The Maeslant barrier is a storm surge barrier and a critical component of the Dutch coastal flood defense system. Its reliability is formally assessed through a Reliability and Availability (RA) analysis, which estimates the probability of non-closure during storm events. However, concerns have been raised regarding the completeness and transparency of this analysis, particularly the potential omission of relevant failure events. This thesis investigates whether a selected set of previously unaccounted for events can be systematically identified and quantified to improve the accuracy of the non-closure probability.

A three-stage methodology was developed. First, a structured inventory of unaccounted-for events was constructed using HAZOP, FMEA, What-If, and external event screening techniques, mapped across four analytical dimensions. Second, the list was filtered based on estimated occurrence probability and quantifiability, resulting in a shortlist of three events: epistemically uncertain events, non-stationary component degradation, and the unverified reliability of human interventions. Third, these events were quantified using structured expert judgment, research into time-dependent fault tree modeling, and human reliability assessment.

Results indicate that these unaccounted-for events can alter the estimated non-closure probability, either increasing it by an order of magnitude or reducing it by up to 50%. Moreover, the analysis revealed limitations in the current RA analysis, including outdated reliability assumptions, a fragmented integration of human interventions, and a lack of empirical data. These findings support the need for a more transparent and adaptable RA framework. The discussion highlights that while completeness in risk assessment is theoretically unattainable, similar to the limitations of physical laws, models should strive for an optimal balance between complexity, traceability, and applicability.

Recommendations include developing a centralized component lifecycle database, maintaining a registry of previously unaccounted-for events, formally integrating the OPSCHEP model into the fault tree structure, and adopting structured human reliability verification. These changes can improve the accuracy, transparency, and credibility of the Maeslant barrier’s non-closure probability and serve as a blueprint for other critical infrastructure systems. ... Read more

The Ramspol Barrier, an inflatable storm surge barrier, is an integral component of the flood protection system for the Zwarte Meer and its hinterland. However, its current operational procedure faces several challenges that compromise flood safety, cause disruptions for the shipping industry, and significantly burden the operation team. The current operational procedure activates specific steps based on predefined water level thresholds, often mobilizing the operation team and disrupting shipping without ultimately leading to a barrier closure. It is also questioned whether closures are always the best strategy for optimizing flood safety in the system.

This thesis aims to improve the operational procedure of the Ramspol Barrier by balancing flood safety, operational burden, and disruptions to vessel navigation.

A comprehensive system and data analysis were conducted to understand the dynamics of high-water events, which are primarily driven by onshore winds, precipitation, and wind-induced water setups. These conditions, coupled with restricted drainage at the Afsluitdijk and increased discharges from the ZwarteWater and IJssel, result in rapid water level rises at the barrier. The current operation protocol triggers closure at +0.50 m NAP and an inland flow, protecting the Vecht Delta. However, this blocks outflow, delays vessel movements, and causes water accumulation in the Zwarte Meer. The analysis further revealed a strong correlation between higher water levels and wind setup. In comparison, the impact of discharges from the IJssel and ZwarteWater on higher water levels was minimal. The findings also highlighted the growing strain on the operation team and the disruptions faced by the shipping industry over recent years.

Nine scenarios were simulated using a developed reservoir model to assess the system’s sensitivity to wind setups and discharges, the effect of adjustments to the operation procedure, and the model’s predictive capabilities. The simulations revealed that closures during receding or stagnating water levels, or when wind setup was already developed, sometimes resulted in higher water levels than non-closure scenarios. In cases where water levels were between +0.40 m NAP and +0.50 m NAP, navigation could often be maintained, provided wind setups were reducing or stagnating, and no significant discharge peaks were predicted. Moreover, minor wind events at initial water levels above +0.50m NAP frequently triggered unnecessary closures, which could elevate water levels, highlighting the need for more robust closure criteria based on sustained flow rather than momentary fluctuations. When multiple peaks
occurred, the timing and proximity of these peaks played a crucial role in determining the impact on water levels. Earlier closures reduced water levels but extended operational disruptions, while higher discharges led to faster water level rises post-closure, requiring earlier openings. Lastly, enhancing the ability to predict critical water levels and closure and opening criteria can significantly benefit both operational teams and the shipping industry. Implementing 24- to 48-hour forecasts would enhance planning by ensuring teams are on-site when needed, minimizing unnecessary disruptions, predicting the timing and likelihood of Ramspol Barrier closures, and enabling the shipping industry to adjust schedules to reduce waiting times.

