SMALL SAMPLE ACCEPTANCE SAMPLING TABLES USING MEAN LIFE CRITERIA BASED ON THE WEIBULL DISTRIBUTION (INCLUDING THE EXPONENTIAL CASE),

Abstract

Acceptance criteria are established for mean life requirements, assuming that a Weibull failure time distribution, with shape parameter 0.6 = or < beta = or < 3.6, applies. This includes the exponential failure distribution as a special case (beta = 1), and when beta = 3.4, a close approximation to the normal distribution. Mean life is defined as the average time to failure of similar items operated under similar conditions. Acceptance sampling plans based on the binomial probability distribution have been developed for the mean life criterion, using sample sizes of 2 = or < n = or < 50, and acceptance numbers of zero = or < C = or < (n-1), c = or < 8, to assure, with minimum confidence of 90% and/or 95%, the acceptance of lots for which the estimated mean life is equal to or greater than some specified operating time. In addition to their use as acceptance sampling criteria, mean life estimates would also be appropriate for establishing correction factors to compare basic environmental test results, with those obtained under other than basic conditions. (Author)

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 1969
Accession Number
AD0694471

Entities

People

  • Donald J. Fisk

Tags

DTIC Thesaurus Topics

  • Binomials
  • Environmental Tests
  • Mathematics
  • Normal Distribution
  • Probability
  • Probability Distributions
  • Random Variables
  • Sampling

Fields of Study

  • Mathematics

Readers

  • Regression Analysis.
  • Software Engineering