OPUS - Publikationenserver der Technischen Hochschule Nürnberg Georg Simon Ohm
Not a member yet
2332 research outputs found
Sort by
Sustainable supply chain models for 3D printed houses
No full textAlthough steps have been taken to improve sustainable practices in the construction industry, progress to date has been slow. Advances in 3D printing technology provide a significant opportunity to reconfigure construction industry supply chains to improve its sustainability credentials. This study investigates the adoption of 3D construction printing (3DCP) as a sustainable alternative and/or to augment traditional construction industry supply chains. By analysing six 3DCP organisations and feeding six critical insights into the Business Model Navigator, this study proposes three supply chain models i.e. generative construction service, hybrid service, and emergency service. A generative supply chain model can be used as an on-demand service for customers with limited knowledge of 3DCP. The hybrid model can integrate 3DCP with a traditional construction supply chain to construct part of the building. Finally, the emergency services model can be used to rapidly construct cost-effective housing (e.g for refugees). The adoption of 3D Printing using these supply chain models in housing construction supply chains has the potential to offer considerable benefits both for construction companies and end-users, while at the same time improving sustainability, affordability and productivity
Sustainability Science Communication
No full textThe Anthropocene, marked by human-induced climate change, necessitates urgent action to address climate goals and respect planetary boundaries. While sustainability research provides knowledge, the first challenge lies in communicating the findings in an adequate manner to the public and several stakeholders, such as economic and political actors. Therefore, this study explores the significance of science communication in sustainability science, focusing on a case study—the True Cost Accounting (TCA) campaign by the University of Greifswald, Technical Institute of Nuremberg, and German retailer PENNY. TCA herein serves as a transparency tool, economic incentive, and discussion basis for sustainable consumption. This study investigates consumer perceptions of ecological prices of foods through a face-to-face survey during the 2023 PENNY campaign, comparing results to an informational campaign carried out in 2021. Findings indicate a high awareness of the true cost campaign in 2023, with 50.8% of participants hearing about it. Consumers’ willingness to pay true costs and potential behavior changes were explored. In comparison to results from the informational campaign of 2021, customers showed a decrease in this WTP when the true prices would actually impact their spending, indicating an attitude–behavior gap. In addition, a willingness to reduce the consumption of animal foods—if TCA was implemented—of 60.5% was determined, suggests that TCA has the potential for sustainable behavior change. This study highlights factors that influence consumer attitudes and preferences regarding the inclusion of TCAs, such as environmental, social, and animal welfare costs. Customers’ understanding of increased prices like,
in this case, the compensation for environmental and social costs—is an argument in favor of true prices. The results emphasize the need for differentiated scientific communication strategies to bridge knowledge and action gaps in sustainability science
Outlier Reduction with Gated Attention for Improved Post-training Quantization in Large Sequence-to-sequence Speech Foundation Models
No full textThis paper explores the improvement of post-training quantization (PTQ) after knowledge distillation in the Whisper speech foundation model family. We address the challenge of outliers in weights and activation tensors, known to impede quantization
quality in transformer-based language and vision models. Extending this observation to Whisper, we demonstrate that these outliers are also present when transformer-based models are trained to perform automatic speech recognition, necessitating mitigation strategies for PTQ. We show that outliers can be reduced by a recently proposed gating mechanism in the attention blocks of the student model, enabling effective 8-bit quantization, and lower word error rates compared to student models without the gating mechanism in place
Maßgeschneiderte KI: Wie Custom GPTs skalierbare Startup- Unterstützung neu definieren
No full textDer Aufstieg von Chatbots auf Basis von Large Language Models (LLMs) hat die generative KI demokratisiert und sie einer breiten Öffentlichkeit zugänglich gemacht. Immer mehr Menschen greifen heute auf Chatbots wie ChatGPT zurück, um alltägliche Aufgaben wie das Verfassen von E-Mails oder die Ideengenerierung zu beschleunigen (Bilgram et al., 2023; Cooper & Brem, 2024; Joosten et al., 2024). Ein besonders bedeutender Fortschritt in diesem Zusammenhang ist die Einführung der „Custom GPTs“ durch OpenAI, die es Nutzern ermöglicht, individualisierte Versionen von ChatGPT zu erstellen (OpenAI, 2023). Dabei können Nutzer ihren GPT durch individuelle Anweisungen und externe Datensätze mit domänenspezifischen Informationen so anpassen, dass er sehr spezifische Aufgaben erledigen kann
Leaching of catalyst platinum from cured silicone elastomers: A preliminary study for comparing reagents
