Theses authorised for defence

DOCTORAL DEGREE IN ARCHITECTURAL DESIGN

• NARANJO SERRANO, MÓNICA GABRIELA: Espacios públicos no programados. Contextos de descompresión y densificación urbana
  
  Author: NARANJO SERRANO, MÓNICA GABRIELA
  Programme: DOCTORAL DEGREE IN ARCHITECTURAL DESIGN
  Department: Department of Architectural Design (PA)
  Mode: Normal
  Deposit date: 17/04/2026
  Reading date: 09/06/2026
  Reading time: 11:30
  Reading place: ETSAB (Escuela Técnica Superior de Arquitectura de Barcelona) - Planta Baja - Sala de GradosAv. Diagonal, 649-651 - 08028 - Barcelona
  Thesis director: GASTON GUIRAO, CRISTINA | SICA PALERMO, NICOLÁS
  Thesis abstract: Starting from the premise that the city is not built as a sum of isolated objects, but rather as a complex system of spatial, historical, and social relationships, this work investigates the role of architecture in the generation of Unprogrammed Public Spaces (UPS) as fundamental components of the contemporary urban fabric. This research focuses on those open spaces that, without having been explicitly conceived as public space within architectural programs or urban regulations, acquire a collective character through design decisions linked to site occupation, volumetric configuration, and the building’s relationship with ground level.This comparative study is based on the analysis of two urban axes: Dearborn Street in Chicago and Avenida Paulista in São Paulo. Both cases make it possible to observe how modern architecture, in different urban and cultural contexts, was capable of producing significant urban voids that expanded the public dimension of the city. While in Chicago architectural strategies were oriented toward introducing spaces of relief within a highly consolidated fabric, in São Paulo they responded to the need to structure and densify an urban landscape in formation.The study of Dearborn Street, as the main case, is supported by a historical reconstruction that shows how, despite repeated reconstruction processes following the fire of 1871, the city maintained rigid morphological parameters for decades—alignment to the building line, fragmented parceling, and full lot occupation—which were only challenged with the insertion of modern projects in the mid-twentieth century. The emergence of buildings such as Inland Steel, the Civic Center, the First National Bank, and the Federal Center enabled the formation of a sequence of plazas and open spaces that, far from being isolated gestures, configured a continuous system of unprogrammed public spaces along the street axis.In São Paulo, the analysis of six buildings reveals a different yet conceptually related approach. Strategies such as the tower-on-podium typology, the liberation of the ground plane, and the perforation of the ground floor allowed not only the creation of open spaces at ground level, but also a volumetric release in height, introducing air, light, and urban continuity into a city marked by accelerated processes of densification. The buildings studied—Quinta Avenida, Pauliceia, Nações Unidas, Conjunto Nacional, Banco Sulamericano, and MASP—demonstrate how these operations gave rise to UPS with a high capacity for permanence, appropriation, and collective representation.Based on the comparative analysis, recurring mechanisms—historical, spatial, morphological, and social—are identified as intervening in the configuration of UPS. It is demonstrated that these spaces are neither residual nor collateral effects of the architectural project, but rather the direct result of conscious decisions that allow architecture to yield space to the city or, alternatively, to allow the city to extend into the interior of the building. Likewise, it is shown that UPS generate itineraries of pauses and movement within the apparent monotony of the urban fabric, acting as articulators of the pedestrian experience.It is concluded that, although there is no single model of UPS, universal strategies can be identified whose adaptation is always local and context-specific. In contemporary scenarios characterized by a scarcity of available public land, UPS acquire a strategic role in the construction of the city. Finally, this work seeks to contribute a conceptual and methodological framework that makes it possible to identify, analyze, and design UPS as part of a collective project for the city, in which architecture consciously assumes its public and urban responsibility.
  

DOCTORAL DEGREE IN ARCHITECTURAL, BUILDING CONSTRUCTION AND URBANISM TECHNOLOGY

• MARTIN FANJUL, VALENTIN: Arquitectura y autismo: consideraciones acústicas para el desarrollo de espacios específicos
  
  Author: MARTIN FANJUL, VALENTIN
  Programme: DOCTORAL DEGREE IN ARCHITECTURAL, BUILDING CONSTRUCTION AND URBANISM TECHNOLOGY
  Department: Department of Architectural Technology (TA)
  Mode: Normal
  Deposit date: 25/03/2026
  Reading date: 08/06/2026
  Reading time: 11:00
  Reading place: ETSAB (Escola Tècnica Superior d'Arquitectura de Barcelona) - Planta Baixa - Sala de GrausAv. Diagonal, 649-651 - 08028 - Barcelona
  Thesis director: LACASTA PALACIO, ANA MARIA | DAUMAL DOMENECH, FRANCESC DE PAULA
  Thesis abstract: This doctoral thesis combines acoustics, autism and architecture with an unequivocal practical focus, so that the conclusions and possible future avenues of research it may generate can be used in specific spaces for people on the autism spectrum.Existing knowledge is recent and scarce, it is mainly based on the experience of technicians who have created specific spaces, whether newly built or adapted. This fact is further emphasised when the disorder itself adds the concept of spectrum, which implies very broad and diverse conditions that further hinder the ability to obtain generalised solutions.Another factor that has influenced this search is the difficulty for potential users to convey their acoustic needs, especially those with more severe disorders, which is why the research has been structured in two ways.The first is a quantitative survey aimed at technical managers and carers of people with ASD, consisting of a set of 10 questions, which was sent to 715 valid email addresses in 30 countries and in three languages: English, French and Spanish.The second is qualitative in nature and is organised through the study of scientific texts, visits to specific centres, relationships with autistic people and their families, as well as professional experience with this group.It has been argued that venues should be designed primarily for those who suffer from hyperacusis, but studies have not been conclusive about how this condition evolves over the course of their lives. For this reason, it has been recommended that acoustic properties be maximised during childhood and youth, when there is no doubt about its prevalence.A significant factor found is that acoustic interventions should prioritise high-pitched sounds over low-pitched ones, as they are more annoying. For this reason, acoustic solutions were presented using existing construction systems based on the data obtained.The state of the art recommends solutions to improve sound environments, including the use of carpets, but after studying other disciplines, it has been found that they should not be used despite their acoustic properties. During visits to specific centres, those in charge reported acoustic problems relating to activities in greenhouses and swimming pools, which is why interventions based on existing construction solutions have been developed.The studies carried out have detected an anomaly between whether high-pitched or low-pitched sounds, depending on the language of the speakers, cause more discomfort. While for English and Spanish it was the high-pitched sounds, for French it was both, but the limitations of the sample obtained do not allow these findings to be irrefutable, although they do generate a line of research to consider the language of people with ASD as a factor in acoustic design.The complexity of acoustic design for autistic people has been demonstrated, as evidenced by the modifications made in the specific centres consulted. Collaboration in its design is necessary between all those involved, including family members, technicians and the staff responsible for their care. Acoustic quality is not exclusive to autistic people; research that improves their sound environments will also improve those of neurotypical people.KeywordsAcoustics, autism, architecture, ASD, acoustic quality, sound environments, design and construction solutions.
  
• XIE, ZEYUE: Development of nonwoven-reinforced ECCCs and multifunctional smart textiles for energy harvesting and thermal management in buildings
  
  Author: XIE, ZEYUE
  Programme: DOCTORAL DEGREE IN ARCHITECTURAL, BUILDING CONSTRUCTION AND URBANISM TECHNOLOGY
  Department: Department of Architectural Technology (TA)
  Mode: Normal
  Deposit date: 06/05/2026
  Reading date: pending
  Reading time: pending
  Reading place: pending
  Thesis director: ARDANUY RASO, MONICA | VENTURA CASELLAS, HEURA
  Thesis abstract: The escalating demand for energy efficiency in modern infrastructure has catalyzed the development of smart building materials capable of active and passive thermal management. While smart textiles and Electrically Conductive Cementitious Composites (ECCCs) offer promising solutions, their practical application is currently hindered by the dispersion limitations of traditional fillers, the mechanical-electrical trade-off, and the lack of multifunctional integration. This thesis addresses these challenges by developing a multifunctional smart fabric and novel nonwoven-reinforced ECCCs, which could culminate in a synergistic, all-weather thermal management system for intelligent building interiors. First, this work addresses the discrepancy in fabric conductivity characterization by establishing a structure-based evaluation framework. Instead of using the same testing method for all materials, contact and non-contact techniques are chosen according to whether the fabric exhibits intrinsic or extrinsic conductivity and its conductivity range, leading to more reliable measurements for different fabric types. Secondly, a multifunctional smart Cotton/Polydopamine/Copper Sulfide/TPU (CO/PDA/CuS/TPU) fabric was fabricated for smart applications. This advanced material integrates anisotropic conductivity, active Joule heating, passive photothermal conversion, and thermoelectric energy harvesting into a single unit. The fabric exhibited rapid active heating (up to 72 °C) and efficient photothermal conversion under solar irradiation. Furthermore, a thermoelectric generator integrated into the fabric demonstrated the capability to harvest energy (350 nW at ΔT = 50 K) and function as a self-powered temperature sensor. Thirdly, a novel ECCC reinforced with stainless steel/polyester (SS/PES) nonwoven fabric was engineered to overcome the agglomeration and loss in mechanical performance issues associated with particulate fillers. The study revealed that the bonding strength of the cement matrix significantly contributes to the generation of the conductive network. An optimal fiber content of 20% was identified near the percolation threshold, which achieved superior electrothermal efficiency compared to higher conductivity formulations (reaching a ΔT of 46.2 °C within 30 minutes at 21 V). The resulting composite demonstrated excellent mechanical properties, durability, and reliable de-icing capabilities. Finally, the functionality of these materials was validated through finite element simulation and experimental testing under varying temperature and humidity conditions. Based on these findings, a closed-loop synergistic ecosystem was proposed, combining the ECCC (as floor heating) and the smart fabric (as curtains). This system dynamically adapts to environmental conditions: utilizing passive photothermal heating and thermoelectric monitoring during the day, and switching to active Joule heating at night, where the curtains recover heat loss from the ECCC elements. This research provides a holistic solution for next generation smart buildings, demonstrating significant potential for energy conservation and intelligent environmental control.
  

DOCTORAL DEGREE IN ARTIFICIAL INTELLIGENCE

• CAZZARO, FRANCESCO: Advancing Text-to-Query Semantic Parsing Systems
  
  Author: CAZZARO, FRANCESCO
  Programme: DOCTORAL DEGREE IN ARTIFICIAL INTELLIGENCE
  Department: Department of Computer Science (CS)
  Mode: Normal
  Deposit date: 07/05/2026
  Reading date: 08/06/2026
  Reading time: 11:00
  Reading place: sala de juntes de la FIB, edifici B6 - Campus Nord UPC (Barcelona)
  Thesis director: QUATTONI, ARIADNA JULIETA
  Thesis abstract: Executable Semantic Parsing is the task of mapping a natural language sentence into a formal meaning representation that can be executed over a knowledge base to retrieve information. This task presents several challenges. Natural language variability makes the translation process inherently difficult and, at the same time, meaning representations are highly compositional structures built from elementary units. As a result, semantic parsers must exhibit strong compositional generalization, a capability that remains challenging for current models. Moreover, annotated data for semantic parsing is scarce, and its collection is labor intensive and costly, making it difficult to train models.In this thesis, our aim is to improve and advance semantic parsing systems. We introduce a pipeline that decouples the semantic parsing process into two steps, a translation stage and a reordering stage, which enhances the compositional abilities of the parser. We also propose a data generation method that recombines existing annotated pairs in novel ways. This approach improves the generalization capabilities of semantic parsers while alleviating the data scarcity problem.Furthermore, to address the challenge of limited annotated data, we design a novel approach to automatically generate data pairs from a given knowledge graph without the need of human intervention. This generated data can then be used to train a semantic parsing system specifically designed for that particular knowledge graph. This thesis is among the first to explore semantic parsing for property graphs, where we not only introduce the data generation method but also provide an annotated benchmark for evaluating parsing performance.
  
• HERNÀNDEZ CARNERERO, ÀLVAR: Generalizable Time Series Classification Models for Complex Real-World Applications
  
  Author: HERNÀNDEZ CARNERERO, ÀLVAR
  Programme: DOCTORAL DEGREE IN ARTIFICIAL INTELLIGENCE
  Department: Department of Computer Science (CS)
  Mode: Normal
  Deposit date: 23/04/2026
  Reading date: 08/06/2026
  Reading time: 12:00
  Reading place: Sala d'actes de la FIB - Edifici B6 planta 0, Campus Nord
  Thesis director: SANCHEZ MARRE, MIQUEL | VAZQUEZ SALCEDA, JAVIER
  Thesis abstract: This thesis develops a comprehensive framework for Time Series Classification (TSC) focused on real-world applications characterized by limited data availability, heterogeneous datasets, and complex temporal structure, which concludes with a generalizable and adaptive modelling approach. This research conducts a comprehensive survey of state-of-the-art methods, identifying their strengths and limitations, with a focus on specific domains such as antimicrobial prediction, astronomy, and physiological monitoring. Building on these insights, preprocessing pipelines are developed for the considered time series, including new temporally informed feature generation grounded in domain knowledge, robust feature importance estimation, and wrapper-based feature selection. Methodologically, windowing strategies are engineered to mitigate data scarcity and temporal distribution shifts, a hybrid modular architecture is proposed to integrate mechanisms suited to the specific needs of the data, and an adaptation of additive attention is formulated. The methodology employs ensembles to enhance predictive stability and evaluates model interpretability by analyzing how the trained models prioritize features and by visualizing attention patterns across the sequence to reveal temporal relevance patterns.A central contribution of the thesis is CLAIM (Content-modulated Low-rank Adaptive Isotropic Mixer), a novel architecture for TSC that synthesizes insights from our research and from recent advances in sequence modelling. CLAIM combines content-modulated dynamic mixing, adaptive inductive biases, and a low-rank isotropic block design, resulting in an efficient and interpretable model. Experiments across heterogeneous datasets (including intensive care unit patient trajectories, exoplanet transit signals, and electrocardiogram waveforms) show that CLAIM frequently achieves the best performance in small-data regimes compared to temporal transformer and Multilayer Perceptron Mixer (MLP-Mixer) baselines. CLAIM also offers practical advantages in interpretability, due to its dynamic weights and explicit inductive-bias components. In addition, it demonstrates strong computational efficiency, particularly at inference time. The model modulates structural priors such as locality, causality, and location invariance in a data-driven manner, aligning with effective contemporary design principles for temporal modelling. Furthermore, it adapts its focus across domains, as reflected in the dynamic weighting of temporal components, which corresponds with established domain knowledge.The thesis contributes a unified perspective on real-world TSC, pairing domain-guided pipelines with a generalizable modelling approach. The resulting CLAIM architecture offers an interpretable, efficient, and adaptive solution that advances the state of the art in small-data TSC.
  
• VILALTA ARIAS, ARMAND: Semantic Embeddings in Deep Convolutional Neural Networks
  
  Author: VILALTA ARIAS, ARMAND
  Programme: DOCTORAL DEGREE IN ARTIFICIAL INTELLIGENCE
  Department: Department of Computer Science (CS)
  Mode: Normal
  Deposit date: 26/03/2026
  Reading date: 15/06/2026
  Reading time: 11:00
  Reading place: Sala d'actes de la FIB - Edifici B6, Campus Nord Barcelona
  Thesis director: GARCÍA GASULLA, DARIO | CORTÉS GARCÍA, CLAUDIO ULISES
  Thesis abstract: One of the fundamental questions in Artificial Intelligence (AI) is how to represent knowledge effectively. The main challenge lies in devising a representation that can be exploited for intelligent computations. Since 2012, when Deep Neural Networks (DNNs) established themselves as the state of the art in Computer Vision, their use has rapidly expanded to a wide range of AI domains, including text and speech recognition, image generation, gaming, and robotics. This widespread success stems from their nature as representation learning techniques. However, DNNs require large volumes of data and considerable computational resources to learn such representations, which significantly limits the range of problems to which these models can be directly applied.In this thesis, we explore the possibility of reusing the knowledge learned by a DNN for a specific problem to solve a different one, which is commonly known as transfer learning. We focus on the representations learned by a DNN, codified in the activations of its neurons in response to an input, namely embeddings. All this work stems from a first analysis to understand where valuable information is encoded in Convolutional Neural Networks (CNNs) within the context of transfer learning. When used for characterizing every class of eleven datasets, we statistically measure the discriminative power of every feature found within a deep CNN. We seek to provide new insights into the behaviour of CNN features, particularly the ones from convolutional layers, which they had not used in previous literature. Our results confirm that low and middle level features may behave differently from high-level features, but only under certain conditions. We find that all CNN features can be used for knowledge representation purposes both by their presence or by their absence, doubling the information a single CNN feature may provide. We also study how much noise these features may include and propose a thresholding approach to discard most of it.This enables the definition of a methodology to improve the generalization capabilities of CNN representations, which we refer to as the Full-Network Embedding (FNE), which successfully integrates convolutional and fully connected features. To do so, the embedding normalizes features in the context of the problem and discretizes their values to reduce noise and regularize the embedding space. Significantly, this also reduces the computational cost of processing the resultant representations. The proposed method outperforms single layer embeddings on several image classification tasks while also being more robust to the choice of the pre-trained model used for obtaining the initial features. The performance gap in classification accuracy between thoroughly tuned solutions and the full-network embedding is also reduced, making the proposed approach a competitive solution for a large set of applications. We consider FNE more semantic as it understands CNN activations as semantic concepts. Similarly to many human languages, it considers if a feature of the input is significant in a given context and if it is because of its presence or absence.A first stem is the use of the FNE as the basis for a network representation of concepts where complex network techniques are applied. We propose the construction of a graph embedding space, introducing a methodology to transform the knowledge coded within a deep convolutional network into a topological space.The second stem, uses the same techniques in the context of multimodal embeddings representing text and images. The FNE provides a multi-scale representation of images, which results in richer characterizations while focusing on the relevant information through its semantic discretization. Results for image annotation and image retrieval tasks show a constant improvement when applied to different existing methodologies.
  

DOCTORAL DEGREE IN AUTOMATIC CONTROL, ROBOTICS AND VISION

• PÉREZ QUINTANA, MARC: Multimodal Data Fusion for Multiple Object Tracking: A Reference Perception System for ADAS and AV Functions Validation
  
  Author: PÉREZ QUINTANA, MARC
  Programme: DOCTORAL DEGREE IN AUTOMATIC CONTROL, ROBOTICS AND VISION
  Department: Institute of Robotics and Industrial Informatics (IRI)
  Mode: Normal
  Deposit date: 21/04/2026
  Reading date: 02/06/2026
  Reading time: 11:00
  Reading place: Sala d'Actes de la Facultat de Matemàtiques i Estadística (FME), Campus Diagonal Sud, Carrer de Pau Gargallo, 14, 08028 Barcelona
  Thesis director: AGUDO MARTÍNEZ, ANTONIO
  Thesis abstract: In 2021, 1.19 million people died from traffic accidents, which are the leading cause of death among people aged 5 to 29 years. Most of these accidents are caused by the driver; therefore, automated vehicles present a unique opportunity to reduce fatalities and improve road safety. But the large-scale deployment and adoption of safe automated vehicles require a robust validation procedure, including open-road tests, which are significantly more challenging than tests on proving grounds or in simulation, because there is no ground-truth data on the objects around the vehicle under test. This thesis studies how to build a reference perception system to enable the validation of automated vehicles and advanced driver assistance systems on the open road. We discuss object detection and present a clustering method to detect class-agnostic objects on point clouds. These detections can be projected to the image plane to be combined with image-based detections to improve robustness and add class information. We also present a method to exploit class prototypes, defined as the mean and covariance of the features of that class, to improve the performance of a learning-based object detector on point clouds. Object detections from different modalities can be combined in the presented multiple object tracking method, which works without assuming sensor synchronization and can be adapted to include object information from any source, such as vehicle-to-vehicle communications, and includes the handling of common errors that are understudied in the literature: Misclassifications and partial bounding box detections. These object detectors and multiple object trackers can be evaluated with a novel proposed methodology to evaluate perception systems focusing on safety, which we use to evaluate the effect on object detection performance of different weather conditions and the robustness layers included in recent testing protocols. We also present a pipeline to extract scenarios in standardized formats from recorded images and point clouds, that can be directly used by a simulator. Finally, we discuss how scenario extraction, combined with other tools, can help in the validation of automated vehicles on the open road.
  

DOCTORAL DEGREE IN BUSINESS ADMINISTRATION AND MANAGEMENT

• DÍAZ MEDIAVILLA, MARIA ESTELA: Estudio comparativo de la percepción del riesgo laboral entre diferentes colectivos dentro del sector de la construcción
  
  Author: DÍAZ MEDIAVILLA, MARIA ESTELA
  Programme: DOCTORAL DEGREE IN BUSINESS ADMINISTRATION AND MANAGEMENT
  Department: Department of Management (OE)
  Mode: Normal
  Deposit date: 05/05/2026
  Reading date: pending
  Reading time: pending
  Reading place: pending
  Thesis director: ABAD PUENTE, JESUS
  Thesis abstract: Objective — The main objective of this study is to empirically evaluate differences in risk perception between construction workers and health and safety managers. Additionally, it aims to determine the effect of work experience and accident experience (both direct and witnessed) on the rational and emotional dimensions of this perception.Methodology — A cross-sectional quantitative design was adopted using a psychometrically validated questionnaire. The final sample consisted of 349 professionals from the construction sector in Catalonia. Data were analyzed using principal component analysis and confirmatory factor analysis to validate the construct structure, followed by multiple linear regression models with interaction terms to test the study hypotheses.Results — The findings empirically demonstrate that, within the analyzed sample, risk perception has a three-dimensional structure composed of rational risk perception, emotional perception of moderate-severity risks, and emotional perception of high-severity risks. A significant safety perception gap was identified: health and safety managers exhibit higher levels of emotional perception regarding high-severity risks compared to workers. In addition, both work experience and accident experience exert different effects on risk perception depending on the dimension analyzed and the professional group considered.Conclusions — Risk perception in the construction sector presents a more complex structure than the traditional two-dimensional conception. The empirical evidence indicates that rational risk perception remains a single dimension, whereas emotional perception is divided into two distinct subdimensions depending on the severity of the potential harm: one related to moderate-severity risks and another associated with high-severity risks. Furthermore, perceptual differences between workers and health and safety managers emerge exclusively in the emotional response to high-severity risks. Finally, experiential factors influence risk perception differently depending on professional role, highlighting the complexity of the cognitive and affective processes involved in risk evaluation in hazardous work environments such as the construction industry.Implications — The findings suggest moving away from “one-size-fits-all” safety approaches. Safety management systems should incorporate training methodologies with a strong emotional impact, such as virtual reality simulations, to reduce workers’ tolerance to risk. For health and safety managers, it is advisable to implement recalibration mechanisms, such as cross-audits and project rotation, to mitigate the overconfidence bias associated with accumulated experience.Limitations — The study has a cross-sectional design, which prevents establishing strict causal relationships in the evolution of risk perception. Furthermore, the data are based on self-reported responses, which may differ from actual on-site perception. Finally, the research context is limited to the construction sector within a specific geographical area, suggesting the need to replicate the model in other high-risk industries.Future Research — Future studies should adopt longitudinal designs to examine how risk perception evolves following exposure to critical events. It would also be valuable to incorporate neurophysiological measurements to capture workers’ immediate emotional responses to risk. Additionally, future research could examine perceptual mimicry within work teams in order to better understand how organizational culture shapes individual risk tolerance.
  
• GAMARRA GAMARRA, MARÍA DEL PILAR: Comportamiento Ético: Identidad Moral, Atención Moral y su papel en la Construcción del Liderazgo Ético
  
  Author: GAMARRA GAMARRA, MARÍA DEL PILAR
  Programme: DOCTORAL DEGREE IN BUSINESS ADMINISTRATION AND MANAGEMENT
  Department: Department of Management (OE)
  Mode: Normal
  Deposit date: 30/04/2026
  Reading date: pending
  Reading time: pending
  Reading place: pending
  Thesis director: OLIVELLA NADAL, JORGE | GIROTTO, MICHELE
  Thesis abstract: When a leader acts without ethics, the entire organization is shaken; when leadership is exercised with integrity, it inspires sustainable cultures and strengthens social trust. This doctoral thesis addresses the challenge of understanding how leaders perceive and enact their own ethical behavior, focusing on two fundamental internal drivers: moral identity and moral attentiveness. These factors, often invisible in everyday organizational discourse, distinguish formal leadership from authentic leadership capable of generating legitimacy, resilience, and commitment in complex contexts.The research is grounded in a clear diagnosis: financial scandals, corporate misconduct, and growing social demands for transparency have placed ethics at the center of the business agenda. However, most previous studies have examined ethical leadership from the perspective of subordinates or within Anglo-Saxon contexts, leaving aside the leader’s own viewpoint and the analysis of specific cultural realities such as the Spanish context. This is the gap that the thesis seeks to address.The study is developed in two phases. The first presents a bibliometric analysis and a systematic review of more than thirty years of scientific literature, identifying three theoretical pillars of ethical leadership: values-based theories, cognitive moral development theory, and social learning theory. This review highlights the fragmentation of existing approaches and the scarcity of empirical studies focused on leaders’ self-perceptions, as well as the need to incorporate variables such as gender, age, and hierarchical level.The second phase builds an empirical model applied to the Spanish context. To this end, the classic ethical leadership scale is psychometrically validated and adapted to leaders’ self-perceptions through exploratory and confirmatory factor analyses, reliability tests, and structural equation modeling. The process includes expert review, pilot testing, and the administration of questionnaires to a sample of managers from different sectors. The result is a robust, context-adapted scale that identifies three dimensions of self-perceived ethical leadership: integrative decision-making, the reinforcement of ethical behaviors, and the leader’s personal example as a role model.The findings show that moral identity and moral attentiveness significantly influence these dimensions, albeit in different ways. The internalization of moral identity predicts the integration of ethical principles into decision-making, while the symbolization of moral identity and reflective moral attentiveness explain the tendency to reinforce ethical behaviors. In addition, variables such as gender, age, managerial experience, and hierarchical position act as moderators, indicating that ethical leadership emerges through the interaction of personal and professional characteristics.Overall, this thesis demonstrates that ethical leadership is not an abstract code or a superficial discourse, but a practice rooted in the leader’s internal coherence, reflected in decision-making, and projected onto organizational culture, contributing to the development of more just and sustainable organizations.
  

DOCTORAL DEGREE IN CIVIL ENGINEERING

• APARICIO URIBE, CARLOS HUMBERTO: EXPERIMENTAL AND NUMERICAL ANALYSIS OF FLOOD-RELATED HAZARD ASSOCIATED WITH THE EVACUATION OF UNDERGROUND STAIRS
  
  Author: APARICIO URIBE, CARLOS HUMBERTO
  Programme: DOCTORAL DEGREE IN CIVIL ENGINEERING
  Department: Barcelona School of Civil Engineering (ETSECCPB)
  Mode: Article-based thesis
  Deposit date: 05/05/2026
  Reading date: 12/06/2026
  Reading time: 12:00
  Reading place: UPC Campus Nord, ETSECCPB, C/ Jordi Girona 1-3, edificio C1, Sala 002, Barcelona
  Thesis director: RUSSO, BENIAMINO | TELLEZ ALVAREZ, JACKSON DAVID
  Thesis abstract: The rapid pace of urbanisation has led to the widespread development of underground infrastructure worldwide. Currently, these infrastructures have become increasingly vulnerable to climate-induced extreme weather events such as flooding, particularly in metro stations. Given the unique characteristics of urban underground clusters, stairs serve a dual and critical role during emergencies: they are often the primary evacuation route while simultaneously serving as the main entry path for the incoming floodwaters. This doctoral research explores the interaction between hydrodynamic forces and human evacuation on flooded stairs within underground environments. The study aims to achieve a better understanding of these interactions to identify stability thresholds, establish safety criteria, and develop effective evacuation strategies. A comprehensive approach to the problem was adopted by integrating both experimental and numerical methods.A thorough state-of-the-art review was conducted to identify the current knowledge landscape, highlighting key findings, limitations and potential further exploratory lines. The experimental component of this thesis was based on a full-scale replica constructed at the hydraulic laboratory of the School of Civil Engineering at the Universitat Politècnica de Catalunya – BarcelonaTech (UPC). This represented the flood-prone access stairs of the Paral·lel metro station, which is part of Barcelona’s Metro Network. The stairs’ hydraulic characterisation was followed by a series of experimental tests involving individual subjects walking the flooded stairs under varying discharge conditions and footwear types. Additional tests were assessed under the facility’s maximum discharge, with subjects simultaneously walking the flooded stairs in groups of two and three. A first individual subject's campaign aimed to identify thresholds for pedestrian instability, while a second one intended to optimise evacuation based on the different studied grouping configurations.To complement the experimental observations, several three-dimensional Computational Fluid Dynamics (3D-CFD) simulations were performed using the commercial software FLOW-3D. These simulations analysed the water-dragging force considering single subjects and pedestrian arrangements, providing a detailed identification of hydraulic parameters and water forces throughout the stairs’ domain. This research offers novel insights into flood risk management in urban areas, taking into account their site-specific constraints, particularly access to underground spaces via stairs. Special attention is given to subjects evacuating simultaneously, as well as optimal strategies for managing group evacuations. The outcomes offer valuable, actionable data for engineers, urban planners, and emergency responders. In this way, it will ultimately contribute to improving evacuation safety and hydraulic understanding of flooded underground stairs through the integration of experimental and numerical analysis, thereby fostering urban resilience and evacuation planning in flood-prone underground infrastructures.
  
• LOPEZ CHACON, SERGIO RICARDO: Enhanced short-term prediction of high streamflow combining physically based and machine learning models
  
  Author: LOPEZ CHACON, SERGIO RICARDO
  Programme: DOCTORAL DEGREE IN CIVIL ENGINEERING
  Department: Barcelona School of Civil Engineering (ETSECCPB)
  Mode: Article-based thesis
  Deposit date: 21/04/2026
  Reading date: 11/06/2026
  Reading time: 12:30
  Reading place: UPC Campus Nord, ETSECCPB, C/ Jordi Girona 1-3, edificio C1, Sala 002, Barcelona
  Thesis director: BLADE CASTELLET, ERNEST | SALAZAR GONZÁLEZ, FERNANDO
  Thesis abstract: Machine learning models have demonstrated a strong potential in the recent decade for streamflow prediction purposes. Even reaching considerable accuracy, machine learning models still present some limitations and little explore aspects on this topic: the accuracy decrease on high streamflow, uncertainty estimation applications, and hydrological interpretation of models. High streamflow values are the most relevant for early warning systems of flood mitigation. However, these records are scarce in the data. Hence, a decrease in accuracy is seen, which gets deeper in extrapolated scenarios. This thesis proposes two methodologies that support machine learning models with the outputs of a physically based model to enhance the prediction capabilities in high streamflow prediction. The first methodology produces a machine learning model trained with a combination of observed and synthetic high streamflow events generated by a physically based model. The second methodology creates a hybrid model that combines outputs based on a physically based model and the prediction of its residuals by a machine learning model. Both methodologies have shown accuracy improvements compared to models trained only with observed data, reaching reductions of root mean square errors of more than 23% for streamflow values larger than the 3-year return period in the study area. However, the second methodology reaches closer approximations to the peak observed value with significant extrapolation capabilities. Despite the accuracy, the uncertainty related to the model is a main topic to explore. A methodology to estimate the uncertainty of the output of a streamflow prediction model by employing machine learning techniques is developed in the thesis. The results of the methodology show that the distributions of the residuals can be acceptably described. Consequently, the uncertainty interval covers 87.9% of the observed values higher than the 3-year return period with a width of interval significantly smaller than the broadly used Box-Cox method. Finally, the hydrological interpretation of a machine learning model for streamflow prediction is undertaken. The results show that the model may identify areas of the catchment whose runoff considerably contributes to the control point, as well as period of previous soil saturation, and the impact of the highest precipitation values in the model’s prediction. When the hydrograph rises steeply, the model acceptably considers the recent accumulated precipitation as the most relevant features along with previous saturation. As the hydrograph peak approaches and during the falling limb, previous streamflow acquires main relevance supported by runoff of distant regions. The machine learning model can suitably interpret the catchment system and provide valuable information.
  

DOCTORAL DEGREE IN COMPUTATIONAL AND APPLIED PHYSICS

• JURADO ROMERO, ARNAU: Self-Propulsion of Molecular Swimmers
  
  Author: JURADO ROMERO, ARNAU
  Programme: DOCTORAL DEGREE IN COMPUTATIONAL AND APPLIED PHYSICS
  Department: Department of Physics (FIS)
  Mode: Normal
  Deposit date: 28/04/2026
  Reading date: 28/05/2026
  Reading time: 11:00
  Reading place: Sala de Juntes de la FIB
  Thesis director: REY ORIOL, ROSENDO | CALERO BORRALLO, CARLOS
  Thesis abstract: Active matter systems have the ability to exhibit self-propulsion by consuming energy to produce mechanical work, staying out of equilibrium. They occur on a vast range of scales, from herding mammals and flocking birds down to bacteria. Understanding these systems requires the study of the mechanisms that drive the system out of equilibrium and the emergent collective phenomena that result from activity. Both these questions require a multidisciplinary approach, the innovative application of classical non-equilibrium physics, as well as the development of new theoretical frameworks.A major goal in the field is the development of artificial micro- and nano-scale particles capable of self-propulsion, so-called swimmers. These swimmers hold the potential for groundbreaking applications, such as targeted drug delivery, non-invasive microsurgery, and water purification.This overarching technological goal is met with a number of theoretical and practical challenges, of which the search for viable propulsion mechanisms at the nano-scale is only one of them. However, achieving propulsion at such small scales is limited by the hydrodynamic regime characteristic of these scales, in which reciprocal deformations cannot produce net propulsion. In addition, rotational diffusion, which severely hinders the self-propelling capabilities of a swimmer, increases rapidly with diminishing size.This thesis demonstrates a propulsion mechanism for a small molecule, nitromethane, via all-atom molecular dynamics simulations. The molecule is subject to a high energy vibrational excitation which is then released anisotropically onto the surrounding water solvent. This results in propulsion velocity bursts that are able to enhance the translational diffusion of the molecule. Making use of the all-atom nature of the simulations, the propulsion is explained mechanistically and in terms of self-thermophoresis, an ubiquitous mechanism in micro-metric self-propelling colloids. The nitromethane model thus constitutes the smallest example of a self-propelling particle.Motivated by these findings, the dynamics of self-propelled particles under periodic time dependent propulsion velocities are investigated. By extending the popular Active Brownian Particle (ABP) model, we obtain the translational diffusion enhancement of exponentially decaying propulsion, similar to the one exhibited by nitromethane. The results show that narrow, intense, peaks of activity, sufficiently spaced in time, are able to greatly enhance the diffusion of the swimmer. This strategy provides new avenues for mitigating the negative effect of rotational diffusion as swimmers are downsized.The collective dynamics of a many-body system of self-propelled particles under exponentially decaying periodic activity are also investigated. A well-known collective phenomenon in active matter is the emergence of Motility Induced Phase Separation (MIPS), consisting in the activity-driven phase separation into dense and dilute phases. The inclusion of time modulation in the system results in a significant alteration of the MIPS phase diagram, which has been thoroughly characterized. In some cases, phase separation is severely suppressed, allowing the system to remain as a single homogeneous phase while maintaining high mobility particles.Finally, a class of state-of-the-art machine learning interatomic potentials, based on neural networks, is presented and utilized for the study of thermophoresis, a phenomenon closely related to the propulsion mechanism of nitromethane and with a marked sensitivity to solute-solvent interactions. Neural network potentials allow the exploration of large scale systems with precision rivaling first-principles quantum calculations. These new algorithms are being developed at a fast pace and allow for the accurate characterization of condensed matter systems, a critical ingredient in the development of functionalized self-propelling particles.
  

DOCTORAL DEGREE IN EARTHQUAKE ENGINEERING AND STRUCTURAL DYNAMICS

• NAVARRO GRANADOS, JORDI: Theoretical and Experimental Study on Cold-Formed Elements for Steel Framing in Seismic Areas
  
  Author: NAVARRO GRANADOS, JORDI
  Programme: DOCTORAL DEGREE IN EARTHQUAKE ENGINEERING AND STRUCTURAL DYNAMICS
  Department: Department of Civil and Environmental Engineering (DECA)
  Mode: Normal
  Deposit date: 20/03/2026
  Reading date: 29/05/2026
  Reading time: 12:30
  Reading place: ETSEIBUPC, Campus SudAula CapellaAv. Diagonal, 647.08028 Barcelona
  Thesis director: CASAFONT RIBERA, MIQUEL | BOVÉ TOUS, ORIOL
  Thesis abstract: This Thesis presents a theoretical and experimental study on cold-formed elements for seismic-resistant steel framing systems. The focus is on evaluating the feasibility of using low- and mid-rise buildings based on cold-formed steel framing with flat strap bracings as diagonal members, and built-up steel profiles or concrete-filled steel sections as boundary (chord) studs, to resist seismic actions in moderate to high seismicity areas. Three representative prototype residential buildings with five, seven, and ten storeys, located in areas with peak ground accelerations (PGA) of 0.2 g, 0.3 g, and 0.4 g have been considered.This Thesis consists of six chapters; the first is an introduction, the second describes the current state of the art, and the last presents conclusions and future research. Chapters 3, 4, and 5 constitute the main body of this Thesis, and their content is described below.Chapter 3 presents a theoretical investigation into the compressive performance of single Cold-Formed Steel (CFS) members with Cee-shaped cross-sections, as well as various built-up sections (back-to-back, toe-to-toe, nested and a stud pack 4) derived from Cee profiles. The initial assessment of compressive performance is conducted using closed-form hand expressions. These results are subsequently complemented by design based on the Finite Strip Method (FSM) and further refined through simulations employing the Generalised Beam Theory (GBT). Detailed results for the compressive performance of the studied sections, considering a range of parameters including section thicknesses, steel grades, and buckling lengths considered, are also provided. Chapter 4 describes the performed experimental investigation into the compressive performance of steel-only and concrete-filled cold-formed steel built-up sections. The study encompasses the design and preparation of the test specimens, the experimental results, the predicted compressive resistance, and a comparative analysis between the predicted and observed values. Particular attention is given to the increasing of the effective area resulting from concrete infill; which is theoretically and experimentally validated through modifications to existing design approaches. Additionally, Chapter 4 analyses the compressive resistance of two CFS built-up sections filled with concrete -referred to as 4-pack Concrete-Filled Cold-Formed Steel (CF CFS) and 6-pack Concrete-Filled Cold-Formed Steel (CF CFS)-, across a range of concrete strength classes, from C20/25 to C50/60.Chapter 5 develops the design of three representative prototype steel framing buildings, 5-, 7-, and 10-storey, in regions of moderate (PGA 0.2 g) and high seismicity (PGA 0.3 g and 0.4 g) areas, by means of a Finite Element Method software, complemented with hand calculations. The main goal is obtained the internal forces in the most critical members of the buildings (namely the chord studs and the diagonal bracings of the wall panels). As described in detail in Chapter 5, the spectral ordinates, the base shear forces, and their distribution along the height of the buildings have been determined by hand calculations in accordance with the provisions of the second-generation Eurocode, since these provisions are not yet implemented in the FEM software. The results of the seismic analyses for each prototype building and seismicity level are presented, as well as a sizing proposal of the chord studs based on both the internal forces and the compressive performance of the concrete-filled in Chapters 3 and 4.
  

DOCTORAL DEGREE IN ELECTRICAL ENGINEERING

• JENÉ VINUESA, MARC: Data-Driven and Generative Methodologies for Enhanced Grid-Edge Visibility in Distribution Grids
  
  Author: JENÉ VINUESA, MARC
  Programme: DOCTORAL DEGREE IN ELECTRICAL ENGINEERING
  Department: Department of Electrical Engineering (DEE)
  Mode: Normal
  Deposit date: 12/03/2026
  Reading date: 05/06/2026
  Reading time: 15:00
  Reading place: Sala de Actes de la FME (Facultat de Matemàtiques i Estadística)Enllaç meet: meet.google.com/fqh-wqyh-jns
  Thesis director: ARAGÜÉS PEÑALBA, MÒNICA | SUMPER, ANDREAS
  Thesis abstract: The rapid deployment of distributed energy resources (DERs), such as residential photovoltaic (PV) systems, heat pumps, and electric vehicles, is accelerating the energy transition while reshaping electrical distribution grids. Although these technologies enable decarbonization and flexibility, their widespread installation behind the meter (BTM), together with limited measurement granularity, restricted data access, and the absence of dedicated metering, creates blind spots that hinder reliable grid operation, planning, and monitoring under high DERs penetration. This doctoral thesis addresses limited grid-edge visibility by developing data-driven methodologies to infer unobserved distribution-level phenomena from low-resolution smart meter data under realistic operational constraints. Grid-edge visibility is framed as the ability to reconstruct both legitimate and illegitimate power exchanges, encompassing BTM DERs behavior and non-technical losses (NTLs). The proposed contributions aim to support actionable and risk-aware decision-making in increasingly decentralized distribution systems. The first part of the thesis focuses on NTL detection and characterization. A comprehensive framework is proposed to identify abnormal losses through energy balances and machine learning using transformer- and customer-level active power measurements. The methodology enables fraud detection and classification, while an unsupervised customer flagging module supports targeted inspection strategies. Validation using real-world data from a Spanish distribution system operator demonstrates robust performance under realistic data availability and class imbalance. The second part addresses customer-level BTM PV disaggregation. A deterministic and adaptive hybrid methodology is introduced to detect PV installations, estimate their capacity, and disaggregate generation from net consumption measurements. By combining data-driven and physics-based models with contextually supervised source separation techniques, the approach captures system-specific and seasonal effects while remaining suitable for practical deployment. Extensive case studies demonstrate robustness to data length, seasonal variability, aggregated metering, and cross-domain generalization. Finally, the thesis extends disaggregation to a probabilistic setting using generative artificial intelligence. A conditional diffusion-based framework is proposed to model the distribution of BTM PV generation conditioned on net consumption and exogenous variables. The methodology provides calibrated and sharp probabilistic estimates and is further extended to jointly disaggregate multiple DERs, such as heat pumps. Validation on real-world datasets demonstrates reliable uncertainty quantification and strong generalization under limited training data.
  

DOCTORAL DEGREE IN ELECTRONIC ENGINEERING

• HERNÁNDEZ URREA, MARC: Design and Implementation of Novel Multiparametric Nonclinical Cardiovascular Assessment Devices Using Only Four Electrodes
  
  Author: HERNÁNDEZ URREA, MARC
  Programme: DOCTORAL DEGREE IN ELECTRONIC ENGINEERING
  Department: Department of Electronic Engineering (EEL)
  Mode: Normal
  Deposit date: 08/05/2026
  Reading date: pending
  Reading time: pending
  Reading place: pending
  Thesis director: CASAS PIEDRAFITA, JAIME OSCAR | CASANELLA ALONSO, RAMON
  Thesis abstract: Cardiovascular diseases remain the leading cause of adverse outcomes and healthcare burden worldwide. The development of extra-hospital cardiovascular (CV) monitoring strategies has become increasingly important, especially after the COVID-19 pandemic exposed healthcare system limitations and the need for alternative screening and monitoring approaches to reduce hospital congestion. Organizations such as the World Health Organization (WHO) highlight that nonclinical devices for cardiovascular disease (CVD) assessment improve treatment adherence, rehabilitation outcomes, survival indices, and contribute to more sustainable healthcare systems. These technologies have also attracted interest from the pharmaceutical industry, as home-based clinical trials and remote patient monitoring provide advantages over traditional trials, including improved adherence, supervision, dose control, and cohort management.Currently, most nonclinical devices rely on electrocardiogram (ECG) or photoplethysmogram (PPG). ECG provides waveform information, time intervals, heart rate (HR), and heart rate variability (HRV), but only electrical heart information. PPG provides arterial pulse wave (APW) morphology, pulse rate (PR), oxygen saturation, and sometimes blood pressure from distal measurements, but only mechanical information limited to superficial arteries. Other systems for deep artery diagnosis, such as SphygmoCor, echocardiography, and cardiac catheterization, remain restricted to clinical environments due to complexity, cost, and need for skilled personnel. Similarly, impedance plethysmogram systems such as impedance cardiography are limited to clinical use, while nonclinical impedance applications are usually restricted to single limbs. Recent advances show that impedance plethysmogram (IPG) measured between limbs can provide information about APW propagation times proximal to the aorta.This thesis presents the design and validation of a novel, easy-to-use cardiovascular assessment device based on simultaneous acquisition of ECG, multiple IPG signals (limb-to-limb and local), and ballistocardiogram (BCG) using only four electrodes in contact with hands or feet and a home weighing scale, together with algorithms to extract cardiovascular health indicators. The device provides information from time intervals between ECG and IPG signals, such as pulse arrival time (PAT), including proximal information from limb-to-limb IPG and distal information from local IPG measurements. These measurements are combined to estimate pulse transit time (PTT), related to arterial elasticity. The combination of hand-to-hand and foot-to-foot IPG PAT enables estimation of a surrogate of aortic PTT (aPTT), typically obtained from carotid–femoral PTT (cf-PTT) using tonometry, a marker associated with ageing and cardiovascular health. The device also enables respiration extraction from limb-to-limb IPG signals without additional sensors.The system was validated through measurement campaigns using reference instruments such as impedance cardiography and tonometry. The devices showed adequate signal-to-noise ratio (SNR: 41 dB and 22 dB) and acceptable CMRR (55 dB and 25 dB). Comparative studies showed good agreement with impedance plethysmography (r > 0.90). Tonometric PAT and PTT estimations agreed with IPG-derived parameters. Wrist-to-wrist IPG PAT correlated with carotid PAT (r = 0.85), and foot-to-foot IPG PAT with femoral PAT (r = 0.86). IPG-derived carotid–femoral PTT showed moderate correlation with tonometer-based cf-PTT (r = 0.67).These results validate the proposed instrumentation and methodology, demonstrating reliable extraction of clinically relevant cardiovascular timing parameters. Overall, a compact system combining ECG and limb-to-limb IPG enables acquisition of relevant proximal and distal arterial information for nonclinical cardiovascular assessment.
  

DOCTORAL DEGREE IN ENVIRONMENTAL ENGINEERING

• HOUCHMAND, LAURA JO: Integrated assessment of passive rooftop strategies and photovoltaic on building energy demand and urban heat island effects under current and future Mediterranean climates
  
  Author: HOUCHMAND, LAURA JO
  Programme: DOCTORAL DEGREE IN ENVIRONMENTAL ENGINEERING
  Department: Department of Civil and Environmental Engineering (DECA)
  Mode: Normal
  Deposit date: 08/04/2026
  Reading date: 02/06/2026
  Reading time: 12:00
  Reading place: UPC ESEIAATTerrassa, TR5, Aula 2.0a 226Carrer de Colom, 15, 08222 Terrassa, Barcelona
  Thesis director: GASSO DOMINGO, SANTIAGO | MACARULLA MARTÍ, MARCEL
  Thesis abstract: Climate change mitigation and adaptation are pressing challenges, particular in Mediterranean cities, where rising temperatures, intensifying urban heat island (UHI) effects, increasing cooling demand, and growing water scarcity intersect. Rooftops represent a critical interface for climate-responsive building transformation, as they offer substantial potential for passive strategies, such as cool roofs and green roofs, and active renewable technologies, particularly photovoltaic (PV) systems. Despite extensive research on rooftop technologies, four key knowledge gaps remain: (1) insufficient year-round assessment of the UHI impact of roof-mounted PV systems combined with passive roofing strategies; (2) limited integrated evaluation of green roofs and PV systems considering both building energy demand and urban climate implications; (3) lack of comparative analysis of the impact of passive roofing strategies under projected future Mediterranean climate scenarios on the buildings’ energy demand; and (4) inadequate quantification of the impacts of PV integration with passive roofing strategies on building energy demand under future climate change conditions. This doctoral research addresses these gaps through a structured, simulation-based methodological framework. The work begins with a targeted literature review to establish the scientific context of passive and active rooftop strategies in Mediterranean climates and to define the research questions. Building on these gaps, dynamic building energy simulations are conducted using DesignBuilder with EnergyPlus as the calculation engine. Barcelona (Csa climate classification) serves as the case study location. Current climate conditions are represented by high-resolution Typical Meteorological Year (TMY) data (1975–2021), while future conditions are assessed using morphed TMY files based on the IEA EBC Annex 80 methodology under RCP4.5 and RCP8.5 scenarios for mid-term (2047–2060) and long-term (2087–2100) horizons.A typical Mediterranean roof as basic roof (BR) serves as the reference case and is compared with passive strategies, including a cool roof (CR), a soil roof without vegetation (SR), and extensive green roofs under different irrigation regimes (EGR, EGRmin, EGRmax), as well as their integration with rooftop PV systems. Key performance indicators include convective heat fluxes from roof and PV surfaces (UHI contribution), conductive heat fluxes through the roof and annual heating and cooling demand (building energy demand), as well as PV electricity generation.The findings of this thesis highlight trade-offs between energy efficiency, urban heat, water use, and future climate impacts, emphasizing the need for climate-sensitive, typology-specific rooftop strategies.
  

DOCTORAL DEGREE IN MECHANICAL, FLUIDS AND AEROSPACE ENGINEERING

• MASCLANS SERRAT, NÚRIA: Scientific Machine Learning in Turbulent Flows: Observability, Reconstruction & Acceleration
  
  Author: MASCLANS SERRAT, NÚRIA
  Programme: DOCTORAL DEGREE IN MECHANICAL, FLUIDS AND AEROSPACE ENGINEERING
  Department: Department of Mechanical Engineering (EM)
  Mode: Normal
  Deposit date: 06/05/2026
  Reading date: 04/06/2026
  Reading time: 11:30
  Reading place: Sala polivalent de l'edifici A (EEBE) del Campus Diagonal-Besòs.
  Thesis director: JOFRE CRUANYES, LLUÍS
  Thesis abstract: The analysis and design of engineering systems governed by wall-bounded turbulent supercritical fluid flows are fundamentally constrained by two distinct barriers: the intrinsic limitations of optical diagnostics in resolving scalar thermodynamic fields under extreme fluid regimes, and the prohibitive computational cost required to achieve fully converged flow statistics in high-fidelity simulations. This thesis addresses these challenges by developing advanced scientific machine learning (SciML) frameworks that rigorously integrate physical domain knowledge into deep learning architectures.To overcome the experimental observability gap inherent to high-pressure transcritical flows, this work first proposes a novel thermophysics-informed neural network (TINN). By embedding the real-gas equation of state directly into the network's optimization loss function as a soft constraint, while enforcing physical boundary conditions through the hard-constrained network architecture, this framework successfully reconstructs hidden thermodynamic state variables, specifically density and temperature, exclusively from available kinematic velocity data. This methodology provides a reliable, non-intrusive alternative to overcome the severe optical distortions that traditionally limit quantitative scalar measurements in supercritical fluid experiments. To address the computational burden of temporal numerical integration, the thesis introduces a paradigm shift in turbulence simulation by adapting deep reinforcement learning (DRL) to accelerate the convergence of flow statistics. Specifically, a Reynolds eigenspace perturbation (REP)-based DRL methodology is formulated. In this approach, a distributed multi-agent DRL framework acts as an active flow control system, iteratively interacting with the numerical solver to optimize the instantaneous flow trajectory. Initially established and validated on a reduced one-dimensional turbulence (ODT) model, the agents apply mathematically constrained perturbations directly to the eigenspace of the Reynolds stress tensor. This strict structural constraint ensures that all dynamic modifications to the Reynolds stress magnitude, shape, and orientation maintain rigorous physical realizability. This hybrid DRL-CFD methodology is subsequently scaled to fully resolved, three-dimensional direct numerical simulations (DNS) of turbulent channel flows. By overcoming complex software engineering barriers to achieve a low-latency, in-memory coupling between the dynamic Python-based REP-DRL framework and a massively parallelized C++ CFD solver, the implementation dynamically manipulates the instantaneous flow fields. While the CFD simulation operates within a statistically stationary state, this active flow control framework steers the system to achieve statistical convergence in a significantly reduced integration time.Collectively, this thesis demonstrates that embedding physical laws, thermodynamic equations, and structural constraints into machine learning algorithms transforms them from passive data interpolators into scalable, physically consistent frameworks capable of both recovering hidden flow physics and actively accelerating numerical simulations. The proposed methodologies establish a foundational pathway towards bridging the gap between experimental measurement limitations and computational feasibility, thereby facilitating both the fundamental study and the practical engineering design of complex turbulent flows.
  

DOCTORAL DEGREE IN NETWORK ENGINEERING

• GUZMÁN ALBIOL, MARC: An Exploration of Constraint Systems in Verifiable Computation
  
  Author: GUZMÁN ALBIOL, MARC
  Programme: DOCTORAL DEGREE IN NETWORK ENGINEERING
  Department: Department of Network Engineering (ENTEL)
  Mode: Normal
  Deposit date: 23/04/2026
  Reading date: 03/06/2026
  Reading time: 11:00
  Reading place: sala Aula Màster del C3 (Sala C3005
  Thesis director: MUÑOZ TAPIA, JOSE LUIS
  Thesis abstract: The accelerated adoption of digital services has highlighted the need for trust-minimized computation, where parties can verify the correctness of computations without re-executing them or revealing sensitive data. Zero-knowledge proof systems, including SNARKs and STARKs, provide cryptographic guarantees of correctness, privacy, and succinct verifiability, enabling applications in scalable blockchains, privacy-preserving identity systems, and verifiable federated learning.This thesis addresses key inefficiencies in constraint-based zero-knowledge proof systems at the arithmetization layer. The research focuses on two complementary problems: optimizing binary comparisons within Rank-1 Constraint Systems (R1CS), and extending the expressiveness of STARKs through an Extended Algebraic Intermediate Representation (eAIR).The first contribution presents a weighted accumulation method for implementing strict binary comparisons in R1CS. Traditional approaches generate a large number of constraints due to the lack of native comparison and control-flow operations in the R1CS model, forcing costly bit-by-bit decompositions and creating performance bottlenecks. The proposed weighted accumulation method significantly reduces constraint overhead without compromising system security or correctness, achieving substantial efficiency improvements over thelexicographic approach.The second contribution introduces the eSTARK protocol, which extends standard STARKs by enabling the concise handling of complex constraints such as lookups, permutations, and copy constraints. These operations are difficult to encode efficiently in standard AIR. The eSTARK protocol integrates vector commitment arguments and polynomial optimizations, providing a flexible and user-friendly framework for representing a broader class of computations without introducing unnecessary arithmetization overhead.Both contributions address practical limitations of current zero-knowledge proof systems. The first focuses on reducing constraint complexity for common operations, while the second expands the expressiveness of the proof system itself. Together, they demonstrate the importance of arithmetization-level optimizations for improving the efficiency and usability of zero-knowledge proofs.
  

DOCTORAL DEGREE IN OPTICAL ENGINEERING

• RAMÍREZ CANO, NATALIA: Medida morfológica de la tortuosidad de los vasos retinianos mediante análisis digital de imagen
  
  Author: RAMÍREZ CANO, NATALIA
  Programme: DOCTORAL DEGREE IN OPTICAL ENGINEERING
  Department: Department of Optics and Optometry (OO)
  Mode: Article-based thesis
  Deposit date: 07/05/2026
  Reading date: pending
  Reading time: pending
  Reading place: pending
  Thesis director: MILLAN GARCIA VARELA, MARIA SAGRARIO
  Thesis abstract: Retinal vascular tortuosity has been widely identified by the medical community as a direct and non-invasive biomarker for the development of ocular, vascular, and systemic diseases. This is because the retinal vascular network is a delicate structure highly sensitive to alterations in blood pressure and other hemodynamic factors. These alterations produce observable changes in its vessels, modifying morphological parameters such as thickness, curvature, deviation, and tortuosity level. This makes the retinal vascular network a prime target organ for the early detection of ocular and vascular pathologies.Quantitative analysis of tortuosity in the retinal vascular network is typically performed by processing digital retinographies using algorithms that allow for the assessment of tortuosity indices. Local (characterizing the tortuosity of vessel segments free of bifurcations) and global (synthesizing the tortuosity information of the entire vascular network) indices are used. However, the calculation of these indices is conditioned by technical factors such as the spatial resolution of the images, the framing used in the acquisition of the retinographies (centered on the macula or the optic disc), as well as the choice of combinations of weighting schemes and local indices for calculating the global tortuosity indices.This doctoral thesis, presented as a compilation of three scientific publications, comprehensively addresses these challenges. First, it studies the impact of the retinographies’ framing (macula or optic disc) on the estimation of local tortuosity. Second, it evaluates the influence of image resolution on the robustness of the calculation of the local indices. Finally, an objective and quantitative validation framework for global indices based on the mathematical composition of local tortuosity index measures is proposed.The results obtained allow for the establishment of usage criteria for the selection of reliable local indices, demonstrate the limitations arising from acquisition conditions, and propose a methodological bridge between local and global measures. Thus, the thesis contributes to the standardization of tortuosity analysis in retinal images, with the aim of improving its clinical applicability and its value as a biomarker for systemic diseases.
  

DOCTORAL DEGREE IN PHOTONICS

• MALLA, ADITYA JAGADEESH: COLLOIDAL QUANTUM DOT EMITTERS IN THE SHORTWAVE INFRARED REGION
  
  Author: MALLA, ADITYA JAGADEESH
  Programme: DOCTORAL DEGREE IN PHOTONICS
  Department: Institute of Photonic Sciences (ICFO)
  Mode: Normal
  Deposit date: 16/04/2026
  Reading date: 29/05/2026
  Reading time: 10:00
  Reading place: ICFO Auditorium
  Thesis director: KONSTANTATOS, GERASIMOS
  Thesis abstract: Shortwave infrared (SWIR) light sources are indispensable for applications including advanced imaging, spectroscopy, and sensing; however, their widespread adoption is hindered by the high cost and limited scalability of epitaxial semiconductor technologies such as InGaAs. Colloidal quantum dots (QDs) offer an attractive alternative owing to their high photoluminescence quantum yield, size-tunable emission, large-area processability, and compatibility with low-cost solution-based fabrication. Among various QD-based emitters employing lead sulphide (PbS), this thesis focuses on two complementary technologies: electrically driven quantum-dot light-emitting diodes (QLEDs) and optically pumped downconverters (DCs).The first part of this thesis addresses performance enhancement in QLEDs (emitting at 1380 nm) through systematic device engineering. Charge imbalance is identified as a key factor limiting QLED efficiency and radiance. By optimising the ZnO electron transport layer via controlled annealing-temperature tuning, electron injection was modulated, leading to a maximum external quantum efficiency (EQE) of 20%. Furthermore, the charge balance within the emissive layer was optimised by controlling its thickness, resulting in an increase in maximum radiance from 5 W.sr-1.m-2 to 17.5 W.sr-1.m-2. Building upon this, a dual electron transport layer architecture was implemented to decouple interfacial quality from bulk electron transport, enabling a further enhancement in maximum radiance to 30 W.sr-1.m-2 while maintaining comparable EQE.Light extraction and Joule heating constitute an additional bottleneck in achieving high-performance QLEDs. To overcome substantial optical losses into substrate modes inherent in conventional bottom-emission devices, top-emission QLED (TQLED) architectures were investigated. These offer improved light extraction and allow for the use of opaque, high-thermal-conductivity silicon substrates to manage Joule heating. A high-performance sputtered indium tin oxide (ITO) electrode was developed, exhibiting optical transmission exceeding 85% at 1400 nm and a low sheet resistance of 33 Ω/□. By utilising optimised architecture with integrated ITO optical spacers and a dielectric/metal/dielectric top electrode, a low-Q microcavity was established. This modified the far-field radiation pattern to a forward-directed profile and narrowed the emission linewidth. The synergy between this resonant optical design and superior thermal dissipation enabled a record radiance exceeding 100 W.sr-1.m-2, and allowed for the first demonstration of active see-through SWIR imaging illuminated solely by QLEDs.The second part of the thesis is focused on lead sulphide QD-based DCs. The QD-DCs suffer from performance degradation under high excitation power densities due to the significant heat generation in the process of light absorption. We have developed high-power, stable, and spectrally tunable narrowband and broadband SWIR DCs (1000 nm - 1600 nm). By mixing two different-sized QDs, we exploit Förster resonance energy transfer and photon reabsorption to realise a binary system with a high photoluminescence quantum yield of 35 %. Embedding the QDs in a poly(methyl methacrylate) host mitigates local thermal stress on the QDs, enabling standalone DCs with a high emission power density (EmPD) of 110 mW.cm-2 at 1380 nm. Further optimisation with a spectrally selective distributed Bragg reflector for enhanced light extraction and a sapphire substrate for efficient heat dissipation, we achieved a record EmPD of 385 mW.cm-2 at 1380 nm with optical power conversion efficiency of 10% and operational stability above 230 hours at an EmPD of 190 mW.cm-2. This demonstrates a scalable route to low-cost SWIR light sources, narrowing the performance gap between solution-processed DCs and conventional epitaxial semiconductors.
  
• NIMJE, KARTIKA NARAYAN: Photonic Strategies for Approaching Fundamental Limits in Thermophotovoltaic Energy Conversion
  
  Author: NIMJE, KARTIKA NARAYAN
  Programme: DOCTORAL DEGREE IN PHOTONICS
  Department: Institute of Photonic Sciences (ICFO)
  Mode: Normal
  Deposit date: 06/05/2026
  Reading date: 18/06/2026
  Reading time: 16:00
  Reading place: ICFO Auditorium and Microsoft Teams Meetinghttps://teams.microsoft.com/meet/396755562864232?p=sfg2ZBdHKIuPmehIJj
  Thesis director: PAPADAKI, GEORGIA THENO
  Thesis abstract: Thermophotovoltaic (TPV) systems convert thermal radiation into electricity by coupling a hot emitter to a photovoltaic (PV) cell. Despite important recent developments in the field, TPV performance is ultimately constrained by a persistent power--efficiency trade-off: broadband radiative exchange between an emitter and a cell yields high electrical power at the expense of efficiency, whereas narrowband exchange leads to high efficiency at the cost of reduced power. A more empirical way to express this constraint is the difficulty of improving both the current and voltage characteristics of a cell simultaneously. Understanding the power--efficiency trade-off is therefore fundamental to optimizing TPV performance. This thesis develops a unified framework for understanding this trade-off and identifying photonic and electronic design strategies that enable optimal operation while remaining anchored to thermodynamic bounds.The first part of the thesis establishes the conceptual and quantitative backbone. A photonic view of thermal radiation connects Planck’s law and the Stefan--Boltzmann limit to the density of optical states, coherence, and surface polaritons, and clarifies the distinction between far-field and near-field exchange. These ideas are combined with the practical building blocks of TPVs — spectral and angular emission control, junction physics, recombination, and quantum efficiency — to establish a practical design toolkit for TPVs. Within this toolkit, detailed-balance and radiative--thermodynamic analyses place TPVs on a power--efficiency landscape bounded by Carnot and exergy (Landsberg) limits and identify spectral bandwidth as a central control knob governing performance.On this foundation, the thesis explores three routes for mitigating the trade-off. First, hot-carrier TPVs introduce an internal thermodynamic degree of freedom by treating the carrier subsystem as a reservoir with its own temperature and chemical potential and by harvesting carriers through energy-selective contacts; in the ideal radiative limit, this enables a single junction to emulate multicolor performance and approach Carnot efficiency at finite power. Second, an analytical framework for a near-field TPV system based on fluctuational electrodynamics derives analytical expressions for photon tunnelling between plasmonic and semiconducting media, establishing scaling laws that show how evanescent modes can deliver super-Planckian, spectrally concentrated fluxes that support high power and high efficiency simultaneously. Third, substrate engineering is shown to be instrumental in near-field TPV design: in an ITO/InAs case study, optimizing thin, low-loss plasmonic films with tuned plasma frequency and thickness reshapes the tunnelling spectrum, concentrating useful above-bandgap transfer while suppressing sub-bandgap losses, and shifts electrical power–efficiency curves outward in the radiative limit.This thesis reframes the TPV power--efficiency compromise as malleable rather than fixed: while the global thermodynamic frontiers are immutable, the effective frontier accessible to practical architectures can be steered through controlled interventions in spectra, modes, and carrier energetics. The analytical models, thermodynamic benchmarks, and optimization strategies developed here provide a principled basis for designing and evaluating TPV systems that combine photonic control, hot-carrier extraction, and near-field coupling, and for assessing these theoretical results in relation to realistic material and device constraints.
  
• POPOV, PAVEL PEYCHEV: Quantum simulation of lattice gauge theories with qudit systems
  
  Author: POPOV, PAVEL PEYCHEV
  Programme: DOCTORAL DEGREE IN PHOTONICS
  Department: Institute of Photonic Sciences (ICFO)
  Mode: Normal
  Deposit date: 29/04/2026
  Reading date: 29/05/2026
  Reading time: 10:30
  Reading place: Elements Room i https://teams.microsoft.com/meet/339611996180202?p=2I9fTirIejfsIaqwoW
  Thesis director: LEWENSTEIN, MACIEJ | KASPER, VALENTIN
  Thesis abstract: The spectacular progress in controlling quantum matter has opened new avenues for studying fundamental physics. Various experimental platforms now host hundreds of quantum units, capable of quantum state engineering, Hamiltonian simulation and universal computation, already surpassing what is classically tractable. Remarkably, the versatility of such quantum simulators allows for investigating the physics from very high to very low energy scales. While the long-term goal is to be able to perform fault-tolerant quantum computation, noisy intermediate scale quantum (NISQ) devices are prone to errors and quantum algorithms need to be tailored to the underlying physical platform by exploiting its advantages. In that regard, qudits offer enhanced Hilbert space dimension per information carrier with respect to qubits, allowing for significant reduction of costly entanglement operations. Moreover, the higher-dimensional Hilbert space of qudits natively accommodates complex many-body models, thereby minimizing algorithmic overhead.In this thesis, we investigate the opportunities that qudit devices offer for the quantum simulation of lattice gauge theories. Being extremely successful nonperturbative framework for studying three of the four fundamental interactions--electrodynamics, the weak and the strong force-- lattice gauge theories can be formulated as many-body systems amenable to quantum simulation. This approach overcomes the intrinsic bottlenecks of classical methods, unlocking the ability to explore out-of-equilibrium phenomena and finite-density equilibrium states.The first part of this thesis is dedicated to the development of encoding procedures for lattice gauge theories with Abelian and non-Abelian symmetry on qudit quantum hardware. Building upon advances in the understanding of the structure of the gauge-invariant Hilbert space for specific symmetry groups, we propose scalable qudit implementation of gauge theory models in arbitrary spatial dimensions and devise variational protocols for their equilibrium and out-of-equilibrium simulation. Crucially, our methods apply to gauge theories with dynamical fermionic matter, without the need for nonlocal encodings for the fermions, as they are unitarily removed in the encoding process. In the second part of this thesis, we use quantum-inspired numerical techniques to reveal some of the plethora of physical phenomena simple many-body models with local symmetry host. Using the multi-flavour Schwinger model (quantum electrodynamics in one spatial dimension) as an example, we show how to identify signatures of fractons — gauge field configurations with fractional topological charge. Furthermore, by examining pure gauge theories with non-Abelian dihedral symmetry, we identify the importance of the central subgroup for the spectrum and the dynamics of the many-body model, relating nontrivial fusion rules to lack of confinement and presence of exotic particle excitations. Most importantly, the lattice gauge models for both examples above, due to their simplicity, are amenable to near-term implementation on qudit quantum hardware.Ultimately, this work takes a significant step toward harnessing qudit quantum devices for the simulation of high-energy and condensed-matter systems. By detailing resource-efficient hardware implementations and outlining near-term applications, our findings provide compelling motivation for the continued symbiosis of theoretical design and experimental realization.
  
• STAMMER, PHILIPP MAXIMILIAN: Photons and Information -- A modern approach to strong-field quantum optics
  
  Author: STAMMER, PHILIPP MAXIMILIAN
  Programme: DOCTORAL DEGREE IN PHOTONICS
  Department: Institute of Photonic Sciences (ICFO)
  Mode: Normal
  Deposit date: 06/05/2026
  Reading date: 09/06/2026
  Reading time: 10:00
  Reading place: ICFO Auditorium
  Thesis director: LEWENSTEIN, MACIEJ
  Thesis abstract: Photons and Information are notions indispensable for modern quantum technologies, and their interplay provides the foundation of understanding the quantum nature of light. In particular, photons are robust information carrier and substantiate the inherent discreteness of the electromagnetic radiation field. The information of photons, and their intrinsic quantum fluctuations, are manifested in the measurement of correlation functions of the field. Central to the understanding of photon fluctuations is the quantum theory of optical coherence, and the photon signature can reveal itself in different properties. Furthermore, the measurement of correlations is needed from an information theoretic perspective to distinguish classical from quantum signatures. These concepts ultimately lead to the original goal of the present Thesis:How can we understand the notion of photons and information in a modern strong field quantum optics perspective?Strong field quantum optics aims for the generation, characterization and the control of quantum light, beyond the conventional wisdom of strong field physics. In the traditional approach to strong field phenomena, in particular the photon up-conversion process of high-order harmonic generation, the field was merely treated classically. And hence, the notion of the photon could not exist in such descriptions. However, recent advances in the field of strong field quantum optics have revealed new signatures unaccountable by classical theory. This includes the generation of genuine non-classical states of light from strong field phenomena, revealing the intrinsic entanglement between all field modes in the process of high-order harmonic generation, or finding quantum signatures in the photon emission process itself via photon anti-bunching. In this Thesis, we reveal all such properties of the generated light field, and establish the underlying theories for the description of these phenomena. This includes the development of the quantum theory of optical coherence for high-order harmonic generation, as well as introducing new quantum information theoretic perspectives to strong field quantum optics. Therefore, the work presented in this Thesis provides the foundation for the modern formulation of quantum optical phenomena in strong field physics.
  

