Improved five-factor Altman evaluation model credit about the enterprise with economic indicators as fuzzy numbers

In this work, we used the Altman model, the apparatus of the theory of fuzzy sets and mathematical simulation in conditions of high uncertainty, in order to give more information to the decision maker about the creditworthiness of the enterprise, as well as the possible impact of the error on the conclusion of bankruptcy of the enterprise when calculating economic indicators.
The improved Altman model, developed initially in two respects (the rms integral approximation is used to accurately calculate a quantitative credit rating and the apparatus of fuzzy sets in order to order sets according to the degree of confidence in the obtained probability), expanded by presenting the input data as triangular fuzzy numbers.
As a result of the work done, it was possible to construct an algorithm for assessing the creditworthiness of a particular enterprise, which is based on the continuous dependence of the probability of bankruptcy on the value of the Altman function. The coefficients of the model can be triangular numbers with additional criteria for pre-reading at critical points of the classical Altman model.
The work carried out a simulation of the assessment of creditworthiness for incoming fuzzy economic indicators in the form of $\alpha$-sections of a fuzzy set to predict the impact of errors in the assessment of economic indicators on the conclusion of bankruptcy of an enterprise. The described improved Altman mathematical model with the procedure of a computational experiment (where the probability of bankruptcy of an enterprise is calculated 1000 times), supplemented by fuzzy indicators, allows you to find leftside and right-side sets of $\alpha$-levels of the fuzzy set k$_{i}$ and calculate the effect of small changes in Altman coefficients on the estimate of the probability (its stability) of bankruptcy enterprises.
This approach helps not only to adequately assess the creditworthiness of the enterprise, but also to enable it to predict the change in the result of the model due to a possible error in the input data.