Probabilistic optimum inspection planning of steel bridges with multiple fatigue sensitive details

Inspection planning of fatigue sensitive bridges is a complex process. This, in part, is due to various uncertainties associated with the damage propagation and the detection technique used to perform inspections. The planning process starts with identifying the most critical locations of fatigue damage occurrence. In this context, non-destructive testing (NDT) techniques, applied at different times along the service life, can be used to monitor the crack growth and support the fatigue life estimation. The next step is to establish an inspection plan which considers the available budget and the target performance level of the investigated bridge. This paper proposes a probabilistic approach for providing effective inspection plans for fatigue critical bridges. The proposed approach considers multiple fatigue critical locations in the bridge and chooses the best inspection option among a set of predefined types of non-destructive inspections for each detail. Uncertainties in the damage propagation are considered in addition to the uncertainties in the damage detection process. A multi-objective optimization problem, which accounts for various critical details in the bridge, is formulated to find the optimum inspection times and the required NDT technique for each inspection. An existing fatigue critical steel bridge is used to illustrate the proposed probabilistic approach.