@misc{13334, abstract = {{Retrieval-augmented generation (RAG) based on large language models (LLMs) has established itself as a key technology for combining domain-specific information with generative language skills, thereby providing transparent, up-to-date information. Many firms are already piloting such LLM-based information systems, but report a high degree of complexity in planning and implementation. A generally accepted regulatory framework that consistently maps key decisions is not yet available to companies. This article therefore presents a multi-level system that organizes design decisions throughout the configuration process. This framework is intended to support users in the planning, realizing, evaluation, and further development of an LLM-based information system. To achieve this goal, a qualitative-empirical research design was chosen. First, publications from the period 2022 to 2025 were identified and selected using a systematic literature search in accordance with the PRISMA guideline. The selected publications were then evaluated using a qualitative content analysis. The result is a system that was reviewed, revised and finalized at an expert workshop.}}, author = {{Ullrich, Dominik and Wallys, Jens and Hinrichsen, Sven}}, booktitle = {{Intelligent Human Systems Integration (IHSI 2026): Disruptive and Innovative Technologies}}, editor = {{Ahram, Tareq and Karwowski, Waldemar and Giraldi , Laura and Benelli , Elisabetta}}, isbn = {{978-1-964867-76-2}}, issn = {{2771-0718}}, keywords = {{Retrieval-Augmented Generation, LLM-Based Information System, Conceptual Framework}}, location = {{Florence}}, pages = {{63--73}}, publisher = {{AHFE International}}, title = {{{Conceptual Framework for Designing Domain-Specific LLM-Based Information Systems}}}, doi = {{10.54941/ahfe1007065}}, volume = {{200}}, year = {{2026}}, } @misc{13120, abstract = {{This paper introduces an approach that leverages large language models (LLMs) to convert detailed descriptions of an Operational Design Domain (ODD) into realistic, executable simulation scenarios for testing autonomous vehicles. The method combines model-based and data-driven techniques to decompose ODDs into three key components: environmental, scenery, and dynamic elements. It then applies prompt engineering to generate ScenarioRunner scripts compatible with CARLA. The model-based component guides the LLM using structured prompts and a “Tree of Thoughts” strategy to outline the scenario, while a data-driven refinement process, drawing inspiration from red teaming, enhances the accuracy and robustness of the generated scripts over time. Experimental results show that while static components, such as weather and road layouts, are well captured, dynamic elements like vehicle and pedestrian behavior require further refinement. Overall, this approach not only reduces the manual effort involved in creating simulation scenarios but also identifies key challenges and opportunities for advancing safer and more adaptive autonomous driving systems.}}, author = {{Danso, Aaron Agyapong and Büker, Ulrich}}, booktitle = {{Electronics}}, issn = {{2079-9292 }}, keywords = {{large language models, generation, Operational Design Domain, autonomous vehicles, simulation, CARLA, ScenarioRunner, prompt-engineering, fine-tuning}}, number = {{16}}, pages = {{3177}}, publisher = {{MDPI}}, title = {{{Automated Generation of Test Scenarios for Autonomous Driving Using LLMs}}}, doi = {{10.3390/electronics14163177}}, volume = {{14}}, year = {{2025}}, } @misc{13291, abstract = {{The application of Large Language Models (LLMs) for the automated generation of assembly instructions shows significant potential for improving work preparation in production processes. However, challenges remain regarding the overall information quality and precision of the generated instructions. In light of these challenges, this study explores how the information quality of automatically generated assembly instructions can be enhanced through the targeted provision of structured input data, such as Assembly and Quantity BOMs (Bills of Materials), as well as the use of optimized prompt chaining techniques. The methodology employs ChatGPT-4o in combination with Retrieval Augmented Generation (RAG) within the Microsoft Azure environment. The results demonstrate that structured data inputs, particularly the use of Assembly BOMs with defined Tool-to-Component relations, significantly improve the precision and relevance of the generated instructions. Despite these advancements, achieving consistent information quality remains a barrier to broader practical implementation. Therefore, feedback loops should be integrated into the assembly instruction generation process to ensure continuous refinement and reliability. Future research should investigate the use of RAG or similar frameworks, focusing on optimizing data structures and implementing feedback mechanisms to enhance the automated generation of assembly instructions.