A Trace Metamodel Proposal Based on the Model Driven Architecture Framework for the Traceability of User Requirements in Data Warehouses - Advanced Information Systems Engineering - page 131

Fig. 7. Generic relationships between CIM elements and PIM Elements
the DW. In turn, this information requirement is decomposed into contexts and
measures. At the right hand of the figure, the corresponding multidimensional
elements from the PIM appear. As aforementioned, the business process is re-
lated to the fact (and the corresponding packages), whereas the first context is
related to the dimension package, the dimension, the base level and its associ-
tation with the dimension. On the other hand, the second context is associated
with the second level in the dimension. With this approach, if we wish to check
the result of the transformations for debugging, we can check which rule created
each element with the information stored in the traces. In addition, these traces
allow us to keep track of which elements in the PIM model match with each
requirement in the CIM model, making requirements validation easier. More-
over, if any requirement is changed, we know which elements are affected and
which rule created them, being able to execute the corresponding rule of the
transformation to regenerate the affected part of the PIM.
Although in this section we have presented the generation of traces from a
goal-based CIM, the traces could be used to trace any element in a different CIM
model from another proposal, as long as it has a unique reference identifier.
5 Example of Application
In this section, we will present an example of application for our traceability
proposal, showing how the traces can be navigated to retrieve useful information.
A University wishes to build a DW in order to analyze the factors which
influence the performance of the students. This university has a transactional
database created for managing the information about professors, degrees, sub-
jects and students, which will serve as data source for the data warehouse. The
analysts wish to analyze the students grades by subject , professor who teaches
them, and student ,takingintoaccounthowmany hours they spend studying