Desert-based solar projects can aid in addressing the global demand for renewable energy due to their high solar irradiance and vast land availability. These projects are increasingly important as the world seeks sustainable energy solutions to combat climate change. However, they face significant risks such as extreme weather conditions, logistical challenges, water scarcity, and socio-political complexities, which must be managed to ensure their success and sustainability. Therefore, susrainable risk management and mitigation strategies need to be employed.

The power grid is becoming more intertwined with communication technologies, forming what is known as the smart grid. This integration allows for more efficient management of the power grid, which can help reduce emissions. However, the increasing connectivity with communication technology also expands the attack surface for cybercriminals or cybercriminal organisations. In 2015, a Ukrainian power station was attacked via a cyberattack, leading to an outage that affected almost 1.4 million people. The growing interconnection between the power grid and communication technologies can thus have severe impacts. Therefore, this research introduces an interdependent communication network and power grid to investigate how failures in the communication network affect the power grid's robustness. For generating communication networks, we use Python’s NetworkX package, and for the power grid, we use pandapower’s IEEE 118-bus test system. We analyse the robustness of the smart grid by examining two different communication networks: a mesh network and a double-star network. We simulate attacks on the communication network by initiating failures based on random selection, degree, betweenness, and closeness centrality, with an increasing number of node failures. We also distinguish between two failure behaviours: one that includes communication network failure propagation (e.g., the spread of malware or the failure of dependency nodes within the communication network) and one without this failure propagation behaviour. In general, the smart grid shows higher robustness with a mesh communication network. However, under different failure behaviours, attack strategies, and numbers of nodes attacked, this can vary. Prioritisation should focus on reducing communication network failure propagation, as this significantly impacts the robustness of the smart grid. Even though degree centrality and betweenness centrality have the most impact on double-star networks, for mesh networks, closeness centrality should also be considered, especially in scenarios without network failure propagation.

The industrial sector is one of the most energy consuming and CO2-emitting end-use sectors. To reach climate goals, decarbonization of these industrial sectors is imminent, however, not so apparent. Especially the hard-to-abate industry sectors, such as the chemical, mineral processing, iron, and steel sectors, are difficult to decarbonize as they require either high temperature heat or use fossil fuels as feedstock.

Hydrogen has the potential to reduce carbon emissions in industries such as chemicals, glass, iron and steel, as well as to serve as a cleaner heat source. To reach net-zero emissions by 2050, sectors that currently use fossil fuels for high-temperature processes and as feedstock will likely need to shift towards blue or green hydrogen. Currently, some industrial hydrogen use relies on gray hydrogen, produced from fossil fuels and contributing to emissions. In contrast, blue hydrogen captures and stores CO2 produced from fossil sources, while green hydrogen is entirely emissions-free, generated from renewable energy. In other words, some processes need to change from gray to green/blue but most of them need to change from other fossil fuel based processes to hydrogen processes.

Hydrogen offers a way to cut carbon emissions in industries like chemicals, glass, iron, and steel, and can also act as a cleaner heat source. Achieving net-zero emissions by 2050 will likely require sectors that currently depend on fossil fuels for high-temperature applications and feedstocks to adopt blue or green hydrogen instead. Today, certain industrial applications still use gray hydrogen, derived from fossil fuels and contributing to carbon emissions. However, blue hydrogen captures and stores the CO2 generated, while green hydrogen is emissions-free, produced using renewable energy. In essence, some processes will need to transition from gray to blue or green hydrogen, while many others will shift from fossil-fuel-based processes to hydrogen-based alternatives.

However, the timing and extent of the hydrogen transition are uncertain as they are heavily influenced by external factors such as hydrogen prices, available subsidies, and alternative decarbonization options. Additionally, industrial plant owners may be reluctant to disclose decarbonization plans due to competitive pressures, adding another layer of demand and participant uncertainty that complicates infrastructure planning.

This thesis addresses the planning of hydrogen infrastructure within an industrial port cluster (IPC). IPCs are defined by their proximity to water and concentration of industrial activities related to a specific sector. In order to effectively address spatial constraints, this thesis will plan the hydrogen networks along the current road network within IPCs. Current infrastructure planning methods have a time horizon of ten years. However, as the expectation is that the hydrogen demand will increase towards 2055, a time horizon of ten years can increase the total costs of the network when the network is implemented over time between 2025-2055. This introduces the following research question:

“How can a cost-efficient, robust pipeline network for an industrial port cluster be developed over time under uncertainty?”

To answer this question, the robust backtracking planning method (RBPM) is developed. This method aims to minimize costs over the 2025-2055 time frame while facilitating the hydrogen to the demanding plants. Because the demand for hydrogen is likely to grow over time, this method finds a robust network that is able to facilitate the demand in many possible future demand scenarios of 2055.