The results indicate that an adaptive, forecast-driven approach to barrier operation could potentially improve flood protection, reduce disruptions for the shipping industry, and alleviate the team’s operational burden. ... Read more

Storm surge barriers are large movable hydraulic structures that close during a storm surge to prevent coastal floods in bays and estuaries. These barriers are continuously maintained to rely on their operation when needed. Due to rising sea levels and coastal urbanization, coastal flood risk is increasing. Which barrier improvements are needed to cope with this risk increase, is unknown. This thesis presents a systematic method to find, select and evaluate barrier improvements focusing on lowering the probability of a failure to close a barrier. The method is applied to the Maeslant barrier to demonstrate its applicability. ... Read more

Many Dutch infrastructure assets were constructed in the first half of the 20th century and by considering an average design life time of roughly 100 years, a lot of these structures will reach the end of their designed life in the upcoming decades. This results in a large replacement and renovation challenge in the nearby future. Assessing the end of life of individual critical hydraulic structures (case studies) and prolonging their lifetime by suggesting adaptive measures is of importance to better define and solve this challenge. In this research, the Haringvliet sluices are used as a case study.

The focus of this research is to apply two existing methods, the framework of Vader et al. (2023) and the Adaptation Pathway Approach of Haasnoot (2013), to the Haringvliet sluices, to make an estimate of the remaining life time and to propose suitable measures to elongate it. Next to this, this research aims to connect both methods mentioned above to create a new integral method. This full method aims to qualitatively and quantitatively analyse hydraulic structures and map out possible paths to elongate its remaining life time. The results of this thesis provide valuable insights in assessing the remaining lifetime of different hydraulic structures in the future. The Haringvliet sluices are an interesting candidate for this research due to the many functions it has to fulfil and the complexity of its interactions within the water system.

The first part of the analysis on the Haringvliet sluices describes the functional analysis as well as the technical decomposition performed on the Haringvliet sluices. The goal of the functional analysis is to define requirements for each of the functions. The goal of the structural decomposition is to identify the most important deterioration mechanisms of the structural elements, which in turn are important in estimating the remaining technical life. The next step in the analysis aims to either qualitatively or quantitatively describe the external drivers which influence the functional performance and technical state of the Haringvliet sluices. The external drivers to consider for the Haringvliet sluices, or hydraulic structures in general, can be subdivided into physical external drivers and economic, political and societal drivers. From this analysis, the drivers which are expected to have the most severe influence on the Haringvliet sluices are defined. For the Tipping Point analysis in the next part of the analysis, existing models are evaluated and used to quantify Tipping Points due to the expected most dominant driver and function combination. A Tipping Point is reached if the system fails to meets its technical or functional requirements due to changes caused by the external drivers. The dominant combination investigated in the analysis is the effect of sea level rise on the flood protection function. The Tipping Point analysis shows that there is a large dependence of the Tipping Points on the evaluation criteria. The analysis gives insight in how other evaluation criteria, like accepting more overflow discharge over a dike section, could impact the Tipping Point. Even though Tipping Points are sensitive to decisions on evaluation criteria and chosen climate scenarios, it does provide valuable insights in the response of the system to sea level rise when investigating different decision criteria. In the last part of the analysis an Adaptation Pathway for the remaining lifetime of the Haringvliet sluices is presented by proposing life elongating measures. Life elongating measures for the Haringvliet sluices can be split into two different categories: Strategies for the bigger water system, i.e. future visions for the Dutch Delta, which could have either a positive or negative effect on the remaining life and life elongating measures focussed on the object or direct system. It is chosen to only investigate mitigation measures focused on the Haringvliet sluices and its direct system.

