Enhancing Intersection Capacity and Reducing Pollutant Emissions: Real-Time Analysis and Insights

Air containing harmful and dangerous substances to human health has long become an integral attribute of cities with developed infrastructure. The significant negative contribution is made by automobile transport. The situation is especially aggravated in the area of regulated intersections, where the driving modes of automobiles change sharply (braking, accelerating, decelerating, idling) and traffic jams often occur. Various measures are taken to improve the air quality, ranging from administrative measures to the use of new environmentally friendly materials. One way to improve the air quality in the intersection area is to increase the intersection capacity (IC), reducing the probability of deep congestion. This article proposes a methodology for analyzing the IC of a regulated intersection based on real-time data processed using neural network algorithms, as well as analyzing the emissions of pollutants (PM 2.5) when changing the queue structure of vehicles. The identified patterns of the passage sequence of different categories of vehicles with levels of pollutant emissions provide valuable information for making appropriate administrative decisions on traffic regulation to reduce the negative environmental impact.