Conclusion - Model-Driven and Software Product Line Engineering - page 247

about how transformation rules must be composed to derive
fine-grained configured products. Given that our approach
takes into account fine-grained variations and fine-grained
configurations, it also copes with the derivation of fine-grained
configured products. Our decision models have been presented
in [ARB 09a, ARB 09b, ARB 08].
9.2.5. Product derivation
Based on our decision models, we propose a mechanism for
selecting transformation rules and modifying their execution
ordering according to selected variants. In our current
implementation, we have used the model transformation
engine of oAW to execute model transformation workflows
derived from our decision models. Our decision models,
however, are independent of model transformation languages
and can be used to support product derivation in contexts
different from the oAW context. For instance, currently, we
explore how our decision models can be used to derive products
by using theATL language and its facilities for transformation
rules composition [ROM 09]. Further work on this field is part
of our future work. Our mechanism for product derivation has
been presented in [ARB 09a, ARB 08].
9.2.6. Comparison table
Table 9.1 presents a summary of this section taking into
account our approach and the related approaches. This table
summarizes the comparaison of MD-SPL approaches (the
rows). The lines represent the various characteristics we
previouslydiscussed(yes/nomeaningpresence/absenceandn/a
for not applicable).
We believe that our MDE mechanisms are scalable to
traditional SPL engineering where models are used only as
artifacts for documentation. Currently, MDE is being used not
only in academia but also in industry. Several international
events,journals,and research projects are concerned about the