Natural disasters pose significant threats not only to the environment and infrastructure but
also to human lives. Therefore, effective disaster risk management must account for physical
and social vulnerabilities. This research focuses on developing a comprehensive tool for assessing social vulnerability to enhance disaster risk management strategies. Social vulnerability is a complex concept that includes various dimensions, such as socio-economic
and cultural factors. Recognising the multifaceted nature of vulnerability, this study aims to
create a multidimensional index that equally represents each dimension of social vulnerability. By incorporating intersectionality theory, the research acknowledges that vulnerability is compounded and varies across different population groups. The proposed multidimensional index method introduces new calculation tools to balance the weights of all social dimensions in the composite index. Utilising socio-economic data from Indonesia, this study calculates the social vulnerability index and compares it with the existing Social Vulnerability Index (SoVI) to identify deviations and improvements. To validate the applicability of this alternative calculation, the study applies the new index to real disaster scenarios, specifically focusing on flood events in Indonesia. The research findings demonstrate that the multidimensional index effectively measures all aspects of social vulnerability in a proportional manner. The research also identified several significant policy implications, including the prioritization of policies, targeted policy development, and customized policies for disaster risk management.

The climate policy domain is critical in addressing the multifaceted challenges posed by environmental degradation, socio-economic factors, and political dynamics. This thesis examines the Stimulering Duurzame Energieproductie (SDE(+)) policy in the Netherlands, evaluating its programmatic, process, and political success using an integrated framework combining dimensions of policy success and criteria for climate policy evaluation. Employing a qualitative research design, the study incorporates structured content analysis, stakeholder interviews, and secondary data analysis to provide a nuanced assessment.

The findings indicate that the policy's success has improved from the transition to and during the SDE+, showing both resilient and conflicted success. While the SDE+ policy has significantly boosted renewable energy production, particularly in solar PV and onshore wind, it has not achieved the 2020 targets and challenges like grid capacity strain and socio-economic inequalities remain. The evaluation highlights the need for a more equitable distribution of burdens and benefits and internalising external costs such as net congestion into the competitive subsidy allocation process.

This comprehensive analysis contributes valuable insights into climate policy effectiveness, supporting the development of more resilient and equitable renewable energy policies in line with international climate targets. Using a new climate-specific evaluation framework, this thesis provides a holistic and nuanced evaluation of the SDE(+) case study, demonstrating its applicability to climate policy. The study underscores the importance of integrating multiple dimensions of policy success to address the complex challenges of climate governance.

The Urban Transportation Network Design Problem is an important challenge in urban planning that aims to design effective transportation networks in urban environments. This research addresses three significant gaps in the existing literature. First, much of the research has remained focused on traditional objectives, such as minimizing costs and travel times or maximizing satisfied travel demand. This study, however, shifts the focus to optimizing for accessibility, thus explicitly incorporates equity into the transportation planning process. Second, while most studies concentrate on designing new networks or enhancing uni-modal networks, this research focuses on existing multi-modal networks, offering a more realistic approach to transportation planning. Finally, the study applies a novel approach to the Urban Transportation Network Design Problem through the use of Reinforcement Learning, specifically Tabular Q-learning.

The research methodology is structured in three parts. We begin by operationalizing equity through accessibility to employment opportunities. This metric is transformed using three notions of justice formalized as Social Welfare Functions: utilitarian, Atkinson, and lowest quintile. The second part involves gathering, cleaning, and integrating data to construct and simplify a comprehensive spatial model of the urban area, forming the foundation for subsequent optimization. The final part applies Reinforcement Learning, using the previously defined equity metrics and spatial model. The model is trained across different equity-driven reward functions to generate optimized network expansions.

A case study in Cape Town, South Africa, demonstrates the practical application of this approach, chosen for its its complex multi-modal transportation network and significant socio-economic disparities. The study's results reveal that each notion of justice leads to radically different expansion generations in entirely separate regions of the urban area. The results indicate that a traditional, utilitarian reward function leads to maintaining existing the existing distribution of accessibility, while the Atkinson reward function improves access for disadvantaged central regions by connecting these vulnerable areas to the Central Business District. The lowest quintile approach focuses on improving access for the most disadvantaged regions, particularly in the central southern areas, ensuring that the expansions provide maximum benefit to areas with the poorest access. The study also highlights the limitations of Tabular Q-learning in handling large state-action spaces and sparse rewards, suggesting the need for Deep Reinforcement Learning methods in future research.

Overall, this research contributes to the growing body of literature on equitable transportation planning by providing a versatile framework that can be adapted to other urban contexts. The study concludes with recommendations for future research and practical applications in urban planning, advocating for the continued exploration of equity-focused approaches in the design and expansion of transportation networks.

