Geometric Tail Approximation for Reliability and Survivability
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
A common problem in developing high-reliability systems is estimating the reliability for a population of components that cannot be 100% tested. The radiation survivability of a population of components is often estimated by testing a very small sample to some multiple of the required specification level, known as an overtest. Given a successful test with a sufficient overtest margin, the population of components is assumed to have the required survivability or radiation reliability. However, no mathematical justification for such claims has been crafted without making aggressive assumptions regarding the statistics of the unknown distribution. Here we illustrate a new approach that leverages geometric bounding arguments founded on relatively modest distribution assumptions to produce conservative estimates of component reliability.
- Research Organization:
- Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA), Office of Defense Science (NA-113)
- DOE Contract Number:
- NA0003525
- OSTI ID:
- 1825217
- Report Number(s):
- SAND2021-12569; 700773
- Country of Publication:
- United States
- Language:
- English
Similar Records
APPROXIMATE CONFIDENCE LIMITS FOR THE RELIABILITY OF SERIES AND PARALLEL SYSTEMS
Poor survival of treatment-related acute nonlymphocytic leukemia