For Flexibility and Adaptability
Marc Rabaey, Eddy Vandijck
MOSI,Vrije Universiteit Brussel,Pleinlaan 2, Brussels, Belgium
Koenraad Vandenborre, Herman Tromp
Software Lab Ghislain Hoffman,Ghent University, Ghent, Belgium
Martin Timmerman
Information Technology,Royal Military Academy, Brussels, Belgium
Keywords: Business processes, embedded information systems, information, strategy, enterprise architecture,
Abstract: In this ever faster changing world, organisations are faced with the need to have flexible processes. This is
only possible if these processes have full control over their supporting information systems, which we
propose to embed into the business processes, in stead of a global, common enterprise information system.
Therefore we are introducing a fifth stage in the architecture maturity model of Ross to implement these
Embedded Information Systems.
Organisations need to continuously adapt to
changing business environments. However, the
justified desire for a flexible business processes
support is impossible without adaptive software.
In this paper we describe a proposal for a
business process embedded information system.
First, we will look at information and describe how
it relates to the business. Information is not only
needed at operational level but it also provides raw
material for intelligence in the decision process.
Next we will discuss the enterprise architecture
where all components (information, business
processes and Information and Communication
Technology) are brought together in one framework
and we state our proposal. Finally, before some
conclusions we give indications for further research.
The U.S. Office of Management and Budget (OMB)
defines information as any communication or
representation of knowledge such as facts, data, or
opinions in any medium or form, including textual,
numerical, graphic, cartographic, narrative, or
audiovisual forms (OMB A-130). Knowledge itself
is the awareness and understanding of information
gained through experience or observation or through
reasoning (deduction and induction).
Burkan (1991) states that information has four
Compulsion causes the first, tactical survival of
the enterprise is the concern of the second. Much of
the attention regarding information is given to these
functions in Information and Communication
Technology (ICT) less to the latter two. The
strategic management use of information reflects to
minimise surprises. Its effectiveness is measured in
the ability to detect and anticipate, both short term
and long term. Contrary to control, which is
inwardly focused, strategic management is more
outwardly focused. If strategic management is the
ability to see the future, then leadership is the ability
Rabaey M., Vandijck E., Vandenborre K., Tromp H. and Timmerman M. (2006).
In Proceedings of the Eighth International Conference on Enterprise Information Systems - DISI, pages 144-150
DOI: 10.5220/0002464501440150
to shape the future. It is measured by how well the
leaders can move the enterprise's resources to new
directions (Burkan, 1991).
Related to this and before discussing these
business processes, Art of War (Bernard, 1976) is
briefly introduced, because its principles prescribe
the way of doing business.
3.1 Art of War
To Bernard (1976), there are three principles of the
Art of War:
Balance between Goals and Means;
Liberty of Action;
Economy of Forces.
The first principle has one rule: the permanent
seeking of intelligence, inside and outside the
organisation (see below). The balance between goals
and means is about determination of the right
objectives, given the environment and the available
and/or possible needed resources related to these
objectives. The result is the Grand Strategy of the
entire organisation. As a consequence of this
balance, two other strategies can be derived:
Business and Resources Strategies. The first is
focused on the creation and the deployment of the
core competences to attain the imposed objectives,
while the latter is focused on the means and the
processes to support the first (Rabaey, 2004, 2004b).
The liberty of action is about security: avoiding,
preventing as much as possible hostile actions of
other organisations and the assuring of the
communication lines (logistics, information (for
intelligence)). The economy of forces treats the
economical and right use of the resources (efficiency
and effectiveness).
The deeper in the hierarchy of the organisation
the less impact have the leaders on the resources
aspects and the scope of their business levels.
Therefore in the ever faster changing world, the
structure becomes less hierarchical, so flexibility is
gained. This implies that these leaders (and/or
managers) need to have access to more information
and have a more extended information system.