The robust network is then implemented incrementally for 2035, 2045, and 2055 using a backtracking approach. In this context, backtracking means that when an industrial plant transitions to hydrogen in one stage, pipelines are installed with the robust networks’ capacity, rather than just the minimum required to meet that plant’s immediate needs. This extra capacity ensures that if other plants transition in later years, the existing network can accommodate the increased demand without needing costly pipeline extensions. By preemptively building capacity, this approach reduces future installation costs and enhances the network’s ability to adapt to evolving demand patterns.

The RBPM is tested on simulations of multiple simplified IPCs. By testing different IPC simulations, it is studied how the difference in industrial plants determines the development of the network. The RBPM is compared to the results of a traditional planning approach which only plans the networks with a time horizon of ten years.

The results show that the RBPM incurs lower costs over 30 years, but it requires a higher investment in 2035 due to the greater capacity installed at that time. This thesis finds that the total potential hydrogen demand and the physical size of an IPC significantly affect the performance of the RBPM compared to the traditional planning approach. Additionally, the projected installation and operational costs over time also impact the RBPM’s performance relative to the traditional planning approach.

For IPCs with comparatively low hydrogen demand — typically clusters with fewer iron and steel facilities, chemical plants, or refineries — the RBPM emerges as the most economical approach. This method requires only slightly higher investment by 2035 but ultimately generates substantial savings by 2055. By installing sufficient pipeline capacity upfront, the RBPM avoids the need for additional pipelines every ten years, leading to long-term cost efficiency through 2055.

For IPCs with high hydrogen demand—typically found in iron and steel plants, basic chemical plants, or refineries—the initial installation costs and ongoing operational expenses of RBPM make it less advantageous. While RBPM may offer slightly better economic profitability over a 30-year period, the substantial investment required in 2035 compared to traditional planning makes implementation challenging due to budget constraints. In these high-demand clusters, the decision between RBPM and the traditional approach for developing a hydrogen pipeline network depends on the cluster’s budget, anticipated future installation costs, and projected operational expenses over time.

Opportunities for further research include the application of the RBPM to a real case study to validate the result, increasing the amount of possible future scenarios by incorporating uncertainty in installation and operating costs and increasing the demand and participant uncertainty range. Lastly, another research direction to explore is the generation of different robust network methods and their performance.

Society is facing a huge challenge in switching the energy sectors dependence on fossil fuels into an energy sector using mostly renewable energy sources. The switch towards using more sustainable energy sources is known as the energy transition. The goal of the energy transition is to lower the greenhouse gas (GHG) emissions emitted by the energy sector. Lowering the GHG emissions helps society limit the global warming caused by GHG [3]. 17.5 % of the global energy usage comes from the energy use in buildings [50]. It is thus very important that the energy use in buildings transitions towards using more sustainable energy sources. One of the renewable energy sources that is ought promising in the energy transition for energy use in buildings is geothermal energy [3]. Geothermal energy is energy that is captured in reservoirs of hot water in the earth’s crust. The hot water captured in the hot water pockets is pumped to the surface, to use it in spatial heating. The return pipe returns the cooled water to the geothermal well, where it can heat up again over a certain period of time [63] [23].

In some cases, geothermal energy is applied using a district heating network. A district heating network is an example of a system that provides heating and/or cooling capacities to a group of buildings [65]. A district heating network is a network of pipelines that transport the hot water from the geothermal well to the buildings in the district. A geothermal well in combination with a district heating network is developed in Delft [27]. The district heating network will deliver energy to the TU Delft campus, two neighborhoods in Delft and industry at the Schieweg in Delft [28].

Besides the district heating network in Delft, it is expected that district heating networks will be applied more often to accelerate the energy transition. Yun-Chao and Chen (2012) concluded that most optimization techniques optimize the whole system with its components. Less optimization techniques are applied to the sole components. Besides the fact that most optimization methods optimize the system as a whole, most optimization objectives only include optimizing the cost of the system. Also, effective optimization techniques are required as optimizing large graphs may be computationally time consuming [36]. In literature there are also clear signals that state that the trade-off between thermal comfort, and efficiency with respect to cost has to be tackled [53]. In this research, optimizing district heating networks for cost is compared to optimizing district heating to maximize thermal comfort or efficiency.

In this research two models are developed: a model that calculates the cost of the district heating network, and a model that calculates the thermal losses of the district heating network. Both models are applied to a district heating networks that is developed in a street network. Furthermore, multiple heuristics are applied to come up with better district heating networks. The optimization technique is tested on 100 small, randomly generated district heating networks. After that, the district heating network in Delft is optimized. The differences in cost, efficiency, etc. will be evaluated. Besides, the performances of the district heating networks are evaluated by introducing energy deficits under different conditions.

Optimizing the district heating networks for cost led to a very consistent result: When compared to their individual starting point, the district heating networks became cheaper and more efficient. A moderate-strong correlation is found between the the increase in efficiency and the decrease in cost while optimizing the district heating networks. In contrast to that, the networks that maximize efficiency are much more expensive than their cost optimized alternative, while the increase in efficiency is in most cases moderate. However, there are rare cases where the efficiency is increased much at a moderate increase in cost. This phenomenon is also found in Delft. Given the result that the efficient district heating network also performed much better than the cheapest alternative during energy deficits, in this research it is shown that choosing an objective function has a very large impact on the characteristics of the network. Therefore it is shown that for future district heating network optimization, it is important to trade off cost against efficiency.

