Abstract
Abstract
This paper outlines a new technique to address the paucity of data in determining a generalized S-N surface on the basis of fatigue reliability design. A new single-point likelihood method (SPLM) is presented for estimating a P-S-N curve, dealing with small sample numbers of data. A confidence level formula for P-S-N curve estimated by SPLM is also given. The concepts are then applied for the determination of a generalized S-N surface. Finally, contrast tests for the P-S curve are conducted to examine the presented method, demonstrating the practical and economic use of the proposed technique. It is shown from the results that the P-S-N curve can be obtained realistically using small sample numbers and that the required number of test specimens in this case is less than those for the conventional method.
Get full access to this article
View all access options for this article.
