Mathematical model of optimizing the arrival of fire units with the use of information systems for monitoring transport logistics of Voronezh city

Abstract

In recent years, the strong pace of construction is increasing in big cities. With their growth becomes a question of the deployment of firefighters and the number of fire stations. The most effective solution is the problem of finding the optimum route of fire departments, taking into account the information transport logistics systems within the city that will allow us to arrive at the scene at any time, regardless of the degree of congestion of city roads. Prompt arrival of fire units provides the most successful fire fighting. The main objective of the study is to develop a preliminary route and the route in case of unforeseen factors affecting the time fire engine arrived. To construct the routes used to develop actively in the current methods of machine learning artificial neural networks. To construct the optimal route requires a correct prediction of the future behavior of a complex system of urban traffic based on its past behavior. Within the framework of statistical machine learning theory considered the problem of classification and regression. The learning process is to select a classification or a regression function of a predetermined broad class of such functions. After determining the prediction scheme, it is necessary to evaluate the quality of its forecasts, which are measured not on the basis of observations, and on the basis of an improved stochastic process, the result of the construction of the prediction rules. The model is verified on the basis of data collected in real departures real fire brigades, which made it possible to obtain a minimum time of arrival of fire units.