Natural disasters, particularly floods, pose a significant threat to cities and communities, and their fre- quency and severity are increasing due to climate change. Data from the EM-DAT database reveals that floods have been the most frequent natural disasters since 2002. Projections from the Intergovern- mental Panel on Climate Change suggest that heavy rainfall events will continue to increase, making floods more likely in the 21st century.
There is an urgent need to address the socio-economic impact of these disasters, particularly on the most vulnerable populations. Economic losses not only disrupt efforts to mitigate climate change, but also worsen inequalities, especially between high and low-income groups. Traditional risk assessments are based solely on property losses. Previous research highlighted that this is not sufficient to measure the socio-economic impact of floods. An interdisciplinary approach is required to integrate methodolo- gies, improve representation of vulnerable populations and fully understand the complex challenges posed by poverty and climate change.
This study uses a modelling framework applied to a case study of Charleston, South Carolina. For the case study, an extreme hurricane was chosen to test the model. The framework aims to measure well-being losses due to flooding and to evaluate the impact of insurance policies on vulnerable groups. The well-being losses are derived from household consumption levels.
Results from the case study show that total well-being losses exceeded total asset losses. Absolute well-being losses were similar across income groups. The relative well-being losses, represented per dollar of damaged assets, were higher for the low-income groups. The implementation of insurance policies reduced overall well-being losses. When the scenario with full insurance coverage is in force, there are still well-being losses. In order to create complete resilience, it is necessary to implement multiple measures.
Further research is recommended to improve the modelling framework. Many assumptions increased the uncertainty of the model. Detailed and additional data would improve the certainty of the results. Ultimately, the research aims to provide useful evidence for policy makers and stakeholders to make more informed decisions to build resilient communities and reduce inequalities in the face of climate change.

In 2050, over 80% of the worldwide electricity demand is expected to be supplied by renewables. The mismatch between supply and demand resulting from these intermittent energy sources and the physical limits of existing electrical grids are challenges in this transition, for which batteries are part of the solution. Therefore, there is a renewed interest to develop advanced and environmentally sound batteries, which requires assessing their environmental impacts by means of life cycle assessment (LCA). However, in current LCA studies the use phase is insufficiently addressed and even oftentimes excluded due to complexity. The aim of this research was to gain insight into current approaches of modelling the use phase in existing LCA and footprinting studies to provide LCA practitioners with recommendations and improved approaches. A literature review was performed which included 26 papers, Annex II of Regulation (EU) No 2019/1020 and the PEFCRs for High Specific Rechargeable Batteries for Mobile Applications. Next to storing renewable energy, batteries can serve different services, also called applications. The implications of incorporating the utilisation of a battery for multiple applications simultaneously, i.e., value stacking, in modelling the use phase are discussed in a qualitative way, since this is emerging as a practical and economically beneficial operational strategy. Finally, the relative effect on a battery’s life cycle impact assessment (LCIA) scores of four issues identified in the literature review is analysed in an illustrative case study about an organic redox flow battery (ORFB).

It appears that many studies do not provide clear information on how the functional unit (FU) is specified, which application(s) the battery is utilised for, application characteristics, modelling assumptions including the electricity and battery inputs or complete LCI data. Overall, the degree of transparency of many battery LCA studies is mediocre which complicates judging the usefulness of results and should be improved to improve comparability and reproducibility for which recommendations are provided. Moreover, the interaction of battery parameters and application characteristics is captured in proposed modelling guidelines for the electricity and battery system input flows. Value stacking results in environmental benefits, particularly when a battery is used to store renewable electricity which is used to serve another application simultaneously. It seems only interesting for battery technologies with high cycle lives such as RFBs and some lithium-ion batteries because these offer the ability to increase battery utilisation without considerably decreasing the battery’s lifetime.

To reach sustainability ambitions, battery applications leading to a reduction in environmental impacts should be promoted for which a general incentive policy is not appropriate. Such policy stimulates all battery applications, which could lead to small or even negative contributions to environmental impact reduction compared to the current situation. Even though this is a temporary transition problem, it could lead to an undesirable interim increase of environmental impacts during the transition. To this end, performing comparative assessments of applications that are expected to be served by batteries in the future, requiring the involvement of transmission network operators, and how these are served in the current situation are useful.

