Supporting system model verification using ontologies

OData support
Supervisor:
Dr. Szatmári Zoltán
Department of Measurement and Information Systems

Model driven development (MDD) is becoming popular in the field of system engineering. The success of this method is based on the strict separation of functional description from implementation details. In a typical MDD development process first, the designer creates the portable, reusable and easily modifiable functional description of the system captured by the platform independent model (PIM). As the second step, implementation and deployment specific details are defined in the platform specific model (PSM) derived from the PIM. Finally, product artifacts (source code, documentation, configuration descriptors, etc.) are automatically generated from the PSM.

The correctness of the PSM depends on the correctness of the PIM, which is ultimately based on the requirements and specification of the system. The costs for the correction of design errors are significantly higher if detected in later phases of the development process. Due to this reason it is highly advantageous to support the precise definition of requirements against the system and support their automatic validation.

In the current paper, I extend the model driven development paradigm with the automatic derivation of PIM models along with their structural constraints from requirement descriptions. For capturing these requirements, I used description logic based ontology, specified in the W3C recommendation [1], as it supports the design of structures and constraints [2] with the liberty of open world assumption and a rich expressivity close to human mind. From these ontology based requirements, I specified a mapping to EMF [3] the de facto industrial standard in metamodeling for describing platform independent models.

This mapping is integrated into a transformation framework that allows to automatically derive a preliminary platform independent model along with its structural constraints defined in the ontology based requirements. As a result, the constraint violations described in the requirements can be directly [4] validated on top of the platform independent model, allowing early and automatic error detection.

References:

1. OWL 2 Web Ontology Language. (2009) URL: http://www.w3.org/TR/owl2-overview/

2. Dobson, G.; Hall, S.; Kotonya, G.: A Domain-Independent Ontology for Non-Functional Requirements. IEEE International Conference on e-Business Engineering, 563-566. (2007)

3. EMF: Eclipse Modeling Framework, http://www.eclipse.org/emf

4. Bergmann, G., et. Al.: Incremental Model Queries over EMF Models. Model Driven Engineering Languages and Systems, 13th International Conference, MODELS'10: Springer, 2010

Downloads

Please sign in to download the files of this thesis.