The Risks of Extrapolations of Metal Fatigue Data

    Volume 16, Issue 3 (May 1988)

    ISSN: 0090-3973

    CODEN: JTEOAD

    Page Count: 4


    Fuchs, HO
    Professor Emeritus of Mechanical Engineering, and Professor of Statistics, Stanford University, Stanford, CA

    Johns, MV
    Professor Emeritus of Mechanical Engineering, and Professor of Statistics, Stanford University, Stanford, CA

    (Received 16 June 1986; accepted 28 September 1987)

    Abstract

    The risks associated with extrapolations are quantified by simulation for the case of metal fatigue data. Extrapolations to 1% failures are made from samples of 10 or 20 or 50 specimens for four assumed distributions of the failures. Two methods of extrapolation are investigated. Method A corresponds to straight line fitting on lognormal probability paper. Method B applies extreme value theory to the 40% smallest values in a sample and gives superior results for samples of 50 specimens. For the small samples usual in metal fatigue, Method A gives equally good (or poor) results. Results are presented in several measures. An empirical risk divisor is introduced as a practical means of reducing the risk.


    Paper ID: JTE10378J

    DOI: 10.1520/JTE10378J

    ASTM International is a member of CrossRef.

    Author
    Title The Risks of Extrapolations of Metal Fatigue Data
    Symposium , 0000-00-00
    Committee G03