Audit Clause Coverage Index (ACCI)

GMI-ISO-007ISO Management SystemsIntegration-Required

Definition

Distribution of internal audit evidence-gathering activity across ISO 55001 clauses over a rolling audit cycle, measuring whether AI-driven scoping provides proportional coverage or creates clause-level blind spots.

The Measurement Gap

ISO 55001 Clause 9.2 requires that organizations conduct internal audits at planned intervals to provide information on whether the management system conforms to its own requirements and to the standard. When AI-driven audit planning tools scope audit programs, they optimize for efficiency by concentrating audit effort on areas with the highest historical finding rates, the greatest risk scores, or the most recent changes. This optimization is rational from an efficiency standpoint, but it systematically underweights clauses that have low finding rates, stable configurations, or less quantifiable requirements. Over successive audit cycles, AI-driven scoping creates clause-level blind spots where entire sections of the standard receive minimal or no audit attention. Legacy audit metrics such as audit completion rates and audit plan adherence percentages measure whether audits happened on schedule and whether the plan was followed, but they do not measure whether the audit program provided proportional coverage across all clauses. An organization can complete 100% of planned audits on time while leaving three or four clauses effectively unaudited because the AI scoping tool consistently deemed them low-priority. The consequence is an assurance mechanism that becomes progressively less able to detect nonconformities in under-covered areas, creating blind spots that compound over time. When a finding eventually surfaces in an uncovered clause, it arrives as a surprise because the assurance system had been optimized to not look there.

Decision Supported:

Is AI audit scoping providing full management system coverage across all clauses, or is it concentrating evidence-gathering on high-finding areas while leaving other clauses as blind spots?

Formula

Plain Language: Count the number of auditable clauses that received at least the defined minimum audit coverage (at least one audit sample per clause per cycle). Divide by the total number of auditable clauses to produce a coverage index.

Notation:

ACCI = 1 - (Clauses_below_minimum / Total_clauses)

Variables:

Variable Definition Unit
Clauses_below_minimum Number of auditable clauses that received less than the defined minimum coverage during the audit cycle. Minimum coverage is defined as at least one audit evidence sample per clause per complete audit cycle. Count
Total_clauses Total number of auditable clauses in the applicable standard. For ISO 55001, this is 7 main clauses (Clauses 4 through 10) or 26 sub-clauses depending on the selected granularity. For ISO 9001, 14001, and 45001, the clause structure follows the same Annex SL numbering with standard-specific sub-clauses. The granularity level should be defined and documented before the first ACCI calculation. Count

Sign Convention: ACCI ranges from 0 to 1. A score of 1.0 means every auditable clause received at least minimum coverage during the audit cycle. A score of 0.70 means 30% of clauses fell below minimum coverage. Higher values indicate more complete clause-level audit coverage.

Granularity Note:

Organizations should select either main-clause or sub-clause granularity and apply it consistently across audit cycles. Sub-clause granularity (26 clauses for ISO 55001) provides a more sensitive measure but requires more detailed audit evidence tagging.

Proposed formula, pending validation pilots.

At a Glance

Unit
Index (0 to 1)
Trend Direction
Higher; approaching 1.0
Reporting Frequency
Per audit cycle (typically annually)
Category
Assurance Integrity
Reporting Representativeness Score (RRS) →

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