Publikationer från Mälardalens högskola
Not a member yet
16964 research outputs found
Sort by
COMPARING SUPERVISED AND UNSUPERVISED APPROACHES FOR ARRHYTHMIA DETECTION IN ECG SIGNALS
No full textThe detection of anomalies in electrocardiogram (ECG) signals is critical in healthcare for early identification of arrhythmias, irregular heartbeats that can indicate serious cardiac conditions. Arrhythmias are common cardiac irregularities that can lead to serious complications if not identified promptly. Early arrhythmia detection can improve patient outcomes and reduce the diagnostic burden on clinicians by facilitating timely intervention. This thesis evaluates and compares supervised and unsupervised machine learning methods for arrhythmia detection in ECG data. Four models were tested on the widely used MIT-BIH Arrhythmia dataset, which provides annotated ECG recordings. A supervised support vector machine (SVM) classifier and three unsupervised approaches, an autoencoder (AE) neural network, an Isolation Forest, and the DBSCAN clustering algorithm. The primary focus is on the SVM (supervised) and the autoencoder (unsupervised) models. Experimental results show that the SVM achieved the highest overall detection performance, confirming the strength of supervised learning for this task. The autoencoder was competitive, demonstrating the potential of unsupervised neural approaches to distinguish normal and abnormal heartbeats. In contrast, the Isolation Forest and DBSCAN algorithms underperformed compared to these methods. These findings highlight the respective strengths and limitations of supervised and unsupervised techniques for ECG anomaly detection. The study provides insights into how different machine learning models perform in a clinical arrhythmia detection task and informs the design of future decision-support tools for cardiovascular care.
Optimizing CSR strategies for Brand equity : A managerial approach in multinational corporation
No full textDate: [2025-08-20] Level: Bachelor Thesis in Business Administration, 15cr Institution: School of Business, Society and Engineering, Mälardalens University Authors: Emilia Jurewicz Parisa Dehghan Manshadi (2001/03/22) (1977/08/26) Title: Optimizing CSR strategies of brand equity: a managerial approach in multinational corporations. Tutor: Aziza Al Ghafri Key words: Brand equity, Corporate Social responsibility, Multinational Corporation, Manager, qualitative, reports. Research question: How can managers best use CSR strategies in a way they maximize brand equity of multinational corporations? Purpose: The purpose of this thesis is to explore how CSR strategies can be optimized by managers in multinational corporations to enhance customer-based brand equity, particularly in ethically conscious markets such as the UK. Method: This qualitative study relies on thematic analysis of secondary data, including CSR reports and podcast interviews with managers from two major multinational companies. The data was analyzed to reveal managerial practices and strategic decisions that align CSR initiatives with the key brand equity components. Conclusion: The study concludes that effective CSR strategies significantly enhance brand equity when they are authentically embedded into the core business model. Managers can achieve this by aligning CSR activities with stakeholder expectations. Firms that treat CSR as a strategic asset see higher consumer trust, loyalty, and brand differentiation in competitive markets
Lean Implementation in 3PL Warehouses: Challenges and opportunities
No full textThe rapid growth of e-commerce and the increasing product variety in the fast-moving consumer goods industry have intensified pressure on third-party logistics warehouse operations. In these environments, speed accuracy, and flexibility are critical, yet many processes remain dependent on manual labour. This reliance increases the risk of human errors such as item misplacement, incorrect data entry, and inefficient workflows, ultimately impacting supply chain performance and customer satisfaction. While Lean principles are widely used in manufacturing to eliminate waste and improve productivity, their implementation in third-party (3PL) logistics warehouses remains limited. Existing research reveals a gap in understanding how Lean can be effectively adapted to 3PL environments. Objective and research questions This study aims to explore the challenges and enablers impacting the adoption of Lean principles to improve internal logistics in 3PL companies. Recommendations