46 research outputs found
Maximising Product Possibilities While Minimising Process Change: A Case of Introducing Light Weight Material in Automotive Manufacturing
No full textAbstractCompanies are challenged to achieve maximised benefits in time, money and novelty when introducing new products or technologies into their existing manufacturing systems. This paper set the focus on the introduction of new materials into existing products and if the adverse effects on manufacturing will outweigh the benefits. An automotive case was studied where discrete event simulation was used as tool to evaluate process consequences when introducing new material and process technologies into the production system. The question concerned if discrete event simulation can verify production system capabilities even in early conceptual design stages. The case analysis is concluded by three challenge areas concerning early stages of production system design. The difficulties of evaluating operational key performance indicators early in design processes become evident and needed future research efforts within the area are pointed out
Semantic knowledge integration for learning from semantically imprecise data
No full textLow availability of labeled training data often poses a fundamental limit to the accuracy of computer vision applications using machine learning methods. While these methods are improved continuously, e.g., through better neural network architectures, there cannot be a single methodical change that increases the accuracy on all possible tasks. This statement, known as the no free lunch theorem, suggests that we should consider aspects of machine learning other than learning algorithms for opportunities to escape the limits set by the available training data. In this thesis, we focus on two main aspects, namely the nature of the training data, where we introduce structure into the label set using concept hierarchies, and the learning paradigm, which we change in accordance with requirements of real-world applications as opposed to more academic setups.Concept hierarchies represent semantic relations, which are sets of statements such as "a bird is an animal." We propose a hierarchical classifier to integrate this domain knowledge in a pre-existing task, thereby increasing the information the classifier has access to. While the hierarchy's leaf nodes correspond to the original set of classes, the inner nodes are "new" concepts that do not exist in the original training data. However, we pose that such "imprecise" labels are valuable and should occur naturally, e.g., as an annotator's way of expressing their uncertainty. Furthermore, the increased number of concepts leads to more possible search terms when assembling a web-crawled dataset or using an image search. We propose CHILLAX, a method that learns from semantically imprecise training data, while still offering precise predictions to integrate seamlessly into a pre-existing application
Determining transition readiness in Swiss childhood cancer survivors - a feasibility study.
No full textBACKGROUND
The successful transition of childhood cancer survivors (CCSs) from pediatric to adult long-term follow-up care is a critical phase, and determining the right time point can be challenging. We assessed the feasibility of the use of existing transition readiness tools in the context of the Swiss health care system, assessed partly transition readiness in Swiss CCSs, and compared our findings with Canadian CCSs for which these tools were originally developed.
METHODS
We officially translated the Cancer Worry Scale (CWS) and Self-Management Skill Scale (SMSS) into German and integrated them into this cross-sectional study. We included CCSs attending the long-term follow-up (LTFU) clinic in the Division of Oncology-Hematology, Department of Pediatrics, Kantonsspital Aarau. We used descriptive statistics to describe transition readiness.
RESULTS
We randomly recruited 50 CCSs aged ≥18 years at participation. The CCSs had a median CWS score of 62 (interquartile range 55-71), indicating a moderate level of cancer-related worry. Despite high self-management skills, some answers showed a dependency of CCSs on their parents. Our experience shows that the CWS and SMSS are easy for Swiss CCSs to use, understand, and complete. The interpretation of the results must take differences in health care systems between countries into account.
CONCLUSIONS
The translated CWS and SMSS are appropriate additional measures to assess transition readiness in CCSs. These scales can be used longitudinally to find the individual time point for transition and the completion by CCSs enables the health care team to individualize the transition process and to support the CCSs according to their individual needs
Investigating the innovation process as influencing factor on creativity : A case study in the manufacturing industry
No full textThe purpose of this thesis is to fill a research gap in the understanding of the involved mechanisms between the innovation process and creativity in a manufacturing company. In particular focus are the two innovation process interfaces between the technology development, product development and production departments and their influence on creativity and information flow. A conceptual framework is proposed, based on the combination of previous research in individual creativity and innovation process, to assess the influence via the innovation process interfaces on product knowledge and creative climate as creativity prerequisites. This study uses an explorative two single case study design with integrated triangulation and an abductive approach to fulfilling the research aim. These case studies took place in two Swedish manufacturing companies with a complete in-house innovation process using semi-structured interviews, two questionnaires and focus groups. The findings imply that the innovation process affects the basic prerequisites of individual creativity knowledge and creative climate due to its structural, strategic and physical implementation. Furthermore, the interfaces between the innovation sub-processes technology development, product development and production mediate the exchange of information between the involved departments. The results of this study adding valuable insights in this research field and propose further research including systems theory to achieve further in-depth understanding
Approaching emergent patterns with Kronecker algebra in industrial agents
No full textCyber-physical systems (CPSs) integrate distributed physical components, software,and monitors. Their behaviour results from the interactions between the parts and is not deducible from the components. In schools of fish or ant colonies, nature knows similar phenomena and is commonly summarised as self-organisation or emergent behaviour. A systematic literature review revealed properties required in multi-agent systems (MASs) to show emergent pattern formations. The dynamically interacting agents need to create, without external control, a robust pattern that is novel w.r.t. the individual parts of the system over time. Nevertheless, there are limited methods to identify such patterns. A potential solution is a formal language approach based on a cooperating array grammar system and Kronecker Algebra. Kronecker Algebra manipulates matrices, representing state machines capable of executing formal language grammars. The new Kronecker Synthesise and Symmetric Skip operations enable scenario synthesis to identify unexpected behaviour while ensuring consistency. Adding execution priorities allow pinpointing priority inversions between agents sharing a common resource. Moreover, it enables worst-case execution time (WCET) analysis of processes executing on one central processing unit (CPU). Applying the operations to a publish-subscribe communication system model results in pattern formations that individual agents cannot execute. Adding priorities affects the pattern formation and the execution time of agent interactions. Experimentations with a time-predictable publish-subscribe environment confirm the findings and the suitability of the proposed approach. Limiting the results is the absence of multiple experiments. Future work includes extending Kronecker Algebra to handle concurrent prioritised agent interactions and linear time model checking
In vitro characterization of a nineteenth-century therapy for smallpox.
No full textIn the nineteenth century, smallpox ravaged through the United States and Canada. At this time, a botanical preparation, derived from the carnivorous plant Sarracenia purpurea, was proclaimed as being a successful therapy for smallpox infections. The work described characterizes the antipoxvirus activity associated with this botanical extract against vaccinia virus, monkeypox virus and variola virus, the causative agent of smallpox. Our work demonstrates the in vitro characterization of Sarracenia purpurea as the first effective inhibitor of poxvirus replication at the level of early viral transcription. With the renewed threat of poxvirus-related infections, our results indicate Sarracenia purpurea may act as another defensive measure against Orthopoxvirus infections
