MDA and RM-ODP: two approaches in modern ontological engineering

Constantly growing number of new system architectures and meta-data standards increases the difficulty of interoperability problems. Fortunately, the fundamental principles that are used by different system modeling frameworks are not so numerous. Understanding of ontological foundations for existing system modeling standards allows for the classification of the standards, and thus, provides system architects with the Ariadne’s thread that helps to pass successfully through the labyrinth of heterogeneous system models. In this article we explain two approaches existing in practical ontological engineering and perform their comparative analysis. The analysis familiarizes the reader with strengths and weaknesses of the approaches, and thus helps to grasp the preferences for their practical applications. The first of the approaches is illustrated with the example of Model Driven Architecture, and the second – with the example of Reference Model for Open Distributed Processing. Model Driven Architecture (MDA) ([3], [9]) proposed by the Object Management Group (OMG) is a recently emerging vision on system modeling that targets integration of different successful industrial solutions for the system architecture. Reference Model for Open Distributed Processing (RM-ODP) [4] is an ISO and ITU standard presenting general framework for modeling of distributed systems. The scope of applications of MDA and RM-ODP and their goals are similar. Particularly, MDA deals with “full lifecycle integration and interoperability of enterprise systems comprised of software, hardware, humans, and business practices” [3]; and RM-ODP considers lifecycle of distributed systems from enterprise, information, computational, engineering and technology viewpoints. Both MDA and RM-ODP present ontologies for system modeling (for details see [2] and [10] correspondingly). However from the ontological engineering point of view, these frameworks employ two fundamentally different approaches.