Abstract
The purpose of the study is to develop, with the help of artificial intelligence, an innovative digital microcontroller network traffic light object capable of adaptively responding to traffic congestion and changing the time phases of enabling the green signal in interaction with neighboring traffic light objects. The study uses a method of system analysis using computer simulation of existing analog and hybrid traffic light controllers. As a result of the study, an assessment of their effectiveness in the adaptive regulation of modern traffic flows in densely populated cities of the country is given. The results of simulation modeling in the AnyLogic PLE environment were used for calculations. The experimental data of complex entrance/exit road junctions of Almaty and Astana cities were taken as a basis, taking into account the time cyclograms of traffic flow characteristics of highways. An assessment was also given of the existing methods of regulating complex road transport hubs to streamline negative congestion in the center of Almaty and its expressways.