Tool Support for Con ict Resolution of Models for Automated Enterprise Architecture Documentation

Abstract

Improving the documentation of models used for Enterprise Architecture (EA) management is essential for the progress of the discipline in practice. Automated EA documentation seeks to increase data quality and e ciency of this process by collecting relevant EA data from existing information sources in the organization. While empirical evidence on the feasibility of the approach and theoretical solutions have already been proposed in research, tool support for the implementation is still an open issue since technical solutions for fundamental problems arising during the automated documentation of EA models are not available. In this paper we demonstrate a collaboration platform to support the resolution of model con icts during automated documentation1. We incorporate stakeholders in the resolution of these con icts using an implementation of a con ict resolution work ow and a novel interactive visualization.

Keywords

• Enterprise Architecture
• automated Enterprise Architecture documentation
• collaboration
• con ict resolution

-

Enterprise Architecture (EA) models typically consist of a plethora of elements that are interconnected with each other. In current practices the documentation and maintenance of these models is performed manually to a large extent. As a result this process is perceived as very time-consuming and error-prone by the majority of the organizations in industry [ 6 ]. Automated EA documentation seeks to gather and maintain relevant EA data from existing information sources in the organization and transform this data to a respective EA metamodel. While theoretical solutions to achieve automation have been proposed in research [ 2 ] and the feasibility of this approach has been evaluated in case studies [ 1 ] as well as on an empirical basis [ 3 ], technical solutions to support automation processes and the newly arising challenges related with this approach 1 http://www.youtube.com/watch?v=HxbAVjsM484, last accessed on: 2013-07-15 are not available. Main contribution of this paper is a demonstration of a solution to collaboratively resolve model con icts to achieve automated documentation. 2

Resolving Model Con icts

The establishment of an automated EA model documentation in practice is faced with a variety of di erent challenges [ 4 ]. These challenges are related with the transformation of models, security concerns, economic considerations, and the collection of the EA data. In particular, complete automation of model transformations are not possible due to data (e.g. from source systems like Network Scanners and Monitors [ 3 ]) that is often too ne-grained for mere EA purposes and missing model coverage. Appropriate tool support for automated model documentation needs to address these issues appropriately to achieve this long-term goal. In previous work, we presented a collaborative work ow to resolve these model con icts by involving stakeholders [ 7 ]. The work ow uses con ict resolutions tasks that are escalated to di erent stakeholders in the organization. These con ict resolution tasks consist of interactive visualizations that can also be used by business users not having a technical background. 2.1

Demonstration The EA model can be accessed and manipulated using a web application in the demo system. Every element is represented with structured information on wiki pages. We exemplify the system using ArchiMate as EA modeling language, although an arbitrary information model can be used with the presented solution. with a particular stakeholder (e.g. EA Coordinator, EA Repository Manager, Data Owner, or EA Stakeholder) and the concerning element(s) in the EA model. In case an EA Repository Manager is not able to resolve a con ict on his own, he can create and con gure a con ict resolution task. The con guration (cf. Fig. 2) of a new con ict resolution task contains information on the con ict model elements (Application Component and Node Device) as well as the mapping model element (Infrastructure Interface). In addition, the responsible Data Owner(s) of the information source that was used to gather the data can be selected to resolve this con ict. The information description shown at the bottom of the window allows the creator of the task to provide some additional information. After this task is submitted in the system, the selected Data Owner receives a new task containing the respective interactive visualization that is shown in Fig. 3 in his worklist. The Infrastructure Interface SAP RD Hosting is shown in the middle of the screenshot. Users can drag elements from the Application Component and the Node Device and drop these elements in the Infrastructure Interface to create a new relationship that could not be established during the model transformation automatically. The dropping of elements is supported with an additional coloring on valid target positions in the Infrastructure Interface. Changes in the visualizations are directly persisted in the EA model as soon as the Data Owner submits this task. In case the Data Owner is not able to resolve this tasks, he can forward it to an EA Stakeholders in the organization. Especially for EA Stakeholders not having a technical background, the given interactive visualization should simplify the con ict resolution. Resolved con icts within major releases are automatically forwarded to an EA Coordinator for approval purposes. This approval uses the same interactive visualization and it provides an additional quality assurance step in the system. As illustrated in Figure 4, we utilized an Enterprise 2.0 framework [ 5 ] and a framework for interactive EA visualizations [ 8 ] to realize an EA repository, which allows us to import data from arbitrary information sources to our EA repository. Con ict tasks can be created by end-users in the EA repository. A con ict task employs the interactive visualization framework on execution, i.e. the interactive visualization for the con ict resolution is generated in a model-driven manner. On execution, the interactive visualization is sent to the browser. The actual rendering within the browser is accomplished using the RaphaelJS2 framework to 2 http://raphaeljs.com/, last accessed on: 2013-07-15 provide cross-browser support. Changes to a con ict visualization are propagated to the EA repository such that underlying data is manipulated. 3

Conclusion

This paper has demonstrated a collaboration platform to resolve arising con icts during the automated documentation of EA models. This solution uses a collaborative work ow that has been developed and evaluated in previous work. The presented system is the rst step towards tool support for automation in this domain and it facilitates the incorporation of stakeholders. Over the next years, we will evaluate the system in practical settings to promote the development of collaborative tools for EA management and automated model documentation in particular.

Acknowledgment

This research has been sponsored in part by the German Federal Ministry of Education and Research (BMBF) with grant number TUM: 01IS12057.

References

1. Markus Buschle , Mathias Ekstedt, Sebastian Grunow , Matheus Hauder, Florian Matthes, and Sascha Roth . Automating enterprise architecture documentation using an enterprise service bus . In Nineteenth Americas Conference on Information Systems (AMCIS) , Seattle, USA, 2012 .

2. Matthias Farwick , Berthold Agreiter, Ruth Breu, Ste en Ryll, Karsten Voges, and Inge Hanschke . Automation processes for enterprise architecture management . In Enterprise Distributed Object Computing Conference Workshops (EDOCW) , 2011 15th IEEE International, pages 340 { 349 . IEEE, 2011 .

3. Matthias Farwick , Ruth Breu, Matheus Hauder, Sascha Roth, and Florian Matthes . Enterprise architecture documentation: Empirical analysis of information sources for automation . In 46th Hawaii International Conference on System Sciences (HICSS) , Maui, Hawaii, 2013 .

4. Matheus Hauder , Florian Matthes, and Sascha Roth . Challenges for automated enterprise architecture documentation . In Trends in Enterprise Architecture Research and Practice-Driven Research on Enterprise Transformation (TEAR) , Barcelona, Spain, pages 21 { 39 . Springer, 2012 .

5. F. Matthes , C. Neubert , and A. Steinho . Facilitating structuring of information for business users with hybrid wikis . In Lecture Notes of Communications in Computer and Information Science (CCIS) , 2012 .

6. Sascha Roth , Matheus Hauder, Matthias Farwick, Ruth Breu, and Florian Matthes . Enterprise architecture documentation: Current practices and future directions . In 11th International Conference on Wirtschaftsinformatik (WI) , Leipzig, Germany, 2013 .

7. Sascha Roth , Matheus Hauder, Felix Michel, Dominik Munch, and Florian Matthes. Facilitating con ict resolution of models for automated enterprise architecture documentation . In Nineteenth Americas Conference on Information Systems (AMCIS) , Chicago, USA, 2013 .

8. M. Schaub , F. Matthes , and S. Roth . Towards a conceptual framework for interactive enterprise architecture management visualizations . In Modellierung, Bamberg, Germany, 2012 .