My research centres on enterprise modelling to support IS
development and operations. I focus on enterprise models that
integrate multiple stakeholder perspectives tightly and which may
embed multiple partial models for different uses. My projects in this
area range from theoretical studies through method- and
tool-development to industrial case studies and consultancies. While I
consider theoretical studies to be the core of my research, I use the
applied projects (a) to gain experience with practical enterprise
modelling in real-life settings, (b) to develop my theoretical
frameworks further, (c) to attract more practically-minded research
students and (d) to acquire funding.
General Motivation and Goals
Enterprise modelling and enterprise models of various kinds have
become widespread in a wide range of areas that are crucial to the
effectiveness and efficiency of modern enterprises, both within
information systems development and evolution and in a wider range of
areas which are not necessarily related to information
systems. Although isolated applications of various kinds of enterprise
models is a hot research topic across a range of disciplines, few
researchers have been developing frameworks for enterprise modelling
that are sufficiently general to accommodate multiple uses of the same
enterprise model or parts thereof. Providing the concepts, theories
and tools for such frameworks has many benefits, including higher
return on model investment, facilitating organisational learning, more
frequent model updates, increased model reliability and completeness,
rapid accommodation of new applications of existing enterprise models
through reuse. Multi-usage enterprise modelling also provides the
machinery needed to accommodate multiple stakeholder perspectives
within a single enterprise model, thereby facilitating
individualisation of large and possibly complex enterprise
models. Finally, multi-perspective multi-usage enterprise models
implicitly allow modelling constructs from different partial languages
to be combined in new ways as needed. As a consequence, elements from
different partial models can also be combined at will, thus providing
richer enterprise modelling languages.
Facet modelling
Facet modelling is an approach to multi-perspective multi-usage
enterprise modelling created to meet the requirements stated in the
previous section. The facet modelling framework is open-ended and
easily extensible. It supports tightly integrated representation of
problem domains from multiple perspectives simultaneously and it
supports integration of partial models with different uses within the
same facet model. The foundation of this work includes a metametamodel
of 32 basic facet-modelling concepts and relationships.
Experience from a realistically sized case study has been
reported. Work in progress includes a semi-formal facet language based
on an extensive and integrated process, agent, rule and information
modelling language (PPP.) Work is also underway on a formalisation
using set theory. I also plan to define facet languages which
incorporate the OPEN Modelling Language (OML); the Framework of
Information System Concepts (FRISCO); and the Unified Modeling
Language (UML).
Ontological analysis and evaluation of information systems modelling languages
Early attempts to apply facet modelling made it clear that tightly
integrated multi-perspective multi-usage modelling is difficult
without an elaborate and solid semantical foundation. I have therefore
recently aligned the ontological assumptions behind facet modelling
with the more elaborate Bunge-Wand-Weber (BWW) model of information
systems. The BWW-model now supports definition of facet languages as
follows: (1) Modelling constructs are first interpreted and defined
precisely in relation to the BWW-model. (2) By inspecting their
BWW-definitions, semantical overlaps between modelling constructs can
be identified at a fine level of granularity. (3) The semantical
overlaps are finally resolved by re-formulating the modelling
constructs as parts of a facet modelling language. The combined
approach has already been used in the analysis and re-formulation of
PPP. Steps (1) and (2) above have even been undertaken for OML and
FRISCO. An ontological analysis and evaluation of the Unified Modeling
Language (UML) is currently in preparation. Another thread of
ontological research looks specifically at the whole-part construct as
it is used in OO modelling. While being an important issue within
OO-modelling semantics, clearly defined whole-part relations are also
a prerequisite for proper integration of perspectives in facet
models. (Different perspectives are often expressed at different
levels of granularity.)
Problem domain representation during requirements engineering
Facet modelling was initially developed to support
multi-perspective modelling during RE and to support tight integration
of RE modelling languages. The earliest papers on facet modelling
therefore use RE as the application domain. In particular, facet
models support progress towards agreement because inconsistencies and
conflicts between stakeholders' views can be readily identified when a
tightly integrated and precisely defined modelling language is used. A
more recent thread of research in modelling languages for RE deal with
elicitation of security requirements using scenarios and a variant of
the Use Case notation.
Representation, Assessment and Improvement of information Systems Architectures
RAISA ("Representation, Assessment and Improvement of Information
Systems Architectures") comprises a position, a method, a
representation framework, a modelling tool and an alignment model for
managing how an enterprise organises its information processes and
information resources. For example, the RAISA methodology addresses to
what extent and according to which principles an enterprise should
centralise or decentralise its applications, databases etc. along with
the associated responsibilities for development, maintenance and
such. RAISA is documented on the web and is currently being written
out in paper form.
Past research interests and projects
- Information systems performance engineering
- Conceptual models for requirements engineering
- Visual programming
Last updated 2005.06.11 by
Andreas.Opdahl@uib.no.