Weibull or Lognormal Distribution to Characterize Fatigue Life Scatter – Which is More Suitable? – Continued

Abstract This paper describes a study to determine whether fatigue test life scatter is best characterised by a Weibull or lognormal statistical distribution for a high strength steel used for landing gear structures. It is a response to “Face 2” of the ICAF 2017 Plantema Memorial Lecture and 2019 follow-up paper with the question; “Weibull or Lognormal Distributions to Characterize Fatigue Life Scatter?” These concluded that a Weibull distribution appears to be more suitable than a lognormal distribution for statistical modelling of fatigue life scatter to define an allowable service life at a specific probability of failure. Those studies used a homogenous dataset of 18 fatigue tests, and a non-homogenous dataset of 86 fatigue tests from a variety of sources. This paper reviewed HBK historical fatigue tests to identify a homogeneous dataset of 371 fatigue tests for a high strength steel used for landing gear structures. Weibull and lognormal statistical modelling of this dataset concluded that its fatigue life scatter is best characterized by the lognormal distribution.