Predicting Traffic Congestion at Urban Intersections Using Data-Driven Modeling

Urban traffic congestion is a significant challenge faced by modern cities, impacting commute times, safety, and overall quality of life for residents (Weisbrod et al., 2003). Intersections, which serve as crucial points of convergence for vehicular traffic, are particularly susceptible to congestion, leading to stop-and-go patterns and increased travel times (Gazis, 2002). Predicting congestion at intersections can provide valuable insights for city planners and governments, enabling them to implement strategies for optimizing traffic flow, enhancing infrastructure, and improving the overall transportation system (Wang et al., 2016). This study aims to develop a predictive model for congestion at intersections in major U.S. cities, leveraging a comprehensive dataset of trip-logging metrics from commercial vehicles. The dataset encompasses a wide range of features, including intersection coordinates, street names, time of day, and traffic metrics, providing a rich source of information for identifying potential congestion patterns.