8.
Assessing Product Reliability
8.2. Assumptions/Prerequisites 8.2.1. How do you choose an appropriate life distribution model?


A model derived from random crack growth occurring during many independent cycles of stress  The derivation
of the Fatigue Life model is based on repeated cycles of stress causing
degradation leading to eventual failure. The typical example is crack growth.
One key assumption is that the amount of degradation during any cycle is
independent of the degradation in any other cycle, with the same random
distribution.
When this assumption matches well with a hypothesized physical model describing the degradation process, one would expect the BirnbaumSaunders model to be a reasonable distribution model candidate. (See the note in the derivation for comments about the difference between the lognormal model derivation and the Fatigue Life model assumptions. Also see the comment on Miner's Rule). 