Abstract:
Researchers and developers in the field of software engineering are always
looking for ways to attain the highest degree of efficiency and quality possible
in software production processes. Today, Method Engineering (ME) is being
pursued in order to achieve this goal.
Requirements Engineering (RE) plays a vital role in method engineering just as
in software engineering processes. Despite the tremendous efforts in improving
the processes of software production, RE practices are weaker than other
practices in the context of software engineering. Regarding this issue and the
relative youth of method engineering, RE practices in method engineering are
also weaker than other practices.
Most research in the field of method engineering includes preparation of
detailed repositories of useful method fragments and providing a way for
selecting the appropriate method fragments and connecting these fragments based
on the requirements. Furthermore, although all of these activities are based on
specific requirements, few explicit ways have been presented to extract the
requirements of a new methodology.
Due to the correspondence and analogy between software engineering and
methodology engineering, it can be concluded that interest has been mostly
focused on methodology design and implementation (even a variety of tools have
been provided); analysis, on the other hand, remains weak. This issue will
create a great amount of risk for the software production process, because
software production using an inappropriate methodology or without compliance
with the requirements is bound to cause confusion and directionlessness in the
middle of the production process.
The aim of this project is to provide an appropriate framework for describing
and extracting the requirements in the context of methodology engineering.
Because of relative immaturity of these areas, appropriate research approaches
are required. In order to achieve this objective, this project will use two
approaches to study and provide the target framework. The two proposed
approaches are not necessarily alternatives, but complementary.
-
In order to evaluate RE methods, they can be categorized
according to their philosophy. Regarding the idea that
"Software processes are software too", the first idea that
comes to mind is that the RE methods used in software
engineering could be transitioned into the context of
methodology engineering. Some of these methods may not be
applicable in the context of methodology engineering, and
will therefore be put aside; others will have to be adapted
for use in the new context.
-
Another approach is based on the notion that method
engineering needs specific RE practices according to its own
characteristics. This raises the idea of ME-specific RE
practices.
After defining the framework, it will be applied to a case study
in order to evaluate its applicability.
|
 |