A Dynamic Signal Timing Control Algorithm for Urban Traffic Evaluation at Complex Signalized Intersections

Abstract

 In this paper address the aforementioned problem by modeling the optimal traffic flow time, as the traditional signal distribution in the current traffic signal system is unpredictable. To determine the optimal signal durations, this research proposes a dynamic signal distribution approach. The primary objective of this study is to optimize traffic flow rates, reduce vehicle waiting times, and transform the conventional signal system into a dynamic model integrated with a multi-agent system. In this work, intersections are modeled as autonomous agents within a multi-agent framework using a comprehensive modeling technique. The proposed algorithms optimize phase sequences and signal timings to ensure traffic efficiency. Simulations were conducted with up to 300 iterations, and the results were compared with those obtained from the traditional model. The average queue length and vehicle waiting time were finally measured. The proposed method was applied to a simulated traffic scenario, and the results showed values of 89, 114, and 83 seconds, respectively.