Kudelina K. Models, methods, and information technology of the control and forecasting of silicon monocrystal production indicators

Object - processes of operations management of monocrystal production; goal - the development of models, methods, and information technology of the control and forecasting of silicon monocrystal production indicators; methods - systems analysis, set theory, mathematical statistics and cluster analysis, theory of artificial neural network, theory of identification and time-series analysis, functional structure modelling and structure modelling of algorithmic techniques of control system, that correspond with object-oriented approach to systems engineering; novelty - the information technology for the operations control and forecasting of the solid Cz-Si-monocrystals production indicators was developed for the first time; the approach to structural synthesis of ARIMA-models of time sequences was presented for the first time; the method of generalized forecast construction based on the generalization of a set of specific forecasts was improved, it is achieved using adaptive estimation of generalization parameters depending on the current values of accuracy degree of the specific forecasts; mutual absorption method for clustering of time series has obtained the follow-up development due to use as a vector of classification signs of a vector of statistical estimations of the basic characteristics of series; results - developed models and information technology of the control and forecasting of silicon monocrystal production indicators were realized in subsystem "Modelling and forecasting" of automated organizational and technological control system of solid Cz-Si-monocrystals production in Silicon Co Ltd (Svetlovodsk) and in educational process; branch of application - enterprises on solid Cz-Si-monocrystals production.