Smart Traffic Management System Using IoT and Real-Time Data Analytics

Abstract

Urban traffic congestion continues to reduce travel reliability, increase vehicle delay, and create operational challenges at signalized intersections where demand can change rapidly. Conventional fixed-time signal timing often cannot respond effectively to short-term fluctuations, motivating the need for traffic management that integrates continuous sensing with timely decision support. This study aimed to design and evaluate a smart traffic management system that combines Internet of Things data collection and real-time streaming analytics to support adaptive signal timing and continuous operational monitoring. An experimental systems research design was applied by implementing an end-to-end pipeline that connected traffic data sources, publish–subscribe messaging, streaming analytics, and a rule-based adaptive decision module for updating signal timing parameters under operational constraints. System performance was evaluated using telemetry delivery reliability and end-to-end latency from data generation to decision output, while traffic performance was assessed by comparing baseline fixed-time control and adaptive control using average delay, queue length, throughput, travel time, and stops. The results showed that the telemetry pipeline maintained high delivery reliability with low message loss, and the streaming layer achieved low end-to-end latency with stable processing across the evaluation period. When labeled congestion states were available, the analytics module produced balanced congestion classification performance suitable for decision support. Compared with fixed-time control, the adaptive strategy reduced average vehicle delay and queue length and increased throughput under comparable demand conditions, with improvements observed across different demand settings. The study concludes that integrating connected sensing and real-time streaming analytics within a modular architecture can provide a practical foundation for responsive traffic management, enabling timely operational actions and measurable efficiency gains. This work contributes an implementable pipeline and evaluation evidence that can guide phased deployment and further development toward larger-scale coordination, improved robustness, and predictive or more advanced decision modules.