Explainable AI (XAI) systems are rapidly gaining significance. While frameworks for XAI interpretability for experts abound, metrics for laypeople’s comprehension are absent. This study addresses this gap by investigating interpretability factors from both developer and layperson perspectives. The core research question is: "How can XAI developers assess to what extent XAI is interpretable for laypeople?".

Applying the Design Science Research Methodology, findings from multiple literature reviews are combined to construct a preliminary XAI interpretability framework for laypeople, featuring crucial factors and their relationships, as well as associated principles. The proposed framework underwent validation through semi-structured interviews with 12 XAI experts, informing revisions and refinement of our key principles. Subsequent layperson surveys, considering a specific use case, offered insights into preferences about interpretability factors, informing further refinement.

The final theoretical framework highlights pivotal factors including simplicity, transparency, comprehensiveness, complexity, clarity, generalizability, trustworthiness, explanation fidelity, model fidelity, intentionality, relevance, affordance, coherence with prior beliefs, and actionability. Surrounding the framework are key principles emphasizing trustworthiness, relevance, simplicity, clarity, coherence, intentionality, actionability, fidelity, contextualization, and ethical considerations, serving as actionable guidelines for XAI developers and researchers.

The implications of the study are profound, offering valuable insights for advancing XAI research and system design. The refined framework and principles act as a foundation for both novice and experienced XAI developers, fostering interdisciplinary research among AI, human-computer interaction, psychology, and philosophy experts. These findings can drive the responsible adoption of AI systems across sectors like healthcare, finance, and transportation, while informing policies and regulations governing AI technologies. Our study promotes responsible AI practices, enhancing user trust and understanding, while facilitating the creation of more effective guidelines and standards.

Freshwater is becoming an increasingly scarce resource globally. The chemical industry is the second largest freshwater consuming industry and the largest wastewater producer. To reduce the freshwater consumption sustainable measures are introduced within the industrial sector, of which Industrial Symbiosis (IS), has gained prominence. Specifically for water, this involves one company utilizing the residual wastewater of another company based on the differences in water quality requirements However, the chemical industry lags behind in embracing IS practices. Two important reasons cause this trend. Firstly, it is technically challenging to recycle water because of all the different water qualities used and produced in chemical parks. And secondly, the industrial symbiosis initiatives have to start with the stakeholders involved in the chemical park themselves in bottom-up regulated countries, while currently, many stakeholders decide not to join an IS water network. This research focuses on the feasibility of implementing IS water networks within chemical parks, considering stakeholder behaviour (wanting to join or not join an IS water network). A model is developed to measure the optimal performance of an IS water network, encompassing technical optimization and stakeholder decisions. The study reveals that the potential for industrial symbiosis in chemical parks relies heavily on existing treatment facilities and stakeholder participation. The type of treatment facility structure, with in-house or separate treatment facilities, impacts costs and water recycling capabilities, with chemical parks with separate treatment facilities exhibiting higher costs and higher recycling potential. Key decision criteria for stakeholders include revenue growth, compliance to regulations, trust among stakeholders, and awareness of water recycling possibilities. Recommendations to enhance industrial symbiosis of water include technical adjustments and construction efforts to the IS water network within the chemical park, and policies addressing stakeholder concerns. The contributions of this research lie in integrating technical and social aspects in the design of optimal IS water networks, providing insights into decision criteria that facilitate or hinder industrial symbiosis in specific chemical parks and how to address them to recycle most water as possible with least amount of costs.

This thesis aims to answer the question "What are the properties of a resilient road network?" The resilience of a road network indicates the magnitude and consequences of a disruption, which can be events such as a traffic accident or a flood. The literature is not in agreement about the exact definition of road network resilience, and the way to quantify it. A literature search is done to look for definitions and resilience metrics. Eleven metrics are found that quantify resilience in road networks to accidents. These eleven metrics are compared based on whether they are on a network level, and how many variables are used. Two metrics are chosen, one based on travel time, and one based on space-mean flow. Another metric, based on the outflow of the network is added.

The three chosen metrics are then compared in small networks to see if they give the same results. The small network consist of five nodes with different link configurations. The conclusion about which network is the most resilient changes with the resilience metrics. The metric based on travel time is deemed the best for this type of research. The metric based on network outflow can have distorted results due to high peaks, and the metric based on space-mean flow is better suited for a different experiment design.

To see how network parameters such as link density influence the resilience, a simulation with random networks is done. 266 networks with nine nodes in random locations, and random links between them are simulated. It is concluded that networks with a lower link density have a higher resilience. This is not only illustrated by the relation between resilience and link density, but also by several other network parameters which are related to link density such as average number of lanes and connectivity. A reason that these networks are more resilient could be that there is less spillback, the links in the high density networks are longer and have more lanes, so congestion will not spread to other links as fast.

