Digital Twin Environment with Reinforcement Learning-Based on Traffic Control

Transportation networks in smart cities are considered as one criteria of quality of life. This paper discusses Digital Twins (DTs) adopted in the Artificial Intelligence (AI) domain, especially traffic management systems in smart cities. DTs is done by collecting and analyzing massive amounts of data from various sources such as cameras, sensors, and Global Position System (GPS), where Machine Learning (ML) algorithms and predictive analytics are performed on these systems. In this paper, we discussed different types of methodologies such as inverted pyramid and snowballing methodologies along with security threats of the data that gathered that faced in this area. This study concluded the comparisons of previous algorithms to get the best results in traffic control management systems based on areal practice to develop a traffic intersection located in Alhasa city, Saudi Arabic. As this inverted pyramid algorithm is more reliable as classified the most important data first, and suitable for network data analysis, than the snowballing algorithm.