In recent years, interest in digital twins (DTs) has increased among both industry and academia. Consequently, various methods have been proposed to enhance the development and maintenance of DTs. The creation of DTs relies on integrating models and data from a variety of sources, including sensors, Internet of Things (IoT) devices, and other data streams. Model-driven engineering (MDE) is considered an effective approach for developing IoT applications. Previous research indicates that current methods focus primarily on software and systems perspectives, often overlooking the importance of organizational integration. Additionally, there is a growing need for approaches specifically designed to meet the needs of small and medium-sized enterprises (SMEs). This thesis project aims to develop (i) a method that includes clearly defined steps, (ii) a tool to assist SMEs in developing IoT-based DTs, and (iii) consider aspects of organizational integration in and following the development process.

Model-based engineering (MBE) is a powerful paradigm that leverages models as essential pillars of the development process, enabling teams to clarify requirements, streamline design, specify behavior, and perform rigorous verification and validation tasks across the entire system life cycle. Digital twins (DTs) represent revolutionary software systems that mirror cyber-physical, socio-economic, or biological entities, systems, or processes. Built from robust models and data, DTs are deployed for high-impact applications such as planning, monitoring, control, and optimization of the twinned entity. The model-centric nature of DTs has naturally ignited recent exploration into harnessing MBE for the engineering and operation of DTs. However, this organic evolution has created a fragmented landscape of (partial) solutions. To confront this challenge, this article presents a rigorous and systematic literature survey on the field of model-based DT engineering (MBDTE), accompanied by a novel taxonomy for categorizing MBDTE approaches. We also introduce crisp definitions of both the field of MBDTE and the models themselves. We conclude by highlighting research gaps and outlining avenues for further exploration.

In many areas of conceptual modeling, large language models (LLMs) can be applied as assistive technology, for example, to gather domain knowledge to support model creation or to generate models from text. However, there is limited work on using LLMs to support enterprise architecture (EA) modeling. An LLM-based approach for generating EA models from text must consistently incorporate the different architectural layers of an EA. The aim of this paper is to contribute to this area by analyzing existing research, designing an approach to generating LLM-based models from text, and evaluating it experimentally. More concretely, an LLM-based approach designed for enterprise models is adapted for EA modeling with ArchiMate. The results of the experiments not only confirm feasibility but also a decent level of model quality. A set of recommendations for practical LLM use in EA modeling is abstracted, and implications for future research are derived.

Enterprise modeling (EM) typically includes multiple perspectives to analyze or design an organization, which is supposed to reduce the complexity of the modeling process and improve the structure of the resulting enterprise models. However, for non-modeling experts, i.e., stakeholders without solid training in EM, the development and integration of different model perspectives remains a challenge. Based on previous research in conceptual modeling, the paper investigates the use of large language models (LLMs) to support non-modeling experts of multi-perspective EM. Positioned in the roadmap for increasing the reach of EM, an LLM-based toolchain and reusable prompts for generating multi-perspective enterprise models from textual descriptions are proposed and evaluated. The results indicate that LLMs can be seen as assistive technology for certain tasks in EM. The main contributions of the paper are (1) an analysis of the state of research in LLM use in EM, (2) a study on LLM use for producing multi-perspective enterprise models, and (3) positioning LLMs in the roadmap for increasing the reach of EM.

Enterprise Architecture (EA) aims to align IT with business goals, but the concept of “future viability” remains academically abstract and practically elusive, creating a significant theory-practice gap. This study addresses this gap through a qualitative content analysis of semi-structured interviews with seasoned EA practitioners. Our findings reveal that practitioners define future viability not as a technical property, but as a direct outcome of business strategy, making its assessment highly context-dependent. While confirming that business capabilities serve as a “central anchor point” for translating strategy into design, the analysis also shows that the primary obstacles are not technical but organizational: the significant impact of technical debt and pervasive institutional inertia. By empirically validating and nuancing theoretical concepts, this paper contributes a practitioner-grounded understanding of future viability, offering actionable insights into the real-world challenges and strategic imperatives for building resilient and adaptable enterprise architectures.

Large language models (LLMs) have been found to be a support for modeling tasks in various application areas, including enterprise modeling (EM). In EM, LLMs can be applied to help domain experts create models efficiently that adhere to the correct syntax of the modeling language. In this context, how to organize the interplay of the domain expert and LLM is an important topic. Should the domain expert get an LLM-generated model and improve it (LLM-first) or should LLMs be used to improve models developed by domain experts (domain expert-first)? The paper investigates the interplay between domain expert and LLM by investigating three different application examples and conducting quasi-experiments. The results also contribute to determining the potential and limits of LLMs in EM.

The interest in digital twins (DTs) has recently increased among various industries. Consequently, various methods have been proposed to enhance their development and maintenance. The creation of DTs relies on integrating models and data from a variety of sources, including sensors, Internet of Things (IoT) devices, and other data streams. Model-driven engineering (MDE) is considered an effective approach for developing DTs. Our previous research indicates that current methods focus primarily on the software and systems perspective, often overlooking the importance of organizational integration. Additionally, we identified a growing need for approaches specifically designed to meet the needs of small and medium-sized enterprises (SMEs). In this paper, we introduce a method for MDE of DTs with the corresponding tool support meeting the specific requirements of SMEs. In three case studies from different domains, we demonstrate the applicability and confirm that domain experts are able to develop DTs with domain knowledge using our method.

