THE USE OF FUZZY COGNITIVE MODELS FOR DIAGNOSTICS OF PROBABILITY OF ENTERPRISES’ BANKRUPTCY

The article researches theoretical and methodical foundations of cognitive analysis and semistructured economic systems’ modeling, problems of improving of existing methodical approaches to diagnostics of probability of enterprises’ bankruptcy through the use of fuzzy cognitive model. Cognitive diagnostic of probability of bankruptcy is aimed to obtaining knowledge of processes of business entity’s activities based on a study of quantitative and qualitative indicators with a purpose to assess both current and future state of an enterprise base on accounting and reporting data as well as expert appraisal. The result of cognitive diagnosis is to determine the probability of enterprises’ bank-ruptcy based on integral indicator of probability of bankruptcy, which reflects the manage-ment level in five directions, namely: the management level of fixed and circulating assets, personnel, financial resources and level of culture of enterprise. Is proposed to conduct cog-nitive diagnostics of probability of enterprises’ bankruptcy in two directions, namely: Q-diagnostics, based on an assessment of financial performance, and V-diagnostics, based on the research of non-financial verbal indicators. Cognitive modeling makes it possible to solve problems of a conceptual nature, to make managerial decisions that will provide the business entity with competitive advantages in future. A characteristic feature of the model of cognitive diagnostics of probability of enter-prises’ bankruptcy is that it is based on anticipative management concept, it can help to not only to determine the probability of bankruptcy of a business entity, but also to investigate the main factor of bankruptcy in the perspective and identify ways to improve the level of this factor. Using this model, enterprises will not only assess the probability of bankruptcy today, but also to prevent the bankruptcy of economic entities in the future.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.