Удосконалена модель розпізнавання рівня небезпеки ситуацій та її застосування при дослідженні функціонування системи охорони кордону

The research of systemic problems of recognizing the level of danger of situations and preventing critical situations in complex systems of various purposes is an urgent task. This is because the partial or total destruction of such systems causes not only catastrophic consequences for it, but also leads to significant losses that often exceed the cost of the systems. The analysis of a number of catastrophes in different fields of human activity shows that the magnitude of losses can be much smaller if in a freelance situation timely form and implement rational decisions. One of the possible approaches to addressing the challenges of recognizing and preventing critical situations in complex systems is an approach based on the implementation of mechanisms for adjusting the awareness of the decision maker. The article deals with the improvement of the the model of recognition and prevention of freelance, critical and catastrophic situations in a complex system and its testing on a specific example. The paper formalizes the problem of increasing the reliability of data to determine the allowable period of time for the formation and implementation of a solution for which the probability of a certain situation in a critical, emergency or catastrophic will not exceed a given value, and a description of the problem of classification and recognition of the level of danger situations. It also presents the author's approach to improving the efficiency of solving such problems, software and algorithmic support for its implementation, and a case study on the analysis of the functioning of the state border protection system in case of impacts of a diversionary intelligence group within the controlled border area. This approach is based on establishing a list of reasons that determine the factors that may change the situation of the investigated system, their meaning, options for influencing the reasons for the appearance of the factor, and based on this refinement of the confidence indicator of the decision maker. It can also be used to improve the accuracy of assessment and other quantitative indicators of a person's awareness, including completeness and timeliness. The approach is implemented by applying methods of systematic and mathematical analysis, expert assessments and analogies. The program and algorithmic implementation of the author's approach to solving the problem under study allows to automate the individual steps of solving the problem and improve the level of objectivity of the solutions. Approbation of the improved model on the example confirms the conclusion about the effectiveness of the proposed approach. The proposed methods may be applicable to the complex solution of system analysis problems, in which recognition and prevention of emergency, critical and catastrophic situations in a complex system is an integral part of the task.