorithm adapted to cells. The obtained results on the city of San Francisco are satisfactory, allowing the identification of clusters with certain patterns that allow identifying areas and times of higher congestion, revealing the temporal variability and highlighting the importance of considering the dynamics of vehicular flow in traffic management.