Ergebnisse - IIP-Ecosphere

Publikationen des Projekts

Architecture Artificial Intelligence Asset Administration Shells Asset-Administration-Shell Datenschutz Digital annealer Digital Twin Edge Envelope Spectrum Flexible job shop scheduling IIoT IIoT-Platform IIP-Ecosphere Industrie 4.0 Industry 4.0 intelligent production KI in der Produktion Künstliche Intelligenz Manual Platform Platform Activities Platform Requirements Produktion Prozessüberwachung Requirements Sennheiser Umfrage UML Verwaltungsschale Virtual Platform
2023
Drumond, Rafael Rego; Brinkmeyer, Lukas; Schmidt-Thieme, Lars
Few-Shot Human Motion Prediction for Heterogeneous Sensors Inproceedings
In: Kashima, Hisashi; Ide, Tsuyoshi; Peng, Wen-Chih (Ed.): Advances in Knowledge Discovery and Data Mining - 27th Pacific-Asia Conference on Knowledge Discovery and Data Mining, {PAKDD} 2023, Osaka, Japan, May 25-28, 2023, Proceedings, Part {II}, 2023.
Abstract | BibTeX | Schlagwörter:
@inproceedings{DrumondBS23,
title = {Few-Shot Human Motion Prediction for Heterogeneous Sensors},
author = {Rafael Rego Drumond and Lukas Brinkmeyer and Lars Schmidt-Thieme},
editor = {Hisashi Kashima and Tsuyoshi Ide and Wen-Chih Peng},
year = {2023},
date = {2023-05-25},
booktitle = {Advances in Knowledge Discovery and Data Mining - 27th Pacific-Asia Conference on Knowledge Discovery and Data Mining, {PAKDD} 2023, Osaka, Japan, May 25-28, 2023, Proceedings, Part {II}},
abstract = {Human motion prediction is a complex task as it involves forecasting variables over time on a graph of connected sensors. This is especially true in the case of few-shot learning, where we strive to forecast motion sequences for previously unseen actions based on only a few examples. Despite this, almost all related approaches for few-shot motion prediction do not incorporate the underlying graph, while it is a common component in classical motion prediction. Furthermore, state-of-the-art methods for few-shot motion prediction are restricted to motion tasks with a fixed output space meaning these tasks are all limited to the same sensor graph. In this work, we propose to extend recent works on few-shot time-series forecasting with heterogeneous attributes with graph neural networks to introduce the first few-shot motion approach that explicitly incorporates the spatial graph while also generalizing across motion tasks with heterogeneous sensors. In our experiments on motion tasks with heterogeneous sensors, we demonstrate significant performance improvements with lifts from 10.4% up to 39.3% compared to best state-of-the-art models. Moreover, we show that our model can perform on par with the best approach so far when evaluating on tasks with a fixed output space while maintaining two magnitudes fewer parameters.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}

Close
Human motion prediction is a complex task as it involves forecasting variables over time on a graph of connected sensors. This is especially true in the case of few-shot learning, where we strive to forecast motion sequences for previously unseen actions based on only a few examples. Despite this, almost all related approaches for few-shot motion prediction do not incorporate the underlying graph, while it is a common component in classical motion prediction. Furthermore, state-of-the-art methods for few-shot motion prediction are restricted to motion tasks with a fixed output space meaning these tasks are all limited to the same sensor graph. In this work, we propose to extend recent works on few-shot time-series forecasting with heterogeneous attributes with graph neural networks to introduce the first few-shot motion approach that explicitly incorporates the spatial graph while also generalizing across motion tasks with heterogeneous sensors. In our experiments on motion tasks with heterogeneous sensors, we demonstrate significant performance improvements with lifts from 10.4% up to 39.3% compared to best state-of-the-art models. Moreover, we show that our model can perform on par with the best approach so far when evaluating on tasks with a fixed output space while maintaining two magnitudes fewer parameters.
Close
Wißbrock, Peter; Richter, Yvonne; Pelkmann, David; Ren, Zhao; Palmer, Gregory
Cutting Through the Noise: An Empirical Comparison of Psycho-Acoustic and Envelope-based Features for Machinery Fault Detection Inproceedings
In: ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1-5, IEEE 2023.
Abstract | Links | BibTeX | Schlagwörter: Assembly Line Inspection, Envelope Spectrum, Gear Fault Detection, IIP-Ecosphere, Industrial Noise, Psychoacoustics
@inproceedings{wibetabrock2023cutting,
title = {Cutting Through the Noise: An Empirical Comparison of Psycho-Acoustic and Envelope-based Features for Machinery Fault Detection},
author = {Peter Wißbrock and Yvonne Richter and David Pelkmann and Zhao Ren and Gregory Palmer },
url = {https://arxiv.org/pdf/2211.01704.pdf
https://ieeexplore.ieee.org/document/10095756},
year = {2023},
date = {2023-05-05},
urldate = {2023-05-05},
booktitle = {ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
pages = {1-5},
organization = {IEEE},
abstract = {Acoustic-based fault detection has a high potential to monitor the health condition of mechanical parts. However, the background noise of an industrial environment may negatively influence the performance of fault detection. Limited attention has been paid to improving the robustness of fault detection against industrial environmental noise. Therefore, we present the Lenze production background-noise (LPBN) real-world dataset and an automated and noise-robust auditory inspection (ARAI) system for the end-of-line inspection of geared motors. An acoustic array is used to acquire data from motors with a minor fault, major fault, or which are healthy. A benchmark is provided to compare the psychoacoustic features with different types of envelope features based on expert knowledge of the gearbox. To the best of our knowledge, we are the first to apply time-varying psychoacoustic features for fault detection. We train a state-of-the-art one-class-classifier, on samples from healthy motors and separate the faulty ones for fault detection using a threshold. The best-performing approaches achieve an area under curve of 0.87 (logarithm envelope), 0.86 (time-varying psychoacoustics), and 0.91 (combination of both).},
keywords = {Assembly Line Inspection, Envelope Spectrum, Gear Fault Detection, IIP-Ecosphere, Industrial Noise, Psychoacoustics},
pubstate = {published},
tppubtype = {inproceedings}
}

Close
Acoustic-based fault detection has a high potential to monitor the health condition of mechanical parts. However, the background noise of an industrial environment may negatively influence the performance of fault detection. Limited attention has been paid to improving the robustness of fault detection against industrial environmental noise. Therefore, we present the Lenze production background-noise (LPBN) real-world dataset and an automated and noise-robust auditory inspection (ARAI) system for the end-of-line inspection of geared motors. An acoustic array is used to acquire data from motors with a minor fault, major fault, or which are healthy. A benchmark is provided to compare the psychoacoustic features with different types of envelope features based on expert knowledge of the gearbox. To the best of our knowledge, we are the first to apply time-varying psychoacoustic features for fault detection. We train a state-of-the-art one-class-classifier, on samples from healthy motors and separate the faulty ones for fault detection using a threshold. The best-performing approaches achieve an area under curve of 0.87 (logarithm envelope), 0.86 (time-varying psychoacoustics), and 0.91 (combination of both).
Close
https://arxiv.org/pdf/2211.01704.pdf
https://ieeexplore.ieee.org/document/10095756
Close
Feichtinger, Kevin; Meixner, Kristof; Rinker, Felix; Koren, István; Eichelberger, Holger; Heinemann, Tonja; Holtmann, Jörg; Konersmann, Marco; Michael, Judith; Neumann, Eva-Maria; Pfeiffer, Jérôme; Rabiser, Rick; Riebisch, Matthias; Schmid, Klaus
Software in Cyberphysischen Produktionssystemen - Herausforderungen zur Umsetzung in der Industrie Journal Article
In: ATP-Magazin, 2023 (4), pp. 62-68, 2023, ISSN: 2190-4111.
Abstract | BibTeX | Schlagwörter: IIP-Ecosphere
@article{FMR+23,
title = {Software in Cyberphysischen Produktionssystemen - Herausforderungen zur Umsetzung in der Industrie},
author = {Kevin Feichtinger and Kristof Meixner and Felix Rinker and István Koren and Holger Eichelberger and Tonja Heinemann and Jörg Holtmann and Marco Konersmann and Judith Michael and Eva-Maria Neumann and Jérôme Pfeiffer and Rick Rabiser and Matthias Riebisch and Klaus Schmid},
issn = {2190-4111},
year = {2023},
date = {2023-04-01},
journal = {ATP-Magazin},
volume = {2023},
number = {4},
pages = {62-68},
abstract = {Um den effektiven und effizienten Betrieb von Cyberphysischen Produktionssystemen (CPPSen) sicherzustellen, spielt Software eine zunehmend wichtige Rolle. Die enormen Fortschritte bei Softwareentwicklungsmethoden, welche in den letzten Jahren erzielt wurden, scheinen jedoch die aktuellen Herausforderungen der Industrie nicht zu erfüllen, weil diese die Industrie nicht oder nur langsam erreichen. In diesem Beitrag werden die Herausforderungen für die Softwareentwicklung in CPPSen aus Sicht von neun Industrievertretern aus acht europäischen Unternehmen unterschiedlicher Größe diskutiert. Um den digitalen Transformationsprozess für eine zukunftsfähige Produktion zu begleiten, wurden aus den beschriebenen Herausforderungen Perspektiven für die Forschung erarbeitet. Die Umsetzung dieser Ziele ist vor dem Hintergrund von ökonomischen, sozialen und Nachhaltigkeitsanforderungen notwendig.},
keywords = {IIP-Ecosphere},
pubstate = {published},
tppubtype = {article}
}

Close
Um den effektiven und effizienten Betrieb von Cyberphysischen Produktionssystemen (CPPSen) sicherzustellen, spielt Software eine zunehmend wichtige Rolle. Die enormen Fortschritte bei Softwareentwicklungsmethoden, welche in den letzten Jahren erzielt wurden, scheinen jedoch die aktuellen Herausforderungen der Industrie nicht zu erfüllen, weil diese die Industrie nicht oder nur langsam erreichen. In diesem Beitrag werden die Herausforderungen für die Softwareentwicklung in CPPSen aus Sicht von neun Industrievertretern aus acht europäischen Unternehmen unterschiedlicher Größe diskutiert. Um den digitalen Transformationsprozess für eine zukunftsfähige Produktion zu begleiten, wurden aus den beschriebenen Herausforderungen Perspektiven für die Forschung erarbeitet. Die Umsetzung dieser Ziele ist vor dem Hintergrund von ökonomischen, sozialen und Nachhaltigkeitsanforderungen notwendig.
Close
Eichelberger, Holger; Palmer, Gregory; Niederee, Claudia
Developing an AI-enabled Industry 4.0 platform - Performance experiences on deploying AI onto an industrial edge device Journal Article
In: Softwaretechnik-Trends, 43 (1), pp. 35-37, 2023, ISSN: 0702-8928.
Abstract | BibTeX | Schlagwörter: IIP-Ecosphere, Industrie 4.0, Industry 4.0, Platform
@article{nokey,
title = {Developing an AI-enabled Industry 4.0 platform - Performance experiences on deploying AI onto an industrial edge device},
author = {Holger Eichelberger and Gregory Palmer and Claudia Niederee},
issn = {0702-8928},
year = {2023},
date = {2023-02-01},
journal = {Softwaretechnik-Trends},
volume = {43},
number = {1},
pages = {35-37},
publisher = {GI},
abstract = {Maximizing the benefits of AI for Industry 4.0 is about more than just developing effective new AI methods. Of equal importance is the successful integration of AI into production environments. One open challenge is the dynamic deployment of AI on industrial edge devices within close proximity to manufacturing machines. Our IIP-Ecosphere platform was designed to overcome limitations of existing Industry 4.0 platforms. It supports flexible AI deployment through employing a highly configurable low-code based approach, where code for tailored platform components and applications is generated. In this paper, we measure the performance of our platform on an industrial demonstrator and discuss the impact of deploying AI from a central server to the edge. As result, AI inference automatically deployed on an industrial edge is possible, but in our case three times slower than on a desktop computer, requiring still more optimizations.},
keywords = {IIP-Ecosphere, Industrie 4.0, Industry 4.0, Platform},
pubstate = {published},
tppubtype = {article}
}

