On the basis of a gust-load distribution, evaluated by applying level crossing-counting method, an 8-step program test and tests with a random sequence of stress cycles were performed. The results of these tests are compared with the results of tests with a random sequence of maxima and minima. The latter are based on a distribution of maxima and minima, which were derived from the gust-load distribution splitted in several stationary Gaussian processes. In further tests, ground-air-ground cycles were included in the study. Programmed and randomized flight-by-flight tests were compared.

The results of the study revealed that in the case of steady mean stress and the applied conditions program loading overestimates life by a factor of about 6. In case of flight-by-flight tests, however, the load-time history (either programmed or randomized) seem to be unimportant.