Purpose: This paper explores the potential of machine learning (ML) algorithms to mitigate uncertainty in early environmental assessments (ex-ante LCA), which are hindered by prospective nature and limited quantitative data availability. Methods: A systematic literature review with keyword searches on Scopus identified three ML categorization groups in ex-ante LCA: streamlined LCA, ex-ante LCA parameter projection, and ancillary models and data. Two following case studies addressed literature gaps in price forecasting for economic allocation and recycling rate projections. Results: In streamlined LCA, 16 studies linked molecular and technical parameters to project production-related emissions of organic chemicals, applied product clustering of product groups, and generated spatially explicit impact category results. The application of ex-ante LCA parameter projection, as evidenced by 14 publications, involves the use of ML to project life cycle inventory (LCI) data, project characterization factors, and integrate natural parameters with LCI data in a comprehensive modeling approach. In nine other papers the applications to ex-ante LCA remained undefined but potentially applicable. For both case studies, best results were obtained with a Recurrent Neural Networks (RNN) algorithm with long-short-term-memory (LSTM). Commodity price forecasting in the first case study achieved a projection accuracy of 0.96 (MSE), 0.98 (RSME), and 10.17% (MAPE) for copper and 88.86 (MSE), 9.43 (RMSE), and 21.23% (MAPE) for molybdenum. Probability modelling is identified as a modeling approach which incorporates uncertainty. The recycling rate forecast case study identified plastic recycling and glass recycling rates as the best suiting covariates and demonstrated multivariate modeling possibilities with 0.22 (MSE), 0.48 (RSME), and 0.38% (MAPE) in a model with 68 covariates. Discussion: A limited yet growing body of literature indicates that ML applications in ex-ante LCA represent an emerging field of science. While streamlined LCA shows promise, it faces constraints related to data precision and a static nature. In the ex-ante LCA parameter projection categorization, the sub-group of similarity clustering of LCI processes suffers from data uncertainty in LCI databases, making the approach more suitable for updates of existing technologies than for emerging ones. On the other hand, LCI generation through ex-ternal parameters represents a highly technology-specific case, showing significant promise. The projection of characterization factors and the sub-group of integrated modeling are identified as promising, but the limited number of scientific studies hinders the generalizability of these findings. Case studies on price forecasting and recycling rate projection demonstrate ML’s applicability in economic allocation and waste treatment projections. Overall, the results suggest that ML holds potential for reducing uncertainty in ex-ante LCA, laying the groundwork for focused research and contributing to a nuanced understanding of uncertainty reduction in this domain. Recommendations: The paper emphasizes the need for targeted research in the goal and scope phase and in End-of-Life (EoL) treatment forecasts, e.g. via the use of time-series multivariate modeling. Furthermore, it encourages further exploration of streamlined LCA into applications with a high degree of technical predictors, along with the extended projection of characterization factors and integrated modeling. Additionally, the use of probabilistic modeling as a tool to incorporate uncertainty into the modeling is recommended, aiming to enhance the applicability and transparency of ML applications for reducing uncertainties in ex-ante LCA.

Energy storage systems can play a vital role in the implementation of a sustainable energy grid by allowing the wide scale implementation of fluctuating renewable energy sources like wind and solar. Therefore it is expected that in the coming years, the total amount of energy storage systems will drastically increase. Analysing the sustainability
of these storage systems before widespread implementation is therefore of great importance. In this thesis such a life cycle assessment has been conducted to assess the environmental sustainability of one type of energy storage system: the Organic Redox Flow Battery produced by the Baliht research group.

To meet global market demands, it will remain important to further scale up photovoltaics (PV) production. During the production of solar cells, several defects can occur. Current approaches in quality inspection are reaching their speed limits. This thesis project evaluates the feasibility of faster quality inspection by using deep learning-based computer vision (CV) algorithms to detect production defects without human supervision at high speeds. The goal is to achieve this while reducing the necessary manual efforts to label (annotate) defects in the training data of such algorithms.

The second goal of the project is to investigate in which ways and to which extent this innovation can impact the sustainability performance of the solar cell production process. Multiple scenarios are investigated using a Life Cycle Assessment (LCA) model. The results are used to estimate the potential large-scale impact of increasing solar cell production throughput.