3.2 Business Processes
An information system supports the organisation and
consequently its business processes. Rabaey (2004a)
defines a business process as a logical set of
activities that consumes resources to attain its
Now, information in the processes is used on
different levels: steering information, operative
information and historical information. These types
of information will be found in BPM-tools. The last
years we can see an effort to merge BPM-aspects
and ICT-aspects into one tool.
However a mental gap exists between both
worlds. BPM-people think process-wise, while ICT-
people think application-wise, meaning only some
parts of the processes are automated and relevant
information are stored in databases. Reporting on
transactions is based on the information kept in this
databases. As a matter of fact, only historical
information is kept in these databases. Information
about steering and executing the processes are not
stored in databases, but in the models of automated
workflows, only a data-query cannot be performed
on these models.
Automating partly or totally a business process
was first done by a workflow system. Kobielus
(1997) defines a workflow as the flow of
information and control in a business process (p.
For the leaders and managers of organisations, it
is useful to be able to query these business processes
and workflows for a better control.
3.3 Business Integration
In the era of merges and virtual organisations
(McHugh, 1995), business integration becomes
important. Business processes and workflows now
span multiple organisations. They start with the
expression of the needs of a customer and end with
the satisfaction of his/her need (and the
administration). Different levels of integration are
possible: Strategy, strategic, operational and
operative (Rabaey, 2005c).
Strategy Integration is when two or more
organisations decide to develop and to adopt one
global strategy for all the concerned organisations.
So everything is integrated and the multiple
organisations are seen as one.
If the common interest of two or more parties is
of such a strategic importance that one or more
business processes are integrated and possibly
business units are integrated, the term Strategic
Integration is used. Its main characteristics are the
set up of common business and resources strategies
and by consequence a common information
management, next to the individual ones of the
involved parties.
Within Operational Integration, no common
business or resources strategies are set up. From the
respective individual strategies, a common
Operational Strategy is deduced. Neither a common
information management is established, only
conventions are defined.
Operative Integration is, looked upon from a
strategic point of view, the weakest of the possible
forms of integration. In this case, no common
strategy and even no common Operational Strategy
are formed. The impact on ICT is that for the
realisation of the integration merely interaction
standards between business processes have to be
defined. For these purposes, no information
management has to be defined, a merely commercial
contract between two or more parties is sufficient.
3.4 Strategy and Capabilities
Kaplan (2005) states that strategy at many
companies is almost completely disconnected from
execution. Therefore Kaplan (2005) proposes a new,
dedicated unit to orchestrate both to help the bridge
the divide. However the military deployment
(communication) of strategy through operation
orders exists since ancient times (Bernard, 1976,
Lidell Hart, 1991, Rabaey, 2004a, 2004b).
Nevertheless, the implementation of the balanced
scorecard (Kaplan, 1996) in the enterprise has
enabled the deployment and feedback of the
strategy. Kaplan (2005) gives an example of the US
Army, which applied the balanced scorecard for the
deployment and communication of the strategy (and
for the feedback). This alignment on the civil
balanced scorecard system enabled the
“demilitarisation” of the strategy deployment, so that
civil partners of the US Army can now better
understand the military context.
Rabaey (2005b) proposes a new type of balanced
scorecard to better suite the principles of the Art of
War (Rabaey, 2004a). The Board of an enterprise
would like the enterprise to have some effects on the
society (outcomes) by using its capabilities, which
will then perform actions (output) to obtain these
effects. The sum of these effects is the vision of the
enterprise. The wanted effects will be described in a
number of scenarios. Therefore the CEO will
configure capabilities to perform actions for all
relevant scenarios. In the ever faster changing world,
these capabilities have to be flexible and easy to
Modules deliver the necessary capabilities,
where one module can serve multiple capabilities.
These modules are composed of resources. In the
process area of capabilities generation, modules
and/or resources are acquired following investment
and recruiting plans (acquisition function). So we
have a schema of outcomes – outputs – capabilities –
modules – resources.
3.5 Enterprise Application
With the hype around Business Process
Reengineering (Hammer, 1993) and the raise of the
Internet, more specifically e-Business, priority was
given to improve the efficiency and effectiveness of
the business processes. The processes have to act
and react more agile in the faster world economy.
In a later stage, when simple information
exchange by itself was no longer sufficient, and the
need for more intelligent integration, for instance on
procedures and the semantics of the data, became
clear, a solution was found in Enterprise Application
Integration (EAI). In related research however
(Vandenborre, 2003), it is shown that EAI by itself
is only an end to the means. A more complete
solution is found in the “logical bus architecture”
and its implementation on a technical, organisational
and informational level.
So, EAI was the response to the interconnection
of information system islands between different
functional domains in a company or network of
companies throughout the Internet. In this context
web services became the way to go for most of those
companies (Zimmerman, 2003, Rabaey, 2003).
Now, Enterprise Application Integration (EAI) and
e-Business are evolving towards Service-Oriented
Architecture (SOA) by using Web Technology as
Web Services. The logical bus architecture became
SOA. The main reason is that flexibility can be
obtained. (Rabaey, 2005a).
However information management is very
difficult in these situations, because of the fact that
organisations want to continue to integrate the new
area into their information model, but still want to be
master of the information model even integrated.
The larger the information model, the tougher
become the semantics.
3.6 Information Model
Rebuilding the information model can be started
from the databases systems. However research has
shown that applications contain a lot of meta-data
and data structures. Research by Rever SA (a spin-
off of the Belgian university Notre Dame de la Paix)
shows that 40 to 60% of the data structure and flow
in ICT are in the applications and are thus not in
databases. The University Notre Dame de la Paix is
developing the software DB-Main to do a reversed
engineering by analysing the databases and the
programs. The result is a conceptual model from
which a logical model for any type of database
system can be derived down to the operative
schemes (Hainaut, 2003, Hick, 2003, Henrard,
Since the environment of the organisation is
permanently changing and thus the organisation
also, the information system of the organisation has
to be adapted to the new situation. As a consequence
part of the databases will change, so that after a
while a new re-engineering has to be performed.
So, if a business process could be fully
automated into workflows and it holds itself the
information, then a consistent part of the producible
and needed information will be embedded in the
business process. Referring to the capability
approach, the housekeeping of the information of a
capability is done by the management module.
4.1 Intelligence
Information is not only needed for the business
processes or functions in an organisation (direction
and control of the uses of resources; the
effectiveness of the business). It is certainly the fuel
of the strategy development process (planning future
activities; linking the business with its environment
(Anderson, 1986). If a business process has to
choose itself the needed resources, then it may have
to seek for information to prepare its decision
making. The needed information to reduce the
uncertainty of the decision maker at a level that is
acceptable to him, is called intelligence (Rabaey,
Intelligence is the product of the intelligence
process, which collects, analyses, integrates, and
interprets information. It disseminates the
intelligence to the customer, with the purpose to
reduce the uncertainty on a problem (decision)
and/or to improve the inference rules. And last but
not least, it makes the organisation “aware” of
information and therefore aware of knowledge.
4.2 Decision and Intelligence Process
As mentioned above, collecting intelligence is the
only rule of the first principle of the Art of War
(Balance between resources and objectives). If we
combine the decision process (on any level of the
organisation) with the intelligence process, then a
two way communication is needed. One way to
express the information need from the decision
making process to the intelligence process. Another
is n the opposite direction with the asked
information (pull) or with spontaneously generated
intelligence (push).
The intelligence process will check if the need
can be covered with information in its intelligence
base (see Rabaey 2005a for more details), if not then
it will give its network (sensors) the order to seek for
the relevant information. If it's found then the
assessment process leading eventually to intelligence
will be started.
Rabaey (2005a) proposes a SOA-based solution
–called Intelligence Bus- because of the fact that the
intelligence capability can be formed of modules,
which correspond with business steps in the
intelligence process. This intelligence bus can be
plugged in the management information system of a
business unit (process). As a matter of fact, it can
serve multiple business units, as long as strategic or
operational needs for it exist. Specific security rules
can restrict access to the intelligence bus.
