Mathematical Model of Intellectualization of Computer Control Systems for Critical Infrastructure Objects

The paper investigates the theoretical, algorithmic, and structural foundations for the development of intelligent computer control systems intended for operation at critical infrastructure facilities with increased reliability and safety requirements. The relevance of the study is determined by the growing complexity of modern technical and transport systems, the increasing volume of heterogeneous data processed in real time, and the necessity to minimize operator risk under conditions of uncertainty, incompleteness, and inconsistency of information. Under such conditions, traditional deterministic data-processing algorithms prove insufficient, which necessitates a formalized approach to the intellectualization of computer control systems. A mathematical model of the intellectualization of computer control systems is proposed, providing a staged transition from informational and analytical levels to knowledge-oriented and adaptive-intelligent levels of operation. Within the proposed framework, the baseline and residual operator error risks are formalized as a functional relationship between the initial probability of error and the efficiency of intelligent decision support. Analytical models for risk reduction at different levels of intellectualization are developed, including linear, polynomial, and logistic dependencies that reflect qualitative changes in the risk compensation mechanism. It is demonstrated that the adaptive-intelligent level, implemented using artificial intelligence methods, machine learning techniques, and contextual analysis, ensures nonlinear and threshold-based regulation of decision-support intensity. This approach enables minimization of residual risk without violating the “human-in-the-loop” principle and ensures consistency between automated analysis and expert operator judgment. The proposed model provides a theoretical basis for the development of adaptive decision-support systems and can be applied in the design of intelligent computer control systems in domains with increased requirements for safety, reliability, and adaptability of computational processes