Abstract
Fatigue assessment of several steel bridge details, having equivalent stress range below the constant amplitude fatigue threshold (CAFT), showed that many of these details are free of cracks, although the remaining life calculations predicted that they should have already been suffering from cracks. This shows that the current methods for predicting the fatigue remaining life of steel bridges can be conservative and may lead to unnecessary retrofit and rehabilitation actions. For a better fatigue life prediction, a bilinear S-N approach has been proposed. In this approach, the bilinear S-N lines have different slopes above and below the CAFT. This paper addresses fatigue assessment and service life prediction of existing fatigue-prone steel bridges by integrating structural health monitoring (SHM) data into a probabilistic bilinear S-N approach. The effect of changing the value of the slope of the AASHTO S-N lines below the CAFT on the fatigue life is investigated through a parametric study. In addition, an existing steel bridge is used to illustrate the proposed probabilistic approach.