Now showing 1 - 10 of 12
No Thumbnail Available
Publication

The Digital Twin for Operations, Maintenance, Repair and Overhaul

2023 , Lünnemann, Pascal , Fresemann, Carina , Richter, Friederike

Looking at digital twins in terms of their information sets (master and shadow models), a significant part of the shadow models is created in the context of product life. Digital twins must be designed accordingly, focusing on their dedicated added value or business model. This concerns not only the information and data models, but also the communication technologies, processing routes and interaction mechanisms used. With appropriately designed digital twins, product life becomes a source of knowledge for optimizing or tracking product systems. MRO processes play a special role in this. Here, the digital twin becomes a monitoring system, information source, process manager or information sink through suitable functions and thus a potential knowledge repository.

No Thumbnail Available
Publication

Automated design of rotationally symmetric components with machine learning and synthetic data

2023 , Brünnhäußer, Jörg , Lünnemann, Pascal , Flachmeier, Florian , Wolff, Mario

No Thumbnail Available
Publication

Methodische Analyse bestehender Wertschöpfungssysteme zur Integration Digitaler Zwillinge

2021 , Seegrün, Anne , Lünnemann, Pascal , Lindow, Kai

No Thumbnail Available
Publication

Qualitätskriterien für den Datenfluss im Advanced System Engineering

2019 , Müller, Helena , Riedelsheimer, Theresa , Lünnemann, Pascal , Blüher, Till , Stark, Rainer

No Thumbnail Available
Publication

Implementing digital twins in existing infrastructures

2023 , Lünnemann, Pascal , Lindow, Kai , Goßlau, Leo

Digital twins can offer various added values for companies. As part of a three-year research project, we are investigating the methodological approach, for building digital twins in existing infrastructures. In particular, the functional requirements of future users will be addressed, as this is less focused in existing approaches. Within the framework of this publication, we discuss the applied methodology as well as the created models and concepts. Initial insights were gained in the simultaneous development of digital twins in parallel projects with use cases for electric motors, production process monitoring and maintenance of gas turbine components. In detail, it becomes clear that software development methods (e.g. use cases, user stories, scenario development) are a good way to describe the expected added value functions. It is essential to involve the future users in the development as early as possible. Transferring the necessary functions identified in this way into a functional architecture shows that this architecture is mostly independent of the use case. Likewise, the IT systems used here hardly vary at all. Overall, it shows that a methodical approach can be followed in the development and the implementation can have a high degree of similarity, even in very different use cases, while the exact design, depending on these use cases, is very diverse.

No Thumbnail Available
Publication

Evaluate Similarity of Requirements with Multilingual Natural Language Processing

2022 , Bisang, Ursina Saskia , Brünnhäußer, Jörg , Lünnemann, Pascal , Kirsch, L. , Lindow, Kai

Finding redundant requirements or semantically similar ones in previous projects is a very time-consuming task in engineering design, especially with multilingual data. Due to modern NLP it is possible to automate such tasks. In this paper we compared different multilingual embeddings models to see which of them is the most suitable to find similar requirements in English and German. The comparison was done for both in-domain data (requirements pairs) and out-of-domain data (general sentence pairs). The most suitable model were sentence embeddings learnt with knowledge distillation.

No Thumbnail Available
Publication

Potentials and challenges of Smart Products and related business models

2020 , Wang, Wei Min , Lünnemann, Pascal , Klemichen, Antje , Blüher, Till , Stark, Rainer

Smart products are increasingly penetrating the market. They extend cyber-physical systems with Internet-based services and thus enable customer-specific adaptations and updates of products in the field. For industrial companies, smart product offering bear the potential to expand their product portfolios and increase their sales. In this article, the results of a survey among German engineers are presented that evaluated the actual state of adoption and future expectations regarding smart products in German industrial companies. The results show that the majority of German companies have recognized the potentials of smart products and also introduced such offerings in their portfolio. In this context, it is also expected that the product-related usage data will lead to a more individual communication with the customer. On the other hand, there is a high degree of uncertainty among companies regarding the adaptation of new business models, cooperation with an increasing number of partners, data management and the necessary technological infrastructure. These uncertainties can result in companies leaving potentials unexploited and opening up opportunities for new market participants. In summary, there seems to be a remaining gap between the willingness to abandon traditional business models and the expectations and strategies for future value creation.

No Thumbnail Available
Publication

Automatisiertes Design von rotationssymmetrischen Bauteilen mit Machine Learning und synthetischen Daten

2023 , Brünnhäußer, Jörg , Lünnemann, Pascal , Flachmeier, Florian , Wolff, Mario

No Thumbnail Available
Publication

Geometric Design Process Automation with Artificial Intelligence

2022 , Brünnhäußer, Jörg , Lünnemann, Pascal , Bisang, Ursina Saskia , Novikov, Ruslan , Flachmeier, Florian , Wolff, Mario

Design tasks are largely performed manually by engineers, while machine learning is increasingly able to support and partially automate this process to save time or costs. The prerequisite for this is that the necessary data for training is available. This paper investigates whether it is possible to use data-driven methods to support the design of jounce bumpers at BASF. Based on the analysis of the use case, the geometry of the jounce bumper is approximated with a spline to generate suitable data for training. Based on this, data for training the machine learning model is generated and simulated. In the training process, the appropriate feedforward neural network and the best combination of hyperparameters are determined. In the subsequent evaluation process, it is shown that it is possible to predict the geometries of jounce bumpers with our proof of concept. Finally, the results are discussed, the limitations are shown and the next steps to further improve ssthe results are reflected.

No Thumbnail Available
Publication

From process to activity in the data flow

2019 , Lünnemann, Pascal , Riedelsheimer, Theresa , Wehking, Sebastian , Lindow, Kai