AI-Driven Adaptive Traffic Control for Tehran: Optimizing Real-Time Traffic Flow in a Smart City Context

This study explores how an intelligent traffic management approach could help address two of Tehran’s major urban challenges: heavy congestion and air pollution. The proposed system relies on real-time information collected from traffic sensors, GPS-enabled vehicles, and roadside cameras to better understand traffic conditions as they evolve throughout the day. By applying machine-learning techniques—particularly reinforcement learning—the system is able to anticipate traffic flow patterns and respond dynamically rather than relying on fixed control plans. Traffic signal timings are adjusted continuously, and drivers receive route guidance aimed at distributing vehicles more evenly across the network. The system is designed to align with the broader goals of Tehran’s Smart City initiative. Simulation results indicate noticeable improvements, including shorter travel times during peak periods, reduced congestion at critical intersections, and lower vehicle emissions. Beyond performance gains, the study also discusses practical considerations such as data coordination, scalability, and compatibility with existing infrastructure, highlighting both the promise and the real-world challenges of deploying AI-based traffic control in a large and complex city like Tehran.

کلیدواژه ها:

Adaptive Traffic Control ، Artificial Intelligence (AI) ، Smart City ، Tehran ، Traffic simulation