Dynamic Model for Assessing the Recoverability of an Information and Communication System in Support of Capability Assessment of the Cyber Forces of the Armed Forces of Ukraine

Abstract

The article addresses dynamic assessment of the recoverability of an information and communication system after cyber impact in support of capability assessment of the Cyber Forces of the Armed Forces of Ukraine. The purpose of the study is to develop a dynamic model that represents system degradation, primary recovery, and controlled return of components to the protected operating state, and to form a normalized indicator for scenario comparison. The research method combines a three-state S–A–R system of differential equations, a conservation law for normalized states, a piecewise-constant description of cyber-impact intensity, numerical integration using an adaptive Runge–Kutta method, and scenario analysis. The model distinguishes components in the protected state, affected components, and components that have completed primary recovery but are still returning to normal operation. The integral recoverability indicator combines the final protected-state level, the maximum degradation depth, and normalized recovery time. A numerical example and four illustrative scenarios were used to examine model behaviour under weak, baseline, intensive, and enhanced-recovery conditions. The calculations show that an increase in cyber-impact intensity deepens degradation and prolongs recovery, whereas higher primary-recovery and controlled-return rates reduce the maximum affected share and shorten the time required to reach the specified operating threshold. In the baseline scenario, the maximum affected share is approximately 0.244, the recovery time is about 6.80 h, and the integral indicator is about 0.756; enhanced recovery provides the best result among the scenarios considered. The proposed model directly supports quantitative comparison of recovery conditions, identification of recovery-capability deficits, and substantiation of redundancy and recovery arrangements. The model parameters and weighting coefficients are illustrative and require calibration using exercise data, incident records, controlled tests, or simulation before practical application.