Volume 16, Issue 6 (November 1988)
Statistical Design of Fatigue Crack Growth Test Programs
An experimental design method for determining statistically significant differences in fatigue crack growth rate data due to the influence of various test conditions (factors) is presented. This method is based on a completely randomized factorial design and a three-way classification Analysis of Variance (ANOVA) procedure. Although the example presented is for fatigue crack growth rate testing, this method is generic to any materials test program where comparisons of properties or behavior are made. The procedure allows decisions regarding these comparisons to be based on statistical tests of significance.