Scientific Journal of KubSAU

In the process of formation of nonlinear dynamics, the scientific society was able to refute the classical mechanisms of Newton-Laplace by justifying the chaotic nature of the phenomena of the world. However, despite the emergence of new mathematical models and tools, forecasting of nonlinear systems is a difficult task, as not only the quantitative and qualitative characteristics of the factors affecting the system are unknown, but also there is a problem of a small amount of information for forecasting. In this article, the authors consider the linear cellular apparatus as a tool for prediction the final state, to which the system will come based only on its output indicators of previous years. Since the use of a linear cellular automaton for prediction of nonlinear systems is an assumption of the authors, it should be tested on the series of stochastic systems exposed to different risk factors, which together give either a positive response of the system or a negative one. An example of such series is the time series of yields, as it is affected by climatic conditions, the appearance of which, in turn, is also difficult to predict. Prediction of stochastic systems using linear cellular automaton really makes it possible to get adequate and visual models. Due to the fact that the forecast model has a discrepancy with the real result of 0-15% (both positive and negative), the conclusion is that the predicted value will help either to take measures to ensure that the real value in the future is not lower, or to make sure that the decisions and measures taken are correct, when a value is higher than the forecast