Definition
CIPA measures whether the AI prioritizer's clinical impact score matches the actual clinical context at the time of dispatch.
The Measurement Gap
The Decision This KPI Enables:
When an AI system prioritizes a maintenance work order based on clinical impact, does the priority score reflect the actual clinical situation in the affected space at the time of the request, or does it reflect a static classification assigned when the space was commissioned?
Why Legacy Metrics Miss This:
Traditional work order prioritization metrics track response time compliance and backlog aging: whether work gets done within target windows. Priority assignment itself is treated as a solved problem: critical areas get high priority, non-critical areas get standard priority, and the algorithm sorts accordingly. AI-driven prioritizers add sophistication by weighting equipment criticality, failure history, and area classification, but the area classification component remains static.
The measurement gap appears when clinical reality diverges from the classification map. During a capacity surge, a hospital may overflow ICU patients into post-anesthesia care units, med-surg rooms, or procedure rooms not classified as critical care spaces. Equipment failures in those overflow areas (nurse call systems, monitoring infrastructure, medical gas outlets) carry ICU-level clinical impact but receive standard-area priority scores. Conversely, an OR suite blocked for construction receives critical-area priority for routine maintenance that could wait.
CIPA measures how often the AI prioritizer's clinical impact score matches the actual clinical situation, validated against real-time operational data: patient census, acuity level, active procedures, and overflow status. Without this validation, the prioritizer is optimizing work order sequence against a building map rather than against patient care delivery, and the organization cannot distinguish between a well-prioritized queue and one that systematically underweights dynamic clinical risk.
Formula
CIPA = (WO_accurate / WO_evaluated) × 100
Where:
- WO_accurate = Count of work orders where the AI-assigned clinical impact priority matches the validated clinical situation at time of request
- WO_evaluated = Total count of work orders evaluated in the measurement period where clinical context data was available for validation
Unit of Measure: Percentage (0-100%)
Measurement Frequency:
Monthly, with weekly monitoring during known surge periods (flu season, post-holiday surgical backlogs, disaster response)
Data Requirements:
- CMMS/EAM work order records with AI-assigned priority scores and clinical impact component
- Real-time or near-real-time clinical context: patient census by location, acuity classification, active procedure status, unit overflow/diversion status
- ADT (Admission-Discharge-Transfer) system feeds or bed management system data
- Space classification master data with room function, criticality tier, and any override history
Calculation Notes: Validation requires matching work order timestamps to clinical context at the time of request, not at the time of completion. Clinical context data latency greater than 30 minutes degrades validation accuracy. Organizations should exclude work orders for spaces with no clinical activity (mechanical rooms, storage, administrative areas) from the evaluation denominator to focus measurement on the clinical impact assessment specifically.
Proposed formula, pending validation pilots.