Chalmers Open Digital Repository
Not a member yet
26247 research outputs found
Sort by
The Future of User Evaluation - exploring how emerging technology can be used in user evaluation to investigate driver interaction
No full textTechnology is advancing rapidly across multiple fields and emerging technologies hold significant potential to shape the future of product development and design. This thesis
explores the intersection of emerging technology and user evaluation in collaboration with Volvo Group Trucks Technology. It is particularly focused on the context of human-machine interaction and how emerging technologies can enhance future user evaluation of driver interaction. The study set out to identify what types of insights and results emerging technologies could offer for user evaluation, along with associated practical limitations and opportunities in an industrial automotive setting.
The work was structured in two phases. The first phase involved technology scouting through literature research and expert meetings to identify relevant emerging technologies. In the second phase, two selected technologies, extended reality and emotion interpretation, were tested in practice to further explore their value for user evaluation. A total of 19 expert meetings were held, an XR workshop was conducted with both a pilot and main session, and three types of emotion interpretation tests were performed. Five technologies were identified as particularly relevant for user evaluation in the considered context: extended reality, emotion interpretation, EEG sensors, eye tracking and artificial intelligence. The results suggested that each explored technology demonstrates distinct strengths and opportunities to contribute to in-depth understanding of users and enhanced user evaluation. However, the results of the study also highlight technology limitations and the importance of complementing, rather than replacing, traditional user evaluation methods. The thesis concluded with specific recommendations for each explored technology area, along with an overarching recommendation to actively monitor ongoing developments in emerging technology for user evaluation. Specific recommendations for user evaluation in the automotive context were also presented, as well as the suggestion that a balanced approach between new tools and established practices would be most effective. While further research
in this area is needed, the findings offer promising directions for how industrial actors, like Volvo Group, can advance user evaluation practices through emerging technologies
Surface characterization and grinding of Duplex 2205 Stainless Steel slabs
No full textThis study investigates the optimization of grinding processes for duplex 2205 stainless steel slabs at Outokumpu Avesta works. The primary objective is to analyze and enhance these processes to achieve cost reductions, increased productivity, and improved material processing efficiency. To fulfill this aim, a combination of theoretical analysis and practical experiments has been conducted, with a particular focus on the effects of different grinding methods on oxidation levels and surface quality.
Through the evaluation of various grinding conditions, the study identifies opportunities for process improvements that can enhance production efficiency. It is shown that implementing coarse grindings process only can reduce total material loss by 21 ton, decrease grinding time by 200 hours, and result in potential savings of 40% in grinding wheel costs during the reference period. The findings highlight the need to balance cost-effectiveness, product quality, and environmental considerations. This work aims to support Outokumpu's long-term development strategy while contributing to the company's competitiveness in the global marke
Fast performance estimations for vector architectures using qemu-plugins
No full textNew applications requiring high computations have arrived (AI, HPC, Cybersecurity etc..) leading to algorithm-hardware configuration exploration side-by-side. This has created the need for tools that can quickly navigate (and thereby prune) the
wide design space to be explored by providing performance estimates in a short time-span. This thesis develops/extends a tool to provide performance estimates for an algorithm on a given hardware configuration having vector processing unit and cache hierarchy. It further validates the accuracy of results upon comparison with golden reference
Restoration through architecture; multisensory pavilions and the adaptive reuse of the harbour office on Gothenburg's coast
No full textIn today´s society individuals are expected to meet many demands, and as a result
overwhelming stress has become increasingly common. These challenges are often
amplified during winter in Sweden, where many self-reports “winter blues”. To ease such
symptoms it is often recommended to spend time outdoors, and research shows that
being in, on, or near the sea has positive impact on wellbeing and reduces stress.
Gothenburg´s west mainland coast features several public bathing spots, all very popular
in summer. However, there are currently no non-commercial public facilities that support
extended stays by the seaside in the colder, more windy and rainy seasons.
The purpose of this thesis is to develop a site-specific design proposal that supports
wellbeing in winter. Using a multisensory and adaptive reuse approach, the aim is to
transform Fiskebäck, located on Gothenburg´s west mainland coast and currently lacking
public facilities for winter, into a restorative public place that offers extended visits during
winter. This thesis explores the following research questions:
How can spatial architectural interventions transform a public seaside site on
Gothenburg’s mainland coast to support wellbeing during winter?
How can the Harbour Office building in Fiskebäck be adapted to provide public use
and sensory engagement during winter?
Theory and literature of multisensory architecture, adaptive reuse and wellbeing formed
the theoretical framework for the thesis. The sensorial concepts explored are Light,
Enclosure, Temperature of Space, Sound of Space and Smell of Space. Site-specific
mappings of boundaries and landscape typologies identified key placements, transparency
and tactile terrain qualities. Reference projects were studied to extract architectural
features, which informed a conceptual design strategy toolbox. These tools were tested
through iterative design development to answer the research questions.
The proposal includes four architectural interventions: The adaptive reuse of the Harbour
Office building, and West, South and East Pavilions, each with distinct experential focus.
