Adaptive Interval Outcome Validation Rate (AIOVR)

GMI-MAR-022Maritime IndustriesPilot-Ready

Definition

Percentage of AI-extended intervals where equipment condition at maintenance validates the extension.

The Measurement Gap

What This Metric Measures

AIOVR measures the percentage of AI-extended maintenance intervals where the equipment condition at the time maintenance was eventually performed validates the extension. Validation means the equipment was in acceptable condition and no accelerated degradation occurred during the extended period. It separates extensions that were genuinely justified by outcome from extensions where the organization got lucky.

The Decision It Supports

Are AI-extended intervals actually producing better outcomes (same equipment condition with less maintenance) or are they producing degraded equipment that just has not failed yet? Each extension is justified by condition data at the time of the decision. AIOVR asks the follow-up question: when the maintenance was finally performed, was the equipment actually in acceptable condition, or did the extension push the equipment into degradation that the condition data did not detect?

Why Legacy Metrics Cannot See This

PMS completion rate measures whether tasks are done on time. If the AI extends an interval and the task is completed at the new due date, PMS completion shows 100%. The metric never asks whether the equipment condition at the extended interval was acceptable. An AI system can extend every interval in the fleet, and PMS metrics will report perfect compliance as long as work is done on the new schedule. AIOVR closes this gap by measuring whether the extension decision was validated by the equipment's actual condition when maintenance was performed.

Formula

Primary Formula

AIOVR = Validated_extensions / Total_extensions

Variable Definitions

  • Validated_extensions = interval extensions where equipment condition at the time of maintenance was within acceptable limits, meaning no unexpected wear patterns and no signs of accelerated degradation noted by the maintaining engineer (count)
  • Total_extensions = all AI-recommended interval extensions in the measurement period (count)
  • Acceptable condition = component within manufacturer's replacement criteria, no unexpected wear patterns, no accelerated degradation noted by the maintaining engineer

Units: Percentage. Higher is better. 100% means every extension was justified by outcome. Values below 80% mean one in five extensions found worse-than-expected equipment.

Edge Cases

Equipment that failed before the extended interval expired should be counted as not validated. This is the most serious failure of the extension decision. Extensions within the manufacturer's allowable range versus extensions beyond it should be tracked separately, as within-range extensions carry lower risk. Minimum sample of 50 extended intervals per quarter fleet-wide is required. Segment by extension magnitude: small extensions of 10-20% beyond original will likely validate at higher rates than large extensions of 50% or more.

Proposed formula, pending validation pilots.

At a Glance

Unit
Percentage
Trend Direction
Higher
Reporting Frequency
Rolling quarterly
Category
Adaptive Scheduling Integrity
PMS Interval Stability Index (PISI) →

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