Digital Twin-Based Intelligent Traffic Management System Using Edge Computing and Reinforcement Learning

Abstract

Urban transportation systems experience increasing congestion due to rapid urbanization and growing vehicle populations. Conventional traffic management approaches are unable to adapt effectively to dynamic traffic conditions. This paper introduces a digital twin-based intelligent traffic management framework that integrates edge computing, Internet of Things (IoT) sensors, and Deep Reinforcement Learning (DRL). A virtual digital twin continuously mirrors real-world traffic conditions and predicts congestion using real-time sensor data. The reinforcement learning agent dynamically adjusts traffic signal timing to minimize vehicle waiting time and improve traffic flow. Simulation results indicate significant reductions in travel delays, fuel consumption, and carbon emissions compared with fixed-time traffic control systems. The proposed framework contributes to sustainable smart city transportation through adaptive and intelligent traffic optimization.