2025 Asian Conference on Communication and Networks (ASIANComNet 2025)

Modern cities' fast expansion of urban populations has resulted in more vehicle traffic, aggravating congestion and delays as well as environmental damage. Often reactive and rule-based, traditional traffic management systems are not enough to manage the dynamic character of metropolitan transportation networks. This work suggests an artificial intelligence (AI)-driven framework for real-time traffic management targeted at increasing mobility, thus lowering congestion, and so improving general traffic efficiency in smart cities. Leveraging data from sensors, traffic cameras, GPS-enabled vehicles, and IoT infrastructure, the system includes advanced AI techniques—such as deep learning for traffic prediction, reinforcement learning for adaptive signal control, and graph-based models for dynamic routing decisions. Using a simulated urban setting with real-world traffic data, the framework is evaluated showing notable advantages in trip time reduction, traffic flow optimization, and emission management over conventional methods. By allowing data-driven, responsive, and sustainable traffic management solutions for the smart cities of the future, this study emphasizes how artificial intelligence might alter urban transportation.
 

Artificial Intelligence; Smart Cities; Real-Time Traffic Management; Intelligent Transportation Systems (ITS); Urban Mobility; Deep Learning; Traffic Optimization; IoT