The TROPIC Project

Model-Driven Engineering (MDE) places models as first-class artefacts throughout the software lifecycle. The main promise of MDE is to raise the level of abstraction from technology and platform-specific concepts to platform-independent and computation-independent modelling. To fulfil this promise, the availability of appropriate model transformation languages is the crucial factor, since transformation languages are for MDE as important as compilers are for high-level programming languages. The basic idea of model transformation is that a source model M_a, which conforms to a metamodel MM_a describing its abstract syntax, is transformed into a target model b, which should conform to the metamodel MM_b. For this, the potentially heterogeneous modelling oncepts defined by the metamodels have to be mapped to each other and, according to this mapping, he actual transformation of model M_a into model M_b is executed by means of a transformation engine.

Although numerous model transformation languages have been already proposed, currently no transformation language, not even the QVT (Query/View/Transformation) standard of the OMG, became accepted as the state-of-the-art model transformation language, i.e., an adoption in practice has not yet been achieved.

The current challenges are

• Abstraction mechanisms: Appropriate abstraction mechanisms are required to deal with the complexity of overcoming structural heterogeneities between different metamodels. 

• Reuse mechanisms: Suitable reuse mechanisms are needed in order to reduce the high and error-prone effort of specifying recurring transformations

• Explicit runtime model: A seamless integration between specification and execution of model transformations is essential in order to allow for appropriate debugging facilities

The overall goal of this project is to provide a framework for developing model transformations, in order to resolve structural heterogeneities between metamodels which fulfills the mentioned requirements.