TOWARDS A META MODEL FOR BUSINESS PROCESS
CONCEPTS
Boriana Rukanova, Mehmet N. Aydin, Kees van Slooten, Robert A. Stegwee
Department of Buisness Information Syatems, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands
Keywords: standard, business process, ontology, meta model, modelling technique
Abstract: Information systems involved in automating parts of a business process need to be process-aware, in order
to become an integral part of it. Before automation is achieved, the part of the business process to be
automated needs to be made explicit and then operationalized. Business process models could be used to
make the process explicit. Domain standards could be used to make it operational. However, there is no
approach available to evaluate to what extent both the chosen modelling technique and standard are able to
cover the actual requirements of the business process. A meta model for business process concepts can help
for such an evaluation. Although there have been attempts to identify business process concepts and to
create a meta model of business process concepts, the current studies do not include an explicit approach on
how to identify these concepts. Further, how to construct such a meta model and how to include new
elements to it remains implicit. This paper presents an approach on how to construct a meta model for
business process concepts. The approach defines how to capture and define business process concepts, how
to construct a meta model using these concepts and how to extend the meta model. The paper also illustrates
how to apply the approach. The actual construction of the meta model for business process concepts is a
subject of further research.
1 INTRODUCTION
When two or more computerized information
systems are involved in automating parts of a
business process, they need to be process-aware in
order to handle complex communication. If we want
to use computer systems to support part of a
business process, this requires the disparate
information systems to be able to express, and
especially interpret a broader range of meanings.
Thus, the applications that support the business
processes, not only have to be able to express what
needs to be said, but also to interpret it, and act upon
this interpretation in an intelligent manner. Thus the
context of the business process needs to be captured
and made operational. One possibility to make this
context operational and embed it in the system is by
using domain standards.
EDI standards promised significant advantages
in facilitating the exchange between business
partners, reducing errors, increasing speed, cutting
cost, and building in competitive advantage (Sokol,
1995; Damsgaardn, 2000). However, the EDI
standards failed to support the complex business
communication. Furthermore, the focus of many IS
professionals on EDI was how to provide technical
tools, rather than to support the way people do
business (Huang, 1998). New standards, which
strive to allow for interoperability between disparate
systems, are currently developed. These standards
try to capture the context of a business process, in
order to allow for meaningful communication (for
example, in the healthcare domain two such
standards are HL7 standard for clinical data
interchange and DICOM for digital images).
Standard development organizations or consortia of
companies develop such standards, based on their
specific interpretation of the domain. However, in
order to have value for a particular business process,
a standard needs to be linked to a particular
situation, which might be different from what the
standard developers had in mind. Thus, the standard
needs to be evaluated (for a specific business
process) whether it can cover the context (of the
particular business process), which needs to be
embedded in the system.
The context of a business process might be to a
large extent implicit. Thus, before making it
operational (by using standards), it first needs to be
made explicit. Models could be used to capture and
373
Rukanova B., N. Aydin M., van Slooten K. and A. Stegwee R. (2005).
TOWARDS A META MODEL FOR BUSINESS PROCESS CONCEPTS.
In Proceedings of the Seventh International Conference on Enterprise Information Systems, pages 373-377
DOI: 10.5220/0002510703730377
Copyright
c
SciTePress
to make this communication context explicit. A
commonly accepted way of modelling does not exist
(Wand, 1989). However, the modelling power of
various methodologies, their weaknesses and
strengths, can be analysed in terms of ontologically
based common set of constructs (Wand et al, 1989;
Söderström et al., 2002). As it is well known
(Brinkkemper, 1989), a meta model can help to
make such evaluation. Further, it has been argued
that meta models based on ontological constructs
can be very valuable in such evaluation (Rosemann
& Green, 2002). As the specific business situation,
the modelling technique and the standard have in
common that they all address some elements of a
business process, a meta model for business process
concepts can be a common ground for comparison.
However, to our knowledge, a complete meta model
which can capture key business process concepts is
currently not available.
Although we need a meta model for business
process concepts, due to the complex nature of
business processes, such a meta model will be
difficult to construct without following a well-
defined method. Thus, within this paper we focus on
developing a method how to construct such a meta
model. For illustration purpose, we will show how
the method can be used.
The remaining part of the paper is structured as
follows. In Part two, we discuss the possible
approaches for the method construction. In part
three, we describe the method concerning the
construction of the meta model. We illustrate the use
of the method in part four and we end this paper
with conclusions.
2 APPROACH
We have identified three different strategies that can
be followed to arrive at a meta model. The first one
is to take an existing ontology as a starting point,
build a meta model based on this ontology and use
the meta model to evaluate the capabilities of
existing methods, tools, techniques (MTTs) for
business process modeling. The benefit of this
approach is that the concepts are formally defined
within the ontology. Although some work is done in
that direction (see Rosemann & Green, 2002), a
complete meta model for business process concepts
is not provided. Furthermore, some of the limitations
of starting from an ontology are that the robustness
of the ontology and its practical usefulness can be
questionable (Green & Rosemann, 2000; Wand &
Weber, 2002).
