Enterprise architecture (EA) is a field involving many different disciplines and it has evolved to a stage where the selection of a model to work with EA has become a project in its own with the myriad of different frameworks, guides and models available to help assist the business transformation process. This thesis poses the question if there is the need for a unification model for establishing an EA programme instead of having so many disparate approaches. This question leads to four different hypotheses: There is a need in theory and in practice, there is a need in theory but not in practice and vice versa, and there is no need for a unified approach at all. To test the underlying hypotheses I examined the theory in a comparative study and did a case study in order to see if there are any discrepancies in what the models prescribes and how they are actually applied in practice. For the theoretical comparison I selected five models from the broad range of different approaches to EA; ranging from governmental issued models(OIOEA & FEA) and industrial developed models(TOGAF), to more traditional academic-type approaches (EA3 & EAP) and I supplemented this with a non-EA tried and tested technology-neutral model for programme management(MSP). While the comparative study of these different models proved that they each had their own different areas of focus, methods, activities and target audiences, it remained inconclusive whether there was a need for a unifying model for establishing an EA programme from a theoretical standpoint. The case study however, proved that while an organisation selects an EA framework to work within, they do not follow the steps in the manner the model actually prescribes. Additionally, they also follow steps prescribed by other models, when their chosen model is deemed inadequate to deal with the challenges faced by the business. This result proved the need in practice, for a more unified and wholesome approach to establishing an EA programme and I have thus created a model which builds upon the best-practices of the existing approaches and takes into consideration the reasons why the case study had discrepancies and chose not to follow their chosen model.
|MSc in Computer Science, (Graduate Programme) Final Thesis
|Number of pages