Digital Library of Gesellschaft für Informatik e.V.
Not a member yet
40758 research outputs found
Sort by
Data-Driven Identification and Analysis of Waiting Times in Business Processes
No full textReducing waiting times in end-to-end business processes is a recurrent concern in the field of business process management. The uptake of data-driven approaches in this field in the past two decades, most notably process mining, has created new opportunities for fine-grained analysis of waiting times based on execution data. As a result, a wide range of approaches for waiting time identification and analysis on the basis of business process execution data have been reported in the literature. In many instances, different approaches have considered different notions of waiting time and different causes for waiting time. At present, there is a lack of a consolidated overview of these manifold approaches, and how they relate to or complement each other. The article presents a literature review that starts with the question of what approaches for identification and analysis of waiting time are available in the literature, and then refines this question by adding questions which shed light onto different causes and notions of waiting time. The survey leads to a multidimensional taxonomy of data-driven waiting time analysis techniques, in terms of purpose, causes, and measures. The survey identifies gaps in the field, chiefly a scarcity of integrated multi-causal approaches to analyze waiting times in business processes, and a lack of empirically validated approaches in the field
Automatisierte Erzeugung eines Trainingsdatensatzes zur bildbasierten Tieridentifikation mittels KI
No full textDieser Beitrag beschreibt das Vorgehen und die erzielten Ergebnisse bei der Erzeugung eines Bilddatensatzes, der zum Training einer bildbasierten, KI-unterstützten Identifikationslösung für Milchrinder verwendet werden soll. Der diesem Beitrag zugehörige Bilddatensatz enthält derzeit mehrere Tausend den jeweiligen Einzeltieren zugeordnete Aufnahmen von insgesamt 170 verschiedenen Tieren aus einer 90° seitlich rechts orientierten Perspektive und kann frei heruntergeladen werden. Der Datensatz umfasst bis auf wenige Ausnahmen Bilder der Rinderrasse „Holstein“
Vorgehen bei der Einführung einer Organisation für das Management von Enterprise Systems
No full textDer vorliegende Beitrag beschreibt die Einführung einer Organisationeinheit für das Enterprise Architecture Management in einem deutschen Logistikunternehmen. Die Anwendungslandschaft besteht aus wenigen Eigenentwicklungen sowie verschiedenen Standardsoftwaresystemen, von denen einige durch Konfiguration, andere aber durch Nicht-Standard-Erweiterungen angepasst werden. Die Einführung der Organisationseinheit erfolgt im Rahmen eines Action-Research-Projects, welches sich über drei Phasen gezogen hat. Ziel des Beitrags ist die Ableitung von Empfehlungen für ähnliche Vorhaben auf Basis der Erprobung von Konzepten aus der Literatur sowie den Erfahrungen aus ihrer praktischen Umsetzung. The article at hand describes the introduction of an organizational unit for Enterprise Architecture Management in a German logistics company. The application landscape consists of a few in-house developments as well as various standard software systems, some of which are customized through configuration, while others are modified with non-standard extensions. The introduction of the organizational unit takes place as part of an action research project, covering three phases. The aim of this contribution is to derive recommendations for similar endeavours based on concepts from the literature and experiences from their practical application
A Demonstration of an Integrated Approach to EAST-ADL and Security Modeling
No full textThe automotive industry’s increasing reliance on software innovations has introduced significant security challenges. With connected vehicles and semi-autonomous systems becoming commonplace, automotive cyberattacks are frequently covered in the media. Public awareness is growing that cybersecurity plays a crucial role, akin to the life-saving innovations of the 1970s, such as mandatory seatbelts and speed limits. It is evident that security must be integrated throughout the entire development lifecycle and operation of automotive software systems, with every stakeholder considering security aspects. However, this is complicated by the vast and often complex nature of the security landscape and the limited accessibility of essential security frameworks like MITRE ATT&CK, which are primarily oriented towards implementation rather than serving as an entry point for comprehensive security analysis. In response to these challenges, this paper presents the Security Abstraction Model (SAM), a metamodel-based approach designed to address security concerns across all stages of automotive software development, focusing on the early phases and on understandability among all stakeholders. SAM offers a structured framework for defining, analyzing, and implementing security requirements within automotive systems, leveraging established metamodeling techniques. This paper demonstrates how SAM can be integrated with EAST-ADL to create a systematic approach for securing complex automotive software architectures. In order to illustrate its applicability, an integrated demonstration is developed, showcasing the combined use of EAST-ADL and security modeling
Comparing workflow tools in a parallel performance analysis environment
No full textWhile the runtime of parallel programs is an important metric and tracing and profiling can help to understand application behavior, it’s not trivial to create a reproducible, low-overhead analysis pipeline whose parts encourage reuse. In this work, we evaluate three workflow tools to improve our existing C++ performance analysis environment. For this use case the results show that Snakemake is slow to start, JUBE is difficult to use and Parsl offers a lot of flexibility in defining analysis pipelines