Close
Maximizing the benefits of AI for Industry 4.0 is about more than just developing effective new AI methods. Of equal importance is the successful integration of AI into production environments. One open challenge is the dynamic deployment of AI on industrial edge devices within close proximity to manufacturing machines. Our IIP-Ecosphere platform was designed to overcome limitations of existing Industry 4.0 platforms. It supports flexible AI deployment through employing a highly configurable low-code based approach, where code for tailored platform components and applications is generated. In this paper, we measure the performance of our platform on an industrial demonstrator and discuss the impact of deploying AI from a central server to the edge. As result, AI inference automatically deployed on an industrial edge is possible, but in our case three times slower than on a desktop computer, requiring still more optimizations.
Close
Sauer, Christian Severin; Eichelberger, Holger
Performance Evaluation of BaSyx based Asset Administration Shells for Industry 4.0 Applications Journal Article
In: Softwaretechnik-Trends, 43 (1), pp. 47-49, 2023, ISSN: 0702-8928.
Abstract | BibTeX | Schlagwörter: Asset Administration Shells, IIP-Ecosphere, Industrie 4.0, Industry 4.0, Platform, Verwaltungsschale
@article{nokey,
title = {Performance Evaluation of BaSyx based Asset Administration Shells for Industry 4.0 Applications},
author = {Christian Severin Sauer and Holger Eichelberger},
issn = {0702-8928},
year = {2023},
date = {2023-02-01},
journal = {Softwaretechnik-Trends},
volume = {43},
number = {1},
pages = {47-49},
publisher = {GI},
abstract = {The Asset Administration Shell (AAS) is an upcoming information model standard, which aims at interoperable modeling of “assets”, i.e., products, machines, services or digital twins in IIoT/Industry 4.0. Currently, a number of IIoT-platforms use proprietary information models similar to AAS, but not a common standard, which affects interoperability.A key question for a broad uptake is if AAS can be applied in a performant and scalable manner. In this paper, we examine this question for the open source Eclipse BaSyx middleware. To explore capabilities and possible performance limitations, we present four experiments measuring the performance of experimental AAS in BaSyx and, within the context set by our experiments, i.e., 10-1000 AAS instances, can conclude good scalability.},
keywords = {Asset Administration Shells, IIP-Ecosphere, Industrie 4.0, Industry 4.0, Platform, Verwaltungsschale},
pubstate = {published},
tppubtype = {article}
}

Close
The Asset Administration Shell (AAS) is an upcoming information model standard, which aims at interoperable modeling of “assets”, i.e., products, machines, services or digital twins in IIoT/Industry 4.0. Currently, a number of IIoT-platforms use proprietary information models similar to AAS, but not a common standard, which affects interoperability.A key question for a broad uptake is if AAS can be applied in a performant and scalable manner. In this paper, we examine this question for the open source Eclipse BaSyx middleware. To explore capabilities and possible performance limitations, we present four experiments measuring the performance of experimental AAS in BaSyx and, within the context set by our experiments, i.e., 10-1000 AAS instances, can conclude good scalability.
Close
Alamoush, Ahmad; Eichelberger, Holger
Adapting Kubernetes to IIoT and Industry 4.0 protocols - An initial performance analysis Journal Article
In: Softwaretechnik-Trends, 43 (1), pp. 41-43, 2023, ISSN: 0702-8928.
Abstract | BibTeX | Schlagwörter: IIoT, IIoT-Platform, IIP-Ecosphere, Platform
@article{nokey,
title = {Adapting Kubernetes to IIoT and Industry 4.0 protocols - An initial performance analysis},
author = {Ahmad Alamoush and Holger Eichelberger},
issn = {0702-8928},
year = {2023},
date = {2023-02-01},
urldate = {2023-02-01},
journal = {Softwaretechnik-Trends},
volume = {43},
number = {1},
pages = {41-43},
publisher = {GI},
abstract = {Kubernetes (K8s) is one of the most frequently used container orchestration tools offering, as it offers a rich set of functions to manage containerized applications, it is customizable and extensible. Container virtualization of applications and their orchestration on heterogeneous resources including edge devices is a recent trend in Industrial Internet of Things (IIoT)/Industry 4.0, where K8s is also applied. However, IIoT/Industry 4.0 is a domain with high standardization requirements. Besides equipment standards, e.g., for electrical control cabinets, there are also demands to standardize network protocols, data formats or information models. Such standards can foster interoperability and reduce complexity or deployment/integration costs. Here, the proprietary communication protocol of K8s and similar orchestrators can be an obstacle for adoption. To explore this situation from an interoperability and integration perspective, we present in this paper an approach to replace the communication protocol of K8s without modifying its code base. We show by an experiment that applying our approach with three current forms of IIoT communication, namely Message Queuing Telemetry Transport (MQTT), Advanced Message Queuing Protocol (AMQP), and Asset Administration Shell (AAS), does not significantly affect the validity and the performance of K8s.},
keywords = {IIoT, IIoT-Platform, IIP-Ecosphere, Platform},
pubstate = {published},
tppubtype = {article}
}

Close
Kubernetes (K8s) is one of the most frequently used container orchestration tools offering, as it offers a rich set of functions to manage containerized applications, it is customizable and extensible. Container virtualization of applications and their orchestration on heterogeneous resources including edge devices is a recent trend in Industrial Internet of Things (IIoT)/Industry 4.0, where K8s is also applied. However, IIoT/Industry 4.0 is a domain with high standardization requirements. Besides equipment standards, e.g., for electrical control cabinets, there are also demands to standardize network protocols, data formats or information models. Such standards can foster interoperability and reduce complexity or deployment/integration costs. Here, the proprietary communication protocol of K8s and similar orchestrators can be an obstacle for adoption. To explore this situation from an interoperability and integration perspective, we present in this paper an approach to replace the communication protocol of K8s without modifying its code base. We show by an experiment that applying our approach with three current forms of IIoT communication, namely Message Queuing Telemetry Transport (MQTT), Advanced Message Queuing Protocol (AMQP), and Asset Administration Shell (AAS), does not significantly affect the validity and the performance of K8s.
Close
2022
Reimer, Jef; YandongWang,; Laridi, Sofane; Urdich, Juergen; SörenWilmsmeier,; Palmer, Gregory
Identifying cause-and-effect relationships of manufacturing errors using sequence-to-sequence learning Technical Report
2022.
Abstract | Links | BibTeX | Schlagwörter:
@techreport{Reimer2023,
title = {Identifying cause-and-effect relationships of manufacturing errors using sequence-to-sequence learning},
author = {Jef Reimer and YandongWang and Sofane Laridi and Juergen Urdich and SörenWilmsmeier and Gregory Palmer},
url = {https://www.nature.com/articles/s41598-022-26534-y#Abs1},
doi = {https://doi.org/10.1038/s41598-022-26534-y},
year = {2022},
date = {2022-12-25},
abstract = {In car-body production the pre-formed sheet metal parts of the body are assembled on fully-automated production lines. The body passes through multiple stations in succession, and is processed according to the order requirements. The timely completion of orders depends on the individual station-based operations concluding within their scheduled cycle times. If an error occurs in one station, it can have a knock-on effect, resulting in delays on the downstream stations. To the best of our knowledge, there exist no methods for automatically distinguishing between source and knock-on errors in this setting, as well as establishing a causal relation between them. Utilizing real-time information about conditions collected by a production data acquisition system, we propose a novel vehicle manufacturing analysis system, which uses deep learning to establish a link between source and knock-on errors. We benchmark three sequence-to-sequence models, and introduce a novel composite time-weighted action metric for evaluating models in this context. We evaluate our framework on a real-world car production dataset recorded by Volkswagen Commercial Vehicles. Surprisingly we find that 71.68% of sequences contain either a source or knock-on error. With respect to seq2seq model training, we find that the Transformer demonstrates a better performance compared to LSTM and GRU in this domain, in particular when the prediction range with respect to the durations of future actions is increased.},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}

Close
In car-body production the pre-formed sheet metal parts of the body are assembled on fully-automated production lines. The body passes through multiple stations in succession, and is processed according to the order requirements. The timely completion of orders depends on the individual station-based operations concluding within their scheduled cycle times. If an error occurs in one station, it can have a knock-on effect, resulting in delays on the downstream stations. To the best of our knowledge, there exist no methods for automatically distinguishing between source and knock-on errors in this setting, as well as establishing a causal relation between them. Utilizing real-time information about conditions collected by a production data acquisition system, we propose a novel vehicle manufacturing analysis system, which uses deep learning to establish a link between source and knock-on errors. We benchmark three sequence-to-sequence models, and introduce a novel composite time-weighted action metric for evaluating models in this context. We evaluate our framework on a real-world car production dataset recorded by Volkswagen Commercial Vehicles. Surprisingly we find that 71.68% of sequences contain either a source or knock-on error. With respect to seq2seq model training, we find that the Transformer demonstrates a better performance compared to LSTM and GRU in this domain, in particular when the prediction range with respect to the durations of future actions is increased.
Close
https://www.nature.com/articles/s41598-022-26534-y#Abs1
doi:https://doi.org/10.1038/s41598-022-26534-y
Close
Eichelberger, Holger; Reimer, Svenja; Niederée, Claudia; Palmer, Gregory
Virtuelle IIoT-Plattform für die Digitalisierung der Fertigung Journal Article
In: Zeitschrift für wirtschaftlichen Fabrikbetrieb, 117 (12), 2022.
Abstract | Links | BibTeX | Schlagwörter: Digitalisierung, IIoT-Platform, IIP-Ecosphere, Verwaltungsschale
@article{ERNP22,
title = {Virtuelle IIoT-Plattform für die Digitalisierung der Fertigung},
author = {Holger Eichelberger and Svenja Reimer and Claudia Niederée and Gregory Palmer},
url = {https://www.degruyter.com/document/doi/10.1515/zwf-2022-1178/html},
doi = {https://doi.org/10.1515/zwf-2022-1178},
year = {2022},
date = {2022-12-16},
journal = {Zeitschrift für wirtschaftlichen Fabrikbetrieb},
volume = {117},
number = {12},
publisher = {de Gruyter},
abstract = {Für die erfolgreiche Digitalisierung in der Produktion ist die IT-Infrastruktur, zum Beispiel zur einfachen Anbindung von Geräten und Steuerung von Datenflüssen, von zentraler Bedeutung.
Bisherige Lösungen basieren jedoch meist auf proprietären Protokollen und bieten wenig Konfigurations- und Kontrollmöglichkeiten. Basierend auf industriellen Standards, wie z. B. Verwaltungsschalen, wird im Projekt IIP-Ecosphere daher eine offene Code-Basis für die ganzheitliche Umsetzung von Digitalisierungsprojekten in der Produktion entwickelt.},
keywords = {Digitalisierung, IIoT-Platform, IIP-Ecosphere, Verwaltungsschale},
pubstate = {published},
tppubtype = {article}
}

Close
Für die erfolgreiche Digitalisierung in der Produktion ist die IT-Infrastruktur, zum Beispiel zur einfachen Anbindung von Geräten und Steuerung von Datenflüssen, von zentraler Bedeutung.
Bisherige Lösungen basieren jedoch meist auf proprietären Protokollen und bieten wenig Konfigurations- und Kontrollmöglichkeiten. Basierend auf industriellen Standards, wie z. B. Verwaltungsschalen, wird im Projekt IIP-Ecosphere daher eine offene Code-Basis für die ganzheitliche Umsetzung von Digitalisierungsprojekten in der Produktion entwickelt.
Close
https://www.degruyter.com/document/doi/10.1515/zwf-2022-1178/html
doi:https://doi.org/10.1515/zwf-2022-1178
Close
Eichelberger, Holger; Palmer, Gregory; Niederée, Claudia
Developing an AI-enabled Industry 4.0 platform – Performance experiences on deploying AI onto an industrial edge device Inproceedings
In: 13th Symposium on Software Performance 2022, 2022.
Abstract | Links | BibTeX | Schlagwörter: IIP-Ecosphere, Industrie 4.0, Platform Activities, Virtual Platform
@inproceedings{eichelberger2022ssp,
title = {Developing an AI-enabled Industry 4.0 platform – Performance experiences on deploying AI onto an industrial edge device},
author = {Holger Eichelberger and Gregory Palmer and Claudia Niederée},
url = {https://www.iip-ecosphere.de/ssp22/},
year = {2022},
date = {2022-11-09},
booktitle = {13th Symposium on Software Performance 2022},
abstract = {Bei der Maximierung des Nutzens von KI für die Industrie 4.0 geht es um mehr als nur die Entwicklung effektiver neuer KI-Methoden. Ebenso wichtig ist die erfolgreiche Integration von KI in Produktionsumgebungen. Eine offene Herausforderung ist der dynamische Einsatz von KI auf industriellen Edge-Geräten in unmittelbarer Nähe von Produktionsmaschinen. Die IIP-Ecosphere-Plattform wurde entwickelt, um die Einschränkungen bestehender Industrie 4.0-Plattformen zu überwinden. Sie unterstützt den flexiblen Einsatz von KI durch einen hochgradig konfigurierbaren Low-Code-basierten Ansatz, bei dem Code für maßgeschneiderte Plattformkomponenten und Anwendungen generiert wird. In diesem Papier messen wir die Leistung unserer Plattform an einem industriellen Demonstrator und diskutieren die Auswirkungen des Einsatzes von KI, ausgehend von einem zentralen Server, auf der Edge.},
keywords = {IIP-Ecosphere, Industrie 4.0, Platform Activities, Virtual Platform},
pubstate = {published},
tppubtype = {inproceedings}
}

