AI ENABLED DIGITAL TWIN SYSTEMS FOR TRAFFIC AND PUBLIC SERVICE OPTIMIZATION SMART

Abstract

The high rate of expansion in smart cities has made their traffic system more demanding and efficient in-service delivery systems. The new technologies of Artificial Intelligence (AI) and Digital Twin (DT) have been introduced to solve these urban challenges. This is a critical review paper on AI-driven digital twin systems that find their application in smart management of traffic and optimization of a public service.

The paper comprehensively conducts the systematic review of the recent sources (published in 2020-2026) covering the theme of intertwining AI methods, including machine learning, deep learning, and reinforcement learning with the concept of the digital twin. These technologies make it possible to monitor, simulate and make data-driven decisions in cities in real-time. The major applications that have been discussed include adaptive traffic signal control, prediction of congestion, optimization of emergency response, and planning urban mobility.

Moreover, the paper assesses the system architectures, data sites like IoT sensors and vehicular networks, and computational models like cloud and edge computing that can support such solutions. The comparative analysis of the existing studies shows the benefit of increased efficiency of traffic

In spite of these advantages, there are also a number of issues such as heterogeneity of data, scalability issues, privacy and high implementation costs. Research gaps and future directions are also established based on emerging technologies in the review 5G/6G, explainable AI and big scale 5G/6G digital twin ecosystem to build sustainable smart cities of the future.