Permit Condition Inventory Completeness (PCIC)

GMI-WT-030Water TreatmentPilot-Ready

Definition

PCIC measures the percentage of actual enforceable permit conditions that are correctly captured in the utility's AI compliance management system, including numeric limits, monitoring frequencies, reporting deadlines, and narrative conditions.

The Measurement Gap

Measurement Gap and Decision Supported

The Invisible Failure Mode: A wastewater utility receives a permit modification that adds seasonal phosphorus monitoring and a narrative condition requiring development of a nutrient management plan within 180 days. The AI compliance system's automated permit parser correctly captures the new phosphorus monitoring limits but fails to capture the narrative condition: it does not appear in the system's condition inventory. Six months later, the state regulatory agency asks for the nutrient management plan status. The utility's compliance officer, who relied on the AI system for deadline tracking, is unaware the requirement exists. The narrative condition has already expired, creating an immediate violation without any compliance activity having occurred.

What Existing Metrics Miss:

Compliance management systems report condition counts and upcoming deadlines but do not verify that their condition inventory matches the actual permit. A system can show 100% compliance against its tracked conditions while missing entire permit sections. Without PCIC, the gap between what the permit requires and what the AI tracks is invisible.

The Decision PCIC Enables:

Is the AI compliance management system tracking every enforceable condition in the current permit, or are there blind spots. Missed conditions, outdated limits from superseded permit versions, or narrative conditions the parser could not interpret?

Cross-References:

PCIC feeds directly into DMRPS (GMI-WT-028). Missing permit conditions mean missing DMR fields, which means incomplete regulatory reports that DMRPS would flag if the fields were included in its check scope. PCIC also feeds into RECA (GMI-WT-029). If the permit adds new reporting triggers the AI does not know about, event classification accuracy degrades for those event types. Together, PCIC, DMRPS, and RECA form the Regulatory Compliance AI Integrity triad: PCIC ensures completeness, DMRPS ensures data quality, and RECA ensures correct classification. On the maintenance KPI side, PCIC parallels Regulatory Margin Index (RMI, GMI-WT-004): RMI tracks how close operations are to permit limits, but that tracking is only meaningful if PCIC confirms all limits are being monitored.

Formula

Plain language: Have a compliance officer manually extract all conditions from the current permit document. Compare against the AI system's condition inventory. Count matches, misses, and extras (AI tracking conditions that are not in the permit, perhaps from a superseded permit version).

Notation:
PCIC = Matched_conditions / Total_permit_conditions

Where:

  • Matched_conditions = permit conditions correctly represented in the AI's inventory (correct parameter, correct limit, correct monitoring frequency, correct reporting deadline)
  • Total_permit_conditions = all conditions in the current enforceable permit (including narrative conditions, seasonal limits, special conditions from modifications)
  • Track separately: Missed_conditions (in permit, not in AI) and Phantom_conditions (in AI, not in current permit)

Units: Percentage. 100% = perfect coverage. Missed conditions are worse than phantom conditions: phantom conditions cause unnecessary monitoring; missed conditions cause compliance failures.

Proposed formula, pending validation pilots.

At a Glance

Category
Regulatory Compliance AI Integrity
← Discharge Monitoring Report Plausibility Score (DMRPS)Regulatory Event Classification Accuracy (RECA-WT) →

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