Optimization of traffic light movement in the city's transport network

Abstract

The article discusses the pressing issue of optimizing traffic flow management in urban environments. The study is relevant in the context of the priority area of scientific and technological development in the field of intelligent transport systems and is devoted to the development and testing of an algorithm for adaptive control of traffic light phases operating uniformly within a selected part of the urban environment. The proposed method is based on the dynamic formation of traffic light phase durations, considering the current traffic situation and predicting the time of lane overload (congestion). The paper presents a detailed methodology for traffic modeling, including the creation of a digital model of the transport network and a series of experiments with different control parameters using the Simulation Of Urban Mobility (SUMO) environment. The study was conducted using the transport networks of six major cities around the world (New York, Moscow, Tokyo, Berlin, Shanghai, and Mexico City). During the simulation, three key indicators were evaluated: the accumulated number of vehicles on the roads during the simulation, the accumulated difference between moving and stationary vehicles during the simulation, and the accumulated level of congestion. The results of the experiments demonstrate an improvement in traffic performance when using the proposed algorithm. The improvement in parameters ranged from 2.41% to 51.07%, depending on the study