- Business process modeling is
a complex, time consuming and error prone task. Model design
requires determining the activities that need to be performed,
ordering of their execution, handling exceptional cases that
might occur, etc. In many cases, variants of process models
need to be created in respond to special business situations. For
instance, there could several insurance claim handling process.
One process is designed for people with special working
environment conditions, another for people over seventy years,
etc. With the rapid growth in the number of process models
developed by different process designers, providing business
process designers with a framework for reusing previously
designed business process models and for making the best use
of them is of great practical value.
-
In order to simplify
and improve the business process modeling task, process
models need to be highly reusable, favoring process flexibility
and minimizing designs made from scratch. Moreover, reusing
can effectively increase the quality and the maturity of the
newly developed process models.
- QBP (Querying Business Process)
is a framework which aims of supportting business process designers to achieve an
effective modeling task by querying and reusing existing process
models. In particular, we summarize the current strengths
of our system as follows:
- The framework is based on a novel, intuitive and visual query language for business process models, BPMN-Q.
It allows users to define their business process models and their queries using a very similar set of notations. Moreover, it supports the notion of paths between nodes in the process graph and supports different way of navigation through the process structures to answer the user queries.
- The framework is enhanced with a semantic query expansion component of BPMN-Q queries. This component
employs an ontological dimension in the query matching
process and tackles the problem of applying different
terminologies when modeling processes. Ontology construction
does not assume a priori semantic tagging or
semantic description of process models.
- To achieve an efficient and scalable performance, the
SQL-backend query processor makes use of the robust
indexing infrastructure typically available by RDBMS
and uses a novel decomposition-based and selectivityaware
translation mechanism of business process graph
queries into efficient SQL scripts.
- The framework architecture is designed in a very flexible
front-end/back-end fashion. On the front-end, it uses
BPMN for modeling business processes, BPMN-Q for
querying business processes and standard SQL relational
processor as its back-end. However, the proposed framework
can be easily adapted to other modeling notations
such as: EPCs, UML ADs.
- Please click here for a live demo