Remote configuration of (meshed) sensor nodes in home and building automation
No full textSensorknoten sind ein integraler Bestandteil der Heim- und Gebäudeautomation. Diese Geräte können über drahtlose Sensornetzwerke verbunden werden. Aufgrund der großen Anzahl von eingesetzten Sensorknoten und der Reichweite, die sie abdecken müssen, ist eine manuelle Aktualisierung zeitaufwändig und mühsam. Diese Diplomarbeit untersucht verschiedene Technologien und Standards, die eine Fernkonfiguration von Sensorknoten in der Heim- und Gebäudeautomation ermöglichen. Ein weiterer Fokus liegt auf den Themen Sicherheit und Zuverlässigkeit eines Over-the-Air-Systems. Basierend auf der Forschung wird eine Reihe von Anforderungen für ein sicheres und zuverlässiges Over-the-Air-System entwickelt. Besonderes Augenmerk liegt auf der Designauswahl der Kombination aus dem Zephyr-Echtzeitbetriebssystem für die Sensorknoten und der drahtlosen Thread-Technologie für das Mesh-Netzwerk. Grund ist die fehlende Forschung zu den Over-the-Air-Fähigkeiten dieser beiden Technologien. Eine Beispielarchitektur wird implementiert und anhand der gegebenen Anforderungen evaluiert. Darüber hinaus werden Bewertungen der Aktualisierungszeiten und der Zuverlässigkeit des entwickelten Systems durchgeführt und Vorschläge für weitere Verbesserungen präsentiert.Sensor nodes present an integral part of Home and Building Automation (HBA). These devices are usually connected with Wireless Sensor Networks (WSN). Due to the large number of deployed sensor nodes and a wide area they have to cover, updating them manually requires time and effort. This thesis investigates different technologies and standards that allow remote configuration of sensor nodes in HBA. Additional focus is given on the topics of security and reliability of an Over-the-Air (OTA) system. Based on the research, a set of requirements for a secure and reliable OTA system is developed. Special focus lies on the design choices of the combination of Zephyr Real-time operating system (RTOS) for the sensor nodes and Thread wireless technology for the mesh network. Reason is the lacking research on the OTA capabilities of these two technologies. An example architecture is implemented and evaluated against the given requirements. Furthermore, evaluations of update times and reliability of the developed system have been carried out and suggestions for further improvements are presented
OPC UA/DDS gateway
No full textIn den vergangenen Jahren ist das Internet der Dinge (Internet of Things, IoT) stetig gewachsen. Neue Herausforderungen entstanden und Verbesserungen in verschiedenen Sektoren, wie in der Automatisierungsindustrie, wurden erreicht. Darüber hinaus wurde aufgrund der zunehmenden Anzahl von Objekten ein neuer Begriff eingeführt: das industrielle Internet der Dinge (Industrial IoT, IIoT). Dieses neue Gebiet ermöglicht die Verbindung von Entitäten, intelligenten Objekten und intelligenten Fabriken, um den Produktionswert zu erhöhen. Dieses Gebiet ist allerdings im ständigen Wandel und neue Technologien werden stetig entwickelt, die darauf abzielen, die verschiedenenSchichten des IIoT zu verbinden. Da sich diese Schichten unterschiedlich entwickeln, sind spezialisierte Middleware-Technologien notwendig, um diese zu verbinden. Zwei bekannte Technologien, OPC UA und DDS, stellen verschiedene Mechanismen zur Verfügung, um die verschiedenen Aspekte des IIoT zu verbinden, bis dato sind diese jedoch noch nicht vereint worden. Das Ziel dieser Arbeit ist es, ein Modell für den Datenaustausch zwischen OPC UA undDDS zu erstellen und Szenarien zu entwickeln, um Daten zwischen OPC UA Umgebungen auszutauschen, die über den globalen Datenraum von DDS gesendet werden können. Weiters sollen Einschränkungen dieses OPC UA/DDS Gateways und der vorgeschlagenenSzenarien diskutiert werden. Die notwendigen Schritte inkludieren die Erstellung der verschiedenen Szenarien, die für den Datenaustausch notwendig sind. Basierend auf diesen Szenarien wird ein Prototyp erstellt und einfache Testläufe werden durchgeführt, um zu zeigen, ob