This small-scale qualitative study explores acceleration within the Swedish school system from a retrospective student perspective. Based on 11 qualitative interviews and content analysis using both deductive and inductive approaches, the study identifies 14 types of acceleration, including gradeskipping and subject-based acceleration. These forms are reflected in research and policy documents, though the term "acceleration" is rarely used explicitly. Findings suggest that acceleration can enhance motivation, learning, self-confidence, and peer interaction, and help students exit negative school situations. However, challenges such as limited access, poor implementation, and social difficulties, especially related to being younger, also emerged. Students who received structured acceleration with teacher or coach support reported the most positive outcomes. Others, who accelerated without instruction, found the process demanding but ultimately beneficial. Access to acceleration often depended on chance encounters with knowledgeable adults, and some students faced resistance. The study concludes that acceleration can be both necessary and beneficial if well-planned, supported, and tailored to individual needs.

This study examines how social workers reason about the provision of support interventions for women exposed to intimate partner violence, and how they perceive organisational frameworks as influencing their professional conditions and discretionary space. The study also aims to highlight which factors social workers perceive as decisive in prioritisation process when working with women exposed to violence. The theoretical framework is primarily based on Michael Lipsky´s theory of street-level bureaucracy and the concept of discretion, complemented by later theoretical judgment. Data were collected through a qualitative research design, consisting of semi-structured interviews with ten social workers employed in eight different municipalities. The empirical material was analysed using a thematic analysis, where recurring patterns were identified and interpreted in relation to the theoretical framework and previous research. The findings indicate that social workers´ decision-making in cases involving women exposed to violence is characterised by a continuous balancing act between the responsibility to provide protection, respect för women´s self-determination, and organisational constrains. Relationships and trust emerge as central prerequisites for the successful implementation of support interventions. Guidelines and policy documents are perceived as both supportive and restrictive, as they enhance legal certainty while simultaneously limiting professional flexibility. Overall, the results suggest that social workers' discretionary space is shaped through an interaction between emotional, relational, and organisational factors.

Syftet med studien är att bidra till en fördjupad förståelse av hur svenska nyhetstidningar genom sina skildringar formar bilden av socialtjänstens arbete med barn och unga. Studien bygger på en kvalitativ analys av 40 tidningsartiklar från lokala, regionala och nationella tidningar publicerade under 2024. Det empiriska materialet har samlats in via databasen Retriever och analyserats med hjälp av kritisk diskursanalys, utifrån ett socialkonstruktionistiskt perspektiv. Analysen fokuserar på hur återkommande språkliga formuleringar och begrepp i nyhetstidningar bidrar till att konstruera socialtjänstens arbete med barn och unga samt hur socialtjänsten, barn och unga positioneras i dessa beskrivningar. Resultatet visar att nyhetstidningars skildringar präglas av fyra övergripande diskurser. Socialtjänsten konstrueras återkommande som ett institutionellt misslyckande, där fokus riktas mot brister, sena insatser och allvarliga konsekvenser för barn och unga. Vidare framträder en diskurs där socialtjänsten framställs som ett hot mot familjen genom språk som betonar tvång, ingripanden och maktutövning. En tredje diskurs konstruerar socialtjänsten som ett administrativt system där byråkrati och administrativa rutiner ges stort utrymme. Samtidigt identifieras en diskurs där socialtjänsten framställs som hjälpande men ifrågasatt aktör, vilket visar på en mer motstridig bild där stödjande insatser samexisterar med misstro och kritik.  Studiens slutsats är att nyhetstidningars språkliga konstruktioner i stor utsträckning bidrar till att reproducera problemorienterade och förenklade bilder av socialtjänstens arbete med barn och unga. Dessa skildringar kan riskera att påverka allmänhetens förståelse av socialtjänstens uppdrag, professionens legitimitet och förtroendet för socialtjänsten, samtidigt som de komplexa villkor och professionella bedömningar som präglar arbetet ges begränsat utrymme i nyhetsrapporteringen. Studien bidrar därmed till en fördjupad förståelse av hur nyhetstidningars språk formar bilden av socialtjänstens arbete i relation till barn och unga i ett samhälleligt sammanhang. 

