Optimum bridge life-cycle management with updating based on inspected fatigue crack under uncertainty

Effective and efficient life-cycle management of fatigue-sensitive bridges should be based on reliable and accurate fatigue crack propagation prediction under uncertainty. In order to reduce the uncertainty and improve the accuracy associated with the fatigue crack propagation prediction, an appropriate updating process integrating the initial information on fatigue crack propagation and outcomes of inspection is necessary. Therefore, life-cycle management of fatigue-sensitive bridges has to include the updating process. This paper presents a probabilistic approach for optimum inspection and maintenance planning with updating based on the inspected fatigue crack size. The initial inspection time is optimized through the bi-objective optimization to minimize the expected damage detection delay and the expected total inspection cost. If a fatigue crack is detected at the initially scheduled inspection time, the existing fatigue crack propagation prediction is updated. After updating, the inspection planning for maintenance is determined. The presented approach is illustrated with a fatigue critical detail of a steel railway bridge.

Automated summary

This paper presents a probabilistic approach for optimum inspection and maintenance planning based on fatigue crack size, aiming to reduce uncertainty and improve accuracy in life-cycle management of fatigue-sensitive bridges. The updating process integrates initial information on fatigue crack propagation and outcomes of inspection to optimize inspection time and maintenance planning. Illustration of the approach is demonstrated with a case study of a steel railway bridge's fatigue critical detail.