Close
Bei der Maximierung des Nutzens von KI für die Industrie 4.0 geht es um mehr als nur die Entwicklung effektiver neuer KI-Methoden. Ebenso wichtig ist die erfolgreiche Integration von KI in Produktionsumgebungen. Eine offene Herausforderung ist der dynamische Einsatz von KI auf industriellen Edge-Geräten in unmittelbarer Nähe von Produktionsmaschinen. Die IIP-Ecosphere-Plattform wurde entwickelt, um die Einschränkungen bestehender Industrie 4.0-Plattformen zu überwinden. Sie unterstützt den flexiblen Einsatz von KI durch einen hochgradig konfigurierbaren Low-Code-basierten Ansatz, bei dem Code für maßgeschneiderte Plattformkomponenten und Anwendungen generiert wird. In diesem Papier messen wir die Leistung unserer Plattform an einem industriellen Demonstrator und diskutieren die Auswirkungen des Einsatzes von KI, ausgehend von einem zentralen Server, auf der Edge.
Close
https://www.iip-ecosphere.de/ssp22/
Close
Alamoush, Ahmad; Eichelberger, Holger
Adapting Kubernetes to IIoT and Industry 4.0 protocols - An initial performance analysis Inproceedings
In: 13th Symposium on Software Performance 2022, 2022.
Abstract | Links | BibTeX | Schlagwörter: IIoT, IIP-Ecosphere, Industrie 4.0, Industry 4.0, Virtual Platform
@inproceedings{alamoush2022ssp,
title = {Adapting Kubernetes to IIoT and Industry 4.0 protocols - An initial performance analysis},
author = {Ahmad Alamoush and Holger Eichelberger},
url = {https://www.iip-ecosphere.de/ssp-kubernetes-cr/},
year = {2022},
date = {2022-11-08},
urldate = {2022-11-08},
booktitle = {13th Symposium on Software Performance 2022},
abstract = {Kubernetes ist eines der am häufigsten verwendeten Container-Orchestrierungstools, denn es bietet
reichhaltige Funktionen zur Verwaltung von containerisierten Anwendungen, ist anpassbar und erweiterbar. Die Virtualisierung von Containern von Anwendungen und deren Orchestrierung auf heterogenen Ressourcen einschließlich Edge-Geräten ist ein neuer Trend im industriellen Internet der Dinge (IIoT)/Industrie 4.0 in der auch verstärkt Kubernetes eingesetzt wird. Allerdings sind, IIoT/Industrie 4.0 Domänen mit hohen Anforderungen an derzeitigen und künftigen Standardisierungsanforderungen. Neben Gerätenormen, z.B. für elektrische Schaltschränke, gibt es auch
Anforderungen an die Standardisierung von Netzwerkprotokollen, Daten Datenformaten oder Informationsmodellen. Hier ist das proprietäre Kommunikationsprotokoll von Kubernetes gegebenenfalls ein Hindernis für die Akzeptanz von Kubernetes. Um diese Situation unter dem Gesichtspunkt der Interoperabilität und Integration zu untersuchen, stellen wir in diesem Papier einen Ansatz zum Austausch des Kommunikationsprotokolls von Kubernetes vor, ohne dessen Codebasis zu verändern. Wir zeigen die Anwendung unseres Ansatzes der Verwendung von Kubernetes mit drei aktuellen Formen der IIoT-Kommunikation: Message Queuing Telemetry Transport (MQTT), Advanced Advanced Message Queuing Protocol (AMQP) und Asset Administration Shell (AAS).},
keywords = {IIoT, IIP-Ecosphere, Industrie 4.0, Industry 4.0, Virtual Platform},
pubstate = {published},
tppubtype = {inproceedings}
}

Close
Kubernetes ist eines der am häufigsten verwendeten Container-Orchestrierungstools, denn es bietet
reichhaltige Funktionen zur Verwaltung von containerisierten Anwendungen, ist anpassbar und erweiterbar. Die Virtualisierung von Containern von Anwendungen und deren Orchestrierung auf heterogenen Ressourcen einschließlich Edge-Geräten ist ein neuer Trend im industriellen Internet der Dinge (IIoT)/Industrie 4.0 in der auch verstärkt Kubernetes eingesetzt wird. Allerdings sind, IIoT/Industrie 4.0 Domänen mit hohen Anforderungen an derzeitigen und künftigen Standardisierungsanforderungen. Neben Gerätenormen, z.B. für elektrische Schaltschränke, gibt es auch
Anforderungen an die Standardisierung von Netzwerkprotokollen, Daten Datenformaten oder Informationsmodellen. Hier ist das proprietäre Kommunikationsprotokoll von Kubernetes gegebenenfalls ein Hindernis für die Akzeptanz von Kubernetes. Um diese Situation unter dem Gesichtspunkt der Interoperabilität und Integration zu untersuchen, stellen wir in diesem Papier einen Ansatz zum Austausch des Kommunikationsprotokolls von Kubernetes vor, ohne dessen Codebasis zu verändern. Wir zeigen die Anwendung unseres Ansatzes der Verwendung von Kubernetes mit drei aktuellen Formen der IIoT-Kommunikation: Message Queuing Telemetry Transport (MQTT), Advanced Advanced Message Queuing Protocol (AMQP) und Asset Administration Shell (AAS).
Close
https://www.iip-ecosphere.de/ssp-kubernetes-cr/
Close
Sauer, Christian; Eichelberger, Holger
Performance Evaluation of BaSyx based Asset Administration Shells for Industry 4.0 Applications Inproceedings
In: 13th Symposium on Software Performance 2022, 2022.
Abstract | Links | BibTeX | Schlagwörter: Asset Administration Shells, IIP-Ecosphere, Industrie 4.0
@inproceedings{sauer2022ssp,
title = {Performance Evaluation of BaSyx based Asset Administration Shells for Industry 4.0 Applications},
author = {Christian Sauer and Holger Eichelberger},
url = {https://www.iip-ecosphere.de/ssp22_aas_paper-crc/},
year = {2022},
date = {2022-11-08},
booktitle = {13th Symposium on Software Performance 2022},
abstract = {Die Asset Administration Shell (AAS) ist ein neuer Informationsmodell-Standard, der auf die interoperable Modellierung von "Assets", d.h. von Produkten, Maschinen, Dienstleistungen oder digitalen Zwillingen im IIoT/Industrie 4.0. Derzeit, verwenden eine Reihe von IIoT-Plattformen proprietäre Informationsmodelle ähnlich dem AAS, aber keinen gemeinsamen Standard, was die Interoperabilität beeinträchtigt. Eine Schlüsselfrage für eine breite Akzeptanz der AAS ist, ob AAS leistungsfähig und skalierbar eingesetzt werden können. In diesem Papier untersuchen wir diese Frage für AAS, die mit der quelloffene Eclipse BaSyx Middleware erstellt werden.},
keywords = {Asset Administration Shells, IIP-Ecosphere, Industrie 4.0},
pubstate = {published},
tppubtype = {inproceedings}
}

Close
Die Asset Administration Shell (AAS) ist ein neuer Informationsmodell-Standard, der auf die interoperable Modellierung von "Assets", d.h. von Produkten, Maschinen, Dienstleistungen oder digitalen Zwillingen im IIoT/Industrie 4.0. Derzeit, verwenden eine Reihe von IIoT-Plattformen proprietäre Informationsmodelle ähnlich dem AAS, aber keinen gemeinsamen Standard, was die Interoperabilität beeinträchtigt. Eine Schlüsselfrage für eine breite Akzeptanz der AAS ist, ob AAS leistungsfähig und skalierbar eingesetzt werden können. In diesem Papier untersuchen wir diese Frage für AAS, die mit der quelloffene Eclipse BaSyx Middleware erstellt werden.
Close
https://www.iip-ecosphere.de/ssp22_aas_paper-crc/
Close
Faubel, Leonhard; Schmid, Klaus; Eichelberger, Holger
Is MLOps different in Industry 4.0? General and Specific Challenges Conference
3rd International Conference on Innovative Intelligent Industrial Production and Logistics (IN4PL), 2022.
Abstract | Links | BibTeX | Schlagwörter: Industrie 4.0, MLOps
@conference{nokey,
title = {Is MLOps different in Industry 4.0? General and Specific Challenges},
author = {Leonhard Faubel and Klaus Schmid and Holger Eichelberger},
doi = {10.5220/0011589600003329},
year = {2022},
date = {2022-11-04},
urldate = {2022-11-04},
booktitle = {3rd International Conference on Innovative Intelligent Industrial Production and Logistics (IN4PL)},
pages = {161-167},
abstract = {An important part of the Industry 4.0 vision is the use of machine learning (ML) techniques to create novel capabilities and flexibility in industrial production processes. Currently, there is a strong emphasis on MLOps as an enabling collection of practices, techniques, and tools to integrate ML into industrial practice. However, while MLOps is often discussed in the context of pure software systems, Industry 4.0 systems received much less attention. So far, there is no specialized research for Industry 4.0 in this regard. In this position paper, we discuss whether MLOps in Industry 4.0 leads to significantly different challenges compared to typical Internet systems. We identify both context-independent MLOps challenges (general challenges) as well as challenges particular to Industry 4.0 (specific challenges) and conclude that MLOps works very similarly in Industry 4.0 systems to pure software systems. This indicates that existing tools and approaches are also mostly suited for the Industry 4.0 context.},
keywords = {Industrie 4.0, MLOps},
pubstate = {published},
tppubtype = {conference}
}

Close
An important part of the Industry 4.0 vision is the use of machine learning (ML) techniques to create novel capabilities and flexibility in industrial production processes. Currently, there is a strong emphasis on MLOps as an enabling collection of practices, techniques, and tools to integrate ML into industrial practice. However, while MLOps is often discussed in the context of pure software systems, Industry 4.0 systems received much less attention. So far, there is no specialized research for Industry 4.0 in this regard. In this position paper, we discuss whether MLOps in Industry 4.0 leads to significantly different challenges compared to typical Internet systems. We identify both context-independent MLOps challenges (general challenges) as well as challenges particular to Industry 4.0 (specific challenges) and conclude that MLOps works very similarly in Industry 4.0 systems to pure software systems. This indicates that existing tools and approaches are also mostly suited for the Industry 4.0 context.
Close
doi:10.5220/0011589600003329
Close
Eichelberger, Holger; Palmer, Gregory; Reimer, Svenja; Trong Vu, Tat; Do, Hieu; Laridi, Sofiane; Weber, Alexander; Niederée, Claudia; Hildebrandt, Thomas
Developing an AI-Enabled IIoT Platform - Lessons Learned from Early Use Case Validation Inproceedings
In: Batista, Thais; Burevs, Tom'avs; Raibulet, Claudia; Muccini, Henry (Ed.): Software Architecture. ECSA 2022 Tracks and Workshops, pp. 265-283, Springer International Publishing, Cham, 2022, ISBN: 978-3-031-36889-9.
Abstract | BibTeX | Schlagwörter: IIoT, IIoT-Platform, IIP-Ecosphere, Industrie 4.0, Industry 4.0, KI in der Produktion, Virtual Platform
@inproceedings{10.1007/978-3-031-36889-9_19,
title = {Developing an AI-Enabled IIoT Platform - Lessons Learned from Early Use Case Validation},
author = {Eichelberger, Holger
and Palmer, Gregory
and Reimer, Svenja
and Trong Vu, Tat
and Do, Hieu
and Laridi, Sofiane
and Weber, Alexander
and Niederée, Claudia
and Hildebrandt, Thomas},
editor = {Batista, Thais
and Bure{v{s}}, Tom{'a}{v{s}}
and Raibulet, Claudia
and Muccini, Henry},
isbn = {978-3-031-36889-9},
year = {2022},
date = {2022-09-19},
booktitle = {Software Architecture. ECSA 2022 Tracks and Workshops},
pages = {265-283},
publisher = {Springer International Publishing},
address = {Cham},
abstract = {For a broader adoption of AI in industrial production, adequate infrastructure capabilities and ecosystems are crucial. This includes easing the integration of AI with industrial devices, support for distributed deployment, monitoring, and consistent system configuration. IIoT platforms can play a major role here by providing a unified layer for the heterogeneous Industry 4.0/IIoT context.},
keywords = {IIoT, IIoT-Platform, IIP-Ecosphere, Industrie 4.0, Industry 4.0, KI in der Produktion, Virtual Platform},
pubstate = {published},
tppubtype = {inproceedings}
}