on how 3PL companies can effectively implement lean and sustain it to optimize their logistics processes are presented in this study. ➢What are the potential challenges faced when adapting and implementing lean philosophy and principles in 3PL companies? ➢What are the enablers that contribute to a successful implementation of lean principles in 3PL companies? ➢How can 3PL companies effectively apply and implement lean principles? Methodology This study uses a combined explorative-descriptive and abductive approach to examine a specific company case. This mixed-methods approach facilitates a deeper understanding of the "how" of the phenomena under investigation. Data analysis focused on identifying patterns and refining theories for future testing.ConclusionThis study aimed to explore lean implementation in 3PL warehouses, identifying challenges, enablers, and proposing a conceptional 4P framework for implementation. Key challenges include lack of understanding, resistance to change, poor communication, and insufficient resources. Enablers of successful lean implementation are a culture of continuous improvement, standardized work, leadership commitment, employee engagement, and tools like 5S and Kaizen. The 4P framework emphasizes a long-term vision, robust systems, and continuous improvement through structured work environments, daily problem-solving, and strong leadership. By addressing the challenges and leveraging the enablers, 3PL companies can effectively implement lean to minimize waste, improve productivity, and achieve sustained performance gains.Den snabba tillväxten av e-handel och den ökande produktvariationen inom snabbkonsumentvaruindustrin har medfört ett ökat tryck på lagerverksamheter inom tredjepartslogistik (3PL). I dessa verksamhetsmiljöer är snabbhet, noggrannhet och flexibilitet avgörande faktorer, men många processer är fortfarande starkt beroende av manuellt arbete. Detta ökar risken för mänskliga fel, såsom felplacering av artiklar, felaktig datainmatning och ineffektiva arbetsflöden, vilket i sin tur påverkar leveranskedjans prestanda och kundnöjdhet negativt. Trots att Lean-principer är väl etablerade inom tillverkningsindustrin för att minska slöseri och förbättra produktivitet, är deras tillämpning inom 3PL-lagerverksamhet fortfarande begränsad. Existerande forskning pekar på ett kunskaps- och tillämpningsgap gällande hur Lean effektivt kan anpassas till 3PL-verksamheter. Syfte och forskningsfrågor Denna studie syftar till att utforska de utmaningar och faktorer som påverkar implementeringen av Lean-principer för att förbättra intern logistik i tredjepartslogistikföretag (3PL). Studien ger också rekommendationer om hur 3PL-företag kan implementera och upprätthålla Lean-strategier effektivt för att optimera sina logistikprocesser. ➢Vilka potentiella utmaningar står 3PL-företag inför vid anpassning och implementering av Lean filosofi och principer? ➢Vilka är de viktigaste faktorerna och möjliggörarna som bidrar till en framgångsrik implementering av Lean principer i 3PL-företag? ➢Hur kan 3PL-företag effektivt tillämpa och implementera Lean principer? Metod Studien använder en kombinerad explorativ-deskriptiv och abduktiv ansats för att undersöka ett specifikt företagsfall. Denna metod kombinerar både kvalitativa och kvantitativa metoder för att skapa en djupare förståelse för "hur" fenomenet undersöks. Dataanalysen fokuserade på att identifiera mönster och teorier för framtida tester. Slutsats Studien syftade till att utforska Lean-implementering i 3PL-lager och identifierade utmaningar, nyckelfaktorer för framgång samt föreslog en 4P-ramverk koncept för implementering. De främsta utmaningarna omfattar brist på förståelse, motstånd mot förändring, kommunikation och otillräckliga resurser. Stödjande faktorer för framgångsrik Lean-implementering inkluderar en kultur av kontinuerlig förbättring, standardiserat arbete, ledarskapsengagemang, medarbetarengagemang samt verktyg som 5S och Kaizen. 4P-ramverket betonar en långsiktig vision, robusta system och kontinuerlig förbättring genom strukturerade arbetsmiljöer, daglig problemlösning och starkt ledarskap. Genom att hantera utmaningarna och använda stödjande faktorer kan 3PL-företag effektivt implementera Lean för att minska slöseri, förbättra produktiviteten och uppnå hållbara resultatförbättringar
Investigation of PV-Battery System Integration with Sports Facility Building : Comparative Analysis of Energy Management Strategies
No full textThis study evaluates the cost-effectiveness of integrating a battery energy storage system (BESS) with a grid-connected photovoltaic (PV) system at Rocklunda sports facility in Sweden. Unlike other commercial or residential buildings, the Rocklunda sports facility is dedicated to sports and recreational activities, leading to higher electricity demand due to energy-intensive operations such as heating and cooling systems, lighting, ventilation, and ice rink maintenance. Therefore, this study focuses on integrating a BESS with an existing grid-connected PV system to evaluate the economic feasibility of three operational strategies; price arbitrage, peak shaving and self-consumption. The study begins with a qualitative literature review to establish theoretical frameworks and derive key equations for cost, revenue, and net present value (NPV) analysis. Subsequently, a quantitative methodology is applied, using MATLAB to simulate energy management strategies and Excel to evaluate economic performance. Then, a sensitivity analysis was conducted to evaluate how variation in PV generation, electricity price and battery capacity affect the economic feasibility of the system. Based on current battery and electricity market prices, the analysis reveals that price arbitrage is the only viable energy management strategy at the moment for the Rocklunda sports facility.This work acknowledges support from KK-stiftelsen under the project ‘Flexergy - Energy flexibility through synergies of big data, novel technologies & systems, and innovative markets’ (Grant No. 20200073).INT
INKLUDERING I SKOLAN : Skolkuratorers och lärares perspektiv - en kvalitativ studie
No full textBackground: Many students in Swedish schools experience exclusion due to various challenges that hinder the creation of an inclusive school environment where all students can feel a sense of belonging and participate. Teachers and school counsellors play an important role in children’s lives and are a fundamental part of promoting inclusion, which is a key aspect of social sustainability. Aim: The purpose of this study is to explore how teachers and school counsellors work with inclusion among primary school students. Method: A qualitative method was used to gain a deeper understanding of how teachers and school counsellors promote inclusion I primary school. A purposive sampling strategy was applied, including six informants from municipal schools in the Mälardalen region. Data was collected through six semi-structed interviews, which were recorded, transcribed, and analyzed in accordance with ethical guidelines. Result: The findings showed that all participants experienced good relationships with their students and actively strived to promote inclusion in school. However, several barriers and challenges, particularly related to resources, were identified as hindering inclusive efforts. The study also highlighted that various factors influence the extent to which students feel involved and prepared in the school environment. Conclusion: Teachers and school counselors have a key role in the work of inclusion, but obstacles such as a lack of resources and clear strategies make it difficult. To succeed, joint, systematic work is required where students' differences are seen as an asset. Key words: Exclusion, Inclusion, Relationships, Sense of coherence (SOC), Teachers and school counselors. Ne
The Importance of Reading Aloud for Students’ Language Comprehension and Vocabulary in Primary School
No full textSyftet med vår studie är att undersöka hur lärare resonerar om och beskriver arbetet med högläsning i undervisning samt hur det kan främja elevers språkförståelse och ordförråd. Som metod till studien har vi använt en kvalitativ undersökning där vi genomförde semistrukturerade intervjuer med sex lärare. Det som vi har kommit fram till i vår studie är att högläsning har en stor betydelse i undervisningen för att främja elevers språkförståelse och ordförråd. Det vi även kom fram till är att lärarens didaktiska val kring bokval, engagemang, aktiviteter och hur de utformar undervisning, har en stor betydelse. Att lärarna väljer böcker som väcker elevernas intresse och engagemang. Det handlar även om att läraren pratar omkring boken och kopplar bokens händelser och syfte till elevernas erfarenhet och vardag. En annan aspekt vi kom fram till är att läraren förbereder sig inför högläsning för att kunna ställa bland annat kontrollfrågor och förklara svåra ord för eleverna vid läsning, för att hjälpa eleverna förstå svåra ord, förstå sammanhanget och för att lättare kunna följa med i läsningen.
Balancing between collaboration and autonomy : leadership insights from a Swedish preschool research-practice partnership
No full textPurpose This study examines whether the research-practice partnership (RPP) model supports preschool education leaders in navigating the new power dynamics of collaborative decision-making introduced by recent public reforms in Sweden and, if so, how it facilitates this process. It examines decision-making processes within a Swedish RPP focused on preschool improvement, with a particular emphasis on the negotiation between autonomy, control, and collaboration. Design/methodology/approach This study is a qualitative case study, drawing on 33 h of recorded meetings from a Swedish RPP initiative between 2022 and 2023. Key data include board and process manager meetings with an analytic focus on identifying key tensions and negotiations surrounding autonomy, control and collaboration. Findings The researchers and preschool practitioners negotiated among autonomy, collaboration and control throughout their work together. While initial agreements emphasized autonomous responsibilities reflecting traditional, separate roles of research and local development, tensions later emerged regarding the level of involvement. The study identifies a shift towards “collaborative autonomy,” where members of the partnership could balance both independence and collective decision-making. Originality/value This study contributes in three key ways. First, it shows how RPPs can support educational leaders in navigating tensions between autonomy and collaboration through engagement with research and researchers. Second, it highlights the evolving negotiation process within the partnership, introducing “collaborative autonomy” as a guiding concept. Finally, it extends the RPP discussion beyond the USA, exploring how the model adapts to different national and policy contexts
“Being there” : detailed ethnography, detective work and a little imagination
No full textPurpose: This methodological paper aims to push the boundaries of workplace ethnographies, where technologies in general and the implementation of new technologies in particular are being studied. Intrigued by the experience that “being there” is not enough, and sometimes not even possible, we discuss complementing aspects of the “how” of the work on the ground that set the foundation for high-quality ethnographic studies. Design/methodology/approach: To this end, we mobilize Latour’s (1992, 1993, 2005) thinking around a sociology of translation and how non-humans bring essential force to social relations, and Leonardi’s (2017) three-step methodology for dealing with material, materiality and how matter comes to matter. We illustrate our argument with ethnographic material telling a story of the “life and death” (c.f. Law and Callon, 1992) of visualization boards at a local site of a global manufacturing company. Findings: Our findings suggest that detailed ethnography is the foundation for allowing the progressive detective and imagination work we argue is required in technology-focused workplace ethnographies. “Being there” is the foundation, and it grants us the privilege of doing the important work of joining the dots where senses-enabled trajectories disappear. Originality/value: The paper puts forward an approach to ethnographic on-the-ground work that encourages detective work and imagination to “join the empirical dots” when “being there” is not enough
A physics-informed neural network-based method for predicting degradation trajectories and remaining useful life of supercapacitors
No full textSupercapacitors are widely used in transportation and renewable energy fields due to their high power density, stable cycling performance, and rapid charge–discharge capabilities. To ensure efficient applications of supercapacitors, accurately predicting their degradation trajectories and remaining useful life (RUL) is crucial. For this purpose, a physics-informed neural network (PINN) model is developed using Long Short-Term Memory (LSTM) as the base architecture. Physical equations are embedded into the loss function to ensure consistency with domain knowledge, allowing the loss function to incorporate both physical and data-driven components. The balance between these two loss components is dynamically determined through Bayesian optimization, to enhance the model's accuracy further. Validation results show a root mean square error (RMSE) of 3 mF (the rated capacity is 1 F) in the degradation trajectory prediction and a RMSE of 269 cycles (the average cycle life is 5180 cycles) for the RUL. Ablation experiments were conducted to validate the effectiveness of integrating physical information into the LSTM framework. Results demonstrate that the proposed model outperforms both the data-driven LSTM method and the empirical equation-based method that the PINN model can reduce the RMSE by 85% and 87.5% for degradation trajectory prediction, and 86.5% and 94.6% for RUL prediction, respectively. In addition, a comparison with advanced models demonstrates that our model reduces the requirement significantly on training data while maintaining comparable prediction accuracy, which favors scenarios where data is scarce.
Adaptive resolution in speckle displacement measurement using optimized grid-based phase correlation and statistical refinement
No full textSpeckle metrology is a powerful optical sensing tool for non-destructive testing (NDT) and advanced surface characterization, enabling ultra-precise measurements of surface deformations and displacements. These capabilities are critical for material analysis and surface assessment in sensing-driven applications. However, traditional correlation methods often struggle to balance resolution and robustness, particularly when simultaneously measuring both small-and large-scale deformations in noisy, high-frequency data environments. In this paper, we present an adaptive resolution approach for speckle displacement measurement that combines grid-based phase correlation with statistical refinement for enhanced accuracy and resolution. Unlike traditional phase correlation techniques that rely on global correlation, our method introduces a flexible grid-based framework with localized correlation and dynamic overlap adjustments. To improve measurement performance, we developed an optimization technique that uses the median absolute deviation of residuals between reference and deformed images, enabling the algorithm to automatically adapt grid sizes based on local deformation characteristics. This allows it to handle both small-and large-scale deformations simultaneously and effectively. The approach resulted in a relative error reduction of up to 14 % compared to the best of the results obtained using a manually fixed grid size. The proposed sensing methodology is validated through a series of numerical simulations and experimental studies, including controlled deformations with a micrometer translation stage and random speckle displacements on water-sprayed surfaces. Results demonstrate that our method can accurately detect both known and unknown deformations with high accuracy and precision, outperforming traditional techniques in terms of adaptability and robustness, particularly for surface deformation analysis