Metamodelling Platforms

Abstract

The elements of an enterprise are managed more and more model-based. The state-of-the-art in the area of modelling of organisations is based on fixed metamodels. Product models are created by using product modelling environments, process models are created in business process modelling tools and organisational models are realised in personnel management tools. Web service models link these business models to information technology. They are created by using standardised languages and common ontologies. Information technology is modelled in tools supporting notions such as workflow or object-orientation. The models of the company’s strategy, goals and the appropriate measurements are described and monitored by using tools supporting management concepts such as Balanced Scorecard.

Major requirements to an enterprise modelling platform are flexibility and adaptability. These are fulfilled by environments providing flexible metamodelling capabilities. The main characteristic of such environments is that the formalism of modelling—the metamodel—can be freely defined. This raises research issues on how to design, manage, distribute and use such metamodels on a syntactic as well as on a semantic level and how to integrate, run and maintain a metamodelling platform in a corporation’s environment.

Platforms based on metamodelling concepts should support the following topics:

• 1. Engineering the business models & their web services

• 2. Designing and realizing the corresponding information technology

• 3. Evaluating the used corporation resources and assets

This paper presents a framework for metamodelling platforms and gives some answers to the research issues stated above. As part of the framework flexible metamodel integration mechanisms, using meta-metamodels (meta2-models) and semantical mapping, are discussed. Additionally, a system architecture and the building blocks of a corporate metamodelling platform are described. Finally, best practices from three EU funded projects—REFINE, ADVISOR, and PROMOTE—all realised with industrial partners, are presented.

The full version of the paper can be downloaded from http://www.dke.univie.ac.at/mmp