Modern digital systems rely heavily on firmware updates that are frequently distributed as compressed binary blobs. In forensic investigations and security audits, these blobs often appear withoutfile headers or metadata, rendering standard signature-based extraction tools ineffective. This thesispresents BinSift, a modular Python-based framework designed for the automatic detection, classification, and “blind” decompression of untagged compressed data.To calibrate the system, a large-scale statistical analysis was conducted on the FirmSec dataset, profiling approximately 34,136 firmware images totaling over 200 GB of binary data. Results indicatethat an average Shannon entropy threshold of 7.1 bits per byte provides an optimal balance for capturing modern compression formats like LZMA and SquashFS while minimizing false positives fromhigh-density uncompressed code.The BinSift framework was evaluated against industrial firmware samples, achieving a 59.0% successrate in “True Blind” mode without any prior knowledge of file headers. This approach maintained an81.5% fidelity retention compared to metadata-assisted baselines. When excluding mathematicallyunrecoverable encrypted payloads, the effective success rate rose to 84.4%. These findings demonstrate that entropy-based stream identification and bit-level refinement are viable solutions for bypassing obfuscation in embedded systems forensics.

Syftet med denna uppsats var att undersöka hur diskurser om sexköp konstrueras i digitala samtal, och hur sexköparen som subjekt konstrueras inom dem. Empirin utgörs av diskussionstrådar från det digitala chattforumet Flashback där sexköp som fenomen diskuteras. Studien utgår från en socialkonstruktionistisk vetenskapsteoretisk ansats och använder diskurs som både teori och metod. Resultaten visar på att det finns nyanser i språket om sexköp som huvudsakligen delas upp i rättfärdigande och orättfärdigande diskurser. Dessa kan förstås i relation till manlighet, makt och samhällsnormer. Uppsatsen fokuserar på att synliggöra vilka diskurser kring sexköp som tar form och reproduceras och därmed se hur det sociala arbetet kan förhålla sig till prostitution.

 Transitions between school stages constitute critical phases in students’ education, particularly in mathematics, a subject characterized by a cumulative and hierarchical knowledge structure. This qualitative study aims to investigate how teachers perceive the information provided during the transition from Grade 3, and how this information can be used in planning subsequent instruction. The study is theoretically grounded in the framework of Mathematical Knowledge for Teaching (MKT), which categorizes mathematics teachers’ subject knowledge into different domains.Data were collected through semi-structured interviews with seven teachers experienced in receiving new student groups in Grade 4. The material was analyzed using thematic analysis, resulting in three main themes: Building Further, Creating One’s Own Picture, and The Power of Habit, reflecting three different approaches to transfer information. The findings can be used to understand how certain types of information are valued as bridges from previous teaching methods, strategies, and levels of knowledge, while at the same time there may be distrust toward transferred information. This distrust may, for example, concern assessments and results from national tests, leading teachers to prioritize their own diagnostic tests and observations.Furthermore, the analysis shows that although teachers’ evaluation of transition information activates the MKT domains of Knowledge of Content and Students and Specialized Content Knowledge, the absence of the domain of Horizon Content Knowledge may reduce opportunities for creating continuity between school stages. In such cases, teaching materials and established instructional routines may have a greater influence on planning than the transferred information. Overall, the study offers a conceptualization of the transfer process in mathematics, and the findings may therefore serve as support for the planning and implementation of transition meetings.

Process automation is often portrayed as a straightforward solution, yet its successful introduction and long-term maintenance are organizationally complex. It requires careful planning, strategic management, and sustained commitment from the outset. Automation initiatives also unfold in multi-stakeholder contexts, where various actors are involved in and affected by automation in different ways. Despite the widespread use of process automation, knowledge remains limited regarding how organizations should govern and maintain such initiatives over time, as well as how automation affects employees’ work situations and work environment. This thesis therefore examines the maintenance of process automation by exploring how automation initiatives are organized and governed across development, implementation, and management phases in a multi-stakeholder context, from an organizational work environment perspective. Empirically, the thesis is based on a qualitative and interpretative single-case study of the automation of several human resource (HR) processes at a large industrial company in Sweden. The study focuses particularly on the automation team responsible for maintaining and managing the automated solutions. The empirical material consists primarily of participatory observations and field interviews conducted during a year-long observational study, supplemented by two rounds of semi-structured interviews. The analysis combines stakeholder theory with the demand-control-support model to explore challenges and tensions in the ongoing maintenance and management of process automation. These include technical debt, time constraints, technical issues released to production, support-channel difficulties, end-user-related problems, competing demands and priorities, and insufficient inter-team collaboration. The analysis further shows how these challenges were shaped by structural work conditions characterized by high job demands, limited control, and weak social support. The combined theoretical framing provides a structured way to analyze the tensions that arise when process automation unfolds in complex multi-stakeholder settings. The thesis concludes that the challenges involved in maintaining and managing process automation were simultaneously technical, organizational, and relational. These tensions were linked to a mismatch between the automation initiative’s  objectives and the organization’s arrangements and capacity. They became particularly visible in stakeholder interactions, consultant dependency, and development-oriented maintenance practices. 

