On the fatigue reliability evaluation of structural components under service loading

The fatigue strength evaluation of structural components under service loading is affected by several uncertainties. Proper statistical tools should be used to manage the large amount of causalities and the lack of knowledge on the actual strength-affecting parameters. This paper deals mainly with the statistical analysis to be performed on experimental load histories in order to evaluate fatigue loading spectra and the problems in extrapolating from measured load history to the whole component life. Then, a methodology for assessing the reliability of structural details under random fatigue loading is analysed. This methodology combines (under a linear cumulative rule assumption) the loading spectra, obtained by the new approach, and the constant-amplitude fatigue strength modelled by means of a conventional log-normal distribution. By so doing it is possible to distinguish many causes of uncertainty and suggest evaluations with proper confidence limits. The theoretical aspects developed are applied to mountain bike stems and they show the presence of a large uncertainty margin in reliability evaluation under service loading when there are no considerable quantities of experimental data available (both in load histories and component fatigue strength).