Keywords: Conceptual Modeling Education, Learning Analytics Dashboards, Social Comparison
Astract: Conceptual modeling education benefits from technological support due to the complex nature of the learning processes required to master modeling skills. Along with existing modeling and prototyping tools, providing feedback to students using Learning Analytics Dashboards (LADs) can enhance their learning experience. To interpret LADs, students are provided with a frame of reference, often peer comparison, although its effectiveness is debated. This study presents two LADs used to provide feedback to students from diverse backgrounds enrolled in a conceptual modeling course: a default-LAD with mastery and progress reference frames, and an extended peer-LAD that also includes a performance reference frame. We examine students’ preferences for LAD visuals, the relationship between their study activity and performance, and the relationship between the use patterns of different LAD versions and student activity and performance. The results show that most of the relationships are significant only for the peer-enhanced LAD and are stronger for students with less modeling experience, underscoring the value of peer LADs for novice modelers.

Keywords: Digital Twin Engineering, Enterprise Architecture, Sustainable Real Estate Management
Astract: In the context of Real Estate Management, Digital Twins (DTs) can act as transformative tools for the integration of sustainability into construction practices by providing dynamic, data-driven virtual representations of physical assets. They enable stakeholders to optimise processes, reduce resource consumption, and align with sustainability objectives. However, effective implementation of DTs, in this context, remains limited by fragmented data ecosystems across different organisations, incomplete integration with regulatory frameworks, and insufficient consideration of sustainability goals. This is especially challenging, as access to high quality, and integrated, data is critical for DTs. Enterprise Architecture (EA) is generally used as an instrument to aid in managing and coordinating such challenges, as it provides a structured approach to addressing these challenges by aligning technical, organisational, and sustainability objectives. In line with this, we propose three primary research objectives: (1) assessing the current state of EA approaches for integrating DTs, (2) developing sustainability-oriented EA modelling concepts and design patterns, and (3) exploring the scalability of DTs from individual buildings to smart cities. Our aim is to position DTs as comprehensive decision-support tools for sustainability, enabling stakeholders to achieve better-informed decisions while addressing regulatory compliance, resource efficiency, and life cycle impacts.

Keywords: Conceptual Models, Domain Models, Modeling, Return on Modeling Efforts, Risk Assessment
Astract: Modelers and organizations often struggle to assess the benefits and drawbacks of modeling activities. This paper proposes addressing this challenge through a risk-oriented lens, leveraging the Common Ontology of ValuE and Risk (COVER) and the Reference Ontology for Security Engineering (ROSE). The proposal focuses on identifying assets at risk throughout the modeling process to clarify: when models mitigate risks and contribute to cost savings (models as risk deterrents), when models introduce risk to other assets (models as risk sources), or when they are vulnerable to risk events themselves (models as assets at risk), potentially generating additional costs. This perspective enables modelers and organizations to evaluate the benefits and costs of modeling practices, aligning investments with organizational goals, while helping researchers identify gaps for enhancing modeling languages, methods, and tools. The proposal is evaluated by analyzing case studies from the literature and interviews with nine professionals and researchers.

Keywords: Process Mining

Keywords: GLSP, Langium, LSP, Modeling tools, UML., Web modeling
Astract: The move towards web technologies has arrived at modeling and especially modeling tools. Several frameworks and platforms have been proposed recently, aiming at the efficient realization of web-based modeling tools, i.e., tools developed with web technologies that natively allow for cloud-based deployment and use in web browsers. Most of these frameworks separate modeling tool functionality like editing, rendering, and management of the model into separate specialized components, often realized in different technologies. For the model management of Ecore-based modeling languages like the UML, the Eclipse Modeling Framework (EMF) is still prevalent. While this may maximize reuse, it comes at a cost: EMF is Java-based, entailing a polyglot technology stack, which complicates the development and hinders an entirely browser-based deployment without any backend. In this paper, we address this problem by introducing a Typescript-only approach for generating Langium-based modeling language grammars and model management servers. Using our approach, developers can build next-generation web-based modeling tools with a homogeneous technology stack, which can be executed as cloud applications or plain browser applications without any backend, if needed. We evaluate our approach by applying it to two distinct modeling tools: a Workflow and a UML modeling tool.

