Benefits of decreasing the uncertainty in the current bridge inspection practice

Numerous bridges worldwide have surpassed their service-life. To ensure the user safety, visual inspections are commonly carried out with the consequent assignment of Defect Grades. On their basis, simplified risk evaluations and maintenance intervention prioritization are formulated. Per the inherent nature of visual inspections, these ones include two kinds of uncertainty: interpretation-related and representation-related one. If to attempt to reduce their influence in the inspective process, the former, being an intrinsic feature of inspections performed by humans, could not be tackled. The latter, instead, can be decreased on the basis of a novel semantics-based inspective methodology. On the grounds of a large set of real-life inspections outputs, the present article measures the inspection quality improvement following said uncertainty reduction. This was achieved through the Expected Utility Theory in terms of utility and costs. On the grounds of these two, a novel Uncertainty-induced Cost curve allows the assessment of the cost evolution as a function of the weight of the representation-related uncertainty. The proposed semantics-based inspective methodology represents an improvement over the current-day directly assigned condition grading one, thus improving the efficiency of structural reliability assessments. This leads to an improved prioritization of bridge maintenance interventions and to an increased user safety.