Close
For a broader adoption of AI in industrial production, adequate infrastructure capabilities and ecosystems are crucial. This includes easing the integration of AI with industrial devices, support for distributed deployment, monitoring, and consistent system configuration. IIoT platforms can play a major role here by providing a unified layer for the heterogeneous Industry 4.0/IIoT context.
Close
Eichelberger, Holger; Ahmadian, Amir Shayan; Dewes, Andreas; Ehl, Marco; Alamoush, Ahmad; Staciwa, Monika; Casado, Miguel Gómez
IIP-Ecosphere Platform Handbook v0.4.0 Whitepaper
In: 2022.
Links | BibTeX | Schlagwörter: Architecture, IIP-Ecosphere, Industrie 4.0, Manual, Virtual Platform
@whitepaper{Eichelbergerdb,
title = {IIP-Ecosphere Platform Handbook v0.4.0},
author = {Holger Eichelberger and Amir Shayan Ahmadian and Andreas Dewes and Marco Ehl and Ahmad Alamoush and Monika Staciwa and Miguel Gómez Casado},
url = {https://www.iip-ecosphere.de/wp-content/uploads/2022/09/PlatformHandbook-final-V0.4.pdf},
doi = {https://doi.org/10.5281/zenodo.7047640},
year = {2022},
date = {2022-08-10},
urldate = {2022-08-10},
keywords = {Architecture, IIP-Ecosphere, Industrie 4.0, Manual, Virtual Platform},
pubstate = {published},
tppubtype = {whitepaper}
}

Close
https://www.iip-ecosphere.de/wp-content/uploads/2022/09/PlatformHandbook-final-V[...]
doi:https://doi.org/10.5281/zenodo.7047640
Close
Eichelberger, Holger; Palmer, Gregory; Reimer, Svenja; Vu, Tat Trong; Do, Hieu; Laridi, Sofiane; Weber, Alexander; Niederée, Claudia; Hildebrandt, Thomas
Developing an AI-enabled IIoT platform - An early use case validation Inproceedings
In: SASI4 @ ECSA'22, 2022.
Abstract | Links | BibTeX | Schlagwörter: Artificial Intelligence, Asset Administration Shells, IIoT, Industry 4.0, Platform
@inproceedings{epr+22,
title = {Developing an AI-enabled IIoT platform - An early use case validation},
author = {Holger Eichelberger and Gregory Palmer and Svenja Reimer and Tat Trong Vu and Hieu Do and Sofiane Laridi and Alexander Weber and Claudia Niederée and Thomas Hildebrandt},
url = {https://arxiv.org/pdf/2207.04515},
doi = {https://doi.org/10.48550/arXiv.2207.04515},
year = {2022},
date = {2022-07-10},
urldate = {2022-07-10},
booktitle = {SASI4 @ ECSA'22},
abstract = {For a broader adoption of AI in industrial production, adequate infrastructure capabilities are crucial. This includes easing the integration of AI with industrial devices, support for distributed deployment, monitoring, and consistent system configuration. },
keywords = {Artificial Intelligence, Asset Administration Shells, IIoT, Industry 4.0, Platform},
pubstate = {published},
tppubtype = {inproceedings}
}

Close
For a broader adoption of AI in industrial production, adequate infrastructure capabilities are crucial. This includes easing the integration of AI with industrial devices, support for distributed deployment, monitoring, and consistent system configuration.
Close
https://arxiv.org/pdf/2207.04515
doi:https://doi.org/10.48550/arXiv.2207.04515
Close
Hildebrandt, Petra
Vortrag “Aufwand und Nutzen – Einsatz künstlicher Intelligenz bei der Prüfung von Leiterplatten“ Presentation
30.06.2022.
BibTeX | Schlagwörter: Artificial Intelligence, IIP-Ecosphere, Sennheiser
@misc{nokey,
title = {Vortrag “Aufwand und Nutzen – Einsatz künstlicher Intelligenz bei der Prüfung von Leiterplatten“},
author = {Petra Hildebrandt},
year = {2022},
date = {2022-06-30},
booktitle = {Rethink! Smart Manufacturing D/A/C/H},
keywords = {Artificial Intelligence, IIP-Ecosphere, Sennheiser},
pubstate = {published},
tppubtype = {presentation}
}

Close
Kirschbaum, Julius; Posselt, Tim; Roth, Angela
Use-Case-based Innovation for Artificial Intelligence - An ontological Approach Inproceedings
In: ECIS 2022 Proceedings, AIS Electronic Library (AISeL), 2022.
Abstract | Links | BibTeX | Schlagwörter: Adoption, Artificial Intelligence, IIP-Ecosphere, Innovation, Künstliche Intelligenz, Ontology, Use Case
@inproceedings{Kirschbaum2022,
title = {Use-Case-based Innovation for Artificial Intelligence - An ontological Approach},
author = {Julius Kirschbaum and Tim Posselt and Angela Roth},
url = {https://www.researchgate.net/publication/361667326_Use-case-based_innovation_for_artificial_intelligence_-_An_ontological_approach},
year = {2022},
date = {2022-06-21},
urldate = {2022-06-21},
booktitle = {ECIS 2022 Proceedings},
publisher = {AIS Electronic Library (AISeL)},
abstract = {Research has primarily focused on process models for AI-use-case-adoption, but neglected the use-cases themselves. In this research, an ontological artifact is developed as the basis for an AI-use-case-description-scheme. It allows practitioners and researchers to systematically describe such use-cases based on their level of abstraction and core characteristics. It enables them to classify, document and communicate these use-cases to support AI-adoption. We ground its development in diffusion of innovation theory and build upon research on AI-adoption. In particular, Rogers’ (2003) innovation decision process is utilised as a framework that explains adoption decisions by organisations. A Design Science Research approach is chosen that integrates the ontology development process by Noy and McGuinness (2001). In this research-in-progress, we conduct one ex ante and one ex post evaluation and plan for a second ex post evaluation that ensure the relevance and rigor of the artifact design.},
keywords = {Adoption, Artificial Intelligence, IIP-Ecosphere, Innovation, Künstliche Intelligenz, Ontology, Use Case},
pubstate = {published},
tppubtype = {inproceedings}
}

Close
Research has primarily focused on process models for AI-use-case-adoption, but neglected the use-cases themselves. In this research, an ontological artifact is developed as the basis for an AI-use-case-description-scheme. It allows practitioners and researchers to systematically describe such use-cases based on their level of abstraction and core characteristics. It enables them to classify, document and communicate these use-cases to support AI-adoption. We ground its development in diffusion of innovation theory and build upon research on AI-adoption. In particular, Rogers’ (2003) innovation decision process is utilised as a framework that explains adoption decisions by organisations. A Design Science Research approach is chosen that integrates the ontology development process by Noy and McGuinness (2001). In this research-in-progress, we conduct one ex ante and one ex post evaluation and plan for a second ex post evaluation that ensure the relevance and rigor of the artifact design.
Close
https://www.researchgate.net/publication/361667326_Use-case-based_innovation_for[...]
Close
Eichelberger, Holger; Ahmadian, Amir Shayan; Dewes, Andreas; Ehl, Marco; Ahmadian, Miguel Gómez Casado; Dewes, Andreas; Ehl, Marco; Staciwa, Monika; Casado, Miguel Gómez
IIP-Ecosphere Platform Handbook v0.3.0 Whitepaper
In: 2022.
Links | BibTeX | Schlagwörter: Architecture, IIP-Ecosphere, IIP-Ecosphere Manual, Manual, UML, Virtual Platform
@whitepaper{Eichelbergerd,
title = {IIP-Ecosphere Platform Handbook v0.3.0},
author = {Holger Eichelberger and Amir Shayan Ahmadian and Andreas Dewes and Marco Ehl and Miguel Gómez Casado Ahmadian and Andreas Dewes and Marco Ehl and Monika Staciwa and Miguel Gómez Casado},
url = {https://www.iip-ecosphere.de/wp-content/uploads/2022/06/PlatformHandbook-final-V0.3.pdf},
doi = {https://doi.org/10.5281/zenodo.6620882},
year = {2022},
date = {2022-06-04},
urldate = {2022-06-04},
keywords = {Architecture, IIP-Ecosphere, IIP-Ecosphere Manual, Manual, UML, Virtual Platform},
pubstate = {published},
tppubtype = {whitepaper}
}

Close
https://www.iip-ecosphere.de/wp-content/uploads/2022/06/PlatformHandbook-final-V[...]
doi:https://doi.org/10.5281/zenodo.6620882
Close
Wilkens, Rainer
Vortrag „Wandlungsfähiges Produktionssystem – Gedanken und Ansätze“ Presentation
02.06.2022.
BibTeX | Schlagwörter: Artificial Intelligence, IIP-Ecosphere, Künstliche Intelligenz, Sennheiser
@misc{nokey,
title = {Vortrag „Wandlungsfähiges Produktionssystem – Gedanken und Ansätze“},
author = {Rainer Wilkens},
year = {2022},
date = {2022-06-02},
booktitle = {REFA Talk „Wandlungsfähige Produktion in einer turbulenten Zeit“ },
publisher = {REFA Institut},
keywords = {Artificial Intelligence, IIP-Ecosphere, Künstliche Intelligenz, Sennheiser},
pubstate = {published},
tppubtype = {presentation}
}

Close
Holger Eichelberger, and Heiko Stichweh; Sauer, Christian
Requirements for an AI-enabled Industry 4.0 Platform – Integrating Industrial and Scientific Views Conference
2022, ISBN: 978-1-61208-946-1 / 2519-8394.
Abstract | Links | BibTeX | Schlagwörter: adaptation, AI, asset administration shell, Edge, Industry 4.0 platforms, intelligent production, Requirements
@conference{Eichelberger2022,
title = {Requirements for an AI-enabled Industry 4.0 Platform – Integrating Industrial and Scientific Views},
author = {Holger Eichelberger,and Heiko Stichweh and Christian Sauer },
editor = {Luigi Lavazza, University of Insubria - Varese, Italy},
url = {https://www.thinkmind.org/index.php?view=article&articleid=softeng_2022_1_20_90004},
isbn = {978-1-61208-946-1 / 2519-8394},
year = {2022},
date = {2022-04-24},
journal = {SOFTENG 2022 The Eighth International Conference on Advances and Trends in Software Engineering Engineering The Eighth International Conference on Advances and Trends in Software Engineering},
pages = {7-14},
abstract = {Intelligent manufacturing is one goal of smart industry/ Industry 4.0 that could be achieved through Artificial Intelligence (AI). Flexibly combining AI methods and platform capabilities, such as dynamic offloading of code close to production machines, security or interoperability mechanisms are major demands in this context. However, recent Industry 4.0 software platforms fall short in various of these demands, in particular in upcoming ecosystem scenarios, e.g., when data or services shall be shared across platforms or companies without vendor lock-ins. The aim of the funded Intelligent Industrial Production (IIP) IIP-Ecosphere project is to research concepts and solutions for ‘easy-to-use’ AI in Industry 4.0 and to demonstrate the results in a prototypical software platform. Core questions are which demands shall drive the development of such a platform and how a feasible set of requirements can be determined that balances scientific and industrial interests. In this paper, we discuss our approach on eliciting requirements in this context for two interlinked requirements perspectives, a usage and a functional view. In summary, we collected 67 usage view activities / scenarios and 141 top-level requirements with 179 detailing sub-requirements. About 35% of the requirements have so far been realized in a prototype and some of the identified concepts are currently being taken up by a standardization initiative for edge devices in Industry 4.0.},
keywords = {adaptation, AI, asset administration shell, Edge, Industry 4.0 platforms, intelligent production, Requirements},
pubstate = {published},
tppubtype = {conference}
}

Close
Intelligent manufacturing is one goal of smart industry/ Industry 4.0 that could be achieved through Artificial Intelligence (AI). Flexibly combining AI methods and platform capabilities, such as dynamic offloading of code close to production machines, security or interoperability mechanisms are major demands in this context. However, recent Industry 4.0 software platforms fall short in various of these demands, in particular in upcoming ecosystem scenarios, e.g., when data or services shall be shared across platforms or companies without vendor lock-ins. The aim of the funded Intelligent Industrial Production (IIP) IIP-Ecosphere project is to research concepts and solutions for ‘easy-to-use’ AI in Industry 4.0 and to demonstrate the results in a prototypical software platform. Core questions are which demands shall drive the development of such a platform and how a feasible set of requirements can be determined that balances scientific and industrial interests. In this paper, we discuss our approach on eliciting requirements in this context for two interlinked requirements perspectives, a usage and a functional view. In summary, we collected 67 usage view activities / scenarios and 141 top-level requirements with 179 detailing sub-requirements. About 35% of the requirements have so far been realized in a prototype and some of the identified concepts are currently being taken up by a standardization initiative for edge devices in Industry 4.0.
Close
https://www.thinkmind.org/index.php?view=article&articleid=softeng_2022_1_20[...]
Close
Denkena, Berend; Bergmann, Benjamin; Becker, Jonas; Blech, Heiko
Sensorlose Überwachung der Einzelteilfertigung Journal Article
In: Wt Werkstattstechnik online, Jahrgang 111 (2021) (Heft 05), pp. 305-308, 2022, ISSN: 1436-4980.
Abstract | Links | BibTeX | Schlagwörter: Einzelteilfertigung, Maschinelles Lernen, Überwachung
@article{Denkena2022,
title = {Sensorlose Überwachung der Einzelteilfertigung},
author = {Berend Denkena and Benjamin Bergmann and Jonas Becker and Heiko Blech},
url = {https://elibrary.vdi-verlag.de/10.37544/1436-4980-2021-05-39/sensorlose-ueberwachung-der-einzelteilfertigung-spindle-current-based-process-monitoring-using-artificial-intelligence-jahrgang-111-2021-heft-05},
doi = {10.37544/1436-4980-2021-05-39},
issn = {1436-4980},
year = {2022},
date = {2022-03-31},
journal = {Wt Werkstattstechnik online},
volume = {Jahrgang 111 (2021)},
number = {Heft 05},
pages = {305-308},
abstract = {Durch die Messung von Spindelströmen lassen sich Informationen aus spanenden Fertigungsprozessen ohne zusätzliche Sensorik erfassen.},
keywords = {Einzelteilfertigung, Maschinelles Lernen, Überwachung},
pubstate = {published},
tppubtype = {article}
}

