Information Technology Reference
In-Depth Information
-*
"#
'(%!$
"##!!
" $
!
)*
)*
)*
"#$
&
!
"#$%
!'(%
!.
)*
,*
,*
,*
!
!
+*
!
#
$
$
+*
+*
!"
%
Fig. 3. A lane departure warning system is modeled with the design stream
func-
tional development and architecture
(illustrated by the bold solid lines). Using infer-
ence techniques, product and process requirements can be reasoned automatically
for the
safety measures
stream (see the bold dashed lines)
4.
Workflow model:
The inference on the formalized process model is used
to derive a custom tailored workflow model for the individual developer
in the development stream
functional development and architecture
. The
tailored workflow model is further transformed into HTML documenta-
tion (more thoroughly described in [4]) and a BPEL (Business Process
Execution Language
5
) workflow which can be instantiated. In the BPEL
workflow, a
sequence
of notification services is
invoked
on the basis of de-
rived ISO requirements and related methods. BPEL workflow monitoring
can be used to track the progress of a workflow instance.
Regarding the implementation of these steps in a prototype toolchain, the
transformation and reasoning steps have been implemented in Java, using
the semantic reasoner
Pellet
6
and the
OWL API
7
. Toolchain integration is
built on top of the
Apache Ant
build tool
8
.
4
Lessons Learned from Applying Semantic Reasoning
On a conceptual level, OWL ontologies offer a convenient way for formally
capturing and analyzing domain knowledge. The ability of defining descrip-
5
BPEL - http://www.oasis-open.org/committees/wsbpel/
6
Pellet - http://clarkparsia.com/pellet/
7
OWL API - http://owlapi.sourceforge.net/
8
Apache Ant - http://ant.apache.org/