The second strategy is to start by identifying the
essential elements of business processes. This can be
done using two types of sources: 1) existing business
process definitions (e.g. Davenport, 1993) and 2)
MTTs for business process modeling (e.g. IDEF0,
REAL). After such elements are identified, they can
be linked within a meta model. The advantage of
such an approach is that the elements identified in
this way are specific to business processes. There is
existing research in that direction (see Lin et al.,
2002; Mayer et al., 2000). This approach can lead to
ambiguities and misinterpretations, since usually the
business process elements are not formally defined.
A third strategy for constructing the meta model
is to follow a hybrid approach- using both an
ontology and MTTs. The advantage of such an
approach is that the business process definitions and
the MTTs will provide essential business process
elements and the ontology can provide formal
definition of these elements. This is the strategy that
we will follow as well. In literature there is an
attempt to provide a meta model for business
process concepts following the hybrid strategy (see
Söderström et al, 2002). However the authors did
not provide clear definitions of the concepts used,
what steps they followed to arrive at that meta model
and how to add new concepts to the meta model. To
overcome such limitations, we use the hybrid
approach in combination with well-defined
procedure to arrive at the meta model. This is further
described in the section below.
3 METHOD
The method to construct a meta model for business
process concepts is schematically represented in
Figure 1. The first step in the approach is the
identification of business process elements from
existing business process definitions. The second
step is to take an existing ontology and to check
whether it is possible to express the business process
elements using concepts from that ontology. If yes,
we include those concepts from the ontology, which
we use to express the business process elements in
what we call “ontology business process concept
base or (OBPCB)”. In this way we include only the
relevant concepts from the ontology.
ICEIS 2005 - INFORMATION SYSTEMS ANALYSIS AND SPECIFICATION
374
MTTs
Definitions of a
business process
Key elements and
relationships
Element 1
Element 2
…….
Element N
Ontology
Concept 1
Concept 2
…….
Concept N
Ontology
B
usiness process
concepts base
(OBPCB)
Ontology
B
usiness process
concepts
extension base
(OBPCEB)
Element can be defined
using ontology
concepts
Ontology extension
definition
Concept N+1
Concept N+2
…….
Concept N+ M
Element can not be
defined using ontology
concepts
Business process
concepts base (BPCB)
Meta model of
business process
concepts
STEP 1 STEP 2 STEP 3
Figure 1: Method for construction of a meta model for business process concepts.
If a business process element cannot be expressed
using concepts from the ontology, then new
concepts are defined in what we call “ontology
extension definition”. After that the newly defined
concepts are included in the “ontology business
process concept extension base (OBPCEB)”. The
“business process concept base (BPCB)” includes all
the concepts from the “ontology business process
concept base” and “ontology business process
concept extension base”. Finally in step three, the
“business process concept base” is used as a basis
for the meta model construction. The whole process
is then repeated for a number of MTTs.
4 ILLUSTRATION OF THE USE
OF THE METHOD
For the identification of business process elements,
we have used a number of business process
definitions, as well as one MTT (ARIS). We have
chosen to use the FRISCO ontology (see Falkenberg
et al., 1998) for defining the business process
concepts. The choice of FRISCO is pragmatic, due
to the familiarity of the authors with that ontology.
Bunge-Wand-Weber ontology can be an alternative.
Within this paper we will not provide the definitions
of the FRISCO concepts. They can be found in the
FRISCO report.
We apply the method described in the previous
section to the definitions, as well as to ARIS. Some
examples of business process definitions that we will
use for the illustration are presented below.
Examples of business process definitions
1) A business process is “a specific ordering o
f
work activities, across time and place, with a beginning,
an end and clearly identified inputs and outputs
”
(Davenport 1993)
2) A business process consists of activities ordere
d
in a structured way with the purpose to provide valuable
results to the customer (Ågerfalk, Goldkuhl et al., 1999)
3) A business process can be viewed as a
s
tructure
of activities designed for action with focus on the en
d
customer (Stock and Lambert 2001)
Figure 2: Examples of definitions.
The outcome of the application of the method is
schematically represented in the figure below. The
figure illustrates that based on the business process
definitions, a number of business process elements
were identified. After that, by looking at the
FRISCO concepts and how they are defined, we
have selected those elements that seemed to be able
to cover the elements presented in the definitions.
