Methods of Assessing the Risk of Bankruptcy of an Enterprise Based on a Set of MDA-Models and the Theory of Fuzzy Sets

Abstract

The aim of the work is to develop a methodology for comprehensive assessment of the risk of bankruptcy of an enterprise based on fuzzy-multiple aggregation of estimates obtained through the use of a set of classical models. The technique is based on the use of fuzzy multi-level classifiers and allows the aggregation of estimates for the three groups of models. In each group, an enterprise is assessed according to several conditions. For example, for Altman and Taffler-Tishou models, the assessment is carried out in two states (“medium risk - high risk”). For the so-called Irkutsk model and the Savitskaya model, the assessment is performed according to five conditions (“very low risk—low risk—medium risk—high risk—very high risk”). It is significant that the analysis applies only those indicators that most reflect the possibility of bankruptcy of the enterprise. At the final stage, the normalized bankruptcy risk estimates obtained in each of the groups are aggregated into a final assessment. This assessment is an integral indicator of the risk of bankruptcy of an enterprise. The novelty of the proposed methodology consists in the possibility of combining the conclusions obtained on the basis of various non-standardized methods using different evaluation criteria. In addition, the technique allows to take into account in the model weights, reflecting the reliability of the models for the studied group of enterprises.