Actionable Prognostic Horizon (APH)

GMI-MRO-002MROIntegration-Required

Definition

The usable lead time a predictive model actually provides: the raw prognostic horizon minus real-world procurement, scheduling, and execution time required to complete an intervention.

The Measurement Gap

Predictive models report a prognostic horizon as if it were operational lead time, but the academic NASA Prognostic Horizon metric isolates model foresight from organizational response. It was designed for research, where "response" means updating a prediction. In a plant, response means procuring parts, negotiating a production window, dispatching a crew, executing work, and returning the asset to service. None of those latencies appear in F1, AUC, MTTR, RUL accuracy, or PM Compliance. The failure mode invisible to legacy metrics is the accurate prediction that arrives inside the response time budget: the model saw it coming and the organization could not move fast enough. Consequence: organizations celebrate prognostic accuracy while reactive failures occur on flagged assets.

Decision Supported:

Given my procurement, scheduling, and execution latencies, did the model's foresight on a specific asset provide enough usable time for the organization to actually intervene?

Formula

Plain language: Start with the time between the model's first alert and its predicted failure date. Subtract procurement lead time, scheduling lead time, and execution time. What remains is the actionable horizon. A normalized ratio (actionable / raw) shows the fraction of the model's foresight that was operationally usable.

Notation:

```
APH = PH − (T_procurement + T_scheduling + T_execution)

APH_ratio = max(0, APH) / PH
```

Variables:

Variable Definition Unit
PH Prognostic Horizon: time between first alert and predicted failure date Days
T_procurement Time from need identification to part available in storeroom (administrative + vendor + transit + receiving) Days
T_scheduling Time from part available to confirmed schedule with assigned crew and production window Days
T_execution Time from crew start to asset returned to service (including commissioning/testing) Days
APH Actionable Prognostic Horizon: usable lead time after response budget Days
APH_ratio Fraction of prognostic horizon that was operationally usable Ratio

Proposed formula, pending validation pilots.

At a Glance

Unit
Time (days)
Trend Direction
Higher
Reporting Frequency
Quarterly
Category
Algorithmic Efficacy & Model Health
Alert Net Economic Yield (ANEY) →

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