Ransomware has evolved over the years, shifting from widespread attacks targeting individuals to focused attacks on businesses and agencies. These attacks are performed by ransomware gangs while establishing interaction within the ransomware ecosystem. In this thesis, the ransomware ecosystem is posited as being constructed of three separate sub-ecosystems: the attacker sub-system, the defender sub-system, and the governance sub-system. Since ransomware gangs put in an effort to hide their internal communication and operation from the outer world, difficulties arise in correctly understanding the ransomware ecosystem and a ransomware gang’s establishment of interactions within this ecosystem. As a result, current interventions are ineffective.

While earlier research has been conducted on ransomware, we observe two knowledge gaps: 1) there is a lack of understanding of how ransomware gangs establish interactions with actors in the ransomware ecosystem, and 2) There has been a lack of research that uses ground truth data due to ransomware gangs keeping their internal communication and operations hidden. This thesis uses the leaked internal communication data of the Conti ransomware gang to fill these knowledge gaps and answer the research question: To which extent can the ransomware ecosystem be reconstructed using ground truth communication data of the Conti ransomware gang?”.

To answer this question, a novel methodology is proposed that uses Latent Dirichlet Allocation (LDA) topic modeling to empirically determine overarching topics in Conti's internal communication. It is then researched how these overarching topics map to Conti’s tactics, techniques, and procedures (TTP) which is a commonly used methodology to better understand how ransomware gangs operate. Subsequently, these TTP are leveraged to reconstruct the ransomware ecosystem while taking the perspective of how the Conti ransomware gang establishes interactions within the ransomware ecosystem.

The findings of this thesis indicate that Conti is a large and professional organization that incorporates and adjusts services of service-providing cybercriminals in the attacker ecosystem rather than developing their ransomware themselves using scarce IT talent. In addition, reconnaissance is one of the most critical activities that ransomware gangs perform to get to a successful ransomware attack. While researching Conti's TTP, this thesis identifies novel TTP of ransomware gangs, such as Conti's attack chain, reconnaissance procedure, and money laundering procedure.

We conclude that the ransomware ecosystem can be reconstructed from the attacker ecosystem, the defender ecosystem, and the governance ecosystem, in which ransomware gangs establish interactions within each sub-ecosystem while operating from the attacker ecosystem. In the attacker ecosystem, ransomware gangs establish interactions with service-providing cybercriminals to outsource sub-commodities of their ransomware value chain. This allows them to strengthen their attack vectors by relying on the expertise of others and have a more varied set of attacks. The defender ecosystem is comprised of defenders that defend themselves against ransomware. Ransomware gangs establish interactions by performing extensive reconnaissance on defender territories and valuable information and open-source tools that strengthen their attack vectors. The governance ecosystem comprises governance actors that create and maintain the governance framework that influences the attacker ecosystem and defender ecosystem. Ransomware gangs establish interactions with actors in the governance ecosystem to observe the regulatory frameworks in place and adjust their TTP based on the involved risks of getting caught.

Energy Communities are initiatives ran by members of the same or different communities with the aim of producing, distributing and consuming cleaner energy. The potential of these initiatives, as identified from this clean energy production, to play a key role in the energy transition is assumed to be vital as recognized by the European Commission’s Clean Energy for All package and the provision of a legislative framework for their operation in European countries. In the light of this, European policy makers assume that promoting the engagement of more citizens in such initiatives will lead to a rise in energy justice and a fairer energy transition. However, several researches show that this assumption maybe flawed as the relation between the two is not explicitly verified. Through the acknowledgement of minority groups, this research aims to investigate whether an increased citizen engagement in energy communities could lead to a rise in energy justice. For this, the research is broken down into three key steps involving: 1) investigating what factors could affect the willingness of citizens to engage in energy communities by conducting a literature review and then evaluating potential influence of the identified factors for the city of The Hague as case study area, 2) investigating the distribution of the identified factors around areas with energy communities to find out whether there are any social groups favoured by the distribution of these factors around neighbourhoods with 24 energy communities. This is done by conducting spatial data analysis to investigate whether the engagements of any particular social groups or classes of people might be favoured by the spatial locations of energy communities in specific areas of The Hague 3) finally by conducting one on one interviews with 8 board members of energy communities in The Hague, the research seeks to investigate the relation between the identified factors potentially affecting citizens' desires to engage and the three tenets of energy justice. The results show that in the city of The Hague, although a majority of neighbourhoods in which energy communities are located are characterized by a high number of minority groups, a high majority of the energy communities interviewed (7 out of 8) acknowledge that minority groups are generally less involved. However for the classes of people engaging in such initiatives, a majority seeks to raise their powers and that of other community members in the decision making process. Hence, this allows to conclude that for the case study area, energy communities might only lead to a limited rise in Energy justice due to the partial or non-acknowledgement of minority groups.