Together they form a site journey and sequence of spaces where architecture becomes a
medium to engage the human senses, support wellbeing and strenghten the relationship
with the natural seaside environment in winter
Analyzing Factors Influencing Performance in LLM Inference Systems
No full textLarge Language Models (LLMs) lead to numerous innovative applications, including virtual assistants, content generation, and recommendation systems, which made a revolution in daily life. LLM inference is crucial as it allows these models to process and generate human-like text in real-time, making them adaptable to a wide range of practical applications. This capability has not only enhanced the functionality of AI-driven technologies but also expanded their accessibility and impact across various industries. Latency and throughput are essential metrics for evaluating the performance of LLMs, as they directly influence user experience and system efficiency. They are critical for optimizing the deployment and operation of LLM-based solutions in real-world scenarios. In this thesis, we evaluate the performance of LLM inference by analyzing how different factors affect LLM inference throughput and latency. We aim to study the current performance and internal working mechanism of Llama 3 and GPT-2 to explore potential methods that could keep improving LLM inferences. The approach of this project includes 2 steps which are data collection and data analysis. Our results show that increasing the batch size significantly improves throughput but can also lead to higher latency, indicating a trade-off between speed and responsiveness in LLM inference. Additionally, a larger model size can provide more accurate outputs. Interestingly, allocating more GPU resources reduced overall GPU utilization for both Llama 3 and GPT-2. Inefficiencies in resource allocation could negatively impact the cost-effectiveness of deploying LLM inference’s performance
Besättningsbyten inom svensk sjöfart: En analys av problematiken för svenska sjömän vid gränsövergångar i Sverige
No full textDen 19 augusti 2024 införde Sveriges Kustbevakning skärpta regler gällande gränspassage i Sverige. Det innebär att fartyg tillåts utföra besättningsbyten i färre hamnar än tidigare, samt under striktare tider om dagarna. Syftet med studien var att undersöka de problem som uppstått för svensk sjöfart på kort och långsikt till följd av de nya reglerna samt identifiera lämpliga lösningar på problemen. Studien fokuserar på svensk sjöfart utifrån agenturers och sjömäns perspektiv.
Det är en kvalitativ studie i form av semistrukturerade intervjuer där fem Agenturer och fyra sjömän intervjuades. Resultatet visar att sjömännen har påverkats negativt av de skärpta reglerna på kort sikt. De upplever ökad stress och oro på grund av osäkerheten kring när de får mönstra av och på. På lång sikt såg sjömännen att det kan påverka antalet som vill jobba till sjöss i Sverige. Agenturerna har behövt lägga ner mer tid för planering och utförande av besättningsbyten. Agenturerna ansåg att problemen kan bli skadliga för Sverige då det lett till ett inkomstbortfall för taxi, hotell och flyg i Sverige. Flera respondenter uttryckte frustration kring att gränspassager är jämförelsevis problemfria i andra EU-länder.
Lösningen som samtliga intervjuade bad om var att utöka öppettiderna och tillämpa likartade rutiner som i andra EU-länder. Sjömännens lyfte även förslag om att införa digitala kontroller genom till exempel Teams alternativt digitala ID-verifieringar. Avslutningsvis fann studien att reglerna hittills tillämpats under en kort period och att det återstår att se hur rutinerna potentiellt sett förändas i framöver, samt vilket utfall som reglerna kommer ha på långsikt
Powder Flowability in Snack Food Coatings, Insights from Rheometric and Morphological Analysis
No full textThe rheological characteristics of powders are crucial in the food industry, affecting production and prod uct quality. This research conducted at Santa Maria AB focused on the characteristics of cheese-based snack
seasoning powders’ flow properties under varying formulations and environmental conditions, utilizing both
rheometric and morphological assessments. It investigated silicon dioxide’s impact on flowability, compared
rheological traits of in-house and external suppliers’ samples, assessed humidity and storage effects on flowa bility, and assessed morphological data of five samples. Supplementary methods such as particle sieve analysis
and water activity testing were also employed for enhanced interpretation of powder behavior. Rheometric
assessments utilized the FT4 powder rheometer® by Freeman Technology, while morphological evaluations
were conducted by the Danish Technical Institute.
Key findings revealed that added silicon dioxide (SiO) improved flowability, agreeing with existing litera ture. In-house samples in the aeration test exhibited rheological similarities to external supplier samples when
SiO was included.However, original in-house samples with 1.3% SiO (SM-1.3%) had lower cohesion values
than external samples. This suggests that shear resistance alone does not capture the flow behavior of the
seasoning but also there is a possible interplay between cohesion and adhesion. Furthermore, PCA plot of the
collected data from the FT4 powder rheometer® displayed no apparent clustering among the samples but
rather indicated that differences in rheological behavior could be attributed to factors that were not captured
in the rheometric assessment. From the morphological assessment, findings revealed that the sample SM-1.3%
was larger than SM-1.75% in terms of diameter, further strengthening the observed effect of silicon dioxide on
flowability. Sample with higher silicon dioxide tended to have smaller and more spherical particles but also
a narrow particle size distribution. In addition, humidity testing confirmed the impact of SiO2 on flowability
where samples subjected to higher relative humidity displayed more cohesive behavior, aligning with previous
research. Lastly high shear mixers on fine powders increased cohesion and aerated energy, without improving
flowability. No significant particle size distribution differences were found between low and high shear mixing,
however, further replicates could provide statistical robustness.
Conclusively, results emphasized the importance of anti-caking agents such as silicon dioxide, environmental
control, and the use of appropriate mixing methods for optimizing powder flow. These findings provide a
framework for improving seasoning performance in industrial applications. Future research could benefit from
added replicates, visual inspection such as image analysis and the influence of fat content on powder behavior
to determine the influence of ingredient selectio
