Blekinge Institute of Technology
Not a member yet
13576 research outputs found
Sort by
Aged to perfection? : Unpacking the curvilinear relationship between founder age and new venture performance
No full textThis study contributes to the growing body of research on the relationship between founders' age and new venture performance, addressing existing gaps regarding founders approaching or surpassing retirement age. By analyzing comprehensive Swedish administrative data covering all newly incorporated firms and their founders, we first document an inverse U-shaped relationship between founders' age at founding and firm survival, as well as between founders' aging and employment growth. However, our findings reveal significant disruptions in these patterns between ages 60 and 70. Specifically, the relationship between founders' age at founding and firm survival temporarily improves at the conventional retirement age of 65, transitioning into a more nuanced, bimodal M-shaped pattern. In contrast, the relationship between founders' aging and employment growth plateaus around age 65. Using abductive analysis and a regression discontinuity design, we show that the increased firm survival observed after age 65 is influenced by a different selection of entrepreneurs, as individuals with enhanced financial security from pension access, along with previous industry experience and high levels of education, are more likely to start businesses at this stage. Ageing and Entrepreneurshi
On Evaluation of Data Stream Clustering Algorithms : A Survey
No full textData stream mining is a research area that has grown enormously in recent years. The main challenge is extracting knowledge in real-time from a possibly unbounded data stream. Clustering, a process in which groupings within the data are identified, is a valuable technique to extract and identify underlying structures of the data. An open question in stream clustering is how to evaluate the proposed algorithms. In this survey, we review the literature in the domain to identify common methodologies, datasets, and evaluation measures, used to evaluate the algorithms. We provide a short summary of the stream clustering algorithms in the literature, but our primary focus lies in the survey of cluster validation relevant to the evaluation of data stream clustering algorithms. We begin our literature review with the inception of clustering incrementally, namely with the introduction of the balanced iterative reducing and clustering using hierarchies (BIRCH) algorithm. We identify that the evaluation methodologies primarily focus on performance, and that aspects such as cluster quality are rarely considered. Performance has been the focal point of all evaluations, both in terms of computational performance and accuracy, since the inception of clustering data streams. We also identify that issues in the conventional clustering domain are present in the data stream clustering. However, minor additions to the evaluation methods can improve both the applicability and usefulness of the algorithms.
Enhancing Maintainability in Robot Framework Test Suites : An AI-Based Refactoring Approach
No full textBackground. As software systems grow, maintaining large test suites becomes increasingly challenging, particularly in keyword-driven testing frameworks like Robot Framework, where semantic test smells often emerge. These include overly long test cases, inconsistent structure, and redundant logic—issues that reduce readability, reusability, and long-term maintainability. A clear gap remains in tools that support automated refactoring for such issues. Objectives. This thesis aims to design and evaluate an automated pipeline that improves the maintainability of Robot Framework test suites while preserving their original behavior. The work was conducted in collaboration with Ericsson, a leading telecommunications company, which sought to address maintainability challenges in its industrial-scale test base. Methods. A hybrid approach was implemented by combining rule-based smell detection with AI-assisted refactoring. Scoped prompts and programmatic insertion techniques were used to ensure precision and preserve behavioral consistency. A custom structural validation tool was developed and used to evaluate the system across ten industrial test cases, and a developer survey was conducted to gather insights into the effectiveness of the refactoring. Results. The pipeline refactored nine out of ten test cases with full behavioral preservation, and achieved 92.6% preservation in the remaining case. Developers generally preferred the refactored versions, highlighting improvements in structure, clarity, and readability. Conclusions. The results demonstrate that combining deterministic detection with scoped AI assistance can enhance test suite maintainability without compromising test logic or developer confidence. This approach provides a practical foundation for future AI-driven tools aimed at scalable and maintainable test development
Växer så det knakar : Vikten av god ljudplanering i expanderade storstadsområden