Transportation systems are pivotal in shaping the economic and social dynamics of contemporary societies, fostering connectivity and opportunities while reducing geographical distances. Despite these benefits, they also contribute to adverse effects such as emissions, congestion, and traffic fatalities. Effectively developing and maintaining transportation infrastructure and services that cater to evolving population needs and align with environmental goals requires accurate forecasting of travel demand. However, due to inherent uncertainty in individuals' behavior and data limitations, forecasting this demand is a complex task.
A common limitation often encountered in transport datasets is class imbalance, with regard to the utilization of the different modes. Class imbalance in this context refers to the uneven distribution of samples among the various modes. Modes with a higher number of samples are termed majority modes, while those with fewer instances are labeled as minority modes. The existence of class imbalance within the dataset has the potential to compromise the performance of classifiers, especially for the minority modes, leading to inaccurate forecasts. This, in turn, may result in insufficient investments and provisions for these modes, ultimately having adverse consequences for the population segments that rely on them. Existing studies in the literature have either entirely overlooked or only partially addressed the impact of class imbalance. Recognizing the significance of precise demand predictions and acknowledging the identified gaps within the literature, the primary research question of this study revolves around systematically identifying and addressing the impact of class imbalance in mode choice forecasting.
To address the main question, a framework was proposed. This framework encompassed various aspects including a) the measurement of class imbalance within a dataset and the assessment of its impact on classification performance, b) the investigation of other challenging factors coexisting in imbalanced datasets, with a specific focus on class overlap, and c) the proper evaluation of classification performance across classes. As an integral part of this framework, the 'Performance Gap Metric’ was introduced - a metric employed to evaluate the difference in classification performance between the majority and minority classes. Establishing a threshold of 20%, favorable classifier performance was determined when this metric fell below the threshold, signifying the classifier’s equitable treatment of both minority and majority classes. Subsequently, this framework was applied using the ODiN data as a case study to predict mode choices in the Netherlands. Mode choices encompassed car, bike, and transit, with car representing the majority and transit the minority class. Two modeling techniques, namely Random Forest and an MNL model, were employed in conjunction with various sampling techniques, including the SMOTENC, the Neighborhood-based Undersampling, and the Separation scheme...

The integration of variable renewable sources and electrification is straining electricity grids worldwide, resulting in congestion events. This research aims to understand the consequences of unmet electricity demand caused by demand congestion, to support informed decision making. Using a mixed methods approach, the study focuses on the medium voltage distribution network in Amsterdam, examining the implications of congestion for different customer segments. The research finds that demand congestion leads to a lack of universal access to electricity, as it becomes spatiotemporal-specific and dependent on the type of connection required. The study identifies industry, commercial business, public services, and urban development as key segments with varying responses to congestion. The societal impacts of congestion hinder sustainable development across environmental, social, and economic dimensions. CO2 emissions, housing delays, and employment losses are highlighted as significant impacts in Amsterdam. To mitigate these impacts, it is recommended to foster collaboration among stakeholders within the multi-actor network and make segment-specific as well as area-targeted investments. The research establishes a foundation for further studies on demand congestion and offers insights for effective congestion management in urban areas.

Due to climate change, there is a push toward a reduction of greenhouse gasses. Through spatial planning, the province of South Holland plays a role in this process. However, this requires insights into relation between the energy system and the built environment that is currently lacking. This thesis shows how spatial clustering using the K-prototype algorithm can be used to find areas with similar characteristics. It also describes how this information can be used in spatial planning by decisionmakers like the province of South Holland.

We live in a society where our decisions are increasingly based on research, data, and modeling. The role of research institutions like Delft University of Technology (TU Delft) is to both educate future engineers and to produce research outputs that support our data-driven society. However, research is an ongoing way of regulating and realizing the underlying codes of colonialism (Smith, 1999). Universities are complicit in perpetuating the asymmetrical power relationships that exist in society by occupying a role of “well-intentioned post-colonial saviors” (Anthony-Stevens & Matsaw Jr, 2020). This positioning is embedded in TU Delft’s strategic vision and mission to educate socially responsible engineers who can contribute to solving global problems in collaboration with societal actors. If we claim to value inclusion and diversity within our institution, then we must also critically consider our role in upholding colonial systems. To decolonize our university is to engage in an ongoing process of 1) understanding the lasting impacts of colonization on our educational systems and research practices; and 2) actively uplifting the voices and perspectives of those who have historically been (and continue to be) excluded and underrepresented in academic spaces. The intention to make our research and education inclusive must be accompanied by concrete action. This thesis attempts to redefine meaningful inclusion within the data science community at TU Delft through a lens of decolonization and social justice. I suggest ways to make inclusion actionable by highlighting how the community is already providing the conditions necessary for meaningful inclusion. I also point out how our academic culture and bureaucracy need to shift to allow for greater participation. Finally, I reflect on how our mindset needs to change so that we (specifically researchers already within academia who already possess a certain amount of privilege) can take greater responsibility to support inclusive community.