The treatment of chronic or hard-to-heal skin wounds remains a significant challenge in healthcare, with an estimated annual cost of approximately 60 billion euros in Europe for wound management.

Cutaneous wound healing requires coordinated activation of stem cells, epidermal cells, and extracellular signaling networks. While most wounds are resolved efficiently, impaired healing remains a major clinical challenge. Increasing evidence suggests that regenerative capacity is governed not only by cellular presence but by dynamic regulation of gene regulatory programs.

This thesis investigates how epigenetic modulation influences regenerative behavior in adipose-derived stem cells (ASCs) and keratinocytes in a wound-healing context. By targeting histone acetylation using the histone deacetylase inhibitor, SAHA, and analyzing microRNA-mediated regulation. In parallel, proteomic analyses were performed to explore the skin proteome using the Total Protein tissue Extraction reagent (TPER) as an efficient isolation method in clinical settings.

The findings demonstrate that SAHA modulates ASCs behavior by altering chromatin structure, DNA replication, and cell fate determination. The microRNA-based approach showed that key signaling pathways, such as those involving epidermal growth factor (EGF) and interleukin-1 alpha (IL-1α), are implicated in promoting skin cell proliferation and repair. Furthermore, skin proteomic analyses, including TPER reagent for protein isolation, may enhance clinical wound treatment.

Together, these results advance the understanding of the interplay between gene regulation and protein function in skin healing and the development of skin regeneration future therapies.

Ethics is a universal subject. Mankind have faced moral problems in all ages and in all cultures. Ethics is the reflection on moral problems.  Ethics is an old discipline. Examples of important teachers of ethics are the Chinese philosopher Confucius and the Greek philosopher Socrates, both living in the 5th century B.C. Socrates took ethics seriously: “we are discussing no small matter, but how to live ” he is said to have remarked and he lived himself accordingly. His teaching challenged the leaders of Athens to the extent that they condemned him to death. We can learn from Socrates that ethics requires both theoretical reflection and personal commitment.

Ethics has become a subject in demand in recent years. There is today in the West a wide public interest in ethics but also widespread uncertainty. Religion used to provide people with an ethical framework, but due to secularisation and modernity, religion has lost much of its influence.  There is in a modern, plural society no given answer to ethical challenges. However, the challenges remain. Let me give some examples:

In health care, the possibilities to save human lives both in the beginning and in the end of life; the neonatal baby born much too early, as well and the very old person with a severe disease, give rise to difficult ethical decisions; who should we save and who should we let die? 

Computerisation and information technology face us with new ethical challenges; how to secure privacy in the Information society?

Environmental problems compel us to re-examine our scale of values; how to balance environmental concerns and economic growth?

These are just some examples of moral problems raised in the modern society. When we reflect on moral problems, we find that ethical arguments rest on various preconditions. Ethics raises questions about what a good human life is and to our relationship to other people, animals and nature. 

We live in an age of globalisation. People travel to different places, news media reports from all over the world, migrants cross borders and companies are multi-national; overall, globalisation implies that our range of responsibility is widened. Thus, globalisation poses a challenge for ethics, and we need more knowledge about values and norms in different parts of the world. This book refers to different moral traditions and compares different ethical outlooks, although the author’s own European heritage puts limits to this ambition. 

Large-scale fluid flow simulations yield multiple coupled scalar and vector fields (e.g., pressure, density, and velocity) together with derived topological structures, making exploratory vortex analysis difficult when raw fields are inspected in isolation. This thesis proposes a multifield visualization workflow for vortex analysis that centers on pressure-derived extremum graphs, whose valley lines are treated as candidates for vortex corelines. The approach is implemented in Inviwo as a coordinated pipeline and combines direct volume rendering (DVR) of the pressure field for scalar context, velocity-field depiction via line integral convolution and tube-based streamlines, and mesh-based rendering of the extremum graph with critical points shown as spheres and valley lines shown as tubes; a custom Inviwo importer was developed to load the extremum-graph format. The DVR transfer function uses multiple narrow opacity peaks placed across the low-pressure range: peak placement resolves the fragmentation-vs.\ occlusion trade-off, while keeping each peak narrow preserves surface-like sharpness and limits accumulated ray opacity. Streamline seeding draws from the critical points of the extremum graph, with 1-saddle seeding emerging as the primary tuned reference for this dataset; a pressure-range filter, a choice between uniform exploratory thinning and histogram-guided thinning, and a randomised off-axis displacement keep streamtubes legible near the interaction region. To complement qualitative structure with quantitative cues, sampled pressure is encoded through an inverse monotonic pressure-to-radius mapping on the graph mesh, with linear and exponential variants compared. The coordinated multifield views support joint interpretation of low-pressure regions, flow orientation, and valley-line geometry, making the transition from two initially parallel low-pressure structures toward the developing horseshoe-like configuration visually interpretable across time, without pinning down an exact reconnection timestamp.