}}, author = {{Herbort, Robin and Green, Dominik and Hinrichsen, Sven}}, booktitle = {{Intelligent Human Systems Integration (IHSI 2025): Integrating People and Intelligent Systems}}, editor = {{Ahram, Tareq and Karwowski, Waldemar and Martino, Carlo and Di Bucchianico, Giuseppe and Maselli, Vincenzo }}, isbn = {{978-1-964867-36-6}}, issn = {{2771-0718}}, keywords = {{Assembly Instruction, Retrieval Augmented Generation (RAG), Large Language Model (LLM)}}, location = {{Rome, Italy}}, pages = {{765--775}}, publisher = {{AHFE }}, title = {{{Automatic Creation of Assembly Instructions by Using Retrieval Augmented Generation}}}, doi = {{10.54941/ahfe1005883}}, volume = {{160}}, year = {{2025}}, } @misc{13292, abstract = {{The application of Large Language Models (LLMs) for the automated generation of assembly instructions shows significant potential for improving work preparation in production processes. However, challenges remain regarding the overall information quality and precision of the generated instructions. In light of these challenges, this study explores how the information quality of automatically generated assembly instructions can be enhanced through the targeted provision of structured input data, such as Assembly and Quantity BOMs (Bills of Materials), as well as the use of optimized prompt chaining techniques. The methodology employs ChatGPT-4o in combination with Retrieval Augmented Generation (RAG) within the Microsoft Azure environment. The results demonstrate that structured data inputs, particularly the use of Assembly BOMs with defined Tool-to-Component relations, significantly improve the precision and relevance of the generated instructions. Despite these advancements, achieving consistent information quality remains a barrier to broader practical implementation. Therefore, feedback loops should be integrated into the assembly instruction generation process to ensure continuous refinement and reliability. Future research should investigate the use of RAG or similar frameworks, focusing on optimizing data structures and implementing feedback mechanisms to enhance the automated generation of assembly instructions.}}, author = {{Herbort, Robin and Green, Dominik and Hinrichsen, Sven}}, booktitle = {{Intelligent Human Systems Integration (IHSI 2025): Integrating People and Intelligent Systems}}, editor = {{Ahram, Tareq and Karwowski, Waldemar and Martino, Carlo and Di Bucchianico, Giuseppe and Maselli, Vincenzo}}, isbn = {{978-1-964867-36-6}}, issn = {{2771-0718}}, keywords = {{Retrieval Augmented Generation, Large Language Model, Assembly Instructions}}, location = {{Rome}}, publisher = {{AHFE}}, title = {{{Automatic Creation of Assembly Instructions by Using Retrieval Augmented Generation}}}, doi = {{10.54941/ahfe1005883}}, volume = {{160}}, year = {{2025}}, } @misc{13293, abstract = {{The performance of large language models (LLMs) has improved significantly in recent years, with the result that they are now used in many companies in various industries. However, the design of a company-specific information system involving an LLM is associated with a large number of decisions. This leads to a high level of complexity in the design task. Against this background, companies need a structured approach that methodically supports the planning, development, implementation and long-term maintenance of LLM-based information systems so that domain- and company-specific requirements are taken into account as a result. This article therefore describes a method that supports the design, introduction and maintenance process of an LLM-based information system. The method consists of a process model and a list of design principles, which are also referred to as success factors. The process model developed is based on the proven six-stage REFA planning system. To identify and describe success factors, a systematic literature search was carried out. Based on an analysis of the contents of individual literature sources, success factors for the design of LLM-based information systems were identified. These success factors relate, for example, to the quality of the data provided, data security, user-centered system design and feedback mechanisms for improving information output.