THE RISE OF MACHINE LEARNING IN TRAFFIC MANAGEMENT: A JOURNEY THROUGH INNOVATION AND URBAN TRANSFORMATION

This systematic review evaluates the implementation and effectiveness of machine learning (ML) techniques in traffic management systems through analysis of 286 peer-reviewed articles published between 2015 and 2024. Our comprehensive analysis encompasses 45 metropolitan implementations across 23 countries, focusing on real-world applications, methodological approaches, and quantifiable outcomes. The findings demonstrate that ML-based traffic management systems consistently outperform traditional methods, achieving travel time reductions ranging from 15% to 40% and operational cost savings between 20% and 35%. This review provides an in-depth analysis of current implementations, technical frameworks, challenges, and future directions in the field.