Law enforcement occurs in a complex environment that contains a variety of actors that interact with one another. These interactions create emerging collective behavior over time. For example, inspectors will try to influence non-compliant actors to become compliant, while inspectees may comply or thwart inspections. Inspection agencies such as the Inspectie Leefomgeving en Transport (ILT) evaluate inspectees’ adherence to regulation and aim to boost compliance within an industry. However, they have limited resources and a wide range of potential societal challenges to address. They face an action dilemma, having to decide on what actions to take without full knowledge of whether their actions lead to higher compliance or improved social outcomes. Previous studies of the inspection environment rely on behavioral theories to investigate the underlying motivations of inspectees’ behavior. However, these theories presume inspectees’ motivations and characterize them homogeneously, often assuming they have perfect rationality. This leads to an inaccurate depiction of inspectees, reducing them to one-dimensional actors when in reality, their behavior is motivated by multiple factors and can be idiosyncratic. Data science techniques provide the opportunity to understand behavioral phenomena with data, leveraging datasets to identify statistical patterns in behavior to help inspectorates make decisions within a degree of certainty. A particular modeling technique that focuses on representing empirical data without assuming behavioral motivations is phenomenological modeling. Coupled with agent-based modeling (ABM), phenomenological modeling allows the researcher to simulate possible outcomes of observed behavior before the underlying motivations are understood. This provides insight into the macro-level behavior produced by micro-level interactions within the complex inspection environment...

Electricity is essential in the modern world. Although India is near reaching 100% electrification, for many of its citizens, reliable access to electricity is still a major issue. It is not clear how, when, and where the reliability issues arise. Previous studies have not shed light on electricity outages at a granular level. Furthermore, identifying outages is difficult without proper metering.

This research aims to acquire insight into how electrical outages vary during the nighttime in the Uttar Pradesh, one of the states experiencing the most outages in India. Remote sensing data in the form of nighttime radiance, wind, precipitation, temperature, air quality, land cover, and population are used to identify outages, as they have been identified in the literature to be influencing or correlated with outages. Using a Random Forest (RF) classification algorithm, outages are identified in India for the year 2018. The RF is trained on Electricity Supply Monitoring Initiative (ESMI) real-time electricity household senors. RF is cross-validated using multiple strategies to accurate measure the performance and to ensure no data leakage. Using the resampling technique Synthetic Minority Oversampling technique (SMOTE), the performance of the RF is increased for outage classification. RF is used to classify three classes: Never Access, Normal Access, and Outage. The distribution of these three classes is highly imbalanced, with the Outage class being in the minority.


To further validate the methodology, the spatial and temporal sampling was done during cross-validation. Using this method, 91% of Never Access samples were categorized as such, with a precision of 84%. 69% of Normal Access samples were categorized as such, but with 84% precision. The performance of the outage classification is the worst. Nonetheless, 55% of the Outage samples were categorized as such, although with a precision of just 39%. A map is created for Uttar Pradesh showing the results of the rates for each of the classes in 2018 between 00:00 and 02:00 with a spatial granularity of 0.1° x 0.1°. The map indicate that the electricity usage during the nighttime in Uttar Pradesh is in large parts of the state non-existent. The map be used as a precursor for future field work and help policymaker and researchers identify reliability and fairness issues.

Cash and Voucher Assistance (CVA), a type of humanitarian aid consisting of giving money instead of products, is being used more frequently because of its effectiveness and efficiency in helping people in need (Cash Learning Partnership, 2020b). The debate on using CVA is currently focusing on improving the quality by better incorporating ’voices’ (needs and preferences) of recipients and by enhancing targeting. In targeting it is a major challenge to quickly identify the individuals and families with the biggest needs, given the lack of data (Aiken et al., 2021). Research on ways of measuring impact on and satisfaction of recipients combined with research on demographic and behavioural characteristics of recipients could lead to deeper insights in recipients of trackable CVA modalities (evouchers and ecash). This research uses the marketing literature on customer segmentation combined with machine learning algorithms to come up with an innovative new approach of categorizing recipients of evouchers, using the case of a Red Cross project on Sint Maarten. The main research question is: How can recipients of cash and voucher assistance be categorized using the field of consumer segmentation by using machine learning methods? The objective of this research is to come up with new methods to better understand recipients of CVA. Theories on customer segmentation pointed to the use of data-driven clustering methods to categorize consumers. Combined with a framework of recency, frequency and monetary aspects, recipients of evouchers could be categorized effectively. A required addition to this clustering method is to use a dimension reduction technique to avoid the negative consequence of the curse of dimensionality. Therefore, a two-step approach of dimension reduction and clustering has been applied in this research. It has been found in this research that a factor-cluster approach can lead to insightful clusters using geo-demographic data and behaviour data. Factor analysis has been used to reduce the dimensions while the k-prototype algorithm has been used to cluster into five distinct groups of recipients. The geo-demographic variables that were the most determining in characterizing distinct clusters consisted of: the age of the main beneficiary, the different household compositions and of a constructed factor ’big families and big receivers’. The most distinguishable variables on behaviour were: the number of supermarket visits (frequency), the time between the first voucher was received and the first transaction (recency) and the variables on the amount of money that was spent with the vouchers (monetary). To be able include the ’voice’ of recipients (needs and preferences), a connection between the registra tion data, behavioural data and survey data is needed. In this research only an exploratory connection could be established, due to the lack of a common identifier between the survey data and the other datasets. However, one crucial finding of this research is that it seems like the combination of these data sources can give meaningful insights in the needs, preferences and behaviour of households of Sint Maarten. With these insights specific clusters can be targeted for additional assistance, based on their needs. Recommendations for future studies include studying the validity of the found cluster results with different validation indices on cohesion and compactness, and by using simulations to determine the cluster stability. Before this factor-cluster approach can be deployed in CVA projects, more research on the treatment of limitations of this approach needs to be conducted. This is critical in communi cating the conditions and constraints of this model to humanitarian aid workers in the field. Another recommendation is to improve the design of surveys to measure the needs of recipients. For insightful factor-cluster results on the needs of recipients, survey data should be linked to geo-demographic and behaviour data. More research on including clusters in retargeting methods using feedback loops have a large potential in minimizing targeting errors and more effectively meeting the needs of recipients. With this research, the humanitarian sector can benefit from new ways to understand the needs of the most vulnerable in need. Decision-makers should build upon the feedback of recipients and move towards a new era of humanitarian assistance.