So, small units (related to the strategic mother-
organisation) can have access to a large knowledge
and intelligence bases of the mother-organisation
and nevertheless have their own independent
information system. This results in an enormous
flexibility for these units (resources and
intelligence). Of course, these units provide the
mother-organisation and other units also with facts
and intelligence.
5.1 Framework
Lots of different interpretations of the term
Enterprise Architecture do exist. These
interpretations range from, at one end of the
spectrum, the list of technological choices made in
an organisation concerning infrastructure and
application design to the other end of the spectrum
wherein Enterprise Architecture encompasses these
technological decisions but also sets guidelines to
information architecture and business architecture.
Therefore it is a necessity to consider Enterprise
Architecture broader than merely infrastructure
architecture and application architecture. This
necessity stems from the fact that applications are
build to support business processes and operate on
information gathered through these business
processes (see above). Hence, architecture only
concerned with infrastructure and application design
is insufficient to support a business because such an
Enterprise Architecture has no view on the business
and its dynamics and hence it cannot take
precautions for changing business requirements or
the reuse of certain artefacts in other business
For these reasons, an Enterprise Architecture
consists of distinguished levels. The naming of the
distinguished levels may be different but at least the
general ideas as described hereunder should be part
of the Enterprise Architecture:
z Business Architecture is about the
description of the business processes as
viewed from a business perspective.
z Information Architecture describes the
information need of the business and its
z Application Architecture is about how to
implement the applications or ICT systems,
the programming paradigms and languages,
the development environment, software
documentation guidelines, release to
production procedures, etc.
z Infrastructure Architecture deals with
guidelines concerning hardware platforms,
network infrastructure, operating systems,
Ross (2003, 2004)) defines Enterprise
Architecture as “the organizing logic for
applications, data and infrastructure technologies, as
captured in a set of policies and technical choices,
intended to enable the firm’s business strategy”.
5.2 The Level of Architecture
Ross (2003) defines four stages:
1. Application Silo Stage in which the
Enterprise Architecture is just the collection
of the architectures of isolated applications,
often implemented in different
2. Technology Standardization Stage as the
first step towards an enterprise wide
Enterprise Architecture in which
technology gets standardized and often
centralization is put in place. The
deployment of resources shifts from
application development into the
development of a shared infrastructure.
This phase is further often characterized by
the introduction of data warehouses and
sporadic and not institutionalized business
management participation.
3. Data Rationalization Stage characterized by
an expansion of the enterprise architecture
to include process and data standardization.
The deployment of resources shifts from
application development into data
management and infrastructure
development. The involvement of senior
business managers becomes
institutionalized and a dialogue between
business managers and IT becomes
common practice. Very important in this
phase is the shift of data ownership from IT
towards the business. This phase is further
often supported by tools like ERP, CRM.
4. A Modular Architecture characterized by
enterprise wide global standards with
loosely coupled applications, information
and technology components to preserve the
global standards while enabling local
differences through modules extending the
core processes.
6.1 The Fifth Stage of Enterprise
This paper proposes to add a fifth stage, business
process embedded information systems. Due to the
ever faster changing environment of an organisation
and increasing interactions with it, a global and
central “steering” becomes quite impossible, if the
organisation aims flexibility and rapid response. So
the organisation delegates to the business units and
their processes. However, if the ICT is not federated
then the autonomy is jeopardised by the ICT-
applications. Therefore we propose to embed
information systems into business processes.
As seen with the Intelligence Bus, mostly all
aspects of the information system can be federated
without losing the consistency of the information
system of the mother-organisation. However before
going to the fifth stage, an organisation and its
underlying (business) units have to perform a
information model as described above,
simultaneously the organisational structure has to be
defined, and the business processes have to be
modelled in the Business Process Management
(BPM-) tool following the capability approach.
Once BPM and the conceptual information
model defined, then they have to be merged into a
global model (information and business processes),
where the capabilities should manage their own
(embedded) information system. But as already
mentioned, it has to start with conceptual
information model to solve the problems of
information management.
