Application of artificial intelligence in traffic management in the city of Jizzakh (Uzbekistan)

Abstract

This paper investigates the application of artificial intelligence methods for traffic management in the city of Jizzakh, including traffic flow prediction, optimization of traffic signal cycles, management of transport interchange hubs, and enhancement of road traffic safety. Simulation experiments were conducted using deep learning and reinforcement learning techniques based on real traffic data collected over a five-month period. The obtained results demonstrate a reduction in the average vehicle waiting time at intersections by 30%, an increase in arterial road capacity by 22%, and a 19% reduction in accident risk following the implementation of AI-based algorithms. The novelty of the study lies in the integration of adaptive AI models that account for the local characteristics of the Jizzakh transport network and the operational features of transport interchange hubs, which have not been previously addressed in regional studies.