Published 14 October 2026 · Source-linked research prepared with AI assistance. Proposed services require an agreed scope and qualified delivery team.
The fixed pilot profile
The current workbench uses general-facility-pilot-0.1 with 30 control families. Each pillar has six controls. Total weight is 100. This is a method demonstration and a working evidence organizer, not a validated sector rating. Do not compare different sectors or change scope mid-assessment. The current interface does not permit exclusions; a qualified sector profile must resolve applicability before a customer score.
Maturity levels
| Level | Meaning | Evidence expectation |
|---|---|---|
| 0 | Known absent or failed | Reviewed evidence supports failure or absence. |
| 1 | Defined | Requirement, owner and plan exist. |
| 2 | Partly implemented | Some capability exists; material gaps remain. |
| 3 | Implemented | Current evidence supports implementation; full demonstration remains open. |
| 4 | Demonstrated | The declared criterion is met through the specified review or test. |
| Unknown | Unassessed | Evidence is absent, expired, contradictory or unsupported. |
A reviewed zero counts toward evidence coverage. It earns no readiness points. A blank answer does neither. For controls requiring a performance test, document review alone cannot support level 4.
Calculate the result
For every assessed control, multiply its weight by its maturity level divided by four. Add those earned points to obtain E. Add the weights of assessed controls to obtain W. Provisional assessed readiness is 100 × E ÷ W when W is greater than zero. Coverage is W percent. Retain full precision until display.
These are possible full-score limits if unknown controls later score 0 or 4. They are not confidence intervals and do not quantify error in reviewed evidence.
Apply the release gates
- At least 90% of all control weight has current usable evidence.
- At least 80% of each pillar’s weight has current usable evidence.
- Every designated critical control has a current assessment.
- A qualified reviewer approves scope, evidence and the resulting assessment.
The browser checks the first three conditions from entered fields. It cannot authenticate the reviewer or inspect evidence. Its result therefore remains provisional. A high average does not cancel critical gaps. Critical controls below demonstrated level remain visibly open, without a hidden numerical penalty.
Keep evidence and exposure separate
The evidence record stores the reference, criterion, reviewer, method, review date, expiry, action owner and next action. Expired or contradictory evidence loses assessed status. Current hazard observations, historical frequency and scenario estimates keep their own definitions. A warning does not mechanically lower readiness; a failed test can.
Validate before claiming predictive value
The weights and gates are product-policy hypotheses. Expert review, duplicate assessments, sensitivity testing and a bounded facility pilot should examine whether assessors agree and owners can act on the output. Later outcome work is needed before relating score changes to losses or service continuity. There are no percentiles, safe badges or insurer-approval claims in this release.
Sources & further reading
- GAO — Disaster Risk Improvements Needed to Enhance FEMA’s National Risk Index ↗10 September 2026. Recent findings on use guidance, sensitivity analysis, user feedback, and NRI migration to RAPT.
- NIST — Community Resilience Planning Guide A Playbook ↗Guide and current planning resources. Recovery targets tied to community functions and supporting infrastructure.
- Gerst Dillard and Loerzel via NIST — Methodological recommendations for content validation of community resilience indicators ↗2025. Research abstract supporting structured expert judgment alongside literature and empirical testing.
- FEMA — National Risk Index Technical Documentation v1.20 ↗December 2025. Technical sections on expected annual loss and percentile scores, reviewed through indexed extracts.