Aggregating Operational Data to the Level of EAM

Introduction

An Enterprise Architecture (EA) describes the business and IT system alignment of an enterprise and is represented by one or multiple EA models. In the last years the use of EAs in companies and so the interest in research and development increased a lot. With an EA being established within a company, many different stakeholders use the information the EA provides, displayed by an EA model and are involved in its evolution. Enterprise Architecture Management (EAM) then aims to adapt changes in the EA model, which occur in business processes, relationships of artifacts in the EA Model or are caused by changes in the IT system landscape and integration of new software.

Motivation

Since an EA model provides a certain level of detail regarding its data, new data that must be integrated during the evolution process has to fit to this level of detail. Therefor data, especially with a too high level of detail, for example too specific implementation details, has to be analysed and integrated into the EA model.

EAM is used for a long-term development of an EA, so an EA should evolve instead of being replaced by a completely new version from time to time. Therefor EAM is thought to establish a consistent and precise EA model, why especially a consistent level of detail is necessary, so one knows instantly whether the searched information can be found in the EA model.

Since software and infrastructure development usually is documented in a very detailed way, such a documentation especially contains the data that belongs into an EA. The most common way to integrate this data into the EA model is analysing the documentation and adding it manually into the EA model. Obviously this is not the most efficient way to do so and leads to unnecessary redundant work, because the same data that has to be integrated is already given, just in another structure.

The motivation of this bachelor thesis is to automate this process and the goal of this thesis is to develop an algorithm and a tool that analyses the documentation of the system that is integrated into the EA and integrates the data on the same level of detail as given into the related EA model.

Implementation

First, I am going to perform a systematic literature review (SLR) to find relevant papers regarding data analysis approaches, also in different fields of study, to detect the level of detail of an EA model and the documentation, papers that deal with integration and aggregation of data into an EA and other relevant papers. If the SLR does not yield sufficient results, especially regarding the aggregation approaches, I am going to ask experts of this domain via a survey or interviews.

Next, with the information collected in the set of considered papers, I will develop an algorithm that analyses a given EA model and documentation of the IT system and automatically integrates data from the documentation into the EA model. The algorithm will also take the structure of the EA model into consideration. The integration process will not just add data into the EA model but modify it and update or alter relationships of artifacts.

Another way to obtain structural data of the EA model is regarding its past evolution, especially in which ways data was integrated.

In a final step, I will implement this algorithm to develop a tool, doing the described task automatically for Archi EA models.