Bayesian model for evaluating the pragmatic efficiency of an information system of technological control

Abstract

The article considers the problem of assessing the effectiveness of an intelligent decision support system for controlling complex technical systems at the level of a technological control system. To build a model for assessing the effectiveness of an information system, a pragmatic approach to measuring the value of information and an intelligent technology for forming a solution based on a Bayesian trust network were used. As an indicator of the pragmatic efficiency of the information system, it is proposed to use the pragmatic efficiency coefficient. It is determined by the increase in the probability of achieving the goal set for the technological control system, achieved through the use of additional information in decision-making. It is proposed to use the probability of error-free and timely execution of control as a target indicator characterizing the quality of control performance. The determination of the values of the target indicator of the control system with and without the use of the information system is carried out using the Bayesian trust network, which takes into account the relationship of the most significant elements of the technological control system. In comparison with the known methods, the proposed approach to assessing the effectiveness of an intelligent decision support system is based on joint use: a model for measuring the value of information and an adapted Bayesian model for calculating a target quality indicator. The approach considered in the article makes it possible to determine the coefficient of pragmatic efficiency, taking into account the value of the information produced to achieve the set goal and the uncertainty of the initial data.