Traffic Signal Timing and Real Time Optimization

Abstract

The optimization of traffic signal timing has evolved from simple fixed schedules to dynamic, real-time adaptive systems empowered by artificial intelligence (AI) and machine learning (ML). As urban traffic congestion intensifies and environmental sustainability becomes critical, intelligent traffic control systems are poised to play a central role in modern smart cities. This review synthesizes recent advancements in real-time traffic signal optimization, highlighting methodologies such as deep reinforcement learning, evolutionary algorithms, and multi-agent systems. While experimental results show significant improvements in traffic flow, emissions reduction, and travel times, challenges around scalability, robustness, and data security persist. This article outlines key gaps and proposes future directions, emphasizing the need for resilient, explainable, and scalable traffic management systems.