Electrocatalytic CO2 reduction to formate/formic acid represents a practical and economically viable route for CO2 conversion, among which Sn- and Bi-based materials are regarded as promising electrocatalysts. This thesis explores the structure–performance relationship of Sn- and Bi-based catalysts for electrochemical CO2 reduction reaction (CO2RR) to formate, focusing on two effective optimization strategies: mesostructural engineering and heterometal doping. Sn- and Bi-based catalysts are attractive candidates due to their high selectivity for formate, low cost, and environmental compatibility.

In the first study, mesoporous SnO2 enriched with oxygen vacancies shows enhanced CO2RR performance compared with bulk SnO2 and improved durability. Mechanistic studies reveal that the mesostructure enhances CO2 adsorption, facilitates charge transfer, stabilizes the *OCHO intermediate, and lowers the reaction energy barrier. Moreover, the mesoporous framework promotes the formation and stabilization of oxygen vacancies, maintaining the Sn oxidation state and catalyst stability.

In the second study, Sn-doped BiOCl (the ratio of Sn is 2-10 at. %) nanoplates synthesized via a sol–gel method act as precatalysts that rapidly reconstruct into Sn-modified metallic Bi during the CO2RR. Among the samples investigated, the catalyst with a Sn doping ratio of ~5% achieves a Faradaic efficiency of 87.7% at −1.0 V vs. RHE, considerably outperforming pristine BiOCl. Structural and spectroscopic analyses show that Sn incorporation stabilizes the *OCHO intermediate, facilitates interfacial water dissociation, and promotes the formation of active Bi 003 planes, clarifying the dynamic evolution of active sites.

Overall, this thesis demonstrates that rational control of mesostructure, defect chemistry, and dopant-induced reconstruction can synergistically enhance both activity and stability in Sn- and Bi-based CO2RR catalysts. These insights provide practical guidelines for designing efficient and durable formate-selective electrocatalysts.

Background: This mixed-methods dissertation examined the implementation and costs of guideline-based biopsychosocial (BPS) management of low back pain (LBP) in Finnish occupational health primary care (OHPC). A multifaceted implementation strategy was used, combining training and booster sessions with materials, clinical champions, and outreach visits. We aimed to 1) evaluate the quality of LBP management by applying predefined criteria to physiotherapists’ documentation; 2) analyze the effects of the intervention on OHPC resource utilization and work absenteeism compared to usual care; 3) examine how capabilities, opportunities, and motivation influence the implementation of guideline-based BPS management of LBP, as perceived by occupational healthcare professionals and 4) tailor a shortened version of The Determinants of Implementation Behavior Questionnaire (DIBQ) to multiprofessional rehabilitation context and cross-culturally adapt a Finnish language version.

Methods: The cluster-randomized trial (ISRCTN11875357) included six national OHPC providers, with 27 units randomized into intervention and control arms. In the intervention arm, physiotherapists and physicians received targeted 3-to-7-day training in guideline-based BPS management in 2017 and 2018, while the control arm continued usual care.

Study I evaluated the quality of LBP care from physiotherapists’ first-visit documentation using predefined quality criteria applied to electronic patient records. In total, 98 records documented by 28 physiotherapists from intervention and controls arms were analyzed.

Study II analyzed healthcare utilization and LBP-related sick leave data collected from electronic patient records over a one-year follow-up for 232 patients in the intervention arm and 80 control arm patients. Costs were estimated using linear mixed models by multiplying unit costs (in euros) by each type of OHPC resource use including visits to physicians, physiotherapists, nurses, imaging and sick leave.

In Study III, focus group interviews (n=12) were conducted one year after the educational intervention. with physicians, physiotherapists and nurses (n=51) treating patients with LBP in occupational healthcare teams. The data were analyzed using deductive and inductive content analysis informed by the Capability, Opportunity, Motivation, Behavior (COM-B) model.

In Study IV, the cross-cultural translation of English DIBQ to Finnish was followed by a two-round Delphi survey involving experts in multiprofessional rehabilitation including physicians, physiotherapists, occupational therapists, psychologists, nursing scientists, social scientists. Altogether, 25 experts participated in Round 1, and 21 in Round 2. Participants rated the importance of each DIBQ item for inclusion in the final scale on a 5-point Likert scale (1 = Strongly Disagree, 5 = Strongly Agree). Consensus to include an item was defined as a mean score of at least 4 by ≥75% of Delphi participants. Open-ended comments were analyzed inductively. Item- and scale-level content validities were assessed using the content validity index.

Results: Training in guideline-based BPS management was associated with more frequent documentation of behaviors aligned with high-quality LBP care at physiotherapists’ first visit in OHPC compared to usual care.

The multiprofessional training intervention shifted care from physician-led management toward a multiprofessional, physiotherapist-driven model without increasing total costs over 12 months (€-1908, 95% CI €-6734–2919).

Facilitators of implementation were mainly related to increased capability and motivation. Participants reported improved confidence on their own and the teams’ capabilities on treating more challenging patients in OHPC, enhanced communication skills, a broader understanding of the multidimensional nature of LBP, and a strengthened professional role. Barriers were related to opportunities at the organizational and system levels, including limited time and resources, unclear care pathways, inconsistent team agreements, and structural constraints.

In the cross-cultural adaptation process, consensus was reached after Round 2 on the inclusion of 21 items in the multiprofessional DIBQ (DIBQ-mp). The final version covers 11 Theoretical Domains Framework domains. Item-level content validity indices were at least 0.78, and the scale-level content validity index average was 0.93, indicating excellent content validity for multiprofessional rehabilitation context.

Conclusion: Although the BPS training intervention improved the quality of LBP assessment and management, influenced OHPC resource allocation, and strengthened professionals’ capabilities and motivation, implementation into routine practice remained incomplete at 12-month follow-up. Complementary implementation strategies for sustained organizational support and clear care pathways should be further developed to embed BPS management in occupational health practice. The DIBQ-mp appears to be a promising tool for evaluating implementation determinants in multiprofessional guideline-based rehabilitation interventions.

Syfte: Syftet var att kartlägga skadeprevalensen bland svenska amatörfotbollsspelare samt att jämföra förekomst av skador mellan män och kvinnor.

Metod: I en retrospektiv tvärsnittsstudie mättes periodprevalens under fotbollssäsongen 2024. En digital enkät besvarades av 1251 amatörfotbollsspelare (519 män, ålder 14 år ± 5, 10–51 år). För frekvensanalys användes Chi-två-test.

Resultat: Cirka 41% av spelarna rapporterade att de drabbats av överbelastningsskada och 31% av akut skada. Vid jämförelse av könen i olika åldersgrupper avseende prevalens av akut skada fanns signifikant skillnad i åldersgruppen 10–14 år (p<0,001), där prevalensen var högre bland pojkar. Resultaten visar att skadeprevalensen generellt ökade med stigande ålder. För överbelastningsskador var knä den vanligaste skadelokalisationen bland både män (47,6%) och kvinnor (58,6%), med en signifikant skillnad mellan könen (p=0,043). De vanligaste lokalisationerna för akuta skador bland männen var knä (30,7%) och bland kvinnor fotled (40%). 

Konklusion: Skador är vanligt förekommande bland amatörfotbollsspelare, främst bland äldre tonåringar och vuxna. Förekomsten av akuta skador var högre bland pojkar i den yngsta åldersgruppen jämfört med flickor, men inga könsskillnader sågs i de äldre åldersgrupperna eller för överbelastningsskador. Skador drabbade främst nedre extremitet med vissa könsspecifika skillnader i skadelokalisation. Förekomsten av tidigare allvarlig skada var högre bland äldre spelare. 

Denna rapport analyserar vilka insatser landsbygdspolitiken och den regionala utvecklingspolitiken bör omfatta för att stödja kommuner med långvarigt vikande befolkningsunderlag i att anpassa sig till en förändrad demografi. Utgångspunkten är att demografisk omvandling inte är ett tillfälligt problem utan ett varaktigt strukturellt tillstånd som kräver nya politiska angreppssätt. Rapportens huvudsakliga slutsats är att det krävs en platsbaserad sammanhållningspolitik som erkänner att kom-muner verkar under skilda demografiska och geografiska förutsättningar och som stärker statens kapacitet att hantera en sådan diversitet. Sammanhållningspolitiken bör också identifiera mönster av territoriell ojämlikhet och tydliggöra när dessa ojämlikheter är orättfärdiga och påverkningsbara och bör bli föremål för politiska insatser. Detta förutsätter en tydlig policyteori, där staten klargör hur skillnader mellan platser ska förstås, när och varför de är problematiska, samt vilka åtgärder som krävs för att antingen anpassa samhällsorganisationen till dessa skillnader eller aktivt motverka dem. Mot denna bakgrund identifierar rapporten fem centrala reformområden: 

Sammanhållningspolitik som eget politikområde 

Sammanhållningspolitik bör etableras som ett tydligt och samlat politikområde med ansvarigt stats-råd, institutionell förankring och en explicit policyteori.  

Utvecklade finansiella verktyg för territoriell utjämning 

Den statliga finansieringen bör reformeras genom mer träffsäkra statsbidrag, alternativa fördelnings-principer och prövning av nya platsbaserade skatteinstrument. De verktyg som redan finns och som används kan göras mer kända och beskrivas som en del av den svenska sammanhållningspolitiken. 

Ett utökat statligt ansvar för samhällskritisk infrastruktur och bostäder 

Staten bör ta ett tydligare ansvar för vatten- och avloppssystem samt hållbara bostadsmarknader i krympande kommuner.  

Ett långsiktigt nationellt kunskapsstöd 

En delegation för demografisk anpassning bör inrättas för att stärka analys, lärande och erfarenhets-utbyte på nationell, regional och lokal nivå.  

Stärkt lokal kapacitet och platsbaserad samverkan 

Kommunernas strategiska kapacitet bör stärkas genom långsiktiga och flexibla styrformer samt bre-dare platsbaserad samverkan.  

Sammanfattningsvis förordar rapporten en politik som har som uttalat mål att hålla samman Sverige genom att minska geografiska skillnader i levnadsvillkor samt att stärka statens kapacitet att hantera olikartade geografiska och demografiska förutsättningar. 

The two purposes of this thesis have been to conduct research in order to highlight the contrasts and changes in political values and ideas surrounding homelessness in Sweden, and also relate these values and ideas to justice in Swedish politics in which John Rawls’ theory of justice has been used as a theoretical framework. The analysis focuses on two different Swedish national strategies from 2007-2009 and 2022-2026. 

The work is designed using a qualitative thematic content analysis in which the strategies have been read and the relevant data has been categorized in four different themes connected to the first research question, which is how values and ideas regarding homelessness have changed in the Swedish national strategies on homelessness of 2007-2009 and 2022-2026. The second research question asks how these changes can be understood in relation to justice in Swedish politics.

The first theme is regarding the knowledge of homelessness, the lack / need of and gained knowledge. Secondly, homeless people as a heterogeneous group, which includes how people within homelessness differentiate and are not categorized as one group of people, for example differences between men, women and children or asylum seekers and migrants as homeless people. The third theme clarifies who the strategies assign responsibility to in the work on homelessness. The fourth theme discusses direct and preventive actions and measures which have been or are to be put in place according to the strategies. The fifth and final theme connects the source material to our chosen theory, which is John Rawls’ Theory of Justice. 

Our findings reveal that gaps of knowledge of homelessness have been filled in the second strategy compared to the first. Differences in groups of homeless people have been clarified and more groups identified. The responsibility of the work with homeless people has been broadened in the second strategy and it also presents a stronger cooperation among different types of authorities, levels of government and organisations in the work against homelessness.

Urban transportation planning faces increasing complexity as cities seek to optimize mobility systems without extensive historical data. This paper presents CI-SGNN (City-Invariant Spatial Graph Neural Network), a novel framework for cross-city bike-sharing demand prediction that leverages Point of Interest (POI) distributions and spatial attention mechanisms. Our approach addresses the critical challenge of predicting categorical mobility demand in new urban environments by learning transferable relationships between urban amenities and travel patterns from source cities. The framework integrates OpenStreetMap POI features with GNNs, enabling zero-shot transfer learning across diverse metropolitan areas. We formulate demand prediction as a multi-class classification problem, categorizing origin-destination pairs into five demand levels. Experimental validation using real CitiBike data from Manhattan and Washington DC demonstrates superior performance, achieving 72.4% accuracy, which overperforms state-of-the-art base-lines. The attention-based spatial aggregation mechanism effectively captures inter-zone dependencies. Our results demonstrate successful zero-shot adaptation capabilities, enabling practical deployment for bike-sharing infrastructure planning in cities lacking historical mobility data using only publicly available urban features. 

Hydrothermal carbonization is increasingly used to convert wet biomass residues into carbon-rich hydrochar, yet fresh hydrochar can exhibit strong odours that indicate retained volatile organic compounds. In this study, hydrochars were produced from wine pomace, olive tree sawdust, and rice husk at two hydrothermal carbonization temperatures (180 and 220 °C) and two residence times (2 and 6 h). The volatile organic compounds emissions were quantified in a microchamber at 23, 40 and 80 °C, collected on Tenax under a controlled nitrogen flow, and analysed by thermal desorption gas chromatography and mass spectrometry. Results are reported as mass-specific emission factors (micrograms per gram per hour) using toluene-equivalent semi-quantification and class-resolved interpretation.Across all conditions, olive tree sawdust hydrochars emitted the highest concentrations of volatile organic compounds, followed by rice husk and wine pomace hydrochars, with grand mean values of 86.1, 33.5, and 9.2 µg⋅g−1⋅h−1, respectively. Peak emissions were observed for olive tree sawdust hydrochar produced at 220 °C for 6 h and extracted at 80 °C (152.7 µg⋅g−1⋅h−1); corresponding maxima for rice husk and wine pomace were substantially lower. Increasing microchamber temperature consistently mobilized semi-volatile compounds and shifted profiles towards less volatile fractions, supporting temperature stepping as a rapid diagnostic for batch comparison.Oxygenated compounds dominated overall. Rice husk hydrochars showed a stronger carbohydrate-derived furanic signature, whereas olive tree sawdust hydrochars additionally released abundant lipid- and phenolic-related compounds. Feedstock selection emerged as the primary lever to reduce emissions, while process severity and handling temperature further shaped both magnitude and composition.

Industrial maintenance workflows rely on Computerized Maintenance Management Systems (CMMS) to manage structured asset data. However, spatial representations of facilities are often maintained separately as 2D visual layouts, with limited support for richer 3D representations. Mapping and updating these visual structures is largely a manual process, making it time-consuming, difficult to maintain, and weakly integrated with hierarchical asset data. This thesis evaluates an AI-assisted approach for identifying functional elements in 2D industrial facility layouts, estimating their spatial placement, and integrating the results into CMMS workflows. The work combines two parts: an empirical evaluation of state-of-the-art multimodal AI-based systems and a proof-of-concept prototype that connects AI-generated layout interpretations to an interactive 3D environment.

The empirical evaluation showed consistently high semantic correctness when functional elements were identified, indicating that multimodal AI systems can interpret visual and textual layout cues reliably. Spatial placement was less reliable, with common errors including imprecise localization, systematic shifts, and oversized bounding regions. These results suggest that current models are useful for layout interpretation, but lack the geometric precision required for fully automated mapping. A user evaluation with industry experts showed that AI-assisted features, such as pre-generated placement suggestions, can reduce perceived manual effort while preserving user control. Their usefulness depended on interpretable and easily adjustable outputs, highlighting the importance of human-in-the-loop interaction design.

Overall, the thesis shows that multimodal AI can provide support in industrial layout mapping, but is not yet sufficient for full automation. Its main value lies in augmenting human workflows rather than replacing manual processes. The thesis contributes an empirical evaluation of AI capabilities in this context and design insights for integrating AI-assisted visual mapping and 3D spatial support into CMMS systems.

Sweden aims to become independent in nitrogen fertilizer production after the international market proved volatile under the Russian invasion of Ukraine. Domestic organic recycling of nitrogen-based fertilizers is insufficient, and new domestic mineral fertilizer production is required to meet demand. The world also faces a climate crisis, and Sweden is working to reduce its climate impact. Knowledge of Nitrogen-Based Mineral Fertilizer (NBMF) production and the factors affecting its environmental performance is limited in Sweden, which lacks domestic production. The objective of this master’s thesis is therefore to provide broad insight for Swedish decision-makers regarding possible variations in urea production, the most common NBMF. To achieve that insight, the study aims to identify key factors shaping urea production’s environmental and energy performance across varying Swedish and European conditions, and with it answer whether Sweden offers climate benefits compared to continued European imports.

This study conducts a Life Cycle Assessment of urea production, with the functional unit of 1 tonne of urea ready for market, produced in generalized locations in Sweden and Europe, which are the two options in the factor Geographical location. Inquired data reflect industry production and are aggregated into a mass and energy flow audit in a Python program, which also uses assessment data from the Ecoinvent3 database to estimate greenhouse gas emissions and primary energy usage. Potential key factors are identified using an uncertainty identification guide and then filtered for relevance, scope, and impact on the basic Python model (meaning without considering other factors). The chosen subset of factors is then implemented to form the factor-based scenario analysis, structured based on the Design of Experiments.

The study´s findings are presented in a table of potential key factors affecting the environmental performance of urea production. All factors listed are estimated in the factor-based scenario analysis to create 36 distinct scenarios or are provided as rough estimates. Additional renewable input flows shift material and energy inputs to the best-performing renewable option from an environmental and energy perspective. It is the most prominent factor that reduces greenhouse gas emissions by an estimated 83% and 91% under Swedish and European conditions, respectively. The same factor also reduces primary energy usage by an estimated 37% and 48% under Swedish and European conditions, respectively. In addition, it has a consistent and significant impact across all scenarios. Geographical location also affects greenhouse gas emissions and primary energy, with Swedish conditions emitting 50% ± 10% greenhouse gases than similar European conditions in 2/3 of the scenarios. The geographical difference diminishes and eventually disappears as the share of Additional renewable input flows increases. Other factors have a rough estimate of their impact on normal Swedish and European conditions, as executed and discussed. The discussion addresses the knowledge gap regarding the impact of general factors on urea production. The validity of the study´s results is discussed, why the conclusions are trustworthy, and what the uncertainties can and cannot affect.

The conclusion is that 16 factors affecting urea production are identified, however, it is challenging to determine which factors significantly affect the environmental performance of urea. However, Additional renewable input flows and Geographical location are identified as critical factors, with Additional renewable input flows as the most important factor in reducing greenhouse gas emissions and primary energy usage. Construction of domestic urea production is in line with Sweden's goals of fertilizer independence and reduced climate impact.

Consumption of textiles has the fourth highest environmental impact in Europe and Europeans generated almost 7 million tonnes of post-consumer textiles in 2022. These textiles must be managed, which requires collection and sorting to prepare them for reuse, recycling or waste treatment. There is a knowledge gap of the environmental impact of post-consumer textile management systems on a municipal level where heterogenous textile flows are considered. Therefore, this study aims to examine the local Swedish system for end-of-life treatment of post-consumer textiles, including its actors and material flows, while also investigating the challenges and opportunities faced by the different actors. Evaluating the environmental impact of different systems, where local conditions are considered, could help guide the development of municipal strategies towards low environmental impact regarding post-consumer textiles. 

This study investigates material flows and environmental impact caused by a municipal post-consumer textile management system in a Swedish city using material flow analysis (MFA) and life cycle assessment (LCA). The amount of textiles collected in the system was about 660 tonnes in 2023, which equals 4.0 kg/person. Most of the collected textiles were exported (71%) while 18% were reused locally and 11% were incinerated locally. Most of the exported fraction was reused, which brings the total reuse in the system to 75% while recycling is 11% and incineration is 14%. 

The LCA showed that reusing and recycling textiles while replacing virgin production offers the most environmental benefits. The recycling process, incineration with CHP and export are the end-of-life activities that contribute most to the environmental impact, mainly global warming, freshwater eutrophication and fossil resource scarcity. Collection and sorting only has small contributions. The normalised results show that freshwater eutrophication contributes relatively the most to the environmental impact in the baseline scenario and fossil resource scarcity contributes relatively the most in the optimised scenarios. The optimisation of scenarios was based on changes to the logistics in the system where end-of-life activities were moved locally to reduce transport distances. Based on this study’s findings, the most effective actions to reduce the environmental impact of the end-of-life activities in the post-consumer textile management system are moving recycling to Sweden, diverting textiles from incineration to recycling and decreasing transportation distances. 

The Vehicle Routing Problem (VRP) is a fundamental combinatorial optimization problem concerned with determining cost-efficient routes for a fleet of vehicles serving a set of customers under operational constraints. In recent years, the electrification of transportation has led to the emergence of the Electric Vehicle Routing Problem (EVRP), where routing decisions must account for battery charging requirements and energy-related constraints. These additional considerations significantly increase the complexity of the problem.

This thesis studies exact solution approaches for the EVRP based on branchandprice algorithms, the state-of-the-art methodology for solving large-scale vehicle routing problems to optimality. Branch-and-price combines branch-and-bound with column generation, where the master problem is solved using linear programming relaxation and new columns (routes) are generated dynamically by solving a pricing problem. For the EVRP, the pricing problem takes the form of an elementary shortest path problem with resource constraints, which is typically solved using labeling algorithms.

The thesis consists of three contributions that address different components of the branch-and-price framework. The first and third contributions focus on efficient pricing to speed up the overall algorithm. In the first contribution, bounding techniques tailored to structural properties of the EVRP are introduced, enabling early elimination of non-promising labels and significantly accelerating the labeling procedure. The third contribution instead considers heuristic pricing. A subgraphbased approach is proposed in which the pricing problem is solved on carefully selected subgraphs constructed using a machine learning procedure. This procedure generates subgraphs likely to contain high-quality routes, thereby reducing computational effort while maintaining solution quality.

The second contribution addresses the master problem and proposes a primal heuristic designed to obtain high-quality integer solutions early in the branch-andprice search. The method constructs routes that are compatible from an integrality perspective, enabling rapid identification of feasible solutions even when proving optimality is computationally demanding.

Overall, the proposed methods improve the computational performance of branch-and-price algorithms for the EVRP and contribute to making exact largescale electric vehicle routing more practically tractable.