UrbanFlow: A Digital Twin-Based Adaptive Traffic Signal Optimization System

Conventional traffic signal systems operate on fixed timing schedules that do not account for real-time variations in vehicle density or queue length. This rigid approach frequently results in avoidable congestion, longer commuter waiting times, and inefficient utilization of road infrastructure, particularly as urban traffic volumes continue to grow. This project proposes a smart traffic signal management system built around a digital twin — a safe, realistic virtual simulation of a road network — that enables adaptive, datadriven signal control to be designed, tested, and refined without any risk to real-world infrastructure. The system incorporates a rule-based adaptive traffic signal control mechanism that dynamically adjusts signal timing based on real-time vehicle density and queue length, and is intended for use by traffic control authorities responsible for monitoring and managing urban signal networks, as well as the general public seeking visibility into real-time traffic conditions. The system uses SUMO (Simulation of Urban MObility) to construct and simulate single and multi-intersection road networks under varying traffic conditions, including low, medium, and peak-hour scenarios. It provides a live simulation view, real-time monitoring and analytics, and an interactive reporting dashboard that enables comparative analysis between conventional fixed-timer signal control and the proposed rule based adaptive approach. Simulation results are stored for ongoing analysis, allowing administrators to evaluate the effectiveness of different traffic management strategies over time. By replacing rigid, fixed-timer control with a responsive, condition-aware signal system, this project is expected to reduce average vehicle waiting time, improve intersection throughput, and provide a validated, extensible foundation for further intelligent transportation research. This paper also reviews thirty studies on adaptive and digital-twin-based traffic signal control, SUMO/TraCI simulation tooling, and learning-based control, and positions the implemented system against that literature.