How FlightReady Calibrates Risk Scores
The statistical methodology behind FlightReady's risk score — NTSB training data, factor weighting, and validation against held-out accident records.
FlightReady's composite risk score is a weighted blend of operational factors — weather, terrain, airport complexity, NOTAM hazards, pilot profile, aircraft familiarity, traffic exposure, and external pressure — normalized to a 0–100 scale. Each factor produces an individual sub-score with documented sources so pilots can see what drove the number, not just the number itself.
Calibration begins with historical accident geography. NTSB reports are geocoded and compared to route corridors to estimate relative hazard density. This does not predict individual accidents; it highlights corridors where past outcomes cluster under similar operational contexts. Weather and NOTAM inputs reflect your last analyze session (not live pulls); accident and airport factors draw on ingested corpora with explicit freshness metadata.
Weights are tuned so that no single missing data source silently zeroes out the assessment. When a factor is unavailable — for example, NOTAMs without a configured provider — its weight is redistributed across available factors and the confidence score is reduced. That prevents a falsely low reading when the system simply lacks data.
Validation uses held-out accident samples and synthetic route scenarios to check monotonicity: as objective hazard signals worsen, the composite score should not decrease. Operational recommendations (GO, REVIEW, MITIGATE, DELAY, NO_GO) map to score bands aligned with FRAT conventions, but they remain advisory. The pilot in command always retains authority; the score is decision support, not a substitute for regulations or personal minimums.
Put it to work