Extreme-Condition Model Coverage (ECMC)

GMI-AGR-008Agriculture and Farm OperationsInfrastructure-DependentPattern O-20: Transient Condition Blindness

Definition

The share of defined extreme or transient condition classes (frost after emergence, flood or saturated planting, heat dome at pollination, derecho or hail damage, early freeze at harvest) for which each agronomic model has validated output, weighted by the frequency of those conditions in the operation's climate.

The Measurement Gap

There is no metric that tells you what fraction of extreme or transient conditions (frost, flood, heat, saturated planting) the agronomic models produce validated output for, which means confident recommendations arrive precisely in the conditions where they were never tested.

Formula

ECMC = Σ_c f_c · v_c / Σ_c f_c, where v_c = 1 if the model has validation evidence for condition class c meeting a minimum sample, else 0

Proposed formula, pending validation pilots.

At a Glance

Unit
Ratio (0 to 1)
Trend Direction
Higher
Category
Agronomic Decision Validation

Decision Supported

Whether a model's recommendation can be followed during an extreme event or must be set aside for human judgment, decided before the event rather than during it.

← Crop Model Fidelity Deviation (CMFD)Optimized Plan Robustness Ratio (OPRR) →

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