Digital GreenTech for good
No full textJunge Forschende arbeiten an der Entwicklung von Technologien, die zur Schonung der natürlichen Ressourcen und zur Verringerung von Umweltbelastungen beitragen. Diese kleine Weltreise gibt Einblicke in die Herausforderungen, vor die sie der Klimawandel in ihren Ländern stellt, und ihre vielfältigen Lösungsansätze
Automated labeling of soybeans for size measurements
No full textSoybeans are crucial in agriculture and industry, serving as a key source of protein and oil for food, feed, and various applications. Measuring individual soybean seed properties, such as size, color, and texture, aids in predicting batch parameters. In a conveyor belt setup, images captured by an RGB camera are analyzed using computer vision techniques, such as object detection and segmentation, to identify whole beans, which are then used for size measurements. For such a supervised machine learning approach, annotated data is essential. To address the challenge of manual data labeling, we propose an automated annotation system for identifying beans on a conveyor belt. Using the Segment Anything Model (SAM), contours are extracted, and ellipses are fitted to approximate bean shapes. In a dataset of 17,386 images, SAM generated over 3.5 million contours, with approximately 265,000 annotated as individual beans. Preliminary results from a deep learning-based ellipse detection model and a panoptic segmentation model, both trained on the generated soybean dataset, are presented
Hop yield forecast using Sentinel-2 images
No full textThis study shows thast satellite-based derived information may support decision-making in hop crop management, especially for yield forecasting. The May-to-July period proved crucial for yield predictions, with two approaches supporting this finding. First, tracking changes in the Red Edge Inflection Point (REIP) slope from May to July revealed Pearson correlations of 0.84, 0.77, and 0.61 for the Tradition, Perle and Herkules hop varieties, respectively. A slightly stronger correlation emerged through a linear regression analysis of average REIP values against yields over the 92-day growing period from May 1 to the end of July. Comparing REIP on the 75th growth day in mid-July with yield achieved correlations of 0.86 for Tradition, 0.69 for Perle, and 0.76 for Herkules. Research continues, with plans to incorporate more farms and management data to further refine these predictions
Von manuell zu maschinell: Erfolgsfaktoren bei der Einführung automatisierter Führungssysteme
No full textUm die erfolgreiche Einführung eines Informationssystems zu gewährleisten, sollten Erfolgsfaktoren in Organisationen berücksichtigt werden. Dies gilt auch für automatisierte Führungssysteme, die als Informationssysteme im Personalbereich betrachtet werden können. Bei der automatisierten Führung übernehmen intelligente Agenten Führungsaufgaben, die bisher von Menschen ausgeführt wurden. Dies können zum Beispiel Führungsaufgaben wie die Aufgabenverteilung oder das Leistungsfeedback an Mitarbeiter sein. Der Einsatz von automatisierter Führung ist in Unternehmen der Gig-Economy wie MTurk weit verbreitet. In traditionellen Organisationen ist der Einsatz bisher weniger verbreitet. Durch die zunehmende Verlagerung der Arbeitsweise hin zum selbstständigen Arbeiten mit digitalen Medien im Homeoffice wird der Einsatz von automatisierten Führungssystemen in konventionellen Unternehmen prinzipiell möglich. Um zukünftig die erfolgreiche Implementierung dieser Technologie in konventionellen Unternehmen zu gewährleisten, untersucht dieser Beitrag die Erfolgsfaktoren der Implementierung der automatisierten Führung. Dazu wurden qualitative, halbstandardisierte Interviews mit 24 deutschen Führungskräften durchgeführt. Die Ergebnisse der Studie zeigen, dass die Akzeptanz der Nutzer eine zentrale Rolle spielt. Durch die praktische Anwendung der gewonnenen Erkenntnisse können Unternehmen ihre Transformation zur Automatisierung durch Führungssysteme erfolgreich gestalten. To ensure the successful introduction of an information system, success factors in organizations should be taken into account. This also applies to automated leadership systems, which can be regarded as information systems in the human resources sector. In automated leadership, intelligent agents take over leadership tasks that were previously performed by humans. These can be, for example, tasks such as task allocation or performance feedback to employees. A significant use of automated leadership can be seen in gig economy companies such as MTurk. In conventional organizations, its use has so far been less widespread. The increasing shift towards independent working with digital media in the home office means that the use of automated leadership systems in traditional companies is fundamentally possible. In order to ensure the successful implementation of this technology in traditional companies in the future, this article examines the success factors for the implementation of automated leadership. To this end, qualitative, semi-standardized interviews were conducted with 24 German leaders. The results of the study show that user acceptance plays a central role. By applying the insights gained in practice, companies can successfully manage their transformation towards automation through leadership systems
Creating a Reusable Execution Environment for WiKoDa
No full textEnhancing reusability is a key challenge in implementing FAIR principles for research software. This case study explores strategies to boost the reusability of the WiKoDa research project by creating a Reusable Execution Environment (REE). The REE facilitates software execution in an independent, reproducible setting