Comments on the Tri-Service Bayesian Approach to Nuclear Weapons Reliability Research.

Abstract

The paper presents comments on the report, 'A Tri-Service Bayesian Approach to Nuclear Weapons Reliability Assessment'. The report proposes a procedure for obtaining point and interval estimates for system reliability using component and system test results. The procedure has a Bayesian foundation but the authors suggest that it can be regarded as simply an algorithm used to obtain a reliability assessment with classical properties. The procedure is appealing in that it is a definable procedure that gives point and interval estimates of system reliability based on component, subsystem, and system data, that can be implemented on available computing equipment, and that gives reasonably accurate results when compared to classical properties. The authors examination of the report has two directions: (1) a general criticism of the Monte Carlo study provided in the report along with a criticism of the procedure provided by our own Monte Carlo study.

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1974
Accession Number
AD0778675

Entities

People

  • Ruslell V. Lenth
  • William J. Zimmer

Organizations

  • University of New Mexico

Tags

Communities of Interest

  • Weapons Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Bayesian Networks
  • Intervals
  • Nuclear Weapons
  • Reliability
  • Weapons

Readers

  • Business Analytics
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms