11 Interfakultäre Einrichtungen
Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/12
Browse
17 results
Search Results
Item Open Access Spatio-temporal and immersive visual analytics for advanced manufacturing(2019) Herr, Dominik; Ertl, Thomas (Prof. Dr.)The increasing amount of digitally available information in the manufacturing domain is accompanied by a demand to use these data to increase the efficiency of a product’s overall design, production, and maintenance steps. This idea, often understood as a part of Industry 4.0, requires the integration of information technologies into traditional manufacturing craftsmanship. Despite an increasing amount of automation in the production domain, human creativity is still essential when designing new products. Further, the cognitive ability of skilled workers to comprehend complex situations and solve issues by adapting solutions of similar problems makes them indispensable. Nowadays, customers demand highly customizable products. Therefore, modern factories need to be highly flexible regarding the lot size and adaptable regarding the produced goods, resulting in increasingly complex processes. One of the major challenges in the manufacturing domain is to optimize the interplay of human expert knowledge and experience with data analysis algorithms. Human experts can quickly comprehend previously unknown patterns and transfer their knowledge and gained experience to solve new issues. Contrarily, data analysis algorithms can process tasks very efficiently at the cost of limited adaptability to handle new situations. Further, they usually lack a sense of semantics, which leads to a need to combine them with human world knowledge to assess the meaningfulness of such algorithms’ results. The concept of Visual Analytics combines the advantages of the human’s cognitive abilities and the processing power of computers. The data are visualized, allowing the users to understand and manipulate them interactively, while algorithms process the data according to the users’ interaction. In the manufacturing domain, a common way to describe the different states of a product from the idea throughout the realization until the product is disposed is the product lifecycle. This thesis presents approaches along the first three phases of the lifecycle: design, planning, and production. A challenge that all of the phases face is that it is necessary to be able to find, understand, and assess relations, for example between concepts, production line layouts, or events reported in a production line. As all phases of the product lifecycle cover broad topics, this thesis focuses on supporting experts in understanding and comparing relations between important aspects of the respective phases, such as concept relationships in the patent domain, as well as production line layouts, or relations of events reported in a production line. During the design phase, it is important to understand the relations of concepts, such as key concepts in patents. Hence, this thesis presents approaches that help domain experts to explore the relationship of such concepts visually. It first focuses on the support of analyzing patent relationships and then extends the presented approach to convey relations about arbitrary concepts, such as authors in scientific literature or keywords on websites. During the planning phase, it is important to discover and compare different possibilities to arrange production line components and additional stashes. In this field, the digitally available data is often insufficient to propose optimal layouts. Therefore, this thesis proposes approaches that help planning experts to design new layouts and optimize positions of machine tools and other components in existing production lines. In the production phase, supporting domain experts in understanding recurring issues and their relation is important to improve the overall efficiency of a production line. This thesis presents visual analytics approaches to help domain experts to understand the relation between events reported by machine tools and comprehend recurring error patterns that may indicate systematic issues during production. Then, this thesis combines the insights and lessons learned from the previous approaches to propose a system that combines augmented reality with visual analysis to allow the monitoring and a situated analysis of machine events directly at the production line. The presented approach primarily focuses on the support of operators on the shop floor. At last, this thesis discusses a possible combination of the product lifecycle with knowledge generating models to communicate insights between the phases, e.g., to prevent issues that are caused from problematic design decisions in earlier phases. In summary, this thesis makes several fundamental contributions to advancing visual analytics techniques in the manufacturing domain by devising new interactive analysis techniques for concept and event relations and by combining them with augmented reality approaches enabling an immersive analysis to improve event handling during production.Item Open Access Concepts and methods for the design, configuration and selection of machine learning solutions in manufacturing(2021) Villanueva Zacarias, Alejandro Gabriel; Mitschang, Bernhard (Prof. Dr.-Ing. habil.)The application of Machine Learning (ML) techniques and methods is common practice in manufacturing companies. They assign teams to the development of ML solutions to support individual use cases. This dissertation refers as ML solution to the set of software components and learning algorithms to deliver a predictive capability based on available use case data, their (hyper) paremeters and technical settings. Currently, development teams face four challenges that complicate the development of ML solutions. First, they lack a formal approach to specify ML solutions that can trace the impact of individual solution components on domain-specific requirements. Second, they lack an approach to document the configurations chosen to build an ML solution, therefore ensuring the reproducibility of the performance obtained. Third, they lack an approach to recommend and select ML solutions that is intuitive for non ML experts. Fourth, they lack a comprehensive sequence of steps that ensures both best practices and the consideration of technical and domain-specific aspects during the development process. Overall, the inability to address these challenges leads to longer development times and higher development costs, as well as less suitable ML solutions that are more difficult to understand and to reuse. This dissertation presents concepts to address these challenges. They are Axiomatic Design for Machine Learning (AD4ML), the ML solution profiling framework and AssistML. AD4ML is a concept for the structured and agile specification of ML solutions. AD4ML establishes clear relationships between domain-specific requirements and concrete software components. AD4ML specifications can thus be validated regarding domain expert requirements before implementation. The ML solution profiling framework employs metadata to document important characteristics of data, technical configurations, and parameter values of software components as well as multiple performance metrics. These metadata constitute the foundations for the reproducibility of ML solutions. AssistML recommends ML solutions for new use cases. AssistML searches among documented ML solutions those that better fulfill the performance preferences of the new use case. The selected solutions are then presented to decision-makers in an intuitive way. Each of these concepts was evaluated and implemented. Combined, these concepts offer development teams a technology-agnostic approach to build ML solutions. The use of these concepts brings multiple benefits, i. e., shorter development times, more efficient development projects, and betterinformed decisions about the development and selection of ML solutions.Item Open Access Data processing, analysis, and evaluation methods for co-design of coreless filament-wound building systems(2023) Gil Pérez, Marta; Mindermann, Pascal; Zechmeister, Christoph; Forster, David; Guo, Yanan; Hügle, Sebastian; Kannenberg, Fabian; Balangé, Laura; Schwieger, Volker; Middendorf, Peter; Bischoff, Manfred; Menges, Achim; Gresser, Götz T.; Knippers, JanItem Open Access Temporally dense exploration of moving and deforming shapes(2020) Frey, SteffenWe present our approach for the dense visualization and temporal exploration of moving and deforming shapes from scientific experiments and simulations. Our image space representation is created by convolving a noise texture along shape contours (akin to LIC). Beyond indicating spatial structure via luminosity, we additionally use colour to depict time or classes of shapes via automatically customized maps. This representation summarizes temporal evolution, and provides the basis for interactive user navigation in the spatial and temporal domain in combination with traditional renderings. Our efficient implementation supports the quick and progressive generation of our representation in parallel as well as adaptive temporal splits to reduce overlap. We discuss and demonstrate the utility of our approach using 2D and 3D scalar fields from experiments and simulations.Item Open Access Simulation model for digital twins of pneumatic vacuum ejectors(2022) Stegmaier, Valentin; Schaaf, Walter; Jazdi, Nasser; Weyrich, MichaelIncreasing productivity, as well as flexibility, is required for the industrial production sector. To meet these challenges, concepts in the field of “Industry 4.0” are arising, such as the concept of Digital Twins. Vacuum handling systems are a widespread technology for material handling in industry and face the same challenges and opportunities. In this field, a key issue is the lack of Digital Twins containing behavior models for vacuum handling systems and their components in different applications and use cases. A novel concept for modeling and simulating the fluidic behavior of pneumatic vacuum ejectors as key components of vacuum handling systems is proposed. In order to increase the simulation accuracy, the concept can access instance‐specific data of the used asset instead of object‐specific data. The model and the data are part of the Digital Twins of pneumatic vacuum ejectors, which shall be able to be combined with other components to represent a Digital Twin of entire vacuum handling systems. The proposed model is validated in an experimental test setup and in an industrial application delivering sufficiently accurate results.Item Open Access Visual analytics of multivariate intensive care time series data(2022) Brich, N.; Schulz, Christoph; Peter, J.; Klingert, W.; Schenk, M.; Weiskopf, Daniel; Krone, M.We present an approach for visual analysis of high‐dimensional measurement data with varying sampling rates as routinely recorded in intensive care units. In intensive care, most assessments not only depend on one single measurement but a plethora of mixed measurements over time. Even for trained experts, efficient and accurate analysis of such multivariate data remains a challenging task. We present a linked‐view post hoc visual analytics application that reduces data complexity by combining projection‐based time curves for overview with small multiples for details on demand. Our approach supports not only the analysis of individual patients but also of ensembles by adapting existing techniques using non‐parametric statistics. We evaluated the effectiveness and acceptance of our approach through expert feedback with domain scientists from the surgical department using real‐world data: a post‐surgery study performed on a porcine surrogate model to identify parameters suitable for diagnosing and prognosticating the volume state, and clinical data from a public database. The results show that our approach allows for detailed analysis of changes in patient state while also summarizing the temporal development of the overall condition.Item Open Access Modell zum maschinellen Lernen von Wirkzusammenhängen bei der Holzverarbeitung auf Basis von online-erfassten Werkzeugmaschinendaten(Stuttgart : Fraunhofer Verlag, 2018) Lenz, Jürgen Herbert; Westkämper, Engelbert (Univ.