No full textAddition curing systems involve two-part silicones which require the mixture of a silicone polymer with a catalyst to initiate the cure. Platinum is the most commonly used metal catalyst for addition curing of silicones by hydrosilylation which involves the crosslinking by the addition reaction of silicon hydride species to unsaturated bonds, mainly C=C, but also C=O or C=N double bonds. After crosslinking of the polymers, the platinum catalyst cannot be recovered but remains in the silicone materials throughout the entire product life. In the end, platinum is disposed of together with the silicones and is thus lost to the value chain. The overall objective of this work was to develop a recycling process for the recovery of platinum from addition-cured silicone elastomers. In the first step, this was achieved by efficient digestion methods and by optimizing the leaching processes for exemplary commercial silicone elastomer products. Two different silicone materials were investigated, both of which were crosslinked with a platinum catalyst. The initial Pt content in the tested samples was 12.6 ± 0.2 mg/kg for a commercial silicone impression material and 6.3 ± 0.5 mg/kg for a silicone baking mold, measured by graphite furnace atomic absorption spectrometry (GF-AAS). Samples were first frozen with liquid nitrogen to improve brittleness and then crushed with a simple food processor to obtain a silicone granule. Various acid mixtures, mainly based on sulfuric acid, were investigated as digestion methods in order to extract platinum from the silicone network. These had different effects on the dissolution behavior of silicone samples and the amount of platinum extracted in each case. The amount of platinum leached from the filtrate of the digested samples in each case was measured by ICP-OES to evaluate the efficiency of different leaching mixtures. In addition, the dissolved platinum species present in the solutions was identified by UV/VIS as tetrachloridoplatinate(II) complex. The best platinum leaching results so far were obtained with two methods, both of which used a leaching mixture based on sulfuric acid and hexamethyldisiloxane (M2). In the presence of hydrochloric acid, 9.6 ± 1.6 mg platinum/kg was leached from the silicone impression material and 4.2 ± 0.8 mg platinum/kg from the silicone baking mold. With the additional use of aqua regia instead of hydrochloric acid, 10.4 ± 2.8 mg platinum/kg was extracted from the silicone impression material and 4.8 ± 1.0 mg platinum/kg was extracted from the silicone baking mold. These methods were replicated with n = 3. Using statistical evaluation methods (F-test, t-test, and confidence interval), no significant difference was found between these two best methods. Recovery of platinum(0) from leach mixtures has not yet been achieved due to high dilution and very low platinum concentration in samples and will be part of another study
Exploring the nexus of industry dynamism, climate risk exposure and mental health and well-being of owners of small and micro-sized suppliers: implications for sustainable supply chain management
No full textAlthough climate change-related risks affect all stakeholders along the supply chain, the potential impact on small and micro-sized suppliers is incredibly excessive. The corresponding toll of these climate risk threats on the mental health and well-being of owners of small and micro-sized suppliers can adversely affect their participation in sustainability efforts, ultimately impacting the firm's performance. This often overlooked dynamic forms the core of our research. We probe into two pivotal aspects: how industry dynamism and climate risk affect the mental health and well-being of owners of small and micro-sized suppliers and how, in turn, dictate involvement and, consequently, supply chain sustainability performance. This is further nuanced by the moderating role of the abusive behavior of buyers.
Our study is built on resource dependency theory and the supporting empirical evidence is fortified by a mixed-methods sequential explanatory design. This study comprises three phases. In the first phase, our experiment examines the effect of industry dynamism and climate risk exposure on sustainable supply chain management performance. Hypotheses H1a and H1b are tested in the first phase. The second phase involves using a survey and structural equation modeling to test the comprehensiveness of the model. Here, the relationship between industry dynamism, climate risk exposure, mental health and well-being of owners of small and micro-sized supplier firms, supplier involvement and sustainable supply chain management (H2–H7) is tested in the second phase. In the third phase, we adopt a qualitative approach to verify and provide descriptive explanations of phase two findings.
Our findings underscore the significance of small and micro-sized suppliers in sustainability, offering invaluable insights for both theoretical understanding and practical implementation. Our study highlights that buyers must allocate sufficient resources to support small and micro-sized supplier firms and collaborate closely to address climate change and its impacts.
The key takeaway from this study is that buyer firms should consider SDG 3, which focuses on the good health and well-being of their employees and the mental health and well-being of owners of small and micro-sized suppliers in their upstream supply chain. This approach enhances sustainability performance in supply chains.
