Answers are grounded in this paper and cite their sources.
Cyber-resilience engineering frameworks organize protective activity into goals, objectives and basic actions but give no formal rules for turning heterogeneous security measurements into auditable scores. Building on earlier set-theoretic data and metric models of such a framework, this paper develops a set-theoretic tuple model for assessing the cyber resilience of critical infrastructure and analyzes, with proofs, the evaluation rules on which those models relied. The hierarchy of the Cyber Resiliency Engineering Framework is encoded as an indexed family of sets with an injective identifier map; admissible threat-scenario projections form a finite distributive lattice; and each basic action carries a tuple of metrics. A clipped affine normalization agrees with a five-level class table if and only if the gaps between class boundaries are proportional to those between score breakpoints, and the two-threshold rule of the earlier models is consistent with no table; an anchored piecewise-linear normalization is consistent by construction. A capped knockout operator, which keeps a missing critical capability from being offset by other metrics, retains veto, idempotency and continuity in the normalized scores, whereas neither the hard gate nor power discounting has all three. In an illustrative bank phishing scenario, 9 of 38 observed metrics receive a linear score outside their tabulated level, and the threat-intelligence action moves from high (0.903) to above average (0.856). A parameter study shows how the cap threshold, the weights and the choice of core metrics affect the verdicts. Constructed degraded states activate the cap, the operator is compared with the geometric and harmonic means and the minimum, and a randomized check of the formal rules over 100,000 synthetic states shows that the effects are not peculiar to the example; it does not validate the policy parameters. The model is a formal scoring method that turns heterogeneous security evidence into reproducible, auditable scores; its rules are mathematically valid, but it has not been validated against expert judgment or field data, and no operational or predictive validity under an adversary is claimed.
No takes yet. Share an insight, caveat, or question.
Корченко et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: