Doctoral School

Reading date: 07/01/2026

• GONZÁLEZ GUTIÉRREZ, CÉSAR: Analyzing and Leveraging the Structure of Pre-trained Embeddings
  
  Author: GONZÁLEZ GUTIÉRREZ, CÉSAR
  Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
  Programme: DOCTORAL DEGREE IN ARTIFICIAL INTELLIGENCE
  Department: Department of Computer Science (CS)
  Mode: Normal
  Deposit date: 27/11/2025
  Reading date: pending
  Reading time: pending
  Reading place: pending
  Thesis director: QUATTONI, ARIADNA JULIETA
  Thesis abstract: Developing models with limited annotation budgets (few-shot learning) is of great importance due to the high costs associated with data annotation.Recent advances in text classification have demonstrated that representations derived from pre-trained language models play a crucial role, especially in few-shot learning settings. These new advancements raise two natural questions:1) What properties of pre-trained representations can explain their effectiveness in few-shot learning?, and2) Can we leverage these properties to further enhance performance under limited annotation conditions? In the first part of this work, we address the first question and show that the effectiveness of pre-trained representations in few-shot scenarios can be explained by the degree of alignment between supervised task labels and the hierarchical structure of the pre-trained embedding space. In the second part, we propose a label propagation method designed to exploit this alignment, leading to improved performance in few-shot classification tasks.
  

Reading date: 08/01/2026

• ALCÓN DOGANOC, MIGUEL: Verification and Validation Solutions for the Safety Compliance of Autonomous Driving Frameworks Performance Aspects
  
  Author: ALCÓN DOGANOC, MIGUEL
  Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
  Programme: DOCTORAL DEGREE IN COMPUTER ARCHITECTURE
  Department: Department of Computer Architecture (DAC)
  Mode: Normal
  Deposit date: 01/12/2025
  Reading date: pending
  Reading time: pending
  Reading place: pending
  Thesis director: ABELLA FERRER, JAIME | MEZZETTI, ENRICO
  Thesis abstract: Autonomous Driving (AD) has rapidly evolved from a research concept into an industrial reality. The increasing computational demands of autonomous vehicles have motivated the use of high-performance Multi-Processor Systems-on-Chip (MPSoCs), which offer both performance and energy efficiency. However, ensuring the safety compliance of such complex systems remains a major challenge. The software frameworks used to implement AD functionalities—typically integrating Artificial Intelligence (AI) algorithms—are not designed following a safety-driven development processes, and their non-deterministic behavior conflicts with the strict determinism required by safety standards. This thesis addresses these challenges by developing Verification and Validation (V&V) solutions that improve the safety compliance of AD frameworks, with a particular focus on performance-related aspects.The thesis begins by analyzing the main sources of non-determinism in AD systems across three layers: algorithmic, software architectural, and hardware platform. While variability exists in all layers, the software architecture layer is identified as a key contributor to the overall unpredictability. It not only introduces its own sources of variability but also amplifies those inherited from the other layers. This makes software architecture an effective focal point to improve system determinism and safety assurance.At the foundational level, the thesis addresses the challenge of unit testing within already-integrated AD frameworks, using the open-source Apollo AD framework as a case study. Due to tight coupling and data dependencies among its modules, Apollo does not easily support independent module validation. To enable proper verification of software units, the thesis proposes a systematic methodology to isolate, modify, and reconfigure Apollo modules into standalone, testable units, thus reintroducing unit-level testing capabilities into a complex, AI-based AD framework.The work advances toward system-level safety assurance through the development of dynamic and execution views of Apollo. Dynamic views describe the interactions among software components, linking safety requirements with their implementation and validation tests. However, these views alone fail to capture the concurrent behavior and execution parallelism of the system, which are crucial for verifying performance-related safety requirements. To fill this gap, the thesis introduces execution views, which complements dynamic views by integrating runtime information gathered from execution tracing on MPSoC platforms. Execution views enhance the observability of resource usage, timing behavior, and concurrency, allowing both improved testing and optimized hardware utilization—key aspects for reducing cost and ensuring safety.Finally, the thesis addresses the timing behavior and variability across software components. It identifies, formalizes, and applies a comprehensive set of timing-related metrics capable of capturing inter-module interactions and end-to-end latency properties in AD applications. Traditional timing metrics, such as worst-case execution and response times, fail to capture the interdependencies between components in systems like Apollo. By adopting complementary metrics such as maximum reaction time and maximum time displacement, the proposed approach provides deeper insights into timing dependencies, enabling early detection of timing anomalies and improving validation confidence.Overall, this thesis provides a set of methodologies and tools to improve the V&V of AD software from a safety-performance perspective. The proposed contributions bridge the gap between high-performance AI-based software and the stringent determinism required by safety standards. These advances support the systematic assurance of safety in AD frameworks, ultimately contributing to the reliable and certifiable deployment of autonomous vehicles on high-performance embedded platforms.
  

Reading date: 09/01/2026

• CASTELLARNAU VISUS, MARIA ANGELES: Cosecha de lo invisible. Paisaje de agua en la Val de Ayerbe
  
  Author: CASTELLARNAU VISUS, MARIA ANGELES
  Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
  Programme: DOCTORAL DEGREE IN ARCHITECTURE, ENERGY AND ENVIRONMENT
  Department: Department of Architectural Technology (TA)
  Mode: Normal
  Deposit date: 26/09/2025
  Reading date: 09/01/2026
  Reading time: 12:30
  Reading place: ETSAB (Esc. Técnica Sup. Arquitect) - Pl.Baja - S.Grados Av. Diagonal, 649 - 08028 - BCN (Videoconf: https://meet.google.com/vfh-ownh-czn; Inicio: 12:00 h)
  Thesis director: CUCHÍ BURGOS, ALBERTO
  Thesis abstract: In a context of environmental crisis evidenced by the loss of biodiversity, drought, percolation in the functional structures of productive landscapes, global warming, and trends towards irreversible positions; of social crisis that reveals the strong depopulation of rural areas, the imbalance, the loss of community social structure, and the loss of linkage of people with the natural environment. And of economic crisis resulting in a loss of land use, changes in the system of ownership, the fragility of agricultural and livestock activities increasingly subjected to the costs of industrialization and inputs in the sector, and the tensions of the markets of production and distribution of food. The inland territories of the northeastern Iberia peninsula are thrown into a critical environmental and social vulnerability that jeopardizes the sustainability of habitability in these territories and their systemic functionality as resource and food producing territories.The present research aims, through the analysis of the management of material flows in the cultural landscapes of the pre-Pyrenean zone of Huesca, to find the keys that in the interrelation between architecture and agriculture reveal the strategies that make it possible for human beings organized in society to inhabit these territories.The systemic analysis of the biophysical matrix on which pre-industrial societies organized in communities manage resources to inhabit the territory provides the foundations for the management of material resources that make possible the sustainability of productive systems and, therefore, of habitability in these arid territories.Knowing and understanding the fundamental principles that govern the systemic functioning of the dynamics of micro-systemic and macro-systemic exchanges and management of material resources applied by pre-industrial societies will allow the development of strategies to achieve habitability in these territories and in territories with a similar biophysical and climatic matrix in a post-oil scenario.The present research analyzes this management in the area of the Ayerbe valley, an eminently agricultural territory typical of the pre-Pyrenees in Huesca. From this analysis of the biophysical matrix, the pre-industrial social group and its structure and dynamics of community management and the strategies of water and soil management, technological strategies that allow maintaining the viability of inhabiting these territories are refined.The methodology used consists of a cartographic study, an interview, the study of an 1856 land survey, the analysis of the internal regulations of the irrigation communities, fieldwork and the case study of the different systems of soil and irrigation management.The research results describe these modeling technologies of the natural hydrological system and soil geomorphology, which are deployed in water and soil harvesting and irrigation systems that govern as fundamental laws in the construction of the cultural landscapes of this territory. Thus, flooding, infiltration, drainage, runoff, catchment, conduction, decanting, storage, evapotranspiration, terracing, etc. technologies are described, which aim to replenish nutrients, maintain soil fertility, prevent erosion, optimize water harvesting, adapted to the local rainfall regime, crop cycles and management, and soil structure.In conclusion, the technological strategies detected are governed by fundamentals based on the laws of nature adapted to this climate and this biophysical matrix and are therefore susceptible to reconsideration for the development of strategies for the management of material resource flows not based on the use of fossil fuels in the future.
  

Reading date: 12/01/2026

• ADAMO, ANGELA: Contribution to the decarbonisation of energy intensive industries in the path of the European Union objectives. Application to the case study of SEAT
  
  Author: ADAMO, ANGELA
  Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
  Programme: DOCTORAL DEGREE IN SUSTAINABILITY
  Department: University Research Institute for Sustainability Science and Technology (IS.UPC)
  Mode: Article-based thesis
  Deposit date: 18/11/2025
  Reading date: 12/01/2026
  Reading time: 16:00
  Reading place: Sala polivalent, EEBE, Edifici A, Campus Diagonal-Besós
  Thesis director: MARTIN CAÑADAS, MARIA ELENA | DE LA HOZ CASAS, JORGE
  Thesis abstract: The urgent need to address climate change is intensifying global efforts to decarbonize all sectors, especially the industrial sector, which remains one of the most challenging due to its high-temperature demands and complex operations. Among the most promising solutions is electrification through High Temperature Heat Pumps (HTHPs), potentially combined with electric boilers.This thesis assesses the decarbonization potential of HTHPs in industrial cogeneration systems, using a real case study: the Combined Heat and Power (CHP) plant at SEAT’s automotive factory in Martorell, Spain. Currently powered by natural gas, the plant provides superheated water (SHW) and is a major source of the site’s CO₂ emissions, while facing increasing environmental and regulatory pressure.Unlike prior studies that use simplified or idealized models, this work develops a high-fidelity hybrid thermodynamic model of the CHP system, based on one year of operational data and realistic constraints of electrification technologies. Two modeling approaches were explored—a purely thermodynamic model and a hybrid model integrating empirical data to compensate for sensor inaccuracies. The hybrid model, with lower error margins, was chosen for further analysis.The model includes all major components: gas and steam turbines, post-combustion heat recovery boiler (HRB), absorption chillers, air coolers, and auxiliary boilers, enabling accurate simulation of the plant under real conditions. The technical and economic viability of replacing gas-based heat production with HTHPs and electric boilers was assessed, considering performance limitations (e.g., efficiency loss at high temperatures), availability of low-temperature heat sources, and electricity market dynamics.A key contribution is the evaluation of how current regulatory and market conditions—especially incentives favoring gas-based CHP—impact the competitiveness of electrified solutions. The thesis concludes by analyzing optimal HTHP sizing under various scenarios, considering CO₂ pricing, thermal demand, and plant dynamics.Findings suggest that, although technically feasible, electrification is significantly influenced by regulatory and economic frameworks. The study highlights the importance of detailed modeling, realistic assumptions, and strategic alignment. It also reveals a broader issue: many industrial players lack the data infrastructure and planning needed to implement deep decarbonization. This work provides a replicable methodology and valuable insights for engineers, operators, and policymakers committed to reducing industrial carbon emissions.
  

• RAMIREZ PEREZ, ALEXIS JOHARIV: Comportamiento a flexión y cortante de un tablero continuo de vigas pretensadas con tendones de polímeros reforzados con fibras (FRP)
  
  Author: RAMIREZ PEREZ, ALEXIS JOHARIV
  Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
  Programme: DOCTORAL DEGREE IN CONSTRUCTION ENGINEERING
  Department: Department of Civil and Environmental Engineering (DECA)
  Mode: Normal
  Deposit date: 22/10/2025
  Reading date: 12/01/2026
  Reading time: 11:00
  Reading place: C1-002
  Thesis director: OLLER IBARS, EVA MARIA | MARI BERNAT, ANTONIO RICARDO
  Thesis abstract: The durability of reinforced concrete structures is mainly compromised by steel corrosion, which generates high maintenance costs and reduces structural safety. Fiber-reinforced polymers (FRP) represent an alternative of great interest, as they provide high specific strength and are not susceptible to corrosion. However, their application as active reinforcement in continuous prestressed members is still very limited, due to the scarce experimental research on their structural performance and the absence of specific design guidelines.The main objective of this dissertation is to analyze the flexural and shear behavior of a two-span continuous bridge at 1/3 scale, built with precast prestressed girders and a cast in situ reinforced concrete slab, using carbon carbon fiber composite cables “CFCC” tendons as active reinforcement. The research was organized into three phases: (1) characterization of carbon fiber (CFRP) bars, glass fiber (GFRP) bars, and CFCC tendons, with the latter selected for prestressing due to their suitability; (2) a flexural test on span 1, with a concentrated load applied at midspan, to study the global flexural behavior at the serviceability and ultimate limit states; and (3) a shear test on span 2, with a concentrated load applied 1.6 m from the end support, to evaluate shear strength, effectiveness of GFRP stirrups, and the influence of CFCC prestressing. The results were compared with numerical simulations using the CONS program and with the CCCM analytical model adapted to FRP tendons. The experimental tests showed that CFCC tendons reached 62–76% of their ultimate strength without anchorage slip in the flexural test, confirming their reliability as active reinforcement. Failure was governed by shear-off at the girder–slab interface. In shear, failure occurred after a characteristic diagonal cracking pattern and progressive redistribution of stresses between spans, while shear-off failure was avoided through a reinforcement added after the flexural test.The overall contribution of this dissertation lies in providing the first comprehensive experimental, analytical, and numerical evidence on a continuous bridge prestressed with CFCC tendons. The findings strengthen confidence in the use of FRP in concrete structures, and open new research avenues aimed at optimizing transverse reinforcement and moving towards the codification of this technology.