Each sub-analysis in the total research design has its individual results and therefore its individual assumptions, limitations and critical remarks. The assumptions and limitations of previous steps could have a significant impact on other steps in the analysis. One of the biggest uncertainties in the results of the remaining life time of the Haringvliet sluices is that there is a possibility that other critical combinations found from the driver analysis would reach a potential Tipping Point sooner than the one investigated in this research. Also, the impact of the proposed mitigation actions for the Adaptive Pathways needs to be revaluated on their impact on other function or technical requirements, i.e. an extra feedback loop is needed. The full analysis must be run through several times to create a more complete overview of the most critical ways a structure could reach its end of life and to indicate which measures could be best applied at what time.

By applying the methodology proposed by this research to other case studies and by conducting more research on assessing the remaining life time of critical hydraulic structures, key insights can be found in how to manage our critical hydraulic infrastructure in an adaptive way to elongate their life time.
... Read more

A Digital twin, a digital replica of a physical entity or system, encapsulated in a software model, represents a promising technology that has demonstrated its effectiveness for asset management in various industries. Rijkswaterstaat has gained interest in investigating the possibilities for digital twin integration in the asset management of storm surge barriers. Therefore, in this Engineering Doctorate (EngD) research, the application and development of a digital twin for a storm surge barrier is investigated, with the Maeslant barrier serving as the case study.

Storm surge barriers are technically complex systems that fulfil a vital role in the flood protection of the Netherlands. A strict form of risk based asset management (ProBO) is applied to keep the barrier operational. However, the complexity and infrequent use of the barrier makes this form of asset management highly knowledge intensive. Currently it is a large asset management challenge for Rijkswaterstaat (the organization responsible for operation and maintenance of the storm surge barriers in the Netherlands) to maintain the necessary expertise at the right level. The goal is therefore to explore the following question:

“How can a prototype digital twin support in investigating the application of a full-scale digital twin for storm surge barriers, to strengthen the knowledge and information in support of their maintenance and operations?”

A prototype of a digital twin for the Maeslant barrier has been designed and constructed by the writer of this report, with a particular focus on the barrier’s retaining walls. For the design of this prototype (see Figure 1), knowledge from three fields had to be integrated: 1) organizational & user needs; 2) functioning of the barrier (civil and mechanical engineering); 3) ICT architecture design.

As a first design step the user needs were investigated by conducting interviews with Rijkswaterstaat employees. These were translated into four applications the digital twin should fulfil to provide added value:
1. Enhancing efficiency in knowledge and information management
2. Avoiding unnecessary costs through better barrier status monitoring
3. Improving risk management with models and data analysis
4. Providing insights into barrier behaviour

The user requirements were translated into design requirements to build the digital twin prototype, which is developed using Unity and Python scripts. The scripts are used for data calibration purposes and to make a connection between the digital twin model and existing models. For this prototype, a connection is made with a hydrostatic force model, a failure probability model and a pump discharge model. The Unity model acts as the user interface for the digital twin model in which a 3D data animation model of the barrier is integrated combined with data visualization panels.

The functional application of the prototype is illustrated by means of three use cases:
• Case 1 - Analysis of closing procedures based on observations.
Quickly accessible and direct available data put users in the position to analyse the barrier closures and monitor its status by means of data visualisation, animation and model calculations.
• Case 2 - Knowledge conservation and transfer: hydrostatic forces example.
Increasing knowledge transfer, illustrated by an example in which hydrostatic forces acting on the barrier are animated and supplemented with visualized data that can be analysed.
• Case 3 - Observing abnormal behaviour and including risk management.
Using the digital twin to observe abnormal pump performances and directly calculate the impact of possible measures on the failure probability, to avoid unnecessary quick and costly decisions.