Since the environment of the organisation is
permanently changing and thus the organisation
also, the information system of the organisation has
to be adapted to the new situation. As a consequence
part of the databases will change, so that after a
while a new re-engineering has to be performed at
organisational level.
So, if a business process could be fully
automated and it holds itself the information, then a
consistent part of the producible and needed
information will be embedded in the business
process. Referring to the capability approach, the
housekeeping of the information of a capability is
done by a (management) module (See also
Vandenborre, 2003).
6.2 The Possibilities with Integration
In the case of strategy integrations, Business Process
Embedded Information Systems (BPEIS) are
recommended for the new processes in the new
organisation. The ‘old’ processes can only receive
their BPEIS, if they have been re-engineered. In
function of the existing architecture and
infrastructure of the different former organisations,
the new organisation may opt to go through an EAI-
project before installing BPEIS. The same reasoning
is applicable for a single organisation.
Strategic integrations of business processes have
the advantage that they can start immediately with
BPEIS, since they have no legacy systems. Certainly
the different “mother-organisations” may have
legacy systems, but they can be interfaced or
integrated through the capabilities of SOA.
As already mentioned, within operational
integration no common business or resources
strategies are set up, only an operational strategy. If
one (or more) of the mother-organisations has the
adequate SOA capabilities then BPEIS can be
realised. Otherwise the duration of the business
integration will be the main decision element to
evaluate the implementation of BPEIS.
In almost all cases of operative integration,
BPEIS will not be implemented, since no common
business processes are existing.
6.3 Virtual Data Federation
The consequence of Business Process Embedded
Information System is that if somebody needs to
collect in formation then that person will have to
query each process management module. This
resembles to Virtual Data Federation or Enterprise
Information Integration (EII) issues.
Friedman (2004a) writes: “Gartner positions EII
as a goal, not a technology. The goal is to achieve a
state where the various data assets of the enterprise
are integrated to best meet the needs of the business:
Delivering a timely and complete view of
critical entities and events
Providing connectivity and accessibility to
data across multiple platforms and databases
Ensuring the consistency of data
underpinning related applications
As such, the goal of EII differs little from the
general goals of data integration, which has been a
focus of enterprises for the last three decades.” At its
core, EII technology performs virtual data federation
based on distributed database queries (Friedman,
2004b, 2005).
However the biggest problem is to know if the
semantic of one item is the same in all the databases.
This is the added value of software like DB-Main
(Hicks 2003) which produces a conceptual or
semantic model of the organisational in formation.
From thereon through logical and physical models
the operative models are defined. Again, this will
only be a snapshot if the maintenance of all models
is not done. But since information system is
embedded in business processes or units, the burden
is less. However consistency with the organisational
level must be maintained.
This approach of business process embedded
information systems needs another way of making
decision in ICT-investments. Earlier works (Rabaey,
2004a, 2004b, 2005d) propose a holistic framework
for global ICT-investments. In the situation of
business process embedded information systems,
ICT is more federated. Business units have more
autonomy but are not independent from the mother-
Research has already been done in the domain of
investments in Service-Oriented Architecture (called
Service-Oriented Investments). Since SOA is the
most suited environment to develop business process
embedded information systems, research on Service-
Oriented Investment will be continued.
With the discussion on information and business in
the context of the need for flexibility, we came to the
conclusion that not only the processes themselves
have to be flexible but also the supporting software.
This is only possible if the business units or
processes can dispose on a autonomous way over
their information systems, more specifically their
“own” ICT-applications. Therefore we call it
business process embedded information systems. It
is however not independent because of the fact that
the ICT-applications and the underlying ICT-
infrastructure need to fit in the global ICT-
framework of the mother-organisation.
Further research is the adapted way of making
investments in the Service-Oriented Architecture.
Anderson R., 1986. Management, Information System and
Computers: an Introduction, MacMillan. London
Bernard H., 1976. Totale Oorlog en revolutionaire oorlog:
Band I, Royal Military Academy. Brussels (in Dutch).
Burkan W.C., 1991. Executive Information Systems: from
proposal through implementation, Van Nostrand and
Reinhold. New York.
