Each Evaluation Criterion is evaluated independently as a Criterion-level State, rather than being assessed collectively through a single uniform method. Individual Criteria may carry different Weight or Criticality depending on the relevant Policy Objective, Risk, Operational Context, and other factors. Accordingly, Overall Evaluation is not derived through simple majority voting or a mechanical average of Criterion-level States.

Weight indicates the relative contribution of a Criterion to the Evaluation of a particular Case. Where a Criterion contains multiple Dimensions, different Weights may also be assigned to individual Dimensions. This allows the same Evaluation Framework to adjust the relative importance of different factors according to Risk, Domain, Organization, Use Context, and other relevant conditions.

However, Weight ≠ Restrictiveness. A Criterion may have High Weight because it is highly relevant to the Case while still result in a White State after Contextual Evaluation. Conversely, a Criterion with relatively low Weight may result in a Black State where a clear Prohibited Condition is identified.

A further distinction must be maintained between Weight and Criticality. Weight represents the relative contribution of a Criterion to Evaluation, whereas Criticality represents the Governance significance of a particular State of that Criterion for the Overall Evaluation. A Black State indicating a Critical Prohibited Condition therefore cannot simply be offset by multiple White States. Aggregation Rules must explicitly account for such Non-compensatory Conditions.

In the Overall Evaluation, Criterion-level States are integrated according to Weight, Criticality, Policy Alignment, and the applicable Aggregation Rules. The Overall State does not erase individual Evaluation results. Rather, it preserves them while representing their Governance Significance for the Case as a whole.

This structure makes it possible to verify retrospectively how each Criterion was evaluated, which Weight and Criticality were applied, and which Aggregation Rules led to the resulting Overall State.

By connecting this Evaluation Process with the Evidence Architecture, the system can preserve not only the final Overall State, but also the Criterion-level Evaluations, applied Weight, Criticality, relationship to Policy, Aggregation Rules, and the Evaluation Process through which the Overall State was derived.

Importantly, Overall Evaluation ≠ Execution Decision. Overall Evaluation forms a Governance Input required for Decision, while Allow / Hold / Deny for Execution is formed through the Decision Process, taking into account Authority, Policy, Context, and other Governance Conditions.

As a result, an Execution Decision is not merely an Allow / Hold / Deny outcome. It is accompanied by verifiable Evidence that enables reconstruction of which Evaluation Evidence supported the Decision and under which Governance Conditions that Decision was formed.