The prototype is tested among 14 potential users within Rijkswaterstaat to investigate the perceived added value and the organizational feasibility of a full implementation. The results of the tests demonstrate that digital twin users recognize clear value to enhance knowledge and information management. The current prototype exhibits potential in this regard, especially for the sake of information sharing and education, suggesting significant possibilities for a full-scale (i.e. of the entire barrier) digital twin. In the long term, it is expected that the digital twin could add value in increasing the barrier’s asset management. An important notification, however, is that the reactions of respondents was diverse. The respondents with a more practical role at the barrier (e.g., the operational technical specialists) were less enthusiastic about the prototype for direct use for asset management than the higher management.

The development of the prototype demonstrated the feasibility of a small-scale digital twin of the barrier. However, it does not guarantee a successful implementation of a full-scale digital twin within the Rijkswaterstaat organization. Therefore, the challenges and obstacles in implementation have been further investigated through a technical, economic, and organizational feasibility study. The prototype indicates that the available technology in the market, both at hardware and software level, is sufficient to develop a full-scale digital twin for the Maeslant barrier. However, the current state of hard- and software within Rijkswaterstaat is insufficient for full-scale implementation. Especially on cyber security, challenges need to be overcome to be able to use the maximum capacity of a full-scale digital twin.

A full-scale digital twin is expected to be financially feasible: a business case study indicates that a digital twin could be cost effective within five years. The economic added value of the digital twin is estimated based on expected savings in time and maintenance cost for Rijkswaterstaat. The costs are estimated by extrapolating the development cost of the prototype.

The Maeslant barrier organization is, resulting from the questionnaire, enthusiastic about implementation of a full-scale digital twin. However, achieving a digital twin with added value to the entire organization necessitates further development of the existing prototype, and a better connection to the user needs of the respondents that rated the prototype less positive. Subsequently, for successful implementation, the technically oriented Rijkswaterstaat employees need to be heavily included in the development process and must be given time to support in the implementation.

All points considered it is concluded that a full-scale digital twin is, under reasonable assumption of several surmountable challenges, feasible to implement for the Maeslant barrier. ... Read more

The Hollandsche IJssel storm surge barrier is expected to last another 40 years, but replacement or renovation may be needed sooner than anticipated due to factors such as climate change and societal developments. Yet, there is relatively little known about what factors and how these factors influence the remaining life, and methods that can be used to estimate the remaining life of the barrier are lacking. This thesis aims to identify the dominant factors and associated uncertainties affecting the remaining life and to provide a method for the estimation of the remaining life of the Hollandsche IJssel storm surge barrier. The focus of the study is on the functional and technical life of the barrier.

First, the functions and structural components of the storm surge barrier were identified. For each function, requirements were formulated to support the evaluation of the impact of the external drivers. The physical decomposition of the storm surge barrier was needed to identify the dominant deterioration mechanisms. This analysis revealed that the technical life is not governing for the total remaining life of the barrier. Therefore, the focus should mainly be on the functional life of the storm surge barrier. Next, the impact of external drivers on the functional performance or technical state of the barrier was evaluated by a semi-quantitative analysis in which scenarios with the projected changes were used. The study identified sea level rise as the dominant (physical) driver affecting the remaining life of the storm surge barrier because of the potentially large future changes and far-reaching consequences.

By combining the required performance levels expressed as critical amounts of sea level rise with various sea level rise projections, estimates of the remaining life of the Hollandsche IJssel barrier were obtained. It was found that even for moderate sea level rise scenarios, there is a significant probability that the end of life of the barrier could be reached within 20 years. The estimates of the remaining life of the barrier are, however, uncertain, and a considerably longer lifespan cannot be ruled out. Accordingly, it is recommended to investigate possible measures to lengthen the lifespan and determine the best moment in time to implement these measures. The decision to implement specific measures will be dependent on the costs and benefits. Economic considerations should therefore be included as well. Furthermore, a more system-oriented approach is recommended for this kind of analysis since modifications to a certain object could impact other elements within the system, e.g. dike reinforcements affect the required closure reliability of the storm surge barrier. Finally, a more general recommendation is to perform similar studies for other critical structures as the remaining life of those structures may also be reached sooner than anticipated if, for example, functional aspects are taken into account.
... Read more