Ambitious decarbonization targets are introduced to combat climate change. According to the European Commission, sector coupling is required to achieve these targets. This includes the inclusion of renewables into sectors that are difficult to decarbonize, but also using resources more efficient, and introducing new clean fuels. Recent research has examined the increased integration of renewables into the energy system, or the coupling with other sectors, such as the heating industry. However, a paucity of studies investigated the integration of the transportation sector with the power sector based on different configurations of hydrogen supply chains, with hydrogen as a nodal connector. Some studies looked into the decarbonization of the transportation sector but diminished to explore how the coupling based on hydrogen supply chains including hydrogen storage could support the grid by providing balancing services. Therefore, the aim of this study was to examine potential designs for renewable hydrogen supply chains to provide joint applications, which are satisfying a hydrogen mobility demand, as well as providing balancing services to the power grid. However, the interconnection of the hydrogen system within the mobility sector, as well as the functioning power grid, based on the supply chain, poses multiple challenges on different levels. This complex problem requires solving in an integrated manner. In that regard, this thesis proposes a modelling approach for the optimal functioning of hydrogen generation and storage supply chains. Hereby the objective of the function is to maximize the profit of the supply chains through the supply of hydrogen fuel to the mobility sector and the contribution of balancing services to the power grid. The proposed optimization model is employed to perform a techno-economic comparison of two supply chain configurations: a distributed on-site supply chain, in which the hydrogen is generated at the fuelling stations, and a centralized off-site supply chain, where the hydrogen is centrally produced and afterwards transported to the dispersed fuelling stations. Both scenarios are quantitatively studied under two operating conditions, which are only selling to the mobility demand, and providing ancillary services to the power grid concurrently. It was found that both distributed and centralized hydrogen generation and storage supply chains can be used to offer hydrogen fuel to the transportation sector as well as provide different ancillary services to the grid. Furthermore, the numerical findings show that when both supply chains are planned and run concurrently for various ancillary services, the highest improvement in terms of financial parameters can be observed. The results also indicate that the profitability of the distributed and centralized supply chain is comparable under various operating conditions. However, while the distributed supply chain system presents better economic performances than the centralized supply chain system under the performed conditions, the centralized hydrogen supply chain has greater
4
technological benefits than the distributed supply chain in terms of flexibility. Whereas the decentralized supply chain is more beneficial for smaller demands and a more isolated operation, the centralized supply chain can provide higher values to investors who must deal with larger demands and more interconnected systems. Because the majority of these systems will be operated by larger system operators, and the supply chain, including fuelling stations, will be increasingly connected to the energy sectors rather than operating independently, the centralized alternative is likely to be the most beneficial investment. The analyses reported in this study are useful to interconnected investors or system operators for assessing the technical and economic viability of the widespread deployment of distributed and centralized hydrogen generation and storage supply chains integrated with the power grid. Future research might investigate other system configurations, such as including the hydrogen backbone into the operation or integrating the hydrogen supply chain with diverse industries, such as the heating sector. Different renewable energy sources, such as biomass or hydropower, might also be included. Other research might involve dynamic hydrogen pricing or the engagement of various distribution modalities with global demand points. Finally, for the supply chain systems to be successful, the social aspect must be considered, in which numerous market regulatory and governance impediments must be identified and exploited.

