Machine Learning for Real-Time Traffic Management in Smart Cities

Abstract

Urbanization and population displacement to cities highlight the increasing importance of providing efficient transportation systems. Congestion has manifold effects on city dwellers, from wasting time and fuel to reducing mean speed, increasing environmental pollution, and various health hazards. Many cities across the globe emphasize initiatives that aim to adopt smart city solutions, including the alteration of the built environment, smart infrastructures, passive systems, and the provision of real-time information services to enhance citizens’ well-being. A smart city is a euphemism for a city of the digital era. Among other facilities, transportation plays a vital role in reshaping a smart city. To improve the functioning of the existing transportation system, advanced traffic management solutions have been promoted. Operations research and management theory, along with information technology, have been successfully used in traffic management systems. The development of new machine learning algorithms has provided opportunities to approach this problem from different perspectives in achieving optimal vehicle coordination at cross-sections in urban traffic settings.