CONFIDENCE LIMITS FOR SYSTEM RELIABILITY BASED ON COMPONENT TEST DATA,

Abstract

System confidence limits and their associated confidence level are obtained by combining the probability density functions of the estimator for the variable parameters of each component reliability function to get the probability density function of the estimators for the variable parameters in the system reliability function. Series systems composed solely of components having estimators with similar mathematical forms can be solved analytically. When the system reliability expression consists of a combination of dissimilar functions, Monte Carlo mathematical simulation must be used to obtain an approximation of the distribution of the system reliability estimator. The procedure consists of sampling the estimator's distributions for dissimilar probability distributions to obtain a sample point in the sample space of the estimator distribution. This procedure is repeated until the system probability distribution is adequately simulated. The accuracy of this method is on the order of 1 over the square root of N where N is the number of sample points obtained. One thousand and sixty sample points will give a 99% confidence that the simulated distribution does not differ from the analytical distribution by more than .05. (Author)

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 1963
Accession Number
AD0425849

Entities

People

  • Oscar Albert Bernhoff

Organizations

  • Air Force Institute of Technology

Tags

DTIC Thesaurus Topics

  • Accuracy
  • Confidence Limits
  • Data Science
  • Estimators
  • Information Science
  • Probability
  • Probability Density Functions
  • Probability Distributions
  • Reliability
  • Square Roots

Fields of Study

  • Mathematics

Readers

  • Computational Modeling and Simulation
  • Regression Analysis.

Technology Areas

  • Space