Industrial clusters are facing major challenges in transitioning away from their carbon-intensive activities in order to accelerate the energy transition away from fossil fuels. Therefore, the industrial clusters; the Port of Rotterdam, Chemelot, North-RhineWestphalia (NRW) and Transition System Operator (TSO) GasUnie are planning to develop a pipeline connection for the transportation of renewable fuels between one another (Port of Rotterdam, 2020b). However, this pipeline connection operates in a very uncertain environment with regard to the technology required to decrease greenhouse gas emissions and the future policies that impact on the investment decisions by the industries that will make use of this connection. Therefore, there is much uncertainty about the requirements of the pipeline, making the way in which this new pipeline connection is to be constructed a complex question. Designing a flexible pipeline with capabilities to pro-actively deal with uncertainty could reduce the financial risk and support investment in the pipeline connection. Even so, there is a lack of knowledge about the specific costs and benefits of flexible pipelines and the question arises of how efficiently these flexible pipelines will perform. Therefore, this research is aimed to analyse the performance of various flexible pipeline designs for the connection of Rotterdam, Chemelot and North-Rhine Westphalia and answers the research question:” How do flexible pipeline designs perform for the transportation of sustainable fuels between the Port of Rotterdam, Chemelot and North-Rhine Westphalia from 2025 until 2050?”.

Since the outbreak of COVID-19, far-reaching measures were introduced to isolate people and reduce the chance of transmission among people. In the Rohingya refugee camps in Bangladesh, the camps were locked down for months. As a result, economic dependencies between the refugees and host communities which were built in previous decades were cut-off for months. So far, no research has been published on how to release the burden of limited refugee-host interactions, which can be a vital source of livelihood generation. In this research, the implementation of different vaccine allocation strategies were analysed and compared, to analyse what prioritization strategy can best be introduced to most effectively revive these interactions and control the COVID-19 outbreak. Results show that either prioritizing elderly or high-risk transmission groups can be effective. Prioritizing elderly distinguishes itself as it most effectively reduces severely & critically symptomatic infections. Prioritizing transmission-groups shows maximize refugee-host interactions. When combining both strategies with a dynamic approach of opening/closing the camp from host communities, a more effective strategy is possible for protecting the vital economic functions of an open-system refugee camp. An implication of this research is reflected by the granularity of the model. As the model has a high resolution and is highly computational, insufficient model experiments could be conducted to cover a complete solution space of the model. However, as this research proposes unique study in the dynamics between vaccine allocation and open-system refugee camp, a foundation is grounded for further research.

The duration of protection that COVID-19 vaccines provide is still uncertain (Murray & Piot, 2021). Therefore, a possible future scenario is that the entire population will need an extra dose of a vaccine. Yearly recurring flu vaccinations take place at general practices (GP), so this might also be the case for COVID-19 vaccinations. General practitioners indicate that this would not be efficient. In fact, many general practitioners don’t want to execute any vaccinations, because their main job is to provide primary healthcare (Buckley, 2021). On the one hand, when considering the insights of general practitioners, the question arises whether it is desirable for GPs to take care of the vaccinations. On the other hand, immunization of an entire country is a huge operation, so it is important to know which contribution GPs can make to accomplish this task. To answer this question, the required performance of the system must be investigated. Therefore, the main research question is: What is the required performance of the vaccination system to keep the vaccination rate above a sufficient level and minimize virus spread, given uncertainties in vaccine supply and negative social influence regarding vaccinations? This research explores the interactions the concepts of social influence, service level (i.e. the ease of getting vaccinated) and willingness to vaccinate. Several experiments were run wherein negative social influence influences willingness to vaccinate in varying degrees. The same applies to the service level of the vaccination system, which is expressed in waiting times, vaccination speed and canceled appointments. Uncertain deliveries impacts the vaccination speed and the number of canceled appointments in a negative way. A lower average willingness to vaccinate leads to lower vaccination rate and stimulates virus spread. The minimum vaccination rate to achieve herd immunity is assumed to be 70% (Lippi & Henry, 2021). Since an epidemic and willingness to vaccinate of a population are driven by social interactions, an Agent-Based Modeling (ABM) approach was chosen to conduct this research. Based on the findings of this research the following lessons learned have been formulated. First, it is recommended to focus on higher vaccination capacities in order to reduce the vaccination times and limit infections. Despite the fact that vaccinations can be executed more efficiently by GGDs and that many GPs are not willing to take care of vaccinations, the additional force of the GPs can be useful. Provided that they accept to spend 1 hour per day for 37 days, GPs can take care of 11% of the vaccinations. Therefore, the options should be discussed with them in order to convince them to take part in the vaccination campaign. Besides the 11% of the population that GPs can possibly vaccinate, another 89% percent of the population should be vaccinated. Therefore, re-establishment of (some of) the current mass vaccination sites would be a useful option to enable GGDs to work efficiently. Second, this research shows that the minimal required performance of the vaccination system to keep the vaccination rate above a sufficient level is low. That indicates that the willingness to vaccinate in the Netherlands is robust. Enormous increases in negative social influence, delays and low service levels didn’t cause significant disruptions of the willingness to vaccinate. Even in situations with 40% chance of delayed vaccine supplies and high values of negative social influence, the vaccination rate remains above 70%. However, it is still important take the effects of social influence and low service levels into account, because they might be stronger when the willingness to vaccinate is lower. That will possibly be the case when we are not in a crisis situation anymore. Suggestions for further research are making improvements in the model and investigating the actual importance of social influence on vaccination decision making. The last suggestion is to conduct experiments with the model proposed in this research, with the amount of patients that currently receive an influenza vaccination, instead of the entire population. With those experiments, it can be evaluated to what extent GPs can take care of the vaccinations when the number of patients is equal to the number of people that already visit the GP for a vaccination every year.

