Scientific and methodological apparatus for processing diverse data in automated control systems

This section of the study proposes the conceptual foundations for the use of artificial intelligence in intelligent decision support systems.

In the course of the research, the authors:
– justified the feasibility of using artificial intelligence theory for processing heterogeneous data in automated control systems;
– developed a methodology for data distribution in automated control systems;
– designed a model for evaluating the process of heterogeneous data processing in automated troop control systems using expert information;
– improved the methodology for configuring an information system to evaluate the process of heterogeneous data processing in automated control systems under conditions of uncertainty.

The analysis conducted in the study established that the application of fuzzy graphs and the mathematical apparatus of fuzzy logic in decision support tasks for data distribution and the evaluation of heterogeneous data processing under various conditions, including uncertainty, enables the distribution of data among elements of automated control systems based on the importance of the elements and the number of features in real-time.

The methodology for rational data distribution based on the importance of automated control system elements and the number of features in such systems under conditions of uncertainty has been improved. This methodology differs from existing ones by combining the mathematical apparatus of information theory, fuzzy logic, and expert evaluation, enabling the formalization of features in a unified parameter space and the intellectualization of information processing processes.

A quantitative assessment of the proposed methodology’s efficiency was conducted. The results of this assessment demonstrated that data distribution among the elements of automated control systems based on importance and the number of features using the proposed methodology improves the timeliness of data processing and decision-making regarding the state of the heterogeneous data processing process by 15–17 %.

An enhanced methodology for configuring the information system for evaluating the heterogeneous data processing process in automated control systems under conditions of uncertainty, utilizing a genetic algorithm, was developed. This methodology addresses limitations of other methods in varying specific features while holding other indicators constant, thus improving the efficiency of the developed information system for evaluating heterogeneous data processing in automated control systems.

The scientific outcome is the improvement of the genetic algorithm for differentiated tuning of the fuzzy knowledge base of the information system for evaluating heterogeneous data processing in automated control systems based on posterior data.

A quantitative assessment of the improved methodology’s effectiveness was performed. The results indicated that the proposed methodology enhances the timeliness of configuring the information system for processing heterogeneous data in automated control systems under conditions of uncertainty.