die entwickelten Szenarien ihren Zweck erfüllen. Die Resultate dieser Arbeit zeigen, dass die entwickelten Szenarien den Datenaustausch zwischen OPC UA und DDS prinzipiell ermöglichen. Allerdings sind bestimmte Einschränkungen zu beachten. Alle involvierten Anwendungen müssen über die Bezeichner jenerServer Bescheid wissen, mit denen sie interagieren wollen. Das impliziert, dass es einen globalen Dienst geben sollte, der alle verfügbaren Server auflistet und zugänglich macht. Außerdem sind nicht alle Szenarien in größeren Umgebungen einsetzbar, weil die Menge der generierten Nachrichten das OPC UA/DDS Gateway überlasten könnte. Da es zurzeit keine offene Java-basierte DDS Bibliothek gibt, die nativ RPC over DDS unterstützt,musste letztlich MQTT im Prototyp eingesetzt werden. Zukünftige Arbeiten könnten sich daher auch einer quelloffenen Java-basierenden DDS Implementierung widmen.Over the years, the field of Internet of Things (IoT) was growing rapidly. New challenges raised and improvements to various sectors, such as the factory automation domain were accomplished. Furthermore, due to the increasing amount of Things, a new term got established: the Industrial Internet of Things (IIoT). Within this field, an interconnection of Things, smart objects and smart factories enables the optimization of production value. However, this field is in active change and new technologies emerge that aim to connect the different layers of the IIoT. As these layers evolve differently, middleware technologies are necessary to connect them. Two prominent technologies, namely OPC UA and DDS,provide different features for connecting various aspects of the IIoT and are yet to be connected. The aim of the work is to develop an architectural design for enabling data exchange between OPC UA and DDS, and further to derive scenarios which allow data exchange between OPC UA environments by using the global data space of DDS. Furthermore,the limitations of this OPC UA/DDS gateway and of the proposed scenarios will be discussed. The steps necessary to enable this involve the creation of different scenarios that will cover the data exchange. Based on these scenarios, a prototype implementation will be realized and simple test runs indicate if the scenarios fulfilled their purpose. The findings of this thesis reveal that the developed scenarios enable the data exchange between OPC UA and DDS. However, certain limitations have to be considered. Foremost, involved applications have to be aware of the IDs of servers they want to interact with, which indicates the need for an environmental-wide service that registers all available servers. Additionally, in large scale environments, not all scenarios are usable, as the amount of messages generated could potentially overload the OPC UA/DDS gateway. Furthermore, the scenario which enables data exchange between OPC UA environments via DDS global data space allows for increased scalability for OPC UA. However, all scenarios are only usable in a controlled environment. As it was necessary for this thesis to simulate DDS with MQTT another implication of this thesis is that there is need for an open source Java-based DDS library, that supports RPC over DDS for exchangingMQTT with DDS within the proof of concept
Eisenbahnverschub mit Kronecker Algebra
No full textDas Rangieren von Zügen ist eine grundlegende und zugleich rechnerisch anspruchsvolle Aufgabe des Bahnbetriebs. Diese Arbeit untersucht, inwiefern sich Kronecker-Algebra als Modellierungsgrundlage für Rangierprobleme eignet, und ob sich der daraus entstehende Zustandsraum effizient durchsuchen lässt, um Rangierlösungen zu finden. Wir formalisieren Gleise, Wagen und zulässige Bewegungen mithilfe matrixbasierter Operationen (Kronecker-Produkte und -Summen) und erhalten so ein Zustandsraummodell. Wir erweitern die Problemstellung auf mehrere Lokomotiven. Insbesondere betrachten wir eineVariante mit zwei Lokomotiven, die jeweils an den beiden Enden des Gleisfelds positioniert sind, eine Anordnung, die nach unserem Kenntnisstand bislang nicht untersucht wurde. Zur Lösungssuche setzen wir eine A-star-Suche mit mehreren, unterschiedlichen Heuristiken ein. Eine experimentelle Auswertung auf zufällig generierten Instanzen, inspiriert durch das Inglenook Shunting Puzzle, und mehrgleisigen Layouts zeigt, dass die Wahl der Heuristik maßgeblich die Suchleistung bestimmt. Die wesentliche Einschränkung unseres Ansatzes liegt in der