No full textI ett samhälle med växande städer och en ökande bullerproblematik blir ljud en allt viktigare aspekt att överväga i fysisk planering. Detta arbete bidrar med en djupare förståelse av hur en inkludering av ett ljudperspektiv och en positiv inställning till ljud kan användas fördelaktigt i fysisk planering, för att skapa hållbara och attraktiva stadsmiljöer. Genom en kvalitativ innehållsanalys av offentliga planeringsdokument undersöks hur ljud omtalas och diskuteras i en samtida svensk kontext, medan en deltagande observation genomförs för att undersöka hur ljudupplevelser påverkar den generella uppfattningen av stadsrum i praktiken. Resultaten från innehållsanalysen visar att det finns en god inkludering av ljud i planeringen, främst genom bullerreglering; samtidigt som rådande statistik visar att andelen bullerutsatt befolkning fortsätter öka. Observationen bekräftar tidigare genomförd forskning och visar att en stor del av hur ljud upplevs på platser kopplas till kontext och visuellt uttryck, att människor visar en preferens för ljud skapat av andra människor, och att trivsamma ljud kan fördelaktigt introduceras på platser som distraktion mot buller
Staden är full av vatten : Ett gestaltande projekt om öppen dagvattenhantering i befintlig stadsmiljö
No full textI takt med att städer förtätas minskar andelen grönområden som ersätts av hårdgjorda ytor, vilket försämrar markens förmåga att absorbera nederbörd. I kombination med klimatförändringar och ökade nederbördsmängder ökar översvämningsrisken. Detta ställer krav på nya förbättrade och mer hållbara lösningar för att hantera dagvatten. För att möta dessa utmaningar undersöker studien vilka hållbara dagvattenlösningar som kan integreras i urbana miljöer för att skapa en mer effektiv dagvattenhantering samt hur dessa lösningar bidrar till en förbättrad livsmiljö. För att driva forskningsprojektet vidare har två specifika fall valts ut, Augustenborg i Malmö samt Norra djurgårdsstaden i Stockholm. Strategier för att ta fram gestaltningsförslag har därav resulterat i kvalitativ forskningsstrategi i form av fallstudie utifrån två forskningsmetoder, vilket är dokumentär forskning samt platsanalys. Vidare syftar studien till att ta fram ett gestaltningsförslag i Kalmar kommun i området Malmen. Studieområdet delas in i fem delområden. Gestaltningsförslagen visar på hur olika öppna dagvattenlösningar kan utformas i redan befintliga tätbebyggda stadsområden. Genom att utgå ifrån platsens lokala förutsättningar har studiens gestaltningsförslag visat på hur öppna dagvattenlösningar kan hantera kraftiga skyfall och därav minska risken för översvämningar.
Ett hållbart lantbruk : Ekonomiska förutsättningar och hinder för elektrifiering och elproduktion
No full textBakgrund. Lantbrukssektorn står inför växande krav på att minska sin klimatpåverkan, samtidigt som energibehovet är fortsatt högt, särskilt för drift av maskiner och anläggningar. Elektrifiering och förnybar energiproduktion lyfts fram som möjliga vägar mot ett mer hållbart lantbruk, men implementeringen bromsas av tekniska, ekonomiska och praktiska utmaningar. För att förstå förutsättningarna för omställning till grön energi inom lantbruket krävs insikt om lantbrukares uppfattningar, behov och teknikval. Syfte. Syftet med studien är att baserat på svenska lantbrukares egna erfarenheter öka förståelsen för de ekonomiska förutsättningar och hinder som påverkar omställningen till elektrifiering och förnybar elproduktion inom lantbruket. Särskilt fokus ligger på att belysa skillnader mellan gårdar av olika storlek och med olika huvudsakliga verksamhetsområden. Studien syftar därmed till att besvara följande forskningsfråga: Vilka ekonomiska förutsättningar och hinder påverkar svenska lantbruksgårdar i deras benägenhet att implementera teknik för egen elproduktion och elektrifiering av verksamheten? Metod. Studien genomfördes med en kvalitativ metod där semi-strukturerade användes som datainsamlingsmetod. De intervjuade är samtliga verksamma inom lantbrukssektorn, främst inom djurhållning och/eller växtodling. Det insamlade materialet har analyserats med hjälp av tematisk analys. Resultat. Studiens resultat visar att lantbrukares ekonomiska strategier är präglade av utmaningar relaterade till låg lönsamhet, kapitalintensiva investeringar och behovet av långsiktig stabilitet. Vidare identifieras att teknikinvesteringar ofta ses som ett medel för att förbättra kassaflödet snarare än att skapa konkurrensfördelar. Lantbrukare tenderar att vara försiktiga i sina investeringar och teknikutveckling bedöms utifrån hur den kan stödja ekonomisk hållbarhet och trygghet. Vid analys av teknikacceptans framkommer att lantbrukare skiljer sig i sin vilja att investera, där faktorer som riskaversion och ekonomisk stabilitet ofta väger tyngre än investeringar i fossiloberoende teknologi. Sammanfattningsvis visar studien på en balansgång mellan teknikinvesteringar, ekonomisk hållbarhet och långsiktig planering inom lantbruket. Slutsatser. Studien visar att svenska lantbrukares omställning till elektrifiering och egen elproduktion formas av både ekonomiska förutsättningar och praktiska hinder. Förutsättningar för investeringar är att tekniken upplevs som driftsäker, tidsbesparande och ekonomiskt motiverad i relation till maskinernas nyttjandegrad. Tydlig kommunikation kring ekonomiska stöd och möjligheten till utbildning och praktisk testning av ny teknik lyfts som viktiga faktorer för att minska osäkerhet och underlätta beslutsfattande. Bland hindren framträder höga investeringskostnader, bristande elnätskapacitet, lång återbetalningstid och komplexa tillståndsprocesser. Studien pekar också på potentialen i samägande av förnybara elproduktionssystem mellan närliggande gårdar. Samägande av exempelvis vindkraft, biogas och energilagring, vilket kan sprida risker, sänka trösklar för tillstånd och stärka lokal självförsörjning.Background. Agriculture is facing increasing demands to reduce its climate impact, while energy demands remain high, particularly for the operation of machinery and facilities. Electrification and renewable energy production are highlighted as potential pathways towards a more sustainable farming sector. However, implementation is constrained by technical, economic, and practical challenges. To understand the conditions for transition, it’s essential to gain insight into farmers' perceptions, needs, and choices regarding technology. Objectives. This study aims to examine the economic conditions and challenges that Swedish farmers encounter when implementing electrification of machinery and electricity production at their farm. The study focuses on evaluating farmer’s financial capacity to adopt these technologies and their potential to scale energy production in line with operational demands. Through a detailed analysis of the economic and business dimensions of technology adoption, the study seeks to enhance the understanding of machinery electrification and renewable energy production within Swedish agriculture. The study addresses the following research question: What economic conditions and barriers influence Swedish farmers’ willingness to adopt technologies for on-site electricity production and machinery electrification? Methods. The study was conducted using a qualitative approach based on semi-structured interviews. The participants interviewed are active within the agricultural sector, primarily involved in animal husbandry and/or crop production. The collected material was analyzed using thematic analysis. Results. The study's findings show that farmers' economic strategies are influenced by challenges related to low profitability, capital-intensive investments, and the need for long-term stability. Technological investments are often viewed to improve cash flow rather than to gain competitive advantages. Farmers generally tend to be cautious in investment decisions, evaluating technological development based on its potential to support economic sustainability and security. When analyzing technology acceptance, it becomes evident that farmers differ in their willingness to invest, with factors such as risk aversion and financial stability often weighing more heavily than investments in renewable energy technologies. In summary, the study highlights a delicate balance between technological investment, economic resilience, and long-term planning within agriculture. Conclusions. The study shows that Swedish farmers’ transition to electrification and self-produced electricity is shaped by both economic conditions and practical obstacles. Investment prerequisites include the perception that the technology is reliable, time-saving, and economically justified in relation to the utilization rate of machinery. Clear communication regarding financial support, as well as opportunities for education and hands-on testing of new technology, are highlighted as important factors for reducing uncertainty and facilitating decision-making. Among the obstacles are high investment costs, limited grid capacity, long payback periods, and complex permitting processes. The study also points to the potential of joint ownership of renewable electricity production systems among neighboring farms. Shared ownership of, for example, wind power, biogas, and energy storage can help distribute risks, lower permitting thresholds, and strengthen local self-sufficiency
Prediction of Mini-Mental State Examination Scores for Cognitive Impairment and Machine Learning Analysis of Oral Health and Demographic Data Among Individuals Older Than 60 Years : Cross-Sectional Study
No full textBackground: As the older population grows, so does the prevalence of cognitive impairment, emphasizing the importance of early diagnosis. The Mini-Mental State Examination (MMSE) is vital in identifying cognitive impairment. It is known that degraded oral health correlates with MMSE scores <= 26. Objective: This study aims to explore the potential of using machine learning (ML) technologies using oral health and demographic examination data to predict the probability of having MMSE scores of 30 or <= 26 in Swedish individuals older than 60 years. Methods: The study had a cross-sectional design. Baseline data from 2 longitudinal oral health and ongoing general health studies involving individuals older than 60 years were entered into ML models, including random forest, support vector machine, and CatBoost (CB) to classify MMSE scores as either 30 or <= 26, distinguishing between MMSE of 30 and MMSE <= 26 groups. Nested cross-validation (nCV) was used to mitigate overfitting. The best performance-giving model was further investigated for feature importance using Shapley additive explanation summary plots to easily visualize the contribution of each feature to the prediction output. The sample consisted of 693 individuals (350 females and 343 males). Results: All CB, random forest, and support vector machine models achieved high classification accuracies. However, CB exhibited superior performance with an average accuracy of 80.6% on the model using 3 x 3 nCV and surpassed the performance of other models. The Shapley additive explanation summary plot illustrates the impact of factors on the model's predictions, such as age, Plaque Index, probing pocket depth, a feeling of dry mouth, level of education, and use of dental hygiene tools for approximal cleaning. Conclusions: The oral health parameters and demographic data used as inputs for ML classifiers contain sufficient information to differentiate between MMSE scores <= 26 and 30. This study suggests oral health parameters and ML techniques could offer a potential tool for screening MMSE scores for individuals aged 60 years and older. SNA
Navigating demand forecasting in make-to-order manufacturing : the role of global models and intermittent time-series
No full textDemand forecasting can optimise production and supply chain practices in manufacturing organisations. However, demand forecasting is not widely adopted among make-to-order (MTO) manufacturers with mass customisation offers. Building effective demand forecasting systems is challenging in such organisations due to the numerous unique manufactured articles and sparse demand patterns. This position paper argues that make-to-order manufacturers should employ demand forecasting to a larger extent, and that the forecasting community should address challenges related to the domain. Key challenges include creating models capable of predicting both demand size and timing of intermittent forecasts, as well as a deeper insight into the effects of global deep learning time-series models. We perform a pilot experiment using demand forecasting in a purchasing decision support system to validate the usefulness of demand forecasting for MTO manufacturing organisations with mass customisation offers. A research roadmap is proposed to address the identified challenges
Aerial Reconfigurable Intelligent Surface-Enabled Sagin With Lstm-Enhanced Drl Model
No full textThis paper introduces a network architecture that integrates the space-air-ground integrated network with mobile edge computing (MEC) and orbital edge computing to advance sixth-generation communication systems. The proposed system employs unmanned aerial vehicles equipped with reconfigurable intelligent surfaces and satellite-based MEC to optimize resource management in complex, dynamic environments. By efficiently managing resources such as bandwidth and computational power at both base stations and low Earth orbit satellites, while making offloading decisions, the system aims to minimize utility costs while meeting stringent performance requirements. We utilize a long short-term memory (LSTM)-enhanced deep deterministic policy gradient (DDPG) algorithm to solve the formulated nonlinear programming problem, enabling dynamic and adaptive resource management. The LSTM-enhanced DDPG improves convergence speed by 44.44% compared to conventional DDPG, significantly enhancing cost efficiency. Simulation results validate the robustness of the proposed method against state-of-the-art approaches.
Design of a Predictive Digital Twin System for Large-Scale Varroa Management in Honeybee Apiaries
No full textVarroa mites are a major global threat to honeybee colonies. Combining digital twins with scenario-generating models can be an enabler of precision apiculture, allowing for monitoring Varroa spread, generating treatment scenarios under varying conditions, and running remote interventions. This paper presents the conceptual design of this system for large-scale Varroa management in honeybee apiaries, with initial validation conducted through simulations and feasibility analysis. The design followed a design research framework. The proposed system integrates a wireless sensor network for continuous hive sensing, image capture, and remote actuation of treatment. It employs generative time-series models to forecast colony dynamics and a statistical network model to represent inter-colony spread; together, they support spread scenario prediction and what-if evaluations of treatments. The system evolves through continuous updates from field data, improving the accuracy of spread and treatment models over time. As part of our design research, an early feasibility assessment was carried out through the generation of synthetic data for spread model pretraining. In addition, a node-level energy budget for sensing, communication, and in-hive treatment was developed and matched with battery capacity and life calculations. Overall, this work outlines a path toward real-time, data-driven Varroa management across apiary networks, from regional to cross-border scales.