Tools › Statistics › Distributions › Distribution Fitting › Example

MIL-HDBK-17-1F Section 8.3.7 Problem 3: lognormal B-basis value of 30 strengths

ToolDistribution Fitting
SourceU.S. Department of Defense (2002). Composite Materials Handbook, Volume 1: Polymer Matrix Composites Guidelines for Characterization of Structural Materials. MIL-HDBK-17-1F, 17 June 2002. Section 8.3.7.3, Problem 3.
LevelProfessional

Problem

Thirty strengths (85.39, 97.12, 92.66, 96.43, 90.72, 95.84, 97.30, 109.47, 101.35, 98.01, 86.18, 100.91, 96.05, 92.20, 90.86, 101.27, 101.23, 93.15, 114.32, 100.14, 91.24, 86.11, 93.42, 92.65, 97.58, 97.75, 97.95, 112.49, 95.75, 110.53) fail the Weibull and normal goodness-of-fit tests. Fit the lognormal distribution, test it, and compute the B-basis value.

Inputs to Enter

  • Paste the 30 values of Problem 3
  • Confidence 95 %, coverage 90 %
Open the Distribution Fitting tool

Answers

  • Weibull OSL 0.001 and normal OSL 0.042 (both rejected); lognormal AD = 0.597, OSL = 0.098 (accepted)
  • Mean of ln x = 4.57, s_L = 0.0734 (the handbook prints 1.6050, a misprint: its own B needs 0.0734)
  • k_B(30) = 1.778; B-basis = exp(x_L − k_B s_L) = 85.09
  • The handbook's selection rule (Weibull → normal → lognormal) picks the lognormal result

Notes

Replace the default sample; the page prints the OSL of each fit and the selected basis value. Not the default case.

More Examples for This Tool