Issues in structural fatigue design and evaluation

Current practices in the design and evaluation of structural components for fatigue are generally semi-empirical. This is because either it is not practical or economical to realistically simulate field conditions in the laboratory testing or the actual conditions are not known precisely. Fatigue life curves based on the constant amplitude cyclic loading may result in unconservative design and evaluations where real load cycles are variable and the loads are of transient nature. There is a need to apply probability-based principles to the fatigue considerations, particularly in the areas of developing representative fatigue loading, and fatigue crack detection. In the area of power transmission lines, there is a need for better understanding of the failure mechanism and rationalize the fatigue design provisions. The state of stress within a stress concentration region depends only on the local boundary geometry, the magnitude of the resultant force, and the material type. This equivalence of stress concentration regions can be exploited to develop design aids for stress concentration factors independent of the global conditions. This paper identifies the above issues in detail and establishes some specific areas where there is a need for improvement in the current practice. Key words: fatigue, structural design, structural evaluation, bridges, power transmission lines, offshore structures.