The potential of using performance information in the assessment of existing quay walls

Abstract

The design of new and assessment of existing quay walls is subject to large uncertainties. Dealing with these uncertainties is a crucial part of the engineering process. The way uncertainties are addressed has a large impact on construction and maintenance costs and on the reliability ultimately obtained. Especially in the assessment of existing structures the uncertainties can be large. An existing structure allows us to use actual performance information in the assessment, such as the structural response to loading. One way to obtain the structural response is test loading assisted by monitoring. In this research Bayesian updating is used to reduce uncertainties and to more effectively use the obtained measurement data. We present a case study of an existing quay structure along with fictitious measurement data to demonstrate the potential effects of test loading on the reliability of the structure. The results show that Bayesian updating successfully reduces the uncertainty (i.e. standard deviation) of the model prediction. Using monitoring data and Bayesian updating provides a more realistic model of the capacity of the existing quay structure and thus a more accurate reliability assessment. Which may lead to extension of the structure's lifetime or that higher loads can be accepted.