Close
Durch die Messung von Spindelströmen lassen sich Informationen aus spanenden Fertigungsprozessen ohne zusätzliche Sensorik erfassen.
Close
https://elibrary.vdi-verlag.de/10.37544/1436-4980-2021-05-39/sensorlose-ueberwac[...]
doi:10.37544/1436-4980-2021-05-39
Close
2021
Arnu, David; Klinkenberg, Ralf
Industrial Data Science Platform and Applications in Electronics and Manufacturing Industries Presentation
14.12.2021.
BibTeX | Schlagwörter: Artificial Intelligence, IIP-Ecosphere, Industrie 4.0, RapidMiner, Sennheiser
@misc{nokey,
title = {Industrial Data Science Platform and Applications in Electronics and Manufacturing Industries},
author = {David Arnu and Ralf Klinkenberg},
year = {2021},
date = {2021-12-14},
booktitle = {AI in Manufacturing },
publisher = {Finnish-German Collaboration Initiatives},
keywords = {Artificial Intelligence, IIP-Ecosphere, Industrie 4.0, RapidMiner, Sennheiser},
pubstate = {published},
tppubtype = {presentation}
}

Close
Denkena, Berend; Bergmann, Benjamin; H., Tobias
Transfer of Process References between Machine Tools for Online Tool Condition Monitoring Journal Article
In: Machines, 9 (11), 2021.
Abstract | Links | BibTeX | Schlagwörter: Knowledge Transfer, Machine Tools; Turning; Process Monitoring
@article{Denkena2021c,
title = {Transfer of Process References between Machine Tools for Online Tool Condition Monitoring},
author = {Berend Denkena and Benjamin Bergmann and Tobias H.},
url = {https://www.mdpi.com/2075-1702/9/11/282},
doi = {https://doi.org/10.3390/machines9110282},
year = {2021},
date = {2021-11-10},
journal = {Machines},
volume = {9},
number = {11},
abstract = {Process and tool condition monitoring systems are a prerequisite for autonomous production. One approach to monitoring individual parts without complex cutting simulations is the transfer of knowledge among similar monitoring scenarios. This paper introduces a novel monitoring method which transfers monitoring limits for process signals between different machine tools.},
keywords = {Knowledge Transfer, Machine Tools; Turning; Process Monitoring},
pubstate = {published},
tppubtype = {article}
}

Close
Process and tool condition monitoring systems are a prerequisite for autonomous production. One approach to monitoring individual parts without complex cutting simulations is the transfer of knowledge among similar monitoring scenarios. This paper introduces a novel monitoring method which transfers monitoring limits for process signals between different machine tools.
Close
https://www.mdpi.com/2075-1702/9/11/282
doi:https://doi.org/10.3390/machines9110282
Close
Niederée, Claudia; Eichelberger, Holger; Schmees, Hans-Dieter; Broos, Alexander; Schreiber, Per
KI in der Produktion – Quo vadis? Whitepaper
In: 2021.
Links | BibTeX | Schlagwörter: IIoT, IIP-Ecosphere, Industrie 4.0, KI in der Produktion, Produktion, Umfrage
@whitepaper{Niederée2021,
title = {KI in der Produktion – Quo vadis?},
author = {Claudia Niederée and Holger Eichelberger and Hans-Dieter Schmees and Alexander Broos and Per Schreiber},
url = {https://www.iip-ecosphere.de/wp-content/uploads/2021/11/IIP-Ecosphere-Whitepaper-zur-Umfrage-KI-in-der-Produktion.pdf},
year = {2021},
date = {2021-11-03},
keywords = {IIoT, IIP-Ecosphere, Industrie 4.0, KI in der Produktion, Produktion, Umfrage},
pubstate = {published},
tppubtype = {whitepaper}
}

Close
https://www.iip-ecosphere.de/wp-content/uploads/2021/11/IIP-Ecosphere-Whitepaper[...]
Close
Niederée, Claudia; Eichelberger, Holger; Schmees, Hans-Dieter; Broos, Alexander; Schreiber, Per
Management Summary zu Whitepaper "KI in der Produktion – Quo vadis?" Miscellaneous
2021.
Links | BibTeX | Schlagwörter: IIoT, IIP-Ecosphere, Industrie 4.0, KI in der Produktion
@misc{nokey,
title = {Management Summary zu Whitepaper "KI in der Produktion – Quo vadis?"},
author = {Claudia Niederée and Holger Eichelberger and Hans-Dieter Schmees and Alexander Broos and Per Schreiber},
url = {https://www.iip-ecosphere.de/wp-content/uploads/2021/10/Management-Summary_IIP-Ecosphere-Umfrage_KI-Produktion.pdf},
year = {2021},
date = {2021-11-03},
keywords = {IIoT, IIP-Ecosphere, Industrie 4.0, KI in der Produktion},
pubstate = {published},
tppubtype = {misc}
}

Close
https://www.iip-ecosphere.de/wp-content/uploads/2021/10/Management-Summary_IIP-E[...]
Close
Hildebrandt, Petra
Vortrag „KI allein genügt nicht!“ Presentation
24.09.2021.
BibTeX | Schlagwörter: Artificial Intelligence, IIP-Ecosphere, Künstliche Intelligenz, Sennheiser
@misc{nokey,
title = {Vortrag „KI allein genügt nicht!“ },
author = {Petra Hildebrandt},
year = {2021},
date = {2021-09-24},
booktitle = {Künstliche Intelligenz in der industriellen Produktion },
publisher = {Deutsche Messe AG},
keywords = {Artificial Intelligence, IIP-Ecosphere, Künstliche Intelligenz, Sennheiser},
pubstate = {published},
tppubtype = {presentation}
}

Close
Denkena, Berend; Bergmann, Benjamin; Becker, Jonas; Stiehl, Tobias
Time Series Search and Similarity Identification Journal Article
In: Production at the Leading Edge of Technology, 2022 , pp. 479-487, 2021, ISBN: 978-3-030-78424-9.
Abstract | Links | BibTeX | Schlagwörter: Barycenter Averaging, Time Series Clustering
@article{Denkena2021d,
title = {Time Series Search and Similarity Identification},
author = {Berend Denkena and Benjamin Bergmann and Jonas Becker and Tobias Stiehl},
url = {https://link.springer.com/chapter/10.1007/978-3-030-78424-9_53},
doi = {10.1007/978-3-030-78424-9_53},
isbn = {978-3-030-78424-9},
year = {2021},
date = {2021-09-05},
journal = {Production at the Leading Edge of Technology},
volume = {2022},
pages = {479-487},
abstract = {Monitoring process errors and tool condition in single item production is challenging, as a teach-in is not possible due to a missing reference process. An approach to this problem is anomaly detection, e.g. based on motor currents or axis position signals from metal cutting processes. However, with no references anomaly detection struggles to detect failures from signals, because failure patterns are often too similar to regular process dynamics. While single items inherently constitute an anomaly by themselves, they do contain repetitive elements, like boreholes or milled pockets. These elements are similar, what provides an anomaly detection with additional information on regular processes.

Hierarchical K-Means clustering combined with Dynamic Time Warping (DTW) and Barycenter Averaging (DBA) enables the identification of similar process elements. The algorithm allows ordering similar process segments by similarity in a tree structure. The introduced method supports querying subsequences from any given cutting process, for which it returns the closest cluster in the tree. This allows to (a) improve the data basis for anomaly detection and (b) to transfer labels with errors between processes. The article demonstrates the transfer of labels (for errors) from a turning process, to a single item milling process.},
keywords = {Barycenter Averaging, Time Series Clustering},
pubstate = {published},
tppubtype = {article}
}

Close
Monitoring process errors and tool condition in single item production is challenging, as a teach-in is not possible due to a missing reference process. An approach to this problem is anomaly detection, e.g. based on motor currents or axis position signals from metal cutting processes. However, with no references anomaly detection struggles to detect failures from signals, because failure patterns are often too similar to regular process dynamics. While single items inherently constitute an anomaly by themselves, they do contain repetitive elements, like boreholes or milled pockets. These elements are similar, what provides an anomaly detection with additional information on regular processes.

Hierarchical K-Means clustering combined with Dynamic Time Warping (DTW) and Barycenter Averaging (DBA) enables the identification of similar process elements. The algorithm allows ordering similar process segments by similarity in a tree structure. The introduced method supports querying subsequences from any given cutting process, for which it returns the closest cluster in the tree. This allows to (a) improve the data basis for anomaly detection and (b) to transfer labels with errors between processes. The article demonstrates the transfer of labels (for errors) from a turning process, to a single item milling process.
Close
https://link.springer.com/chapter/10.1007/978-3-030-78424-9_53
doi:10.1007/978-3-030-78424-9_53
Close
Denkena, Berend; Dittrich, Marc-André; Fohlmeister, Silas; Kemp, Daniel; Palmer, Gregory
Scalable cooperative multi-agentreinforcement-learning for order-controlled on schedule manufacturing in flexible manufacturing systems Conference
2021.
Abstract | Links | BibTeX | Schlagwörter:
@conference{Denkena2021b,
title = {Scalable cooperative multi-agentreinforcement-learning for order-controlled on schedule manufacturing in flexible manufacturing systems},
author = {Berend Denkena and Marc-André Dittrich and Silas Fohlmeister and Daniel Kemp and Gregory Palmer },
url = {http://www.asim-fachtagung-spl.de/asim2021/papers/Proof_108.pdf},
year = {2021},
date = {2021-08-26},
abstract = { To operate flexible manufacturing systems efficiently, a robust and optimal
production control is crucial. With an increasing number of workpieces being
processed in parallel, ensuring guaranteed lead times represents a complex
optimization tasks, better known as the flexible scheduling problem. Cooperative
multi-agent reinforcement learning approaches have recently shown their potential in
production control. However, ensuring guaranteed lead times in flexible
manufacturing systems with these approaches remains an open problem. In this work,
an existing cooperative multi-agent framework for flexible job-shop scheduling is
transferred and modified to optimize production control in flexible manufacturing
systems. Using a centralized training for decentralized execution multi-agent deep
reinforcement learning approach, the goal is to optimize order agents to ensure
guaranteed lead times. Furthermore, a comprehensive simulation study investigates
the effect of common knowledge on facilitating cooperation, and empirically evaluate
the frameworks scalability to a range of challenging scenarios. },
keywords = {},
pubstate = {published},
tppubtype = {conference}
}

Close
To operate flexible manufacturing systems efficiently, a robust and optimal
production control is crucial. With an increasing number of workpieces being
processed in parallel, ensuring guaranteed lead times represents a complex
optimization tasks, better known as the flexible scheduling problem. Cooperative
multi-agent reinforcement learning approaches have recently shown their potential in
production control. However, ensuring guaranteed lead times in flexible
manufacturing systems with these approaches remains an open problem. In this work,
an existing cooperative multi-agent framework for flexible job-shop scheduling is
transferred and modified to optimize production control in flexible manufacturing
systems. Using a centralized training for decentralized execution multi-agent deep
reinforcement learning approach, the goal is to optimize order agents to ensure
guaranteed lead times. Furthermore, a comprehensive simulation study investigates
the effect of common knowledge on facilitating cooperation, and empirically evaluate
the frameworks scalability to a range of challenging scenarios.
Close
http://www.asim-fachtagung-spl.de/asim2021/papers/Proof_108.pdf
Close
Eichelberger, Holger; Ahmadian, Amir Shayan; Dewes, Andreas; Ehl, Marco; Staciwa, Monika; Ahmadian, Miguel Gómez Casado; Dewes, Andreas; Ehl, Marco; Staciwa, Monika; Casado, Miguel Gómez
IIP-Ecosphere Platform Handbook v0.20 Whitepaper
In: 2021.
Links | BibTeX | Schlagwörter: Architecture, IIP-Ecosphere, Manual, Rationales, UML, Virtual Platform
@whitepaper{Eichelberger,
title = {IIP-Ecosphere Platform Handbook v0.20},
author = {Holger Eichelberger and Amir Shayan Ahmadian and Andreas Dewes and Marco Ehl and Monika Staciwa and Miguel Gómez Casado Ahmadian and Andreas Dewes and Marco Ehl and Monika Staciwa and Miguel Gómez Casado},
url = {https://www.iip-ecosphere.eu/wp-content/uploads/2021/08/PlatformHandbook-final-V0.2.pdf},
doi = {10.5281/zenodo.5168946},
year = {2021},
date = {2021-08-21},
urldate = {2021-08-21},
keywords = {Architecture, IIP-Ecosphere, Manual, Rationales, UML, Virtual Platform},
pubstate = {published},
tppubtype = {whitepaper}
}

