Intelligent Traffic Management Using Fuzzy Logic and Machine Learning

Abstract

fic violations, and inefficient signal con-trol. Conventional traffic management systems rely on manual monitoring and fixed signal timings, making them ineffective in handling dynamic real-time traffic conditions. This research proposes an Intelligent Traffic Management System (ITMS) that integrates real-time traffic monitoring, adaptive signal control, traffic violation detection, and accident risk prediction through a unified analytical dashboard. The system is designed for an IoT-based four-way junction where sensors and cameras detect vehicle density and dynamically prioritize lanes with higher traffic volume. Using video-based vehicle detection, the system measures vehicle speed in real time and identi-fies violations such as over-speeding, red-light violations, and illegal parking. Drivers receive notifications through a mobile application where they can check violation details and pay fines calculated based on predefined traffic rules. Additionally, a dynamic accident risk scoring mechanism combines real-time vehicle speed data with historical violation records to identify high-risk driving behavior. Analytical dashboards visualize traffic density, violations, and risk levels to support data-driven decision making. The proposed system demonstrates how real-time traffic monitoring, violation detection, and dynamic signal control can improve road safety and traffic efficiency, contributing to the development of advanced smart city traffic management solutions.