Моделирование неопределенности при помощи нейронных сетей

Abstract

Introduction: modern political and economic conditions of uncertainty dictate the need for a systematic strategic development of industrial enterprises, and, consequently, new principles of production management. The functioning of enterprises, which are a complex economic system, is accompanied by the behavior uncertainty of both the elements of the system itself and external factors. The elimination of uncertainty from the modeling of economic processes does not allow to consider the behavior of an object in real conditions, which affects the system development forecasting. The use of neural networks makes it possible to obtain a quantitative assessment of uncertainty factors influence on enterprises. Aims: review of methods for quantifying the uncertainty level as well as development of a neural network model of uncertainty in decision-making. Methods: neural network modeling. Results: methodological approach has been developed to the construction of a neural network model for assessing uncertainty for production transfer to innovative materials at an aviation industry enterprise. Conclusions: the use of a neural network for applicability is expedient in the absence of explicit mathematical dependencies. The neural network model can be easily adapted to various uncertainties in different enterprises.