This is one of the first studies that shows that industry dynamism and climate risk exposure can negatively impact small and micro-sized suppliers in the presence of a contextual element, i.e. abusive behavior of buyers, and ultimately, it negatively impacts sustainable supply chain performance dimensions
Schuldensituation und Schuldenregulierung bei Inhaftierten in Bayern
Get PDFDer vorliegende Bericht basiert auf einer Umfrage unter Fachkräften des Sozialdienstes (n = 47) und der externen Schuldenberatung (n = 28) in bayerischen Justizvollzugs-anstalten. Mit sieben Schuldenberater:innen wurden zusätzlich qualitative Interviews geführt. Dadurch konnte Expert:innenwissen sichtbar gemacht und Erkenntnisse geschaffen werden über die Schuldensituation von Inhaftierten, deren Ursachen sowie die typischen intramuralen Bearbeitungsweisen dieser Problemlagen. Die individuelle Schuldensituation von Inhaftierten in Bayern ist heterogen. So unterscheiden sich die Schuldensummen stark abhängig von Alter und Haftdauer. Schulden resultierend aus der Straftat (z.B. Gerichtskosten sowie Geldstrafen) und der Inhaftierung (z.B. Kosten für eine Zwangsräumung oder aufgelaufene Krankenversicherungsbeiträge) zeigen sich als besonders relevant, noch vor Schulden resultierend aus privatem Konsum. Dabei kommt der psychosozialen Belastung, welche aus der Überschuldungssituation resultiert, eine doppelte Funktion zu: kann sie einerseits eine (teilweise notwendige) extrinsische Motivation zur Schuldenregulierung fördern, kann sie auch zu Hoffnungslosigkeit führen und die Motivation zur Regulierung und auch den Resozialisierungsprozess negativ beeinflussen. Hier wurde insgesamt deutlich, dass die Motivation der Inhaftierten von Expert:innen als ein zentraler Gelingensfaktor gesehen wird. Neben den negativen Auswirkungen der Inhaftierung, wie zusätzliche Schulden oder erschwerte Regulierungsbedingungen (z.B. fehlenden Unterlagen, kein/kaum Einkommen), werden auch positive Auswirkungen auf die Überschuldungssituation gesehen, da sie einen Rahmen für eine gelingende Regulierung und Anlass zur Auseinandersetzung damit darstellen kann.
Die professionelle Bearbeitung ist besonders durch die Dauer der Inhaftierung geprägt. Abhängig von dieser wird entweder ausschließlich die Vermeidung weiterer Schulden, oder aber die Gläubigersuche zur Vorbereitung eines Regulierungsprozesses angestrebt. Dabei ist die Kooperation mit verschiedenen internen und externen Akteuren wichtig, allen voran mit dem Sozialdienst in Haft. Gerade in der praktischen Arbeit selbst wurde deutlich, dass nicht nur bürokratische Tätigkeiten, sondern auch die psycho-soziale Beratung der Inhaftierten und des sozialen Umfeldes im Vordergrund stehen.
Einer potenzielle Erhöhung der Gefangenenvergütung in realistischem Umfang (z.B. um 50%) wird von den Expert:innen wenig Potenzial hinsichtlich der Schuldensituation zugeschrieben, da die finanziellen Mittel für eine effektive Regulierung weiterhin deutlich zu niedrig wären. Dennoch seien damit gewisse Spielräume in der Schuldenbereinigung verbunden aber auch die Möglichkeit einer erhöhten Wertschät-zung für die Inhaftierten. Wenn die Gefangenenvergütung einen substanziellen Beitrag zur Schuldenregulierung darstellen soll, müsste diese wohl eher um ein Mehrfaches des aktuellen Umfangs erhöht werden. Entsprechend sollte über zusätzliche Alternativen nachgedacht werden, die vom Bundesverfassungsgericht geforderte Resozialisierung und Anerkennung der Inhaftierten sicherzustellen, wie nicht-monetäre Vergütung oder auch die Reduktion von Gerichtskosten. Auch die gängige Praxis, dass laut Aussage der Expert:innen ausschließlich die Landesjustizkasse dieses Geld pfändet, sollte in diesem Kontext kritisch diskutiert werden. Zudem wurden Veränderungsbedarfe deutlich hinsichtlich der Klärung der Krankenversicherung bei Inhaftierung, der Gestaltung des Übergangs bei Entlassung und Ver-legung, sowie der Finanzierung der externen Schuldenberatung
Training a Computer Vision Model for Commercial Bakeries with Primarily Synthetic Images
No full textIn the food industry, reprocessing returned product is a vital step to increase resource efficiency. [SBB23] presented an AI application that automates the tracking of returned bread buns. We extend their work by creating an expanded dataset comprising 2432 images and a wider range of baked goods. To increase model robustness, we use generative models pix2pix and CycleGAN to create synthetic images. We train state-of-the-art object detection model YOLOv9 and YOLOv8 on our detection task. Our overall best-performing model achieved an average precision [email protected] of 90.3% on our test set