AUTOMATED MODELING AND FORECASTING SYSTEM OF ENTERPRISE ACTIVITIES

Abstract

Introduction. Mathematical methods and models are an effective tool for studying complex economic systems at different levels of enterprise management. Economic mathematical modelling is actively developing not only as a scien-tific field but also as a means of substantiating management decisions in business, in the analysis and forecasting of socio-economic processes and phenomena. In the arsenal of economic and mathematical modelling now are using- modern computing methods and computer technology. Libraries of economic and mathematical models are an integral part of the architecture of decision support systems in specific areas of the economy. The rapid development of computer technology stimulates the emergence and formation of new theoretical volumes and applied areas of modelling.Purpose. Like any large and complex field of knowledge, mathematical modelling is evolving in different direc-tions, acquiring new flexible research methods. Therefore, based on new hardware, technology and software platforms it is necessary to create new information systems using economic and mathematical models in particular for forecasting of enterprise activities.Results. The article analyses the methods of modelling and forecasting the enterprise, considers the principles of software design. Using systems analysis, the design problem was analytically divided into parts. Also were investigated the connections and relationships between these parts, in particular, were implemented the problem tree and the goal tree. Implemented business process modelling performs based on created structural-logical diagrams, namely the IDEF0 dia-gram, which helps to visually display data and information that affect software development, a server part, input data and users. Using the results of research, the authors developed an automated information system for modelling and forecasting the activities of the enterprise, which uses models of Holt, Brown, exponential smoothing and Box-Jenkins for modelling. Part of the developed system is a designed software product that implements the objectives of the research. The obtained program results allow creating a clear forecast of the future activities of the enterprise.Conclusions. Based on the built graphs of modelling and forecasting of the Cisco Systems company financial activity with using of the developed automated system, we established that the Brown model is the best for providing educational sampling and a forecast of activity. The development of the automated system in the future involves the expansion of functionality, improvement and increasing of quality, as well as the creation of powerful analytics for more detailed forecasting.