Novelty search is a state-of-the-art approach focusing on behavioural novelty, rewarding diverging, as opposed to pursuing static objectives. This is relevant for exploratory modelling and analysis, which focuses on exploration of model through open exploration or directed search. Novelty search as an open exploration strategy is being tested against proven methods such as latin hypercube sampling. Using existing evolutionary algorithms and a developed novelty function, the experiments focus on comparisson, impact of the number of functional evaluations and the impact of the goals of the evolutionary algorithm. Finally it can be concluded that novelty search finds novelties in the lake problem, which makes it a relevant search strategy, but not suited for indiviual exploration. That means that it would still be advised to use latin hypercube sampling for earlier exploration.

This research introduces a method for using global data sets to conduct a flood risk and cascading failures analysis in countries where high-quality data does not exist. As the goal is to support policy-making on a national level, an impacts analysis is additionally proposed to provide a systematic way of comparing system performance against multiple objectives and under potential policy interventions. The method proposed in this study provides policy makers with an avenue to conduct such an analysis in countries where the requisite data may not exist or is unavailable, thus paving the way for a data-driven approach to decision-making.

In the future, the energy supply will primarily consist of energy from renewables, which makes energy producers and grid operators fear that grid security and reliability could be threatened. On top of a more fluctuating supply, energy demand is also expected to increase in the coming years, especially if heating will be electrical when gas is phased out. Demand-side management, and more specifically demand response, could be a way to make demand match supply better and limit the foreseen energy shortages. A smart grid and smart metering technology can be implemented to overcome the hurdles of grid integration of renewables. Consequently, the implementation of a smart grid could potentially steer consumer electricity usage and allow for automated demand shifting. This thesis analyzes consumer participation in demand response in an electricity grid with a high share of renewables. The effect of multiple price- and incentive-based interventions are tested for different consumer populations and supply scenarios. The aim is to improve the understanding of the effect of consumer heterogeneity on the participation in demand response. An agent-based model was constructed that represents the energy system as a copper plate model and incorporates the rule-based behavior of different consumer types. This model was used to answer the main research question: How can a simulation model be used to analyze the influence of consumer heterogeneity on the success of demand response enhancing policies in future electricity networks? The consumer types that were included are the green, the cost-conscious, the convenience-oriented and the indifferent energy consumer. Subsequently, five personal values were identified as important moderators for consumer participation, namely the values of price, environment, comfort, safety and social norm. Different interventions and combinations of interventions were compared based on the three model KPIs energy shortage, total electricity cost and required storage. The comparison of the interventions and intervention combinations showed that the differences in energy shortage between the policies were limited and were mainly determined by supply. For a constant supply, the results revealed that the simultaneous implementation of all policies was most effective to reduce shortages, followed by the policy combinations that included the cost comparison with a reduction. Naturally, these combinations also resulted in lower electricity costs for the consumers. The cost comparison and reduction policy and default shifting could increase consumer participation most when used as a separate intervention. Some main conclusions can be drawn from the analysis of the interventions. First of all, supply was the most important influence on energy shortages. Depending on the shortage period characteristics, shortages could not always be solved completely by demand response, but a higher amount of demand shifters can improve the situation significantly. However, not all consumer types could be motivated to switch to a shifting contract, which means that the consumer population is an important influence on the effectiveness of policies. From this thesis it can be concluded that a simulation study is a successful way to create an inclusive model of the technical, economic and behavioral system components. The model was able to provide a system-level overview of the system dynamics and improved the understanding of interactions between interventions and consumers. Therefore, this model can be used in future case studies to analyze the potential consumer participation and recommend suitable interventions to increase this.

With the increasing research of autonomous vehicles coming closer to introduction in society, it is necessary to research the effects on traffic safety for the introduction of these types of vehicles for rural roads. A lot of research has been done for the introduction of AVs on highway roads, researching the introduction for rural roads is the next step for the implementation of AVs in society. With this thesis the goal is to explore the traffic safety effects on rural roads with the help of agent-based modeling. An agent-based model was created to examine the effects of the penetration rate of AVs on traffic safety and how the characteristics of rural roads would affect traffic safety in this regard. Two main conclusions were drawn from the results. The first is that the introduction of AVs would lead to a lesser heterogeneity of speed between vehicles, this would in turn result in fewer casualties. The second conclusion is a conclusion drawn based on the emergent behavior of the model, the introduction of AVs would lead to less headway for AVs. Future research could be undertaken to study effects of overtakes on vehicle platoons.