It is shown that one of the directions for increasing the efficiency of managing corporate systems (CS) under the influence of a large number of destabilizing fa-tors ("shocks", threats) is the development of a set of models of estimation and analysis of the long-term stability of CS in proactive contour of management, which allow timely diagnosing a decrease in the company's security level and adopting effective preventive management decisions. A review of existing approa-ches to the formation of such a set of models showed a number of limitations, the result of which is a low forecasting accuracy. The proposed approach, unlike the existing ones, allows to: 1) determine the optimal dimension of the information space of diagnostic factors; 2) find the optimal number of classes of situations for which differentiated management strategies can be developed; 3) determine the period of pre-emption, which does not require updating the models of retrospective diagnostics. This makes it possible to identify the class of not only current, but also forecast situations for a given horizon of proactive management and to choose an adequate preventive strategy.

криз у корпоративних системах на основі нейро-нечітких моделей У статті вирішено актуальну проблему загрози формування фінансових криз у корпоративних системах на основі нейро-нечітких моделей, які дозволяють своєчасно спрогнозувати загрозу банкрутства та попередити його. Зазначено, що понад 50% ВВП України виробляють корпоративні підприємства, і, крім того, є чітка тенденція до поглинання корпораціями суб'єктів малого та мікробізнесу. Наведене доводить необхідність підвищення уваги до проблематики банкрутства саме корпоративних підприємств як визначальної ланки національної економічної інфраструктури. Щороку загострення фінансових криз на корпоративних підприємствах набуває все більшого розголосу, що потребує негайного вирішення даної проблематики.

The paper is devoted to the problem of preventing financial crises in corporate systems, whose activities are becoming more and more complex in the context of globalization. The mechanism of early informing and crisis prevention in corporate systems is proposed, and includes five main modules: an analysis of the financial condition of the corporation, an analysis of the financial condition of subsidiaries, an evaluation of the impact of the financial crisis on a subsidiary on the threat of bankruptcy of the corporation as a whole, forecasting the financial condition of subsidiaries and corporation as a whole, anti-crisis management. The first four modules of the mechanism are the modules of implementation of proactive crisis management in the corporation, aimed at preventing the emergence of a crisis state, both in individual elements and in the corporate system as a whole. The fifth module is used in conditions of the current negative estimation of the financial condition of the corporation, and it is a "response" to existing crisis processes and phenomena in the corporation. After its implementation during the process of monitoring of the financial condition, proactive control modules are started to be used to allow early diagnosis and to prevent a crisis state. Particular attention is paid to such modules of proactive management as the evaluation of the impact of financial crises of subsidiaries on bankruptcy of the corporations as a whole and forecasting financial crises. A model basis for these two modules was developed. Neural networks, the mathematical apparatus of fuzzy logic, and the Caterpillar method were used for developing the models of estimation of the crisis threat in the corporate system. The developed set of models allowed to estimate the threat of financial crises in the parent enterprise and in the subsidiaries of the corporation, not only in the current but also in the perspective periods. The obtained results indicate that the financial condition of the investigated corporation is characterized by low level of the bankruptcy threat. Along with this, there is an increase in the threat of bankruptcy in a number of subsidiaries in the perspective period and the strong impact of local crises on the financial position of the corporation as a whole. The latter leads to the need of implementation of the anti-crisis measures in the corporate structure. An adequate tool for choosing anti-crisis measures and developing scenarios for the implementation of the anti-crisis management strategy is simulation modelling based on the concept of system dynamics.

The analysis of indicators that reflect changes in the social, economic and political spheres in recent years has shown their significant deterioration and the possibility of growing social tensions in the regions of Ukraine. The purpose of the study is to classify the regions of Ukraine according to the level of formation of social tensions and to determine anticipative measures aimed at preventing the creation of crisis situations. The article proposes a methodical approach to the classification of regions using the methods of cluster, discriminant analysis and analysis of variance according to the level of social tension, which includes two main stages: substantiation of the system of socio-economic indicators characterizing the level of social tension; selection and substantiation of models of classification of the regions. Within the first stage of the methodical approach the system of indicators which reflect changes in social, economic and political spheres of Ukraine in modern transformational conditions was constructed. Within the framework of the second stage of the methodical approach on the basis of cluster analysis the classification of regions according to the level of formation of social tension was carried out. The classes of regions were selected: with a low level of formation of social tension; with an intensified level of formation of social tension; with a high level of formation of social tension. The results of the study showed that the number of regions in the class with a high level of social tension is constantly growing and, unfortunately, the number of regions with high socio-economic development is decreasing. The classification of regions made it possible to determine the list of preventive measures that can reduce the losses of the state associated with the containment of possible crises in the social sphere. However, the article also states that such a list of activities should take into account the specifics of the region that is part of each class

No abstract