Close
https://www.iip-ecosphere.eu/wp-content/uploads/2021/08/PlatformHandbook-final-V[...]
doi:10.5281/zenodo.5168946
Close
Casado, Miguel Gomez; Eichelberger, Holger
Industry 4.0 Resource Monitoring - Experiences With Micrometer and Asset Administration Shells Inproceedings
In: CEUR-WS Proceedings of Symposium on Software Performance 2021 (SSP'21), CEUR-WS.org, 2021.
Links | BibTeX | Schlagwörter: Asset Administration Shells, IIP-Ecosphere, Industrie 4.0
@inproceedings{41924,
title = {Industry 4.0 Resource Monitoring - Experiences With Micrometer and Asset Administration Shells},
author = {Miguel Gomez Casado and Holger Eichelberger},
url = {http://ceur-ws.org/Vol-3043/short8.pdf},
year = {2021},
date = {2021-08-13},
urldate = {2021-08-13},
booktitle = {CEUR-WS Proceedings of Symposium on Software Performance 2021 (SSP'21)},
publisher = {CEUR-WS.org},
keywords = {Asset Administration Shells, IIP-Ecosphere, Industrie 4.0},
pubstate = {published},
tppubtype = {inproceedings}
}

Close
http://ceur-ws.org/Vol-3043/short8.pdf
Close
Bonhage, Malte; Wilkens, Rainer; Denkena, Berend; Kemp, Daniel
Der digitale Zwilling als Basis für ein intelligentes und skalierbares Produktionssystem Magazine
2021.
Abstract | Links | BibTeX | Schlagwörter: Asset-Administration-Shell, Digital Twin, Verwaltungsschale
@magazine{Bonhage2021,
title = {Der digitale Zwilling als Basis für ein intelligentes und skalierbares Produktionssystem},
author = {Malte Bonhage and Rainer Wilkens and Berend Denkena and Daniel Kemp},
url = {https://cdn.tedo.be/tedo-ecms/4/SPS-MAGAZIN_6_(Juli)_2021.pdf},
year = {2021},
date = {2021-07-12},
journal = {SPS Magazin},
number = {6},
pages = {61-63},
abstract = {Der digitale Zwilling ist der interdisziplinäre Kern zahlreicher I4.0-Anwendungen. Implementierungsansätze sind allerdings oft noch individuell und kostenintensiv. Abhilfe verspricht an dieser Stelle die Verwaltungsschale, als standardisierter digitaler Zwilling. Im Rahmen von IIP-E entsteht eine Implementierung der Verwaltungsschale bei Sennheiser electronic},
keywords = {Asset-Administration-Shell, Digital Twin, Verwaltungsschale},
pubstate = {published},
tppubtype = {magazine}
}

Close
Der digitale Zwilling ist der interdisziplinäre Kern zahlreicher I4.0-Anwendungen. Implementierungsansätze sind allerdings oft noch individuell und kostenintensiv. Abhilfe verspricht an dieser Stelle die Verwaltungsschale, als standardisierter digitaler Zwilling. Im Rahmen von IIP-E entsteht eine Implementierung der Verwaltungsschale bei Sennheiser electronic
Close
https://cdn.tedo.be/tedo-ecms/4/SPS-MAGAZIN_6_(Juli)_2021.pdf
Close
Jalowski, Max; Roth, Angela; Oks, Sascha J.; Wilga, Matthäus
Innovation KI-basierter Dienstleistungen für die industrielle Wertschöpfung – Ein artefaktzentrierter Ansatz Book Chapter
In: Bruhn, Martin; Hadwich, Karsten (Ed.): Künstliche Intelligenz im Dienstleistungsmanagement. Forum Dienstleistungsmanagement., pp. 158-183, 2021, ISBN: 978-3-658-34324-8.
Abstract | Links | BibTeX | Schlagwörter: Dienstleistungsmanagement, Künstliche Intelligenz
@inbook{nokey,
title = {Innovation KI-basierter Dienstleistungen für die industrielle Wertschöpfung – Ein artefaktzentrierter Ansatz},
author = {Max Jalowski and Angela Roth and Sascha J. Oks and Matthäus Wilga},
editor = {Martin Bruhn and Karsten Hadwich},
url = {https://link.springer.com/chapter/10.1007/978-3-658-34324-8_7},
doi = {https://doi.org/10.1007/978-3-658-34324-8_7},
isbn = {978-3-658-34324-8},
year = {2021},
date = {2021-06-29},
urldate = {2021-06-29},
booktitle = {Künstliche Intelligenz im Dienstleistungsmanagement. Forum Dienstleistungsmanagement.},
pages = {158-183},
abstract = {Dieser Beitrag zeigt auf, wie insbesondere für kleine und mittlere Unternehmen die Gestaltung und Innovation von KI-basierten Dienstleistungen in der industriellen Wertschöpfung unterstützt werden kann. Dazu werden Herausforderungen bei der Entwicklung KI-basierter Dienstleistungen erhoben und anschließend Artefakte präsentiert, die zur Innovation von Dienstleistungen zum Einsatz kommen. Diese werden basierend auf den zuvor identifizierten Herausforderungen adaptiert, um die Innovation von KI-basierten Dienstleistungen zu ermöglichen.},
keywords = {Dienstleistungsmanagement, Künstliche Intelligenz},
pubstate = {published},
tppubtype = {inbook}
}

Close
Dieser Beitrag zeigt auf, wie insbesondere für kleine und mittlere Unternehmen die Gestaltung und Innovation von KI-basierten Dienstleistungen in der industriellen Wertschöpfung unterstützt werden kann. Dazu werden Herausforderungen bei der Entwicklung KI-basierter Dienstleistungen erhoben und anschließend Artefakte präsentiert, die zur Innovation von Dienstleistungen zum Einsatz kommen. Diese werden basierend auf den zuvor identifizierten Herausforderungen adaptiert, um die Innovation von KI-basierten Dienstleistungen zu ermöglichen.
Close
https://link.springer.com/chapter/10.1007/978-3-658-34324-8_7
doi:https://doi.org/10.1007/978-3-658-34324-8_7
Close
Hildebrandt, Petra
Vortrag “Vollautomatischer Funktionstest bestückter Leiterplatten – Kann KI bei der Fehlerdiagnose helfen?“ Presentation
28.06.2021.
BibTeX | Schlagwörter: Artificial Intelligence, IIP-Ecosphere, Künstliche Intelligenz, Sennheiser
@misc{nokey,
title = {Vortrag “Vollautomatischer Funktionstest bestückter Leiterplatten – Kann KI bei der Fehlerdiagnose helfen?“ },
author = {Petra Hildebrandt},
year = {2021},
date = {2021-06-28},
publisher = {KI & Data Analytics in Manufacturing },
keywords = {Artificial Intelligence, IIP-Ecosphere, Künstliche Intelligenz, Sennheiser},
pubstate = {published},
tppubtype = {presentation}
}

Close
Wilga, Matthäus; Jalowski, Max; Kirschbaum, Julius; Roth, Angela
A Systematic Characterization of Artificial Intelligence Business Models as a Fundament for Business Model Innovation and Strategic Decision-Making Conference
21st European Academy of Management (EURAM) Conference 2021, 2021.
Links | BibTeX | Schlagwörter: Artificial Intelligence, Business Model
@conference{nokey,
title = {A Systematic Characterization of Artificial Intelligence Business Models as a Fundament for Business Model Innovation and Strategic Decision-Making},
author = {Matthäus Wilga and Max Jalowski and Julius Kirschbaum and Angela Roth},
url = {https://cris.fau.de/converis/portal/publication/260644612?lang=en_GB},
year = {2021},
date = {2021-06-17},
urldate = {2021-06-17},
booktitle = {21st European Academy of Management (EURAM) Conference 2021},
keywords = {Artificial Intelligence, Business Model},
pubstate = {published},
tppubtype = {conference}
}

Close
https://cris.fau.de/converis/portal/publication/260644612?lang=en_GB
Close
Kemp, Daniel; Bonhage, Malte; Wilkens, Rainer
Der digitale Zwilling als Basis für ein intelligentes und skalierbares Produktionssystem Journal Article
In: SPS Magazin, 6 (June) 2021 , 2021.
Links | BibTeX | Schlagwörter: Artificial Intelligence, Digital Twin, IIP-Ecosphere, Sennheiser
@article{nokey,
title = {Der digitale Zwilling als Basis für ein intelligentes und skalierbares Produktionssystem},
author = {Daniel Kemp and Malte Bonhage and Rainer Wilkens},
url = {https://www.sps-magazin.de/komponenten-fuer-die-automatisierung/es-wird-konkret/},
year = {2021},
date = {2021-06-15},
urldate = {2021-06-15},
journal = {SPS Magazin},
volume = {6 (June) 2021},
keywords = {Artificial Intelligence, Digital Twin, IIP-Ecosphere, Sennheiser},
pubstate = {published},
tppubtype = {article}
}

Close
https://www.sps-magazin.de/komponenten-fuer-die-automatisierung/es-wird-konkret/
Close
Wilkens, Rainer
[No title] Presentation
10.06.2021.
Links | BibTeX | Schlagwörter:
@misc{nokey,
title = {[No title]},
author = {Rainer Wilkens},
url = {https://event.phoenixcontact.com/icc2021},
year = {2021},
date = {2021-06-10},
publisher = {Industrial Communication Congress},
keywords = {},
pubstate = {published},
tppubtype = {presentation}
}

Close
https://event.phoenixcontact.com/icc2021
Close
Graf, Walter; Wilmsmeier, Sören
Quantum Technology in Flexible Job Shop Scheduling? – A Field Report Using Digital Annealer Conference
2021.
Abstract | Links | BibTeX | Schlagwörter: Digital annealer, Flexible job shop scheduling, Quantum Algorithms
@conference{Graf2021,
title = {Quantum Technology in Flexible Job Shop Scheduling? – A Field Report Using Digital Annealer},
author = {Walter Graf and Sören Wilmsmeier},
url = {https://youtu.be/doylgwUXy-I
https://www.iip-ecosphere.eu/wp-content/uploads/2021/06/Wil21_Quantum-Technology-in-Flexible-Job-Shop-Scheduling.pdf},
year = {2021},
date = {2021-05-26},
journal = {Quantum Summit 2021 (26. - 27.05.2021)},
abstract = {One area of current research is focusing on the further development of quantum computers in order to achieve the so called "Quantum Advantage". Another activity stream is concerned with the algorithms or solution scenarios required for this. A major challenge here is that the algorithms that can run on today's quantum computers are of a very rudimentary nature and thus have not yet a real solution character, or they are too complex for existing systems and therefore provide only limited insights. In the meantime, Fujitsu has created a bridging technology in the form of the Digital Annealer, which makes it possible even today to process problems of relevant size that would normally be expected to run on a quantum computer, or more precisely here, on a quantum annealer. Together with the Institute of Production Engineering and Machine Tools of the Leibniz University of Hannover, a job shop scheduling scenario was implemented based on a real production environment for a tape dispenser. In this environment, the total production time could be reduced from about 300 hours to 200 hours in less than 1 minute of computing time on the Digital Annealer.},
keywords = {Digital annealer, Flexible job shop scheduling, Quantum Algorithms},
pubstate = {published},
tppubtype = {conference}
}