DOCTORAL DEGREE IN SIGNAL THEORY AND COMMUNICATIONS

• AREIAS FANZERES, LEONARDO: Sound-to-Image Translation Through Direct Cross-Modal Learning: An Exploratory and Architectural Study
  
  Author: AREIAS FANZERES, LEONARDO
  Programme: DOCTORAL DEGREE IN SIGNAL THEORY AND COMMUNICATIONS
  Department: Department of Signal Theory and Communications (TSC)
  Mode: Normal
  Deposit date: 16/04/2026
  Reading date: 18/06/2026
  Reading time: 11:00
  Reading place: Aula de Teleensenyament, edifici B3, Campus Nord, Barcelona
  Thesis director:
  Thesis abstract: Environmental sound conveys rich semantic and contextual information about events, objects, and spatial dynamics. However, prevailing computational approaches to environmental audio analysis, such as Acoustic Event Detection (AED), typically reduce this complexity to discrete textual labels. While effective for automated monitoring tasks, such representations oversimplify acoustic scenes and become inadequate when auditory information must be communicated across modalities. Sound-to-image (S2I) translation offers an alternative approach in which a model synthesizes images that visually depict sound-emitting sources and their surrounding environments.This thesis introduces and advances direct sound-to-image translation, a paradigm that establishes a connection between audio and visual modalities without relying on textual mediation, class supervision, or cluster-based alignment during training. The central hypothesis is that higher-level abstractions learned by deep neural networks provide a shared semantic space in which heterogeneous modalities can connect directly, enabling the generation of images that are interpretable and semantically coherent with the source sound. Such outputs are referred to as informative, meaning that they visually communicate meaningful aspects of the acoustic event.The first part of the thesis presents, to the best of our knowledge, the first study dedicated to direct S2I translation. A densely connected generative adversarial network (GAN), conditioned on audio embeddings, is developed to synthesize images directly from sound. Because multiple plausible images may correspond to a single acoustic event, translation quality cannot be evaluated through pixel-level reconstruction. To address this challenge, an informativity-based evaluation framework is proposed, employing classifiers to determine whether generated images are interpretable and semantically coherent with the source audio. Experiments reveal that, despite the inherent difficulty of the task, the model generalizes to unseen sounds and produces informative outputs for a meaningful portion of translations. Analysis further reveals that latent bottleneck dimensionality influences translation behavior, exposing a trade-off between pixel-space convergence and informativity.Building on this foundation, the second part investigates whether attention mechanisms can strengthen cross-modal alignment. Self-attention and cross-attention modules are integrated into the generator and evaluated across multiple configurations. Results show that attention improves translation performance when applied at early stages of the network, increasing informativity relative to the purely convolutional baseline while preserving the fully direct audio–visual connection.Together, these contributions formally define, validate, and extend direct S2I translation as a distinct research paradigm. Beyond architectural design and training strategies, this thesis advances methodological principles for quantitative evaluation of cross-modal translation in the absence of deterministic visual ground truth. The findings contribute to a broader understanding of multimodal representation learning and highlight the potential of direct S2I translation for applications in multimodal interaction and accessibility-oriented technologies, particularly for enhancing situational awareness in deaf and hard-of-hearing individuals.
  
• MOSAHEBFARD, MOHAMMADREZA: Resource Management in Sliced Converged Optical-Wireless 6G Networks: From Strategic Dimensioning to Dynamic Service Provisioning
  
  Author: MOSAHEBFARD, MOHAMMADREZA
  Programme: DOCTORAL DEGREE IN SIGNAL THEORY AND COMMUNICATIONS
  Department: Department of Signal Theory and Communications (TSC)
  Mode: Normal
  Deposit date: 10/04/2026
  Reading date: 29/05/2026
  Reading time: 11:00
  Reading place: Room C4-021B at the Castelldefels Campus
  Thesis director: VERIKOUKIS, CHRISTOS | VARDAKAS, JOHN
  Thesis abstract: The evolution toward the 6th Generation (6G) networks relies on converged optical-wireless infrastructures and virtualization technologies like Network Functions Virtualization (NFV) to support diverse services through Network Slicing. While enabling flexibility, these paradigms introduce profound complexity in managing shared computational and com-municational resources. This thesis addresses this challenge by identifying a fundamental temporal duality in resource management, spanning the strategic need for agile capacity planning to the operational necessity of real-time service provisioning. This complexity is further aggravated by the presence of various network slices, each with distinct Quality of Service (QoS) requirements. To resolve this dichotomy, distinct methodologies are developed within this PhD thesis.To address the strategic domain, where Mobile Virtual Network Operators (MVNOs) require rapid dimensioning tools for flexible resource leasing, this PhD thesis develops a computationally efficient analytical framework based on one-dimensional Markov chains. Uniquely, this model captures resource occupancy at two levels: physical resources governing Service Function Chain (SFC) instantiation, and virtual resources governing user admission. This stratified modeling avoids state-space explosion while accurately calculating admission ratios, achieving relative errors typically below 2% compared to simulations. The framework is applied to determine the minimum resources required to achieve target admission ratios under varying arrival rates, as well as to identify the necessary capacity to meet different target admission rates under constant traffic loads. This capability enables rapid offline dimensioning to guarantee Quality of Service (QoS) thresholds, offering a scalable alternative to time-consuming simulations and resource-exhaustive optimization approaches.Complementing this, the operational challenge of online SFC Embedding (SFCE) is addressed. Recognizing the NP-hardness of the embedding problem with the objective of jointly minimizing holistic power consumption and blocking probability, HORIZON, a novel holistic heuristic designed to optimize Mobile Virtual Network Operator (MVNO) operational efficiency, is developed. It employs a proactive backward placement strategy coupled with power and latency-aware segmental routing, and integrates comprehensive power models for both servers and Reconfigurable Optical Add-Drop Multiplexers (ROADMs). By enforcing strict inter-slice isolation while maximizing intra-slice efficiency, HORIZON achieves power efficiency within 15% of optimal Integer Linear Programming (ILP) bounds under resource-constrained scenarios. Evaluations across four realistic network topologies demonstrate execution times of 10–18 ms per service request (38–67 times faster than optimal solvers), power savings up to 23.6% compared to state-of-the-art heuristics, and negligible blocking rates, enabling MVNOs to minimize OPerational EXpenditure (OPEX) while strictly adhering to Service Level Agreements (SLAs).
  
• RUIZ CARREGAL, GERARD: High-Resolution Drone-Based Repeat-pass SAR Interferometry for 3D displacement estimation
  
  Author: RUIZ CARREGAL, GERARD
  Programme: DOCTORAL DEGREE IN SIGNAL THEORY AND COMMUNICATIONS
  Department: Department of Signal Theory and Communications (TSC)
  Mode: Article-based thesis
  Deposit date: 15/04/2026
  Reading date: 27/05/2026
  Reading time: 11:00
  Reading place: Sala D4 012, UPC Campus Nord.
  Thesis director: LOPEZ MARTINEZ, CARLOS | IGLESIAS GONZÁLEZ, RUBÉN | LORT CUENCA, MARC
  Thesis abstract: The precise characterization of ground deformation processes is essential for risk assessment and early-warning applications, since displacement is often a precursor to catastrophic failures. Over the past two decades, Synthetic Aperture Radar (SAR) and Multi-Temporal Differential SAR Interferometry (MT-DInSAR) have enabled millimetric displacement estimation over wide areas with dense spatial sampling. Within this context, satellite-based SAR offers regular, long-term global coverage, while Ground-Based SAR (GBSAR) provides near real-time monitoring, enabling the observation of localized rapid deformation processes. Despite these strengths, both systems present inherent limitations. Satellite SAR cannot capture rapid displacements due to multi-day revisit intervals and provides no sensitivity to displacement along the North-South (NS) direction due to its near-polar orbits. GBSAR provides excellent temporal resolution, but its deployment is limited in inaccessible areas, and its fixed geometry often leads to displacement underestimation.Airborne SAR systems mitigate several of these limitations by offering controlled revisit times and multi-view imaging, allowing the retrieval of the three-dimensional (3D) displacement vector. In this context, multirotor drones have emerged as a cost-effective and adaptable platform for airborne SAR. Drone-borne SAR inherits airborne challenges related to platform motion and navigation uncertainties, together with additional constraints such as payload limitations, and requires the adaptation of MT-DInSAR algorithms to low-altitude platforms and non-regular revisit times.The objective of this dissertation is to demonstrate the capability of drone-based SAR systems to estimate ground displacement time-series using multi-temporal interferometric stacks in real operational scenarios. To this end, the thesis develops an end-to-end framework encompassing sensor development, interferometric processing, and MT-DInSAR methodologies. A Ku-band dual-channel Frequency-Modulated Continuous-Wave (FMCW) radar is developed and integrated into a multirotor platform. An interferometric processing chain is further proposed to generate phase-calibrated repeat-pass interferograms, combining Digital Elevation Model (DEM) refinement from single-pass interferometry with a dedicated coregistration strategy, where the MTCD-MSQ algorithm is introduced as a new coregistration approach designed for high-frequency airborne acquisitions. Furthermore, two complementary MT-DInSAR workflows are proposed to address distinct deformation regimes. SD-MT-DInSAR retrieves cumulative displacement time-series when interferometric coherence is preserved between consecutive acquisitions, while SDVEL-MT-DInSAR exploits the flexibility of drones to perform repeated intra-day flights to estimate displacement velocity time-series in rapidly deforming areas, where decorrelation occurs in a few hours. Finally, the thesis employs multi-geometry acquisitions to retrieve the full 3D displacement vector.The framework is validated in controlled experiments with Corner Reflectors (CR) and in real operational conditions over an active open-pit mine, demonstrating submillimetric sensitivity, meter-scale displacement monitoring over several days, and 3D displacement retrieval in complex scenarios.The dissertation confirms that drone-based SAR is a reliable deformation monitoring tool that complements satellite and GBSAR systems, opening new opportunities in geotechnics, mining, and natural hazard assessment.
  

DOCTORAL DEGREE IN SUPPLY CHAIN AND OPERATIONS MANAGEMENT

• MAQUIRRIAIN ANTOÑANZAS, JAVIER: Algoritmos matheurísticos para la programación de mantenimientos preventivos
  
  Author: MAQUIRRIAIN ANTOÑANZAS, JAVIER
  Programme: DOCTORAL DEGREE IN SUPPLY CHAIN AND OPERATIONS MANAGEMENT
  Department: Department of Management (OE)
  Mode: Normal
  Deposit date: 05/05/2026
  Reading date: 18/06/2026
  Reading time: 11:30
  Reading place: Seminari de l'IOC, planta 11, ETSEIB.
  Thesis director: GARCÍA VILLORIA, ALBERTO | PASTOR MORENO, RAFAEL
  Thesis abstract: This doctoral thesis addresses the problem of cyclic preventive maintenance scheduling in industrial systems, which is NP-hard due to its combinatorial complexity. The problem consists of defining a maintenance policy P that schedules the maintenance of M machines over a cyclic sequence of T time periods, where at most one machine can be serviced in each period. The objective is to determine a policy that minimizes the total costs incurred. To this end, mathematical models and matheuristic algorithms are developed and evaluated.A cost structure combining linear operating costs and stepwise maintenance costs is proposed, providing a more realistic representation of behaviour observed in industrial environments. Based on this structure, two variants of the problem are defined: • Cyclic preventive maintenance scheduling with a predetermined cycle length (SECIMAP_Tpred), in which the maintenance cycle length T is fixed in advance.• Cyclic preventive maintenance scheduling with a variable cycle length (SECIMAP_Tvar), in which T is treated as a decision variable to be determined within the problem.For the SECIMAP_Tpred variant, an exact mathematical model (MMat) is formulated, together with a relaxed version (MPR) that enables the computation of lower bounds, which are subsequently used to assess the quality of the designed (mat)heuristic algorithms. Since MMat is computationally inefficient for industrial-sized instances, three matheuristics (MH1P, MH2P, and MHFI) integrating mathematical programming and heuristic strategies are developed. A sequential procedure (MHSEC) that executes the three matheuristics and retains the best solution obtained is also proposed, achieving an average deviation of at most 3.19% from the optimal solution. In addition, a matheuristic cocktail (COMH) is developed which, by applying an early termination technique based on adaptive prioritization of subalgorithms and cancellation of non-promising solutions, obtains the same results as MHSEC while reducing execution times by 53.37%.The study is then extended to the SECIMAP_Tvar variant, in which the cycle length T is optimized. For this case, a new mathematical model for the relaxed problem (MTI) is developed, together with four hybrid algorithms combining metaheuristic procedures and matheuristic methods (MTI+MHFI, ALG1P, ALG2P, and ALGFI), a sequential algorithm (ALGSEC), and an algorithm cocktail (COALG) designed to minimize computational times. The best resolution procedures for the variable-T case (ALGSEC and COALG) achieve an average deviation of at most 1.43% from the optimal solution. Furthermore, COALG attains the same solutions as ALGSEC while reducing execution times by 33.2% through the use of early termination and intelligent pruning techniques that prevent the exploration of solution configurations unable to improve upon the best solution found so far.The comparative analysis of the results obtained for both problem variants (SECIMAP_Tpred and SECIMAP_Tvar) shows that treating T as a decision variable expands the solution space and, consequently, enables the derivation of maintenance policies that achieve a 2.68% reduction in costs.Finally, the thesis presents the general conclusions, the main methodological contributions (including the introduction of stepwise cost structures or the systematic use of lower bounds computed via mathematical programming to guide heuristic search) and directions for future research.
  

