| With the continuous increase in traffic density in modern cities, traffic signal control systems have become an essential component for managing vehicle flow. However, traditional fixed-time traffic signal systems are often insufficient to effectively manage traffic flow and unable to adapt to dynamic traffic conditions, leading to increased waiting time and longer queues at intersections. This limitation has led researchers to explore advanced optimization techniques for Traffic Signal Control Systems (TSCS). This paper provides a comprehensive review of recent research on traffic signal optimization techniques published between 2016 and 2025. The re-viewed studies are categorized into three main groups: Metaheuristic algorithms which have been widely applied to obtain near-optimal signal timing solutions. In recent years, artificial intelligence approaches, particularly, reinforcement learning and deep reinforcement learning have gained increasing attention for real-time traffic signal optimization. Moreover, several analytical and mathematical models have been proposed to enhance traffic signal performance. The analysis of the reviewed studies indicates that metaheuristic approaches gained the largest portion of the literature, followed by artificial intelligence techniques and analytical optimization models |
*** Title, author list and abstract as submitted during Camera-Ready version delivery. Small changes that may have occurred during processing by Springer may not appear in this window.