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.

Over the last couple of decades, refugee flows have been telling us a story of increasing numbers that are unlikely to go down anytime soon. The region of Eastern Africa only is home to 2.7 million people finding refuge in refugee settlements that are not often associated with permanency. UNHCR handbooks and many competent authorities state that refugee camps are of temporary character and as a result we observe a significant similarity in post-disaster food assistance and food assistance in a quasi-permanent settlement. The subsequent disparity between the way of organising food distribution and the needs of the population is the main motivation for the development of Food ATMs. This novel method of food distribution is based on allowing beneficiaries to determine how food is withdrawn to suit their own lives. A key point on how beneficiaries will use the Food ATM is a population's perception of the Food ATM. For instance, alleged food shortages might move people to hoard food and consequently undermine whole-system performance. Assumed similarities in literature on panic-buying events suggest that a combination of behavioural factors is a key vulnerability. Since a system of Food ATMs has not been implemented before, a conceptual design was made by 1) acquiring information through expert consultation on the design of an individual Food ATM and 2) the creation of an optimisation model forming an optimal network of Food ATMs. In order to analyse the interaction between a system of Food ATMs and beneficiaries, the current set of Operation Research (OR) models with stochastic or game theory-based formulation of food demand had to be expanded. For this purpose, concepts from research on panic-buying and emotion-related social interaction were used. This research contributes with the union of these concepts in a geo-spatially explicit Agent-Based Model, replacing current OR models' top-down formulation of demand with an agent-based bottom-up formulation. This enabled 1) increasing the understanding of interaction between a supplier of recipient of food assistance and 2) analysis of proposed operational policy in the context developing the Food ATM. Following 2), this thesis provides policy recommendations that have been made based on subsequent quantitative and qualitative investigation of model results. This thesis represents a first step to investigate the effects of individual behavioural factors in the realistic context of improving food distribution in the Bidi Bidi refugee camp in Uganda. New insights taught the relevance for further research, most notably on the relation with the bigger humanitarian supply chain, information diffusion between beneficiaries and the strategy for a population's adoption of the Food ATM.

The focus of this work is the creation and application of an experimental framework that tests the effectiveness and the impacts of micro-targeted ads. Micro-targeted ads have been used extensively by both political and retail carriers to sway peoples’ opinion, but there haven’t been any academic studies (in an environmentwhich is not controlled) that explore the impacts and the effects of micro-targeted ads. To this end the idea is to use a measurable variable (in this case the misplaced trash in The Hague) and throw generalized (non micro-targeted) ads along with micro-targeted ads, both of them aiming to reduce misplaced trash-bags. The ads will be sent to specific neighborhoods in The Hague through Facebook. The ad itself is considered a "nudge"(an intervention that hints towards behavioral change without forcing it).

To address the increasing passenger demand in the coming years and make public transport less crowded and delayed, insights into predicted passenger flow are needed. A wide range of studies has used and validated that smart card data can be one of the sound bases for predicting short-term passenger demand. However, it also has several disadvantages, such as the relatively long collection time, the insufficiency to reflect the relationship between passenger behavior and ridership. Trip planner data, which emerged as a type of real-time transit information, could reduce the perceived waiting time of passengers and increase the transit ridership due to the improved satisfaction. Combining these two types of data could potentially cater to the interest of operators in matching the vehicle supply and passenger flow demand at an operational level. Our results show that it is novel and useful to incorporate trip planner data in short-term ridership prediction, however, entirely based on this kind of data would be inaccurate. Random Forest Regression outperforms the other six models that we have selected. The request-related features (variables) can take up 20% of the importance of short-term ridership prediction.

In this study, a data-driven, generic and transfer-based methodology for separation and ranking the PTNs has been put forward. With the hierarchy of a network, this is beneficiary for the management and operation of operators for focusing on the higher level network layer and in turn provide better service for passengers. The study introduces three steps to rank the hierarchy of a PTN: (1) using the passenger journey and ride data to derive transfer flow matrix; (2) applying C-space network representation with community detection method to separate and visualize the PTN layer; (3) performing ranking method, regarding inner- and intra- transfer flow. To this end, the hierarchy of a PTN could be presented with temporal attributes. Different day of week and various time period of a day could potentially yield different hierarchy. The proposed unsupervised learning algorithm is based on passenger transfer flow data, independent from geographic location and the mode of transportation. The study shows that the level is changing based on the selected time slot and can be a mixture of different modes, which is dissimilar from the hierarchy purely based on qualitative method.