A new publication from the SUBLIME project examines how monitoring and inspection data can improve the assessment of fatigue-critical details in ageing steel bridges. Fatigue cracks can develop gradually under repeated traffic loads. Reliable information about these cracks is therefore essential for deciding whether a bridge can continue to be used safely and when maintenance is needed.
Elena Zancato from Eindhoven Technical University developed a reliability-based framework that combines a predictive fracture-mechanics model with three types of information: destructive material tests, sensor measurements and non-destructive crack inspections. The framework was demonstrated using a welded detail in a steel girder.
For this case study, material testing alone had only a limited effect on the assessment. Sensor measurements and crack inspections provided much more useful information about structural reliability. Their value also depended on the stage of fatigue damage and the type of inspection used. Inspections that measure crack depth were particularly informative at earlier and intermediate stages, while other inspection data remained useful later in the structure’s life. At high numbers of load cycles, sensor data that reduced uncertainty in the structural model became especially valuable. Overall, combining sensor measurements with inspection results produced the strongest improvement in the reliability assessment.
In practice, the framework can help bridge owners compare possible monitoring strategies before investing in equipment or inspections. It can show which data source is likely to provide the most useful information, whether several methods should be combined, and when an inspection would add the greatest value. This supports better-informed decisions about follow-up inspections, repairs, use restrictions and maintenance planning. It may also help avoid collecting overlapping data that add little new information.
The study contributes directly to SUBLIME’s objective of integrating structural health monitoring with predictive models. More accurate estimates of reliability and remaining safe life can help asset owners use steel infrastructure safely for longer, while avoiding unnecessary interventions, reducing costs and limiting material use and CO₂ emissions. Although the numerical results relate to one representative bridge detail, the framework can be adapted to other structures and information sources.
The article by Elena Zancato, Davide Leonetti, Andrés Martinez Colán and Johan Maljaars is available here: Impact of Monitoring Strategies on the Reliability of a Fatigue-Critical Detail in Steel Bridges.