Skalierbarkeit. Der Zustandsraum wächst mit der Anzahl der Gleise und Wagen exponentiell, wodurch große Konfigurationen schwer durchsuchbar werden. Abschließend skizzieren wir Ansatzpunkte für zukünftige Arbeiten, darunter stärkere Heuristiken, parallele Suche und alternative Zustandskodierungen.Railway shunting is a fundamental and computationally challenging task in rail operations. This thesis investigates whether Kronecker algebra can provide a modeling foundation for shunting problems and whether such models can be searched efficiently to produce shunting solutions. We formalize tracks, railcars, and admissible movements using matrixbased operations (Kronecker products and sums) to obtain a state-space model. The framework is extended to multiple locomotives, specifically for the two-locomotive version, with one locomotive at each end of the tracks, which to our knowledge has not been studied before. To search for a solution, we ended up using A-star search with several different heuristics. An experimental evaluation on randomized instances inspired by theInglenook Shunting puzzle and multi-track layouts shows that the choice of heuristics is crucial for the performance of the search. The main limitation is scalability. The state space grows exponentially with tracks and railcars, which makes large configurations challenging to search. We outline directions for future work, including stronger heuristics, parallel search, and alternative state encodings
Verifikation von nebenläufigen Programmen in schwachen Speicher-Modellen
No full textModerne Processoren nutzen schwache Speichermodelle, um die Ausführung von Code zu optimieren. Im Vergleich zu sequentiell konsistenten Speichermodellen erlauben schwache Speichermodelle die Umordnung von Codeanweisungen. Während dies das Verhalten eines Single-Thread-Programmes nicht verändert, kann es in Multicore-Systemen Fehler verursachen indem Ausführungsreihenfolgen zwischen Threads ermöglicht werden, die ansonst nicht möglich wären.Diese Fehler sind für Entwickler oft schwer zu erkennen und zu verstehen.Diese Thesis schlägt vor, mithilfe von Kronecker Algebra mögliche Ausführungsreihenfolgen von Multithread-Programmen innerhalb schwacher Speichermodelle zu modellieren. Dies kann dann verwendet werden, um Race Conditions im System zu erkennen. Der Fokus liegt dabei speziell auf schwachen Speichermodellen die Release-Acquire Semantiken nutzen. Im Rahmen der Arbeit wird ein Prototyp entwickelt, der das System zur Erkennung von Race Conditions in Multithread-Programmen einzusetzen, die in LLVM-Bytecode kompiliert wurden, um zu zeigen, dass das System in der Lage ist, in Programmen Probleme zu identifizieren, die unter Verwendung von Release-Acquire Speichermodellen auftreten, obwohl das Programm unter sequentiellen Speichermodellen korrekt war. Der Prototyp bewies die Korrektheit der sequentiellen Versionen von Petersons und Dekker's Algorithmus und erkannte auch die Fehler, die auftreten, wenn sie nahezu unverändert unter der Verwendung von Release-Acquire-Speichermodellen ausgeführt werden.Modern Central Processing Units (CPUs) utilize weak memory models to optimize code execution. Compared to sequential consistent memory models, weak memory models allow the reordering of code instructions. While this does not alter the behavior of a single-thread program, it can cause bugs in multicore systems by allowing execution orders between threads that would otherwise not be possible. These errors are often problematic for the developer to detect and understand. This thesis proposes to use Kronecker Algebra to model possible execution orders of multithreaded programs within weak memory models.The output of such a toolchain can then be used to detect race conditions within the system.It focuses explicitly on weak memory models using release-acquire semantics. Alongside the thesis, a prototype is developed for the proposed system to detect race conditions in multithreaded programs compiled into LLVM-bytecode.This is done to show that the system can detect problems caused when running a program using release-acquire memory models. However, the program was correct under sequential memory models. The prototype correctly proved the correctness of the sequential versions of Peterson's and Dekker's Algorithm. Moreover, the prototype detected the errors that arise when Peterson's and Dekker's Algorithms run without modifications using release-acquire memory models
