Criticality Governance Rate (CGR)

GMI-ISO-003ISO Management SystemsInfrastructure-Dependent

Definition

Percentage of AI-driven asset criticality reclassifications processed through the defined risk governance workflow before taking operational effect.

The Measurement Gap

ISO 55001 Clause 6.1 requires organizations to identify risks and opportunities that could affect the achievement of asset management objectives. Clause 6.2 requires that objectives account for risk assessment outcomes. Clause 8.2 requires a management of change process when changes affect asset management activities. When an AI risk scoring engine reclassifies an asset from High criticality to Medium, that single change cascades through the management system: maintenance strategy changes, inspection intervals shift, spare parts stocking levels adjust, and resource allocation patterns realign. The standard requires governance over changes of this significance. When a human analyst reclassifies an asset, the change typically passes through a formal review involving the risk owner, a risk register update, and a strategy impact assessment. When an AI model produces the same reclassification as part of a continuous scoring cycle, the governance workflow often does not fire because the change was not initiated through the human process pathway. Legacy metrics such as risk register currency and management review frequency count governance activities but cannot detect that a class of risk-significant changes is bypassing governance entirely. The consequence is that AI-driven criticality changes take operational effect without the risk owner review, risk register update, or strategy impact assessment that the management system requires. This is particularly acute for downward reclassifications (High to Medium), which reduce attention and resources without the scrutiny that upward reclassifications typically receive.

Decision Supported:

Do AI-driven criticality reclassifications go through the same risk governance workflow as human-initiated changes, or has AI created a bypass pathway for risk-significant decisions?

Formula

Plain Language: Count the number of AI-driven criticality reclassifications that completed the full risk governance workflow (risk owner review, risk register update, and strategy impact assessment) before any downstream operational changes took effect. Divide by the total number of AI-driven criticality reclassifications.

Notation:

CGR = (Governed_reclassifications / Total_AI_reclassifications) x 100

Variables:

Variable Definition Unit
Governed_reclassifications Number of AI-driven criticality reclassifications where the complete governance workflow was completed before downstream operational changes. The complete workflow requires: (1) risk owner review, (2) risk register update, and (3) strategy impact assessment. All three must be completed before the reclassification takes operational effect. Count
Total_AI_reclassifications Total number of asset criticality classification changes initiated or recommended by AI risk scoring engines during the reporting period. Excludes score changes within the same criticality band. Count

Sign Convention: CGR is a percentage from 0% to 100%. Higher values indicate stronger governance coverage over AI-driven criticality changes. A score of 100% means every AI reclassification completed the full governance workflow before taking operational effect.

Directional Analysis:

Track CGR separately for upward reclassifications (e.g., Medium to High) and downward reclassifications (e.g., High to Medium). Organizations typically find CGR near 100% for upward changes but near 0% for downward changes.

Proposed formula, pending validation pilots.

At a Glance

Unit
Percentage
Trend Direction
Higher; approaching 100%
Reporting Frequency
Quarterly
Category
Governance Continuity
← Competence Scope Adequacy (CSA)

Assess Your Organization

The AI-GPS Assessment evaluates your AI governance maturity and recommends KPIs like this one for your implementation roadmap.

Take the Free AI-GPS Assessment