-Prof. a. D. Dr.-Ing. Prof. E.h. Dr.-Ing. E.h. Dr. h.c. mult.)Aufgrund des immer härter werdenden globalen Wettbewerbs müssen produzierende Unternehmen, die auch in der Zukunft profitabel produzieren wollen, ihre Leistungsreserven nutzten. Die Möbelfertigung, die größte holzverarbeitende Industrie, besteht im Hauptprozess aus dem Fräsen von Holzwerkstoffen. Hierbei gibt es Leistungsreserven in der Einsatzplanung der Fräswerkzeuge. Gute Einsatzplanung ist die Voraussetzung für eine hohe Verfügbarkeit des Produktionssystems. Die Einsatzplanung wird durch Entwicklungen wie individuelle Möbelstücke, kleinere Losgrößen und neue Schneidstoffe erschwert. Die Herausforderung der Planungsunsicherheit beim Werkzeugeinsatz in der Holzbearbeitung wächst zusätzlich durch die größere Anzahl an industriell hergestellten Holzwerkstoffen mit jeweils unterschiedlicher Abrasivität. Dadurch wird die Bestimmung der Reststandzeit eines Werkzeuges erschwert. Zielsetzung dieser Arbeit ist die Planungssicherheit des Werkzeugeinsatzes durch eine exakte Planung des Werkzeugwechselfensters sowie durch Prognose der Reststandzeit zu erhöhen. Mithilfe dieser Prognose kann das gesamte Standvermögen des Werkzeuges verwendet werden. Das führt dazu, dass die Verfügbarkeit des Produktionssystems erhöht wird, da durch das Überschreiten der Werkzeugeinsatzgrenze bedingte Stillstände vermieden werden. Hierfür wurde ein Modell erstellt, das online erfasste Daten aus der Werkzeugmaschinensteuerung mit kontextbezogenen Informationen aus Datenbanken wie dem ERP-System und der Werkzeugverwaltung kombiniert. Aus diesen Informationen wird eine werkzeugspezifische Einsatzhistorie gebildet und mit gemessenen physikalischen Werten über den Werkzeugverschleiß und Kantenqualität des Werkstückes in Verbindung gebracht. Diese Verbindung von Bearbeitungshistorie und echten physikalischen Messgrößen bilden die Datenbasis für das maschinelle Lernen von Wirkzusammenhängen. Durch das Erlernen dieser Zusammenhänge kann die Reststandzeit eines Werkzeuges prognostiziert werden und somit die Planungsgenauigkeit des Werkzeugeinsatzes durch exakte Festlegung von Werkzeugwechselfenstern gesteigert werden. Zur Erprobung wurde das entwickelte Modell implementiert und seine Funktionsfähigkeit anhand einer Werkstoff-/Schneidstoffpaarung validiert. Diese Erprobung zeigte dass die Wirkzusammenhänge erlernt werden können.Item Open Access IDEA - towards an interactive tool that supports creativity sessions in automotive product development(2023) Kaschub, Verena Lisa; Wechner, Reto; Krautmacher, Lara; Diers, Daniel; Bues, Matthias; Lossack, Ralf; Kloos, Uwe; Riedel, OliverItem Open Access Coordinating with a robot partner affects neural processing related to action monitoring(2021) Czeszumski, Artur; Gert, Anna L.; Keshava, Ashima; Ghadirzadeh, Ali; Kalthoff, Tilman; Ehinger, Benedikt V.; Tiessen, Max; Björkman, Mårten; Kragic, Danica; König, PeterRobots start to play a role in our social landscape, and they are progressively becoming responsive, both physically and socially. It begs the question of how humans react to and interact with robots in a coordinated manner and what the neural underpinnings of such behavior are. This exploratory study aims to understand the differences in human-human and human-robot interactions at a behavioral level and from a neurophysiological perspective. For this purpose, we adapted a collaborative dynamical paradigm from the literature. We asked 12 participants to hold two corners of a tablet while collaboratively guiding a ball around a circular track either with another participant or a robot. In irregular intervals, the ball was perturbed outward creating an artificial error in the behavior, which required corrective measures to return to the circular track again. Concurrently, we recorded electroencephalography (EEG). In the behavioral data, we found an increased velocity and positional error of the ball from the track in the human-human condition vs. human-robot condition. For the EEG data, we computed event-related potentials. We found a significant difference between human and robot partners driven by significant clusters at fronto-central electrodes. The amplitudes were stronger with a robot partner, suggesting a different neural processing. All in all, our exploratory study suggests that coordinating with robots affects action monitoring related processing. In the investigated paradigm, human participants treat errors during human-robot interaction differently from those made during interactions with other humans. These results can improve communication between humans and robot with the use of neural activity in real-time.Item Open Access The ethics of sustainable AI : why animals (should) matter for a sustainable use of AI(2023) Bossert, Leonie N.; Hagendorff, ThiloTechnologies equipped with artificial intelligence (AI) influence our everyday lives in a variety of ways. Due to their contribution to greenhouse gas emissions, their high use of energy, but also their impact on fairness issues, these technologies are increasingly discussed in the “sustainable AI” discourse. However, current “sustainable AI” approaches remain anthropocentric. In this article, we argue from the perspective of applied ethics that such anthropocentric outlook falls short. We present a sentientist approach, arguing that the normative foundation of sustainability and sustainable development - that is, theories of intra- and intergenerational justice - should include sentient animals. Consequently, theories of sustainable AI must also be non-anthropocentric. Moreover, we investigate consequences of our approach for applying AI technologies in a sustainable way.