Glossary · Reliability, calibration, and statistics
Rule of three
After n independent trials with zero failures, a rough 95% upper bound on the failure probability is 3/n. Zero failures is not zero risk.
Defined in
Handbook · ValidationRare-event assuranceZero failures in n trials is not zero risk. Bound it, then prevent, search, oversample and monitor.Where it comes up
- Falsification and failure regionsHandbook · Validity under pressure
- Guarantee before you estimateHandbook · Practice
- Notation and formula cardHandbook · Reference
- What zero failures buys youCalculator