The implementation of artificial intelligence (AI) in the public sector offers great potential. Repetitive and labor-intensive tasks can be automated to improve overall efficiency. Generative AI, in particular, opens up new possibilities for structuring and integrating heterogeneous data sources. At the same time, AI introduces challenges such as technical complexity and ethical issues that must be addressed during development and implementation. This paper investigates the potential and challenges of using AI in the extract, transform, load (ETL) process in a public sector study. Our findings demonstrate that open-source large language models (LLMs) can efficiently transfer over 5,000 unstructured documents into the structured format of a relational database, achieving a success rate of approximately 96%. The quality of the results was significantly improved through optimization measures, particularly in terms of prompt engineering and post-processing. While the results are encouraging, challenges remain, including processing extensive documents and adapting the data model to greater complexity.

Modeling is a crucial aspect of the development process in various engineering disciplines. The application of modeling in these fields typically encompasses various phases, including target setting, requirement elicitation, architecture specification, system design, and test case development. This paper focuses on the initial stages of systems development, with a specific focus on requirements engineering (RE) in enterprise modeling (EM). In particular, we examine the potential for domain experts to be replaced by artificial intelligence (AI) usage. The objective of this research is to contribute to a more comprehensive understanding of the limitations of large language models (LLMs). To this end, we employ a process from hospitality management and contrast the output of ChatGPT with that of a domain expert in an experiment. A second experiment was subsequently conducted in light of the assumption that the quality of responses can be enhanced by defining the expected output modeling language notation in the prompt and a metamodel for prompt engineering. The findings of this paper indicate that LLMs cannot replace domain experts in modeling the current situation of an enterprise. However, they can be employed as a supporting tool in EM. Moreover, the development of reusable prompts has emerged as both a viable and promising avenue for future research.

Recently, researchers have explored whether large language models (LLMs) can be used as a substitute for domain experts to elicit information that should be represented in an enterprise model. This paper examines a slightly different application purpose, assessing an existing model's quality using an LLM. We will analyze which aspects of model quality can be evaluated using an LLM in principle, referring to the established model quality framework SEQUAL. We will present a first test of assessing perceived semantic quality using ChatGPT. To examine the effect of different prompting strategies, we compared our results to the assessments of human domain experts. Our results suggest that LLMs are suitable for assessing the perceived semantic quality of models and provide a basis for considering further quality dimensions in future work.

There is still a lack of approaches to easily help users cope with the complex and challenging tasks related to Internet of Things (IoT) application development. The absence of standardized procedures and the complexity and heterogeneity of the IoT landscape are perceived as the main challenges. To address the complexity of IoT application development, Model-Driven Development (MDD) has emerged as an effective technique. By means of a Systematic Literature Review (SLR), we provide an overview of the current state of research in MDD for IoT applications. It shows that current approaches often neglect the role of organizational factors. This paper presents a method for MDD of IoT applications that considers both the stated challenges and organizational integration. The method includes a Domain-Specific Modeling Language (DSML) and several functionalities that assist in the modeling and development processes. This eliminates the necessity for any particular IT expertise at the application level. An industrial use case in the field of air conditioning facilities where the method has been implemented is described. Requirements for the methodological and technical support for IoT development were derived from a Small and Medium-Sized Enterprise (SME) during the development process.

Model-Driven Development (MDD) is considered an effective technique for Internet of Things (IoT) application development. Our observation is that existing model-based approaches for IoT solutions focus on the software and systems perspective and show a need for more integration with organizational and business model aspects. Therefore, we developed a method and tool support for developing IoT applications in the field of air conditioning facilities. In this work, we applied quality criteria to evaluate the included Domain-Specific Modeling Language (DSML). To practically validate the modeling language as such and also the way it can be used and supported by the tool, we performed a real-world use case. The main contributions of this paper are a quality evaluation of the DSML and the tool support and lessons learned from both.

The Internet of Things (IoT) is a significant trend in the field of information technology and encourages the implementation of cyber-physical systems, smart connected products, and new business models. Many enterprises struggle to create business value from IoT technology because they have difficulties defining their organizational integration. Model-driven engineering (MDE) is considered an effective technique to address the complexity of IoT application development. Existing approaches focus on requirements from a technical perspective and exhibit a lack of integration with organizational and business model aspects. The paper proposes a modeling approach and a tool to support the development and configuration of IoT solutions in the example of air conditioning and cleanroom technologies (ACT) facilities. The main contributions of this paper are (a) an architecture for the IoT application, (b) a modeling language and tool support for IoT modeling, and (c) the expected practical benefits in the industrial case.

The development of Internet of Things (IoT) applications is a multifaceted and demanding exercise that requires developers to navigate through several complicated tasks and challenges. Primary issues in this area stem from the absence of standardization and the complexity and heterogeneity inherent to the IoT landscape. Model-Driven Development (MDD) is considered an effective technique for IoT application development. We argue that there is a need for methodologies of MDD that encompass system development and integration, as well as organizational aspects. To compensate for this lack, this work introduces MIoTA (Modeling IoT Applications for Air Conditioning Facilities), a modeling method supporting IoT application development in the field of air conditioning facilities. The ADOxx-based MIoTA tool was developed in an industrial use case. It contains a Domain-Specific Modeling Language (DSML) and various functionalities to support the modeling and development processes, so no specific IT skills are required at the application level. The approach has been evaluated practically within the case study and with quality criteria in previous work.