Outcome
This project developed novel methods for decisions on the life-cycle reliability of flood defence systems. By optimising flood defence reinforcements at a system level, the cost of reinforcement projects can be reduced significantly. Uncertainty reduction through monitoring and proof load testing lead to lower reinforcement and risk costs, both in short and long-term. It has also been demonstrated that imperfect inspections and maintenance of flood defences lead to a failure probability increase. This contribution demonstrates the importance of including inspection and maintenance in flood risk assessments.
Figure 1. Example of the activities to ensure life-cycle performance: 1) and 4) reinforcement, 2) and 3) inspection (Photos accordingly by Pascal Ogink, Wouter Jan Klerk, Mark van der Krogt and the Flood Protection Programme).
Motivation and practical challenge
The asset management of flood defences in the Netherlands (figure 1) has been built upon centuries of experience. We have taken great leaps in quantifying the performance of these structures based on failure probabilities and increased our understanding of many potential failure modes. However, as a researcher and advisor on flood risk asset management at Deltares, I saw some missing links in the translation of this knowledge to decisions. In this project, I considered three key topics that I’m convinced would help us to take the next step in flood defence asset management. Optimising flood defence reinforcement design at a system level can lead to more effective and efficient reinforcement projects. Countering large reducible uncertainties in dike strength and pore pressures by monitoring and proof load testing can lead to a more cost-effective dike design. And quantifying the accuracy of inspections and accounting for this in flood defence reliability estimates will greatly improve reliability estimates and the decisions based on them.
Research challenge
Therefore, the challenge is to develop methods for addressing the missing connections between dike reinforcement, maintenance, monitoring, and inspection at different spatial and temporal scales. By doing so, how does, for example, the reinforcement of a diaphragm wall at one spot (Figure 1, bottom-left) help the safety targets for a larger dike section in the coming decades? How does inspection help in maintaining sufficient reliability?
Figure 2. Components of the research relating decisions at a dike section to a whole segment. (sources: figure based on schemes of Wouter Jan Klerk, including dike segment scheme prepared by Richard Marijnissen).
Innovative components
This research uses smart optimisation techniques to relate measures at different spatial scales. For instance, such a technique was applied to a dike reinforcement project to derive optimal planning of different measures (figure 2.1 and 2.2). By using this technique, the system reliability requirement, incorporating all possible strengthening measures, was met optimally when looking at costs.
I use decision trees and Bayesian decision analysis (figure 2.3) to translate monitoring outcomes into uncertainty reduction in dike reliability
estimates. However, the accuracy of the inspections (i.e. the probability of detection) is unclear and to determine it, a field experiment was conducted (figure 2.4).
The insights have been used to determine the impact of damage and imperfect inspections on failure probabilities. To achieve this, degradation rates were based on data analysis of past inspection reports, and a Dynamic Bayesian Network was used to determine the life-cycle cost of different inspection and maintenance strategies.
Relevant for whom and where?
Within the Netherlands, this research is of relevance to the regional water authorities, the Dutch Flood Protection program and the Ministry of Water and Infrastructure. In an international context, anyone with a keen interest in risk/performance-based asset management of flood defences can use it in defining projects and optimising inspections and maintenance.
The research includes key locations for the reinforcement of a dike section and the field experiment to assess the quality of inspections.
Progress and practical application
It is demonstrated that optimisation of flood defence reinforcements at a system level can reduce reinforcement costs by about 40%. For studies on the effectiveness of proof loading and dike monitoring, cost savings were in the range of ~25% due to the achieved uncertainty reduction leading to more efficient designs. However, not only investment costs but also risk costs can be avoided. For example, when accounting for damage to grass revetments, the estimated failure probabilities differ several orders of magnitude from the estimates from the safety assessment. By including this, effective investments in, for instance, in more frequent inspections can be properly valued, leading to more effective and efficient asset management. Further improvements can be achieved by improving the collection of inspection data to more accurately estimate degradation rates and further investigating the impact of damage to, for instance, revetments on their failure probability. For details about findings, see the related outputs.