Close
One area of current research is focusing on the further development of quantum computers in order to achieve the so called "Quantum Advantage". Another activity stream is concerned with the algorithms or solution scenarios required for this. A major challenge here is that the algorithms that can run on today's quantum computers are of a very rudimentary nature and thus have not yet a real solution character, or they are too complex for existing systems and therefore provide only limited insights. In the meantime, Fujitsu has created a bridging technology in the form of the Digital Annealer, which makes it possible even today to process problems of relevant size that would normally be expected to run on a quantum computer, or more precisely here, on a quantum annealer. Together with the Institute of Production Engineering and Machine Tools of the Leibniz University of Hannover, a job shop scheduling scenario was implemented based on a real production environment for a tape dispenser. In this environment, the total production time could be reduced from about 300 hours to 200 hours in less than 1 minute of computing time on the Digital Annealer.
Close
https://youtu.be/doylgwUXy-I
https://www.iip-ecosphere.eu/wp-content/uploads/2021/06/Wil21_Quantum-Technology[...]
Close
Denkena, Berend; Fritz Schinkel, Jonathan Pirnay; Wilmsmeier, Sören
Quantum Algorithms for Process Parallel Flexible Job Shop Scheduling Journal Article
In: CIRP Journal of Manufacturing Science and Technology, 33 , pp. 100-114, 2021.
Abstract | Links | BibTeX | Schlagwörter: Digital annealer, Flexible job shop scheduling, Process parallel optimization, Production planning and control
@article{Denkena2021,
title = {Quantum Algorithms for Process Parallel Flexible Job Shop Scheduling},
author = {Berend Denkena and Fritz Schinkel, Jonathan Pirnay and Sören Wilmsmeier},
url = {https://www.sciencedirect.com/science/article/pii/S1755581721000432},
doi = {10.1016/j.cirpj.2021.03.006},
year = {2021},
date = {2021-03-23},
journal = {CIRP Journal of Manufacturing Science and Technology},
volume = {33},
pages = {100-114},
abstract = {Flexible Job Shop Scheduling is one of the most difficult optimization problems known. In addition, modern production planning and control strategies require continuous and process-parallel optimization of machine allocation and processing sequences. Therefore, this paper presents a new method for process parallel Flexible Job Shop Scheduling using the concept of quantum computing based optimization. A scientific benchmark and the application to a realistic use-case demonstrates the good performance and practicability of this new approach. A managerial insight shows how the approach for process parallel flexible job shop scheduling can be integrated in existing production planning and control IT-infrastructure.},
keywords = {Digital annealer, Flexible job shop scheduling, Process parallel optimization, Production planning and control},
pubstate = {published},
tppubtype = {article}
}

Close
Flexible Job Shop Scheduling is one of the most difficult optimization problems known. In addition, modern production planning and control strategies require continuous and process-parallel optimization of machine allocation and processing sequences. Therefore, this paper presents a new method for process parallel Flexible Job Shop Scheduling using the concept of quantum computing based optimization. A scientific benchmark and the application to a realistic use-case demonstrates the good performance and practicability of this new approach. A managerial insight shows how the approach for process parallel flexible job shop scheduling can be integrated in existing production planning and control IT-infrastructure.
Close
https://www.sciencedirect.com/science/article/pii/S1755581721000432
doi:10.1016/j.cirpj.2021.03.006
Close
Eichelberger, Holger; Sauer, Christian; Ahmadian, Amir Shayan; Schicktanz, Michael; Dewes, Andreas; Palmer, Gregory; Niederée, Claudia
IIP-Ecosphere Plattform – Anforderungen (Funktionale und Qualitäts-Sicht) Whitepaper
In: 2021.
Abstract | Links | BibTeX | Schlagwörter: Functional, Quality, Requirements, Virtual Platform
@whitepaper{Eichelberger2021,
title = {IIP-Ecosphere Plattform – Anforderungen (Funktionale und Qualitäts-Sicht)},
author = {Holger Eichelberger and Christian Sauer and Amir Shayan Ahmadian and Michael Schicktanz and Andreas Dewes and Gregory Palmer and Claudia Niederée},
url = {https://www.iip-ecosphere.eu/wp-content/uploads/2021/03/IIP-2021_002.pdf
https://www.iip-ecosphere.eu/wp-content/uploads/2021/03/IIP-2021_002-eng.pdf
},
doi = {10.5281/zenodo.4485774},
year = {2021},
date = {2021-03-15},
abstract = {Dieses Dokument beschreibt die Anforderungen an die IIP-Ecosphere Plattform. Die Anforderungen basieren auf Diskussionen mit den Partnern und den Arbeitspaketen bzw. Teilprojekten (bzw. deren Repräsentanten) von IIP-Ecosphere, wie etwa den Demonstratoren. Als Grundlagen wurden eine im Projekt erstellte Übersicht aktueller Industrie 4.0 Plattformen als Grundlage sowie eine Anforderungserhebung auf Benutzersicht einbezogen. },
keywords = {Functional, Quality, Requirements, Virtual Platform},
pubstate = {published},
tppubtype = {whitepaper}
}

Close
Dieses Dokument beschreibt die Anforderungen an die IIP-Ecosphere Plattform. Die Anforderungen basieren auf Diskussionen mit den Partnern und den Arbeitspaketen bzw. Teilprojekten (bzw. deren Repräsentanten) von IIP-Ecosphere, wie etwa den Demonstratoren. Als Grundlagen wurden eine im Projekt erstellte Übersicht aktueller Industrie 4.0 Plattformen als Grundlage sowie eine Anforderungserhebung auf Benutzersicht einbezogen.
Close
https://www.iip-ecosphere.eu/wp-content/uploads/2021/03/IIP-2021_002.pdf
https://www.iip-ecosphere.eu/wp-content/uploads/2021/03/IIP-2021_002-eng.pdf
doi:10.5281/zenodo.4485774
Close
Stichweh, Heiko; Sauer, Christian; Eichelberger, Holger
IIP-Ecosphere Platform Requirements (Usage View) Whitepaper
In: 2021.
Abstract | Links | BibTeX | Schlagwörter: AI Services, Application Building, Artificial Intelligence, IIoT, IIoT-Platform, IIP-Ecosphere, Platform Activities, Platform Requirements, Usage View
@whitepaper{Stichweh2021,
title = {IIP-Ecosphere Platform Requirements (Usage View)},
author = {Heiko Stichweh and Christian Sauer and Holger Eichelberger},
url = {https://www.iip-ecosphere.eu/wp-content/uploads/2021/03/IIP-2021_001_IIP-Ecosphere_Platform_Requirements_Usage_View.pdf},
doi = {10.5281/zenodo.4485801},
year = {2021},
date = {2021-03-11},
abstract = {This Whitepaper describes a shared view on the IIP-Ecosphere platform, which was developed as a core technical contribution of the IIP-Ecosphere Think Thank “Platforms”, to foster and complement the requirements collection of the platform, based on this shared view on envisioned platform functionality. Following the Industrial Internet Reference Architecture (IIRA), this Whitepaper describes the IIP-Ecosphere platform from the Usage Viewpoint. The Usage View on the IIP-Ecosphere platform that we discuss in this document represents the common view of all partners involved in the design, the subsequent implementation and, finally, the operations of the platform based on the voice of the prospective users of the platform in the IIP-Ecosphere community. The shared Usage View was collected in terms of a series of workshops with all interested project partners. The shared Usage View established in this document therefore provides a basis for deriving/validating the functional and quality requirements of the overall platform and, thus, enables the subsequent work on the development of the concepts and solutions established in the shared Usage View. For this current version of the Usage View, we jointly decided to focus on application building, distribution and AI services, as these topics strongly correlate with the technical foundations of the platform to be developed. For this focus, we describe a System under Consideration with 18 entities, 19 roles and 67 activities in this Whitepaper.},
keywords = {AI Services, Application Building, Artificial Intelligence, IIoT, IIoT-Platform, IIP-Ecosphere, Platform Activities, Platform Requirements, Usage View},
pubstate = {published},
tppubtype = {whitepaper}
}

Close
This Whitepaper describes a shared view on the IIP-Ecosphere platform, which was developed as a core technical contribution of the IIP-Ecosphere Think Thank “Platforms”, to foster and complement the requirements collection of the platform, based on this shared view on envisioned platform functionality. Following the Industrial Internet Reference Architecture (IIRA), this Whitepaper describes the IIP-Ecosphere platform from the Usage Viewpoint. The Usage View on the IIP-Ecosphere platform that we discuss in this document represents the common view of all partners involved in the design, the subsequent implementation and, finally, the operations of the platform based on the voice of the prospective users of the platform in the IIP-Ecosphere community. The shared Usage View was collected in terms of a series of workshops with all interested project partners. The shared Usage View established in this document therefore provides a basis for deriving/validating the functional and quality requirements of the overall platform and, thus, enables the subsequent work on the development of the concepts and solutions established in the shared Usage View. For this current version of the Usage View, we jointly decided to focus on application building, distribution and AI services, as these topics strongly correlate with the technical foundations of the platform to be developed. For this focus, we describe a System under Consideration with 18 entities, 19 roles and 67 activities in this Whitepaper.
Close
https://www.iip-ecosphere.eu/wp-content/uploads/2021/03/IIP-2021_001_IIP-Ecosphe[...]
doi:10.5281/zenodo.4485801
Close
Fabian Bruckner,; Jahnke, Nils
Datenschutz und Datensicherheit in Datenökosystemen Whitepaper
In: 2021.
Abstract | Links | BibTeX | Schlagwörter: Data Ecosystems, Datenökosysteme, Datenschutz, Datensicherheit, IIP-Ecosphere, International Data Spaces, Usage Control
@whitepaper{Bruckner2021,
title = {Datenschutz und Datensicherheit in Datenökosystemen },
author = {Fabian Bruckner, and Nils Jahnke},
url = {https://www.iip-ecosphere.eu/wp-content/uploads/2021/03/IIP-2021-003-Whitepaper-Datenschutz_Datensicherheit.pdf},
doi = {10.5281/zenodo.4588330},
year = {2021},
date = {2021-03-08},
abstract = {Dieses Whitepaper stellt Probleme und Lösungsansätze im Bereich des Schutzes und der Sicherheit von Daten in Datenökosystemen dar. Dabei wird insbesondere die Perspektive von IIP-Ecosphere betrachtet. Im Rahmen des Whitepapers werden durch eine umfassende Literaturanalyse identifizierte unmittelbare und mittelbare Probleme aus den Bereichen Datenschutz und –sicherheit in Datenökosystemen und dazugehörige Lösungsansätze präsentiert. Insbesondere im Fokus der Betrachtungen steht dabei der in IIP-Ecosphere zu gestaltende Datenmarktplatz. Die ermittelten theoretischen Erkenntnisse werden durch die Befragung von Partnern und Assoziierten des Projekts auf ihre Praxisrelevanz geprüft und konsolidiert sowie mögliche Lösungsansätze für Datenschutz und -sicherheit eingeordnet.},
keywords = {Data Ecosystems, Datenökosysteme, Datenschutz, Datensicherheit, IIP-Ecosphere, International Data Spaces, Usage Control},
pubstate = {published},
tppubtype = {whitepaper}
}

Close
Dieses Whitepaper stellt Probleme und Lösungsansätze im Bereich des Schutzes und der Sicherheit von Daten in Datenökosystemen dar. Dabei wird insbesondere die Perspektive von IIP-Ecosphere betrachtet. Im Rahmen des Whitepapers werden durch eine umfassende Literaturanalyse identifizierte unmittelbare und mittelbare Probleme aus den Bereichen Datenschutz und –sicherheit in Datenökosystemen und dazugehörige Lösungsansätze präsentiert. Insbesondere im Fokus der Betrachtungen steht dabei der in IIP-Ecosphere zu gestaltende Datenmarktplatz. Die ermittelten theoretischen Erkenntnisse werden durch die Befragung von Partnern und Assoziierten des Projekts auf ihre Praxisrelevanz geprüft und konsolidiert sowie mögliche Lösungsansätze für Datenschutz und -sicherheit eingeordnet.
Close
https://www.iip-ecosphere.eu/wp-content/uploads/2021/03/IIP-2021-003-Whitepaper-[...]
doi:10.5281/zenodo.4588330
Close
Wilmsmeier, Sören
Taktzeitoptimierung mithilfe von künstlicher Intelligenz Booklet
2021.
Abstract | Links | BibTeX | Schlagwörter: Optimierung, Taktzeit, Ursache-Wirkungs-Analyse
@booklet{Wilmsmeier2021,
title = {Taktzeitoptimierung mithilfe von künstlicher Intelligenz},
author = {Sören Wilmsmeier},
url = {https://www.phi-hannover.de/forschung/artikel/detail/taktzeitoptimierung-mithilfe-von-kuenstlicher-intelligenz/},
year = {2021},
date = {2021-03-04},
journal = {phi – Produktionstechnik Hannover informiert},
abstract = {Die Ursache von Taktzeitschwankungen in verketteten Fertigungslinien zu ermitteln und die Auswirkung von Verzögerungen vorherzusagen ist eine komplexe Aufgabe. Wie sich künstliche Intelligenz dafür nutzen lässt, untersucht das IFW im Verbundprojekt IIP-Ecosphere.},
month = {03},
keywords = {Optimierung, Taktzeit, Ursache-Wirkungs-Analyse},
pubstate = {published},
tppubtype = {booklet}
}

