Enhancing Urban Mobility through Adaptive Traffic Analysis: A Case Study in Singapore

Abstract Traffic congestion is a significant issue in most urban landscapes, causing delays and affecting the daily commutes of thousands. This study, "Adaptive Traffic Learning and Analysis System" seeks to leverage the capabilities of Large Language Models (LLMs) to analyze data from urban traffic systems and provide detailed insights into congestion points. By using the Simulation of Urban MObility (SUMO) software, we replicate street layouts and traffic signals for specific locations in Singapore, facilitating real-time traffic simulations. The project employs SUMO’s APIs to extract vehicle positioning information, which is then used to compute traffic flow statistics. These insights enable LLMs to propose informed optimizations for traffic management, aiding urban planners and traffic managers in decision-making. The recommendations can be further evaluated and refined using SUMO, aiming to mitigate traffic congestion and improve urban mobility in Singapore.