Keywords: Conceptual modeling, Customer journey map, Modeling costs, Multi-method research approach, Pain points
Astract: Conceptual models offer numerous benefits but require significant investments, requiring modelers to strive to balance costs and benefits. Understanding the modeling process and the frustrations experienced by modelers can provide valuable insights for this assessment. While research acknowledges certain instances of modelers’ dissatisfaction, its scope often limits detailed examination. This study seeks to identify and analyze the main pain points associated with domain modeling through a five-phase empirical study using a multi-method approach. We identified 71 pain points, synthesized them to 41, and prioritized 16 as the most significant and prevalent in domain modeling. We then refined, documented, and exemplified the prioritized pain points, analyzed their potential causes, and discussed their practical implications. Our findings provide valuable insights for improving modelers’ experiences and optimizing the modeling process.

Keywords: Business capability, Data valuation, Data value, Taxonomy
Astract: In our data-centric society, the imperative to determine the value of data has risen. Therefore, this paper presents a taxonomy for a data valuation business capability. Utilizing an initial taxonomy version, which originated from a systematic literature review, this paper validates and extends the taxonomy, culminating in four layers, twelve dimensions, and 59 characteristics. The taxonomy validation was accomplished by conducting semi-structured expert interviews with eleven subject matter experts, followed by a cluster analysis of the interviews, leading to a taxonomy heatmap including practical extensions. This paper's implications are manifold. Firstly, the taxonomy promotes a common understanding of data valuation within an enterprise. Secondly, the taxonomy aids in categorizing, assessing, and optimizing data valuation endeavors. Thirdly, it lays the groundwork for potential data valuation standards and toolkits. Lastly, it strengthens theoretical assumptions by grounding them in practical insights and offers an interdisciplinary research agenda following the taxonomy dimensions and characteristics.

Keywords: ArchiMate, Artificial intelligence, Conceptual modeling, Dataset, Enterprise architecture, Enterprise modeling, FAIR, Machine learning
Astract: Conceptual modeling research is increasingly investigating the application of artificial intelligence (AI) and machine learning (ML) to automate tasks like model creation, completion, analysis, and processing. This trend also applies to enterprise architecture (EA) research. In contrast to its neighboring disciplines, such as business process management, EA lacks proper guidelines, patterns, and best practices to create high-quality EA models. A currently limiting factor for conducting AI-based research to bridge these gaps is the scarcity of openly available models of adequate quality and quantity. With this paper, our aim is to address this limitation by introducing the extended EA ModelSet, a curated and FAIR repository of enterprise architecture models represented in the ArchiMate modeling language that can be used by the research and practitioner community. We report on our efforts to build the EA ModelSet and elaborate on exemplary future empirical and ML-based research that can facilitate the dataset. We hope that this paper sparks a community effort toward the further development and maintenance of the EA ModelSet.

Keywords: heavy equipment, residual value, machine learning, construction, regression
Astract: The residual value of heavy equipment is essential for financial and economic considerations in the construction industry. In practice, empirical methods are frequently used to determine the residual value of a given piece of equipment. Here, various regression methods are compared based on a real-world dataset of used heavy equipment sales from a construction company. The results show that the prediction performance of traditional methods is clearly worse when compared to machine learning models not yet employed for this purpose. For the latter, preprocessing and parameter tuning are essential, and the article guides through these steps. Further, the article demonstrates how a variable importance value comparable across all methods can be obtained. These findings may also be useful in other applications.

Keywords: Domain Modeling, Large Language Models, Tree of thoughts
Astract: Domain modeling is typically an iterative process where modeling experts interact with domain experts to complete and refine the model. Recently, we have seen several attempts to assist, or even replace, the modeler with a Large Language Model (LLM). Several LLM prompting strategies have been attempted, but with limited success. In this paper, we advocate for the adoption of a Tree-of-Thoughts (ToT) strategy to overcome the limitations of current approaches based on simpler prompting strategies. With a ToT strategy, we can decompose the modeling process into several sub-steps using for each step a specialized set of generators and evaluators prompts to optimize the quality of the LLM output. As part of our adaptation, we provide a Domain-Specific Language (DSL) to facilitate the formalization of the ToT process for domain modeling. Our approach is implemented as part of an open source tool available on GitHub.