Close
Die Ursache von Taktzeitschwankungen in verketteten Fertigungslinien zu ermitteln und die Auswirkung von Verzögerungen vorherzusagen ist eine komplexe Aufgabe. Wie sich künstliche Intelligenz dafür nutzen lässt, untersucht das IFW im Verbundprojekt IIP-Ecosphere.
Close
https://www.phi-hannover.de/forschung/artikel/detail/taktzeitoptimierung-mithilf[...]
Close
Sauer, Christian; Eichelberger, Holger; Ahmadian, Amir Shayan; Dewes, Andreas; Jürjens, Jan
Aktuelle Industrie 4.0 Plattformen – Eine Übersicht Whitepaper
In: (DE: IIP-2020/001, EN: IIP-2020/001-en), 2021.
Abstract | Links | BibTeX | Schlagwörter: Artificial Intelligence, Customizability, Ecosystem, Edge, Industry 4.0, platforms, Protocols
@whitepaper{Sauer2020,
title = {Aktuelle Industrie 4.0 Plattformen – Eine Übersicht},
author = {Christian Sauer and Holger Eichelberger and Amir Shayan Ahmadian and Andreas Dewes and Jan Jürjens},
url = {https://www.iip-ecosphere.eu/wp-content/uploads/2021/02/IIP-2020_001.pdf
https://www.iip-ecosphere.eu/wp-content/uploads/2021/02/IIP-2020_001-en.pdf
https://zenodo.org/record/4485756

},
doi = {10.5281/zenodo.4485756 },
year = {2021},
date = {2021-02-15},
number = {DE: IIP-2020/001, EN: IIP-2020/001-en},
abstract = {Dieses Whitepaper gibt eine Übersicht über aktuelle Industrie 4.0 Plattformen, insbesondere aus dem Blickwinkel des IIP-Ecosphere-Projekts, das im KI-Innovationswettbewerb vom Bundesministerium für Wirtschaft und Energie (BMWi) gefördert wird. Dabei stehen Themen wir Interkonnektivität, digitale Zwillinge, Offenheit, Sicherheit und die Nutzung von Künstlicher Intelligenz im Kontext der intelligenten Produktion im Mittelpunkt. Das Dokument beschreibt sowohl die Vorgehensweise der Datenermittlung, die Detailergebnisse für einzelne industrielle Plattformen als auch eine zusammenfassende Übersicht. Es werden insgesamt 21 industrielle Plattformen basierend auf öffentlich verfügbaren Dokumenten anhand von 16 Themenfeldern analysiert. Sowohl Plattformen als auch Analysethemen entstammen intensiver Diskussionen der Projektpartner in IIP-Ecosphere.
Die untersuchten Plattformen decken insbesondere die benötigten Grundfunktionen ab. Beispielsweise wird oft eine Vielzahl an Kommunikationsprotokollen bereitgestellt und verschiedenste Cloud-Dienste integriert. Selbst neuere Trends wie Künstliche Intelligenz sind inzwischen in den Plattformbeschreibungen zu finden. Allerdings ist der Funktionsumfang zwischen den Plattformen auch sehr unterschiedlich. Neuere Standards wie OPC-UA, UMATI oder die Industrie 4.0 Verwaltungsschale werden oft nur zurückhaltend, wenn überhaupt eingesetzt, was teilweise der Entwicklungshistorie aber auch strategischen Erwägungen geschuldet sein mag.
Basierend auf der Plattform-übergreifenden Analyse der 16 Themenfelder leiten wir Herausforderungen für zukünftige Plattformen und insbesondere für unsere Arbeit in IIP-Ecosphere ab. Diese umfassen Themen wie offene Ökosysteme, erweiterbare Architekturen mit standardisierten Schnittstellenbeschreibungen, flexible und dynamische Unterstützung für KI-Verfahren, sicherer und vereinheitlichter Datenaustausch (für Data Sharing, Ressource Sharing und Data Usage Control) wie auch durchgängige und konsistente Konfigurierbarkeit, die das Vertrauen des Nutzers in die jeweilige Plattform stärkt. Eine Standardisierung von (einigen) dieser Themen wäre wünschenswert um den Austausch und die Interoperabilität zwischen Plattformen und Plattformökosystemen zu verbessern und Lock-ins zu vermeiden.},
keywords = {Artificial Intelligence, Customizability, Ecosystem, Edge, Industry 4.0, platforms, Protocols},
pubstate = {published},
tppubtype = {whitepaper}
}

Close
Dieses Whitepaper gibt eine Übersicht über aktuelle Industrie 4.0 Plattformen, insbesondere aus dem Blickwinkel des IIP-Ecosphere-Projekts, das im KI-Innovationswettbewerb vom Bundesministerium für Wirtschaft und Energie (BMWi) gefördert wird. Dabei stehen Themen wir Interkonnektivität, digitale Zwillinge, Offenheit, Sicherheit und die Nutzung von Künstlicher Intelligenz im Kontext der intelligenten Produktion im Mittelpunkt. Das Dokument beschreibt sowohl die Vorgehensweise der Datenermittlung, die Detailergebnisse für einzelne industrielle Plattformen als auch eine zusammenfassende Übersicht. Es werden insgesamt 21 industrielle Plattformen basierend auf öffentlich verfügbaren Dokumenten anhand von 16 Themenfeldern analysiert. Sowohl Plattformen als auch Analysethemen entstammen intensiver Diskussionen der Projektpartner in IIP-Ecosphere.
Die untersuchten Plattformen decken insbesondere die benötigten Grundfunktionen ab. Beispielsweise wird oft eine Vielzahl an Kommunikationsprotokollen bereitgestellt und verschiedenste Cloud-Dienste integriert. Selbst neuere Trends wie Künstliche Intelligenz sind inzwischen in den Plattformbeschreibungen zu finden. Allerdings ist der Funktionsumfang zwischen den Plattformen auch sehr unterschiedlich. Neuere Standards wie OPC-UA, UMATI oder die Industrie 4.0 Verwaltungsschale werden oft nur zurückhaltend, wenn überhaupt eingesetzt, was teilweise der Entwicklungshistorie aber auch strategischen Erwägungen geschuldet sein mag.
Basierend auf der Plattform-übergreifenden Analyse der 16 Themenfelder leiten wir Herausforderungen für zukünftige Plattformen und insbesondere für unsere Arbeit in IIP-Ecosphere ab. Diese umfassen Themen wie offene Ökosysteme, erweiterbare Architekturen mit standardisierten Schnittstellenbeschreibungen, flexible und dynamische Unterstützung für KI-Verfahren, sicherer und vereinheitlichter Datenaustausch (für Data Sharing, Ressource Sharing und Data Usage Control) wie auch durchgängige und konsistente Konfigurierbarkeit, die das Vertrauen des Nutzers in die jeweilige Plattform stärkt. Eine Standardisierung von (einigen) dieser Themen wäre wünschenswert um den Austausch und die Interoperabilität zwischen Plattformen und Plattformökosystemen zu verbessern und Lock-ins zu vermeiden.
Close
https://www.iip-ecosphere.eu/wp-content/uploads/2021/02/IIP-2020_001.pdf
https://www.iip-ecosphere.eu/wp-content/uploads/2021/02/IIP-2020_001-en.pdf
https://zenodo.org/record/4485756
doi:10.5281/zenodo.4485756
Close
Hildebrandt, Petra
Vortrag „Lohnt sich KI?“ Presentation
11.02.2021.
BibTeX | Schlagwörter: Artificial Intelligence, IIP-Ecosphere, Künstliche Intelligenz, Sennheiser
@misc{nokey,
title = {Vortrag „Lohnt sich KI?“},
author = {Petra Hildebrandt},
year = {2021},
date = {2021-02-11},
urldate = {2021-02-11},
publisher = {KI in der Produktion - organisiert durch Deutsche Messe AG },
keywords = {Artificial Intelligence, IIP-Ecosphere, Künstliche Intelligenz, Sennheiser},
pubstate = {published},
tppubtype = {presentation}
}

Close
Jomaa, Hadi S.; Schmidt-Thieme, Lars; Grabocka, Josif
Dataset2Vec: Learning Dataset Meta-Features Journal Article
In: Data Mining and Knowlege Discovery, 10618 (0737), pp. 22, 2021.
Abstract | Links | BibTeX | Schlagwörter: Hyperparameter Optimization, Meta-feature Learning, Meta-learning
@article{HadiS.Jomaa,
title = {Dataset2Vec: Learning Dataset Meta-Features},
author = {Hadi S. Jomaa and Lars Schmidt-Thieme and Josif Grabocka},
url = {https://arxiv.org/abs/1905.11063},
doi = {10.1007/s10618-021-00737-9},
year = {2021},
date = {2021-01-01},
journal = {Data Mining and Knowlege Discovery},
volume = {10618},
number = {0737},
pages = {22},
abstract = {Meta-learning, or learning to learn, is a machine learning approach that utilizes prior learning experiences to expedite the learning process on unseen tasks. As a data-driven approach, meta-learning requires meta-features that represent the primary learning tasks or datasets, and are estimated traditionally as engineered dataset statistics that require expert domain knowledge tailored for every meta-task. In this paper, first, we propose a meta- feature extractor called Dataset2Vec that combines the versatility of engineered dataset meta-features with the expressivity of meta-features learned by deep neural networks. Primary learning tasks or datasets are represented as hierarchical sets, i.e., as a set of sets, esp. as a set of predictor/target pairs, and then a DeepSet architecture is employed to regress meta-features on them. Second, we propose a novel auxiliary meta-learning task with abundant data called dataset similarity learning that aims to predict if two batches stem from the same dataset or different ones. In an experiment on a large-scale hyperparameter optimization task for 120 UCI datasets with varying schemas as a meta-learning task, we show that the meta-features of Dataset2Vec outperform the expert engineered meta-features and thus demonstrate the usefulness of learned meta-features for datasets with varying schemas for the first time.},
keywords = {Hyperparameter Optimization, Meta-feature Learning, Meta-learning},
pubstate = {published},
tppubtype = {article}
}

Close
Meta-learning, or learning to learn, is a machine learning approach that utilizes prior learning experiences to expedite the learning process on unseen tasks. As a data-driven approach, meta-learning requires meta-features that represent the primary learning tasks or datasets, and are estimated traditionally as engineered dataset statistics that require expert domain knowledge tailored for every meta-task. In this paper, first, we propose a meta- feature extractor called Dataset2Vec that combines the versatility of engineered dataset meta-features with the expressivity of meta-features learned by deep neural networks. Primary learning tasks or datasets are represented as hierarchical sets, i.e., as a set of sets, esp. as a set of predictor/target pairs, and then a DeepSet architecture is employed to regress meta-features on them. Second, we propose a novel auxiliary meta-learning task with abundant data called dataset similarity learning that aims to predict if two batches stem from the same dataset or different ones. In an experiment on a large-scale hyperparameter optimization task for 120 UCI datasets with varying schemas as a meta-learning task, we show that the meta-features of Dataset2Vec outperform the expert engineered meta-features and thus demonstrate the usefulness of learned meta-features for datasets with varying schemas for the first time.
Close
https://arxiv.org/abs/1905.11063
doi:10.1007/s10618-021-00737-9
Close
2020
Jalowski, Max; Schymanietz, Martin; Möslein, Kathrin M.
Supporting Participants in Creative Processes: Opportunities for Persuasive Technology in Participatory Design Conference
2020.
Links | BibTeX | Schlagwörter: Creative Process, Creativity, Participant Support, Persuasive Technology, User Behavior
@conference{MaxJalowski,
title = {Supporting Participants in Creative Processes: Opportunities for Persuasive Technology in Participatory Design},
author = {Jalowski, Max and Schymanietz, Martin and Möslein, Kathrin M.},
url = {https://aisel.aisnet.org/icis2020/user_behaviors/user_behaviors/4/},
year = {2020},
date = {2020-12-13},
journal = {Proceedings of the Forty-First International Conference on Information Systems, India 2020},
keywords = {Creative Process, Creativity, Participant Support, Persuasive Technology, User Behavior},
pubstate = {published},
tppubtype = {conference}
}

Close
https://aisel.aisnet.org/icis2020/user_behaviors/user_behaviors/4/
Close
Denkena, Berend; Bergmann, Benjamin; Reimer, Svenja; Schmidt, Alexander; Stiehl, Tobias; Witt, Matthias
KI-gestützte Prozessüberwachung in der Zerspanung Journal Article
In: ZWF Zeitschrift für wirtschaftlichen Fabrikbetrieb, 115 (5), pp. 295-298, 2020, ISBN: 0947–0085.
Links | BibTeX | Schlagwörter: Industrie 4.0, Künstliche Intelligenz, Produktion, Prozessüberwachung
@article{Denkena2020,
title = {KI-gestützte Prozessüberwachung in der Zerspanung},
author = {Berend Denkena and Benjamin Bergmann and Svenja Reimer and Alexander Schmidt and Tobias Stiehl and Matthias Witt},
url = {https://www.degruyter.com/document/doi/10.3139/104.112282/html},
doi = {https://doi.org/10.3139/104.112282},
isbn = {0947–0085},
year = {2020},
date = {2020-05-05},
journal = {ZWF Zeitschrift für wirtschaftlichen Fabrikbetrieb},
volume = {115},
number = {5},
pages = {295-298},
keywords = {Industrie 4.0, Künstliche Intelligenz, Produktion, Prozessüberwachung},
pubstate = {published},
tppubtype = {article}
}

Close
https://www.degruyter.com/document/doi/10.3139/104.112282/html
doi:https://doi.org/10.3139/104.112282
Close