}}, author = {{Hinrichsen, Sven and Herbort, Robin and Green, Dominik and Adrian, Benjamin}}, booktitle = {{Human Interaction and Emerging Technologies (IHIET 2025)}}, editor = {{Ahram, Tareq and Motschnig, Renate }}, isbn = {{978-1-964867-73-1}}, issn = {{2771-0718}}, keywords = {{Large language model, Information system, Retrieval augmented generation}}, location = {{Vienna}}, publisher = {{AHFE}}, title = {{{How to Design an Operation-Specific LLM-Based Information System}}}, doi = {{10.54941/ahfe1006709}}, volume = {{197}}, year = {{2025}}, } @misc{8384, abstract = {{ynamic simulation models are widely utilized to evaluate complex technical components and systems like electric drives or machines. They can support the development process of a production machine by avoiding an inadequate layout of components or an erroneous control design. However, the effort for building them is often too high for this purpose (lot size one). An automated model generation can be utilized to overcome the gap between efforts and advantages of dynamic simulations. This contribution presents an approach for simplifying the dynamic model generation of production machines by using the so-called Asset Administration Shell defined by the initiative Platform Industrie 4.0. The Asset Administration Shell was developed to aggregate all data necessary for maintaining the product across its life cycle. This includes component data and models as well as structural information about a machine. The generation process is performed by using the common FMI standard and a two-step procedure which allows the linkage of different simulation tools. The model generation is demonstrated by an example layout of a machine's internal direct current grid.}}, author = {{Göllner, D. and Pawlik, Thomas and Schulte, Thomas}}, booktitle = {{2021 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM)}}, isbn = {{978-1-6654-3772-1 }}, issn = {{2157-3611}}, keywords = {{Digital Twin, Asset Administration Shell, Dynamic Simulation Model, Industry 4.0, Automated Model Generation}}, location = {{Online (Singapore)}}, pages = {{808--812}}, publisher = {{IEEE}}, title = {{{Utilization of the Asset Administration Shell for the Generation of Dynamic Simulation Models}}}, doi = {{10.1109/IEEM50564.2021.9673089}}, year = {{2021}}, } @inproceedings{4094, abstract = {{Projection-based assitive systems that guide users through assembly work are on their way to industrial application. Previous research work investigated how people can be supported with such systems. However, there has been little work on the question on how to generate and author sequential instructions for assitive systems. In this paper, we present a new concept and a prototypical implementation of an assitive system that can be taught by demonstrating an assembly process. By using a combination of RGB and depth cameras, we can generate an assembly instruction of Lego Duplo bricks based on the demonstration of a user. This generated manual can later on be used for assisting other users in the assembly process. By our prototype system, we show the technological feasibility of assistive systems that can learn from users.}}, author = {{Büttner, Sebastian and Peda, Andreas and Heinz, Mario and Röcker, Carsten}}, booktitle = {{22nd International Conference on Human-Computer Interaction}}, isbn = {{978-3-030-50343-7}}, keywords = {{Assitive system, Authoring, Instruction generation, Computer vision, Teaching by demonstration}}, location = {{Copenhagen, Denmark}}, pages = {{153--163}}, publisher = {{Springer}}, title = {{{Teaching by Demonstrating – How Smart Assistive Systems Can Learn from Users}}}, doi = {{https://doi.org/10.1007/978-3-030-50344-4_12}}, volume = {{12203}}, year = {{2020}}, } @misc{11444, abstract = {{Generation Y marks the transition between a world with and without fully implemented Internet: A new kind of virtual space is formed and is impacting the way we live in this world, the way we perceive it, and the way we interact with it. Mobile devices, such as phones, build a new form of electronic technology type we interact with. This thesis investigates how this new form of relation can give insights to the way we are situated in this world with all its complexities and levels. In a case study, the author focuses on the relationship with mobile devices in the context of memory (making and recalling), materiality (haptic and metaphoric) and the human body itself (perception and sensory system). By developing a practice and contextualizing it in terms of space-theoretical and phenomenological concepts, this thesis aims to start a discourse on our human, (un-)conscious relation to mobile devices in place and time. Is it time for an imperfect, humane view towards the era of information from a Millennial perspective?}}, author = {{Pusch, Lisa}}, keywords = {{Memory Making, Generation Y, Millenials, Human-Computer-Relationship, Mobile Phone, Spatial Theory, Perception}}, pages = {{264}}, publisher = {{ProQuest}}, title = {{{On memory: Body, devices, material. Towards a new practice of MediaArchitecture}}}, year = {{2017}}, }