Search

Search Results

• TRANSPORT FLOW FORECASTING MODEL BASED ON NEURAL NETWORKS FOR TRAFFIC PREDICTION ON ROADS

  Alamir Haider Sagban Hussein, Е.V. Zargaryan, Y. А. Zargaryan

  124-132

  2021-08-11

  Abstract ▼

  In connection with the industrialization of modern society, the growth of the transport sys-tems of our country, an increase in certain necessary for the development of the needs of the citi-zens of our country, the number of vehicles of various types continues to increase every year very fast, causing huge traffic jams on transport roads, especially in large cities and megacities. Thus, forecasting traffic flows is an important and necessary component of optimal traffic control in modern conditions of transport network development. As a solution to this problem, this article aims to analyze and describe the application of artificial intelligence methods, in particular neural networks, which represent a modern approach to modeling in complex and nonlinear situations that arise when predicting a traffic flow model. The shown accuracy method is based on the devel-opment of a neural network to predict the daily traffic flow. The expected traffic flow is then com-pared with the actual dataset recorded on the road section and provided by the infrastructure manager. In fact, neural networks are able to learn from past situations and predict future situa-tions on the transport network. In this study, various neural network structures were examined,and the simulation results showed that the best predictions were obtained using the multilayer perceptron architecture, which has a good generalization system with a root mean square error of 0.00927 with the current set of vehicles. The first part of the article is devoted to defining various concepts related to the current research area, including a review of the literature on traffic predic-tion and neural networks. The second part is devoted to describing the problem of traffic conges-tion using forecasting problems and presenting the proposed solution method with an emphasis on artificial neural networks as a means of forecasting demand and its various structures. Then, nu-merical experiments are illustrated by analyzing the forecast results after the formation and test-ing of various neural network architectures.