TOWARDS A META MODEL FOR BUSINESS PROCESS CONCEPTS
375
Elements from definitions
Activity
Structured order of activity
Business process
Beginning
End
Time
Input
Purpose
Valuable result
Customer D1++
FRISCO BPCB
Action
State-transition structure
Composite action
Pre-state
Post-state
Time (relative, absolute)
Input actand
Goal
Output actand
Actor
Basic state transitions (sequence,
choice, concurrency)
1
ARIS concepts
Function
Function flow
Implicit
Implicit ( initial event+ input of a function
Implicit ( initial event+ output of a function
Time (relative)
A1++
Goal
A2++
A3++
Connector (or, and)
FRISCO BPCEB
Specialization of FRISCO actor:
D1++ customer
….A3+++ Organizational unit
Specialization of FRISCO input actand
A1++
Information services, Other services
Material output, Financial services, Human output, Machine
resources, Computer hardware,
Software, Environmental
Specialization of FRISCO input actand
A2++
Information services, Other services, Material output, Financial
services, Result event
2
Figure 3: Illustration of the use of the method.
When reviewing the ARIS concepts, we have
identified that there is a concept connector (or, and),
which seems closely related to what FRISCO refers
to as “basic state-transition structure”. As a result,
we have included that concept in the OBPCB.
If we look at the elements from definition box,
we will see that there is an element D1++ and when
we look at the ARIS concepts box we will see
elements A (number)++. This means that the
concept defined in the definitions or in ARIS can fall
under a general category of FRISCO, however it can
also be included in the BPCEB as a specialization of
the general concept. This would provide for richness
of the business process concepts. That is why, we
have included all the element that were marked with
++ in the BPCEB. Within this paper we will not
formally define these new concepts, however they
can be defined as specializations of already defined
FRISCO concepts.
5 CONCLUSIONS
We stared this paper with the problem, that we need
to identify to what extent a modelling technique and
a standard are capable to capture the business
process requirements of given situation. We have
further argued that a meta model for business
process concepts can be useful for such an
evaluation.
Within this paper, we have outlined a method
how to construct such a meta model, we have
provided an illustration of how to use the method
and we have arrived at initial set of business process
concepts.
Although the further elaboration of the business
process concepts and the construction of the meta
model will be subject of further research, we
consider that the method described in this paper is a
necessary step towards achieving the final goal.
REFERENCES
Brinkkemper, S., M. Geurts, et al. (1989). On a Formal
Approach to the Methodology of Information
Planning. First Dutch Conference on Information
Systems, Amersfoort.
Damsgaard J., Truex, D. (2000), Binary Trading
Relationships and the Limits of EDI Standards: The
Procrustean Bed of Standards. European Journal of
Information Systems, vol. 9, no 3, pp. 173-188.
ICEIS 2005 - INFORMATION SYSTEMS ANALYSIS AND SPECIFICATION
376
Davenport, T. H. (1993). Process innovation.
Reengineering work through information technology.
Boston, Harvard Business School Press.
Ellison, M. and G. M. McGrath (1998). "Recording and
Analyzing Business Processes: an Activity Theory
Based Approach." The Australian Computer Journal
30(4): pp. 146-152.
Falkenberg, E. D. et al. (1998), A Framework of
Information System Concepts: The FRISCO Report,
IFIP (1998), available on- line at:
http://www.liacs.nl/~verrynst/frisco.html
Green, P. and M. Rosemann (2000). "Integrated process
modeling: an ontological evaluation." Information
Systems 25(2): pp. 73-87.
Huang K. (1998), Organizational Aspects of EDI: a Norm-
oriented Approach, PhD thesis, ISBN 90-3651095-3
Lin, F. R., M. C. Yang, et al. (2002). "A Generic Structure
for Business Process Modeling." Business Process
Management Journal 8(1): 19-41.
Luttinghuis, P. O., M. Lankhorst, et al. (2001).
"Visualizing Buisness Processes." Computer
Languages 27: pp. 39-59.
Mayer, R. J., P. C. Benjamin, et al. (2000). Methods for
Business Process Reengineering. Process Think:
Winning Perspectives for Business Change in the
Information Age. V. Grover and W. Kettinger.
Hershey, Idea Group Publishing: 150-189.
Rosemann, M. and P. Green (2002). "Developing a Meta
Model for the Bunge-Wand-Weber Ontological
Constructs." Information Systems 27: 75-91.
Sheer, A. W. (1999). ARIS- Business Process Framework.
Heidelberg, Springer-Verlag Berlin.
Sokol, P. K. (1995), From EDI to Electronic Comemrce- a
Business Initiative, McGraw-Hill, Inc., New York,
ISBN 0-07-059512-7.
Söderström, E., B. Andersson, et al. (2002). Towards a
Framework for Comparing Process Modeling
Languages. CAiSE 2002, Springer- Verlag Berlin
Heidelberg, 2002.
Wand, Y. and R. Weber (2002). "Research Commentary:
Information Systems and Conceptyal Modeling- A
Research Agenda." Information Systems Research
13(4): 363-376.
TOWARDS A META MODEL FOR BUSINESS PROCESS CONCEPTS
377
