Enterprise Architecture (EA) is a discipline that tries to achieve an alignment between the overall business strategy and its realization in information systems. One important means in this discipline are EA models, which are used to understand the actual state of the organization as well as to plan for the future evolution of the organization. However, the modeling languages used for the creation of these models leave room for interpretation how to use certain element types. Based on this observation, we expect that modelers use EA modeling languages in different ways leading to a similar phenomenon in natural languages: dialects.
With this thesis, a student elaborates on this phenomenon by investigating a set of existing EA models (in the notation of ArchiMate). Therefore, the student will e.g., analyze the use of element types in different settings considering their context. Depending on the capabilities and interests of the student, this can be performed in a qualitative (i.e., analyzing a rather small set of models by hand, but intensively), or quantitatively (e.g., using means of machine learning to analyze the entire set of models for reoccurring patterns). A group of two students could also collaborate on this topic, one focusing on qualitative and the other on quantitative analysis techniques.
The thesis will be positioned within a larger outset of research conducted at different universities in different countries and, thus, will allow the student to exchange with other thesis students conducting their thesis at the same time.