On evaluating Weibull fits to mechanical testing data

The use of coefficient of determination, R 2, and the Anderson–Darling ( A 2) hypothesis test to evaluate the goodness-of-fit to the two-parameter Weibull distribution was investigated. Results of Monte Carlo simulations for sample sizes between 5 and 100 indicated that guidelines provided previousl...

Full description

Saved in:
Bibliographic Details
Published in:Materials science & engineering. A, Structural materials : properties, microstructure and processing Structural materials : properties, microstructure and processing, 2009-12, Vol.527 (1), p.397-399
Main Authors: Tiryakioğlu, Murat, Hudak, David, Ökten, Giray
Format: Article
Language:English
Subjects:
Applied sciences
Castings
Exact sciences and technology
Extreme value
Foundry engineering
Metals. Metallurgy
Other casting methods. Solidification
Production techniques
Reliability
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The use of coefficient of determination, R 2, and the Anderson–Darling ( A 2) hypothesis test to evaluate the goodness-of-fit to the two-parameter Weibull distribution was investigated. Results of Monte Carlo simulations for sample sizes between 5 and 100 indicated that guidelines provided previously in the literature are too conservative for sample sizes up to 80. New guidelines for the use of R 2 and A 2 for sample sizes between 5 and 100 have been developed. The two measures of goodness-of-fit were found to agree more than 95% of the time, regardless of sample size. The use of the new guidelines has been demonstrated on two datasets from the casting literature.
ISSN:0921-5093
1873-4936
DOI:10.1016/j.msea.2009.08.014