Friedman T. (2004a). Clarifying the Meaning of
Enterprise Information Integration, Gartner.
Friedman T. (2004b). 'EII' Vendors Offer Virtual Data
Federation Technology, Gartner
Friedman T. (2005). Acquisitions Expand Sybase's Data
Integration Offerings, Gartner
Hainaut J.-L. & Henrard J., 2003. The Nature of Data
Reverse Engineering, FUNDP. Namur
Hammer, M. & Champy J.A., 1993. Reengineering the
corporation: A manifesto for business revolution.
Harper Collins. New York
Henrard J. & Hick J.-M. & Thiran P., 2003. Strategies for
Data Reengineering. In proceedings of FNRS contact
day on Software (re-)engineering, Louvain-la-Neuve,
22 May 2003
Hick, J.-M. & Hainaut, J.-L., 2003. Strategy for Database
Application Evolution: The DB-MAIN Approach. In
Proceedings of 22th International. Conference on
Conceptual Modelling (ER 2003), Springer Verlag
LNCS 2813
Kaplan R.S. & Norton D.P.,1996. The balanced scorecard:
Translating strategy into action. Boston: Harvard
Business School Press.
Kaplan R.S. & Norton D.P., 2005. The Office of Strategy
Management. Harvard Business Review, October
Kobielus J.G., 1997. Workflow Strategies, IDG Books
Worldwide. Foster City, CA
Liddell Hart B.H., 1991. Strategy. London: Meridian
McHugh P. & Merli G. & Wheeler III W.A., 1995.
Beyond Business Process Reengineering, Wiley. West
OMB A-130: US Office of Management & Budget, 1996.
Memorandum for Heads of Executive Departments
and Establishments: Management of Federal
Information Resources
Rabaey M. & Vandijck E. & Tromp H. (2003). Business
Intelligent Agents for Enterprise Application
Integration. In ICSSEA 2003, Proceedings of the 16th
International Conference on Software & Systems
Engineering and their Applications. CMSL/CNAM.
Rabaey M. & Hoffman G. & Vandenborre K., 2004(a).
Aligning Business- and Resourcestrategy: an
Interdisciplinary Forum for Investments. In
IAMOT2004, Proceedings IAMOT 13
Conference on Management of Technology,
Washington DC. IAMOT
Rabaey M., 2004(b). Investment in ICT: a holistic
business approach. In Proceedings the 15
International Conference, New Orleans (USA). IRMA
Rabaey M.& Leclercq J-M. & Vandijck E. & Hoffman G.
& Timmerman M., 2005(a): Intelligence Base:
Strategic instrument of an Organisation, NATO
Research and Technology Agency; Information
Systems Technology Panel Specialists’ Meeting, The
Hague, The Netherlands, November 2005 (NATO
Rabaey M., 2005(b). CAF and e-Cost. In Proceedings of
the 2nd CAF Event, Luxembourg. European Institute
of Public Administration.
Rabaey M. & Vandeborre K. & Tromp H. & Hoffman G.,
2005(c). Classification of IT-integration based on
Business Collaboration. In Proceedings of the IPSI
2005 Montenegro conference. IPSI BgD.
Rabaey M. & Vandijck E. & Hoffman G. 2005(d). An
Evaluation Framework for Enterprise Application
Integration. In Proceedings the 16th IRMA
International Conference, San Diego (USA). IRMA
Ross J., 2003. Creating a Strategic IT Architecture
Competency: Learning in Stages, MIT Sloan School
of Management, April 2003
Ross J., 2004. Generating Strategic Benefits from
Enterprise Architecture, MIT Sloan Research Briefing,
October 2004
Vandenborre K. & Heinckiens P. & Hoffman G. & Tromp
H., 2003. Coherent Enterprise Information Modelling
in Practice. In Proceedings of the 13th European
Japanese Conference on Information Modelling and
Knowledge bases. Amsterdam. IOS Press
Zimmerman O. & Tomlinson M. & Peuser S.,2003.
Perspectives on Web Services, Springer. Berlin.