Recommendations All-Risk
- Take a system perspective towards flood defence reinforcement projects to achieve more cost-efficient and transparent reinforcement decisions.
- Ensure that reduction of uncertainty is considered properly within and outside the context of dike reinforcements, and ensure that funding arrangements facilitate this.
- Consider uncertainty reduction as effective starting point for longterm adaptation strategies of flood defence systems.
- Improve the collection of inspection data to better understand degradation behavior of flood defences.
- Aim for continuous and targeted improvement of visual inspection of flood defences.
Last modified: 27/02/2022
Main researcher
Dr. Wouter Jan Klerk
Contributing partners
Proof loading and monitoring to optimize flood defence asset management
We calculate the expected value of extra information obtained from proof loading and monitoring. The additional information improves failure probability estimates for slope stability.
Mark van der Krogt
Delft University of Technology
Dr. Wouter Jan Klerk
View storyline Project outputs
Accuracy of visual inspection of flood defences
Insights into whether and how we can improve visual inspections, and starting points where other techniques such as drones can be of added value.
17/11/2021
Bevat: Publication open access journal
Optimal planning of flood defence system reinforcements using a greedy search algorithm
We develop and validate greedy search algorithm that can find (near-)optimal combinations of reinforcement measures for dike segments. The approach was applied to an ongoing reinforcement project, and it was found to reduce investment costs by ~40% compared to the conventional approach.
01/03/2021
Bevat: Publication open access journal
A framework for assessing information quality in asset management of flood defences
Methodology for scoring quality and use of available information for decision-making and assessing of the value of additional information.
28/10/2018 by Dr. Wouter Jan Klerk et al.
Bevat: Conference proceedings
Risk based inspection of flood defence dams: An application to grass revetments
Method to derive risk-based inspection intervals using visual inspection data and a combined degradation and failure model.
28/10/2018 by Dr. Wouter Jan Klerk et al.
Bevat: Conference proceedings
Value of Information of monitoring in the management of flood defences
Analysis of the Value of Information for different monitoring strategies combined with dike reinforcements to identify the conditions under which monitoring pore pressure is most beneficial for long-term asset management.
30/08/2019 by Dr. Wouter Jan Klerk et al.
Bevat: Publication open access journal
Time-dependent reliability in flood protection decision making in the Netherlands
Exploration of the failure probability definition and the influence of including temporal correlation on failure probability estimates.
17/06/2018 by Dr. Wouter Jan Klerk et al.
Bevat: Conference proceedings
Influence of monitoring on investment planning of flood defence systems
Exploring the value of monitoring for the reinforcement of a dike segment.
11/12/2019 by Dr. Wouter Jan Klerk et al.
Bevat: Conference proceedings
Also applicable to this project
Value of information (VoI) of combinations of proof loading and pore pressure monitoring for flood defences
We calculate the expected value of extra information obtained from proof loading and monitoring. The additional information improves failure probability estimates for slope stability.
18/12/2020
ID)->name?>
Contains: Publication open access journal
Check the ResearchGate profile for all other research publications.
Blogs
Bringing All-risk to practice: Martin’s recommendations
08/10/2020 by Wim Kanning
This blog is based on an informal interview between Martin Schepers (manager flood safety projects) and Wim Kanning (All-Risk researcher) on the development of...
Events
04/10/2021
Reflection: Risk-based inspection and interactions between failure mechanisms
Safety standards of flood defences are based on probabilities of flooding. Using a variety of failure models and statistical methods, we try to assess these probabilities as good as possible, in order to take effective measures for flood risk reduction. The question is how well we actually have insight into the flooding probability, and whether we take all relevant factors into account.
10/12/2020
NCR Lecture: A system perspective on flood defence reinforcement projects
Wouter Jan Klerk talked on the NCR December's lecture about the Dutch flood probability standards, and how he works to optimise the way we reach our risk-based target levels. In the dike reinforcement Streefkerk-Ameide-Fort Everdingen a method was developed and applied to optimize flood defence investments at a system level.
