Methodology for determining optimal terms of transport repair using neural networks technology

Modern data technologies contain such industry 4.0 tools that allow you to accurately and efficiently work with a large stream of data. The use of neural networks in various technical service management tasks is due to some advantages over classical statistical methods. So, compared to linear methods, a neural network is able to effectively create non-linear dependencies and more accurately describe data sets. In the classical methods of statistics, a common method is the Bayesian classifier, which builds a quadratic separating surface, a neural network is able to build a surface of a higher order. In addition, the neural network is trained on the entire data sample without fragmenting it, which increases the adequacy of the network settings. Thus, within the framework of the classical approach, it is not possible to obtain a significant improvement in the quality of technical service management; it is important to improve the technical service management methodology, use and adapt neural network modeling tools. Identification of parameters that reflect the dependence of technical service management and the efficiency of the enterprise as a whole, determination of optimal values will allow to develop a methodology based on the use of neural networks in the management of technical service of an enterprise to obtain optimal solutions in the field of maintenance and repair of equipment, which will improve the efficiency of equipment, reduce costs, reduce losses from sudden failures.