DOCTORAL DEGREE IN SUSTAINABILITY

• RHOUMA, ALI: Operationalizing the Water–Energy–Food–Ecosystems Nexus for the Sustainability Assessment of Mediterranean Farming Systems
  
  Author: RHOUMA, ALI
  Programme: DOCTORAL DEGREE IN SUSTAINABILITY
  Department: University Research Institute for Sustainability Science and Technology (IS.UPC)
  Mode: Normal
  Deposit date: 24/04/2026
  Reading date: 16/07/2026
  Reading time: 12:30
  Reading place: Salón de Grados de la EEABB Campus of Castelldefels
  Thesis director: GIL ROIG, JOSE MARIA | BROUWER, FLOOR
  Thesis abstract: Agricultural systems face increasing pressures from water scarcity, climate change, and environmental degradation, particularly in the Mediterranean region where resource constraints are intensifying. Addressing these interconnected challenges requires integrated analytical approaches capable of capturing the complex interactions between water, energy, food production, and ecosystems. The Water–Energy–Food–Ecosystems Nexus has emerged as a promising framework for supporting sustainable resource management; however, its operationalization at the farm level remains limited. This thesis aims to advance the application of the WEFE Nexus for the sustainability assessment of farming systems by developing an integrated analytical framework and practical evaluation tools.The research adopts a progressive methodological approach. First, water-related sustainability indicators specifically the water footprint and water scarcity footprint are applied to assess the pressures of agricultural production on water resources. These indicators provide an initial understanding of resource use efficiency and highlight critical water-related challenges in farming systems. Building on this foundation, the thesis develops a WEFE Nexus assessment framework based on system dynamics modelling to capture interactions between water use, energy consumption, food production, and ecosystem impacts. The framework is implemented through a user-friendly decision-support tool designed to support sustainability assessments at the farm level.The developed framework is subsequently applied to evaluate WEFE Nexus solutions in agricultural systems, with a particular focus on agroecological practices. By integrating multiple indicators related to resource efficiency, environmental performance, and agricultural productivity, the analysis explores how agroecological approaches influence the performance of farming systems within the WEFE Nexus. The results demonstrate that agroecological practices can improve resource use efficiency, reduce environmental pressures, and enhance the resilience of farming systems.Finally, the outcomes of the WEFE Nexus assessment are translated into key performance indicators linked to the Sustainable Development Goals (SDGs). This step enables the quantification of how WEFE-based agricultural practices contribute to broader sustainability objectives and global development targets. By linking farm-level sustainability assessment with the SDG framework, this research provides a novel methodological contribution for evaluating the sustainability impacts of agricultural practices.This thesis contributes to bridging the gap between WEFE Nexus theory and practical agricultural sustainability assessment. The proposed framework offers a robust approach for evaluating sustainable farming systems and provides valuable insights for policymakers, researchers, and stakeholders seeking to promote resilient and resource-efficient agriculture in the Mediterranean region and beyond.
  
• VILLANUEVA ESCOBEDO, BRENT: Design and Application of an AI-Enabled Adaptive Framework for Assessing Circularity in Socio-Ecological and Technical Systems (SETs) within the WEFE Nexus
  
  Author: VILLANUEVA ESCOBEDO, BRENT
  Programme: DOCTORAL DEGREE IN SUSTAINABILITY
  Department: University Research Institute for Sustainability Science and Technology (IS.UPC)
  Mode: Normal
  Deposit date: 14/04/2026
  Reading date: 17/07/2026
  Reading time: 10:30
  Reading place: UPC ESEIAAT edifici TR1Planta 0 Porta 085
  Thesis director: MORATO FARRERAS, JORDI
  Thesis abstract: Sustainability and circular economy transitions in contemporary socio-environmental systems are increasingly challenged by complexity, uncertainty, and strong interdependencies between natural resources, infrastructures, and governance structures. Traditional sectoral and indicator-based assessment approaches have proven insufficient to capture these interactions, often overlooking systemic trade-offs, emergent behaviors, and context-specific constraints. In response to these limitations, this doctoral thesis develops, applies, and validates integrated, participatory, and adaptive assessment frameworks for sustainability and circularity, grounded in systems thinking and the Water–Energy–Food–Ecosystems (WEFE) Nexus.The thesis is structured as a compendium as follows. The first chapter is the introduction, including the research context of the main concepts, the background and rationale, and the design, starting with the research questions, hypothesis, identification of the research gap, the research lines and objectives. The second chapter, dives into the state of the art of the main concepts, presenting the bibliometric analysis as well as the institutional context and the projects that provided the scene for the applied research. The third chapter is dedicated to the methodology, starting with the global methodology framework and then focusing on the WEFE Nexus and integrated assessment. The fourth chapter, proposes an integrated multi-criteria framework for assessing sustainability and circular economy performance in public water utilities operating in Mediterranean contexts, combining environmental footprints, ecosystem services, governance, and operational performance within a decision-oriented analytical structure. The fifth chapter applies an artificial intelligence (AI)–enabled WEFE Nexus assessment tool to evaluate circular bioeconomy (specifically biochar and agroforestry) highlighting trade-offs, synergies, and context-dependent outcomes under future scenarios in a dynamic way. The sixth chapter consolidates these insights into a transferable methodological framework for participatory, AI-enabled circularity assessment across heterogeneous socio-environmental systems. The seventh chapter is focused on the discussion of the theoretical and the applied research, specifically to proof whether circularity was implemented as a static set of material loops, or if enough evidence is found that circularity was implemented as an emergent system property, resulting from interactions between resource systems, ecosystems, institutions, and stakeholder values in the frame of this thesis. The eighth and final chapter gathers conclusions to measure the degree in which the methodology and its derived applications help integrate multi-criteria decision analysis, participatory processes, artificial intelligence, and digital twin concepts to support transparent, context appropriate and adaptive sustainability assessment. Finally, limitations and future research lines are drawn to set some basic principles for replication and adaptation of the analysis tool in different contexts. Overall, the thesis aims to contribute in advancing the state of the art in sustainability science by operationalizing the WEFE Nexus through AI-enabled and participatory assessment frameworks, bridging the gap between conceptual integration and practical support for decision-making. The obtained results, demonstrate that systemic, adaptive, and governance-aware approaches are essential for informing transitions towards sustainability and circular economy in complex socio-environmental systems with resource constraints
  

DOCTORAL DEGREE IN TEXTILE AND PAPER ENGINEERING

• LEZECK, HENDRICK: Application of essential oils microcapsules on the fabric surface to get antibacterial properties.
  
  Author: LEZECK, HENDRICK
  Programme: DOCTORAL DEGREE IN TEXTILE AND PAPER ENGINEERING
  Department: Department of Engineering Graphics and Design (DEGD)
  Mode: Normal
  Deposit date: 28/04/2026
  Reading date: 04/06/2026
  Reading time: 15:00
  Reading place: INTEXTER Conference Room
  Thesis director: LIS ARIAS, MANUEL JOSÉ
  Thesis abstract: Essential oils (E.O.) are widely used in traditional medicine, pharmacy, food, and cosmetic applications due to their natural origin, biodegradability, and broad antimicrobial activity. Despite these advantages, their high volatility and hydrophobic nature significantly limit their direct application in textile substrates. In recent years, the increasing demand for sustainable, functional textiles has driven research to integrate bioactive compounds into fabrics while preserving their efficacy and durability.This doctoral thesis investigates the application of essential oil microcapsules onto textile substrates as a strategy to overcome the intrinsic limitations of E.O. and to impart bio-functional properties to fabrics. By employing microencapsulation techniques, this work aims to enhance the stability, retention, and controlled release of essential oils on textiles, enabling the development of sustainable materials with long-lasting bioactive performance.
  

DOCTORAL DEGREE IN URBAN AND ARCHITECTURAL MANAGEMENT AND VALUATION

• ESPINOZA ZAMBRANO, PAÚL ANDRÉS: Libro del Edificio Electrónico (LdE-e). Una herramienta para impulsar la rehabilitación de edificios residenciales en España
  
  Author: ESPINOZA ZAMBRANO, PAÚL ANDRÉS
  Programme: DOCTORAL DEGREE IN URBAN AND ARCHITECTURAL MANAGEMENT AND VALUATION
  Department: Department of Architectural Technology (TA)
  Mode: Article-based thesis
  Deposit date: 29/04/2026
  Reading date: 01/06/2026
  Reading time: 16:30
  Reading place: ETSAB (Escuela Técnica Superior de Arquitectura de Barcelona) - Planta Baja - Sala de GradosAv. Diagonal, 649-651 - 08028 - Barcelona
  Thesis director: MARMOLEJO DUARTE, CARLOS RAMIRO
  Thesis abstract: The climate emergency and the European Green Deal call for transforming the building sector, a key area in energy consumption and emissions. In Spain, the challenge is greater due to an aging and inefficient residential stock, compounded by the complex management of horizontal property regimes. Despite regulatory progress, energy rehabilitation remains stagnant. This thesis argues that the problem is not solely technical or economic, but a market failure caused by information asymmetry and documentary fragmentation. "Information vortices" prevent the proper valuation of energy efficiency, perpetuating inaction.To address this, the e-Building Logbook (LdE-e) is proposed as an integrated information management model. The model combines two tools from Directive (EU) 2024/1275: the Digital Building Logbook (DBL), as a static technical archive, and the Building Renovation Passport (BRP), as a dynamic roadmap for staged renovations. Its key innovation lies in its technological development using BIM and Blockchain. An architecture is designed to extract data from IFC files to avoid redundancies, alongside a tokenization gateway ensuring data immutability, traceability, and legal validity.The methodology combines qualitative and quantitative phases. First, the model was validated with 21 experts from the sustainable real estate sector, refining its structure, management, and governance —including the proposal of an LdE-e Consortium— and adapting the technology for professional usability.In the quantitative phase, a social perception study was conducted through a survey of 4,041 households across 14 metropolitan areas. Advanced techniques were applied, including Latent Class Analysis (LCA) to profile adoption patterns, and Bayesian-optimised Dense Neural Networks (DNN) to model the complexity of household decision-making.The results reveal two key findings. First, technology is necessary but insufficient: the decisive factor for LdE-e adoption is support through One-Stop Shops (OSS), which transform hesitant users into active adopters. Second, financial stress shapes motivations: households with low energy costs prioritize environmental or lifestyle values, while vulnerable ones act out of economic pragmatism, seeking subsidies and property aesthetic improvements.In conclusion, this thesis provides a robust LdE-e model aligned with European regulations, resolving the information fragmentation in the Spanish market. Its operational viability depends on integration into a socio-technical ecosystem where BIM and Blockchain provide technical security, and One-Stop Shops (OSS) offer the trust and human support needed to mobilize citizens toward decarbonization. The combination of advanced technology and personalized support emerges as the essential formula to overcome current barriers and activate the virtuous circle of energy rehabilitation in Spain's residential stock.