Optimization of Transportation Traffic using Agile Approach

Traffic congestion and its related delays present significant challenges to urban infrastructure and societal efficiency. This project demonstrates an intelligent traffic management system that combines real-time vehicle detection, adaptive signal control, and predictive traffic modeling to solve the problem of urban congestion. The implementation of YOLOv8 for the identification and counting of vehicles with accurate precision in low-quality surveillance footage will feed into the production of the real-time data important for the efficient management of traffic flow. Such information feeds into a Signal Switching Algorithm, which continually adjusts signal timings according to the instantaneous level of traffic to optimize intersections for smoother flow. Validation of its effectiveness is done with a computer simulation of a 4-way intersection via Pygame where different scenarios affecting the reduction of congestion are tested upon the algorithm. With all this being said, another model makes use of historical data about traffic to foretell such patterns of flow, allowing pre-optimized signal timings to filter into traffic management to optimize flow even more. Such a system would provide an overall city traffic management approach that operates in real time and has data-driven insight into predicting more sustainable city transportation based on predictive analysis.