Deferral Portfolio Risk Accumulation Index (DPRAI)

GMI-TEL-047TelecommunicationsIntegration-RequiredPattern O-10: Deferral Compounding

Definition

Aggregate estimated risk of all currently deferred preventive maintenance events, computed as the sum of individual deferral risk scores weighted by asset criticality and deferral age, compared to the organization's defined acceptable deferral risk threshold. DPRAI is expressed as a ratio relative to the threshold: below 1.0 means within acceptable risk, above 1.0 means the deferral portfolio has exceeded the organization's risk tolerance.

The Measurement Gap

AI-driven condition-based PM scheduling defers preventive maintenance when sensor data indicates the equipment does not yet need servicing. Each individual deferral is justified by condition data: the equipment looks fine, so the PM is pushed out. But the portfolio of deferred PMs across the entire infrastructure accumulates risk that no individual deferral justification captures. Twenty individually justified deferrals on twenty high-criticality assets may collectively represent a catastrophic exposure that a single severe weather event or load increase could convert into a cascade of failures. Without a portfolio-level risk accumulation metric, the optimizer generates deferrals that are individually rational but collectively dangerous.

The Decision

Is the portfolio of condition-justified PM deferrals collectively safe, or has it crossed into a risk zone where the next severe weather event, equipment failure, or load increase could trigger a cascade of failures across the deferred population? If DPRAI exceeds 1.0, the deferral portfolio should be reduced by executing the highest-risk deferred PMs immediately, starting with the highest-criticality assets.

Research Foundation

Condition-based PM deferral programs in infrastructure sectors accumulate portfolio risk at rates that typically exceed linear extrapolation because of correlated failure modes (equipment classes that can be deferred simultaneously tend to share environmental stressors, creating correlated failure risk that compounds nonlinearly (Reliability Engineering & System Safety) Portfolio Risk in Condition-Based Maintenance, 2024). Analysis of telecom site failures following Hurricane Ian found that 34% of failed sites had three or more deferred PMs at the time of the event, compared to 8% of sites that maintained service (CTIA Disaster Recovery Lessons Learned, 2023).

Formula

DPRAI = Σ (P_fail_i × C_fail_i × T_escalation_i) / Risk_threshold

Variable Definitions (every variable resolved)

  • P_fail_i = estimated probability of failure before next maintenance opportunity for deferred PM i (0 to 1, derived from condition data and degradation models)
  • C_fail_i = consequence of failure for asset i (criticality score or dollar impact)
  • T_escalation_i = time-escalation factor = Days_since_deferral_i / Original_PM_interval_i: captures that a 30-day deferral on a quarterly PM (T = 0.33) is less concerning than a 30-day deferral on a monthly PM (T = 1.0)
  • Risk_threshold = organization-defined acceptable total deferral risk (set during pilot calibration)

Units:

Dimensionless ratio relative to threshold. DPRAI below 1.0 = within acceptable risk. Above 1.0 = the deferral portfolio has exceeded the acceptable risk threshold.

Edge Cases

Deferred PMs where condition data shows improvement (transient temperature exceedance that resolved): set P_fail to near-zero but keep in the portfolio count for visibility. Multiple deferred PMs on the same asset: sum their contributions. Correlated risk on a single asset is more dangerous than distributed risk across the portfolio. No sampling: compute across all active deferrals; this is a risk accumulation metric.

These are proposed starting points: calibrate to your operational context during pilot: DPRAI below 0.7 is comfortable. Between 0.7 and 1.0 warrants heightened monitoring. Above 1.0 means the deferral portfolio should be reduced by executing the top 20% of deferred PMs by individual risk contribution. Track DPRAI over time, a steadily rising trend, even if below 1.0, means the optimizer is accumulating deferrals faster than they are being resolved.

Proposed formula, pending validation pilots.

At a Glance

Unit
Ratio
Trend Direction
Lower; Below 1.0
Reporting Frequency
Monthly
Category
Maintenance Deferral and Scheduling Quality
← Resource Borrowing Cascade Depth (RBCD)Schedule Condition Quality Correlation (SCQC) →

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