Enterprise Information System, Agility and Complexity
What is the Relationship?
Hakima Mellah
1
and Habiba Drias
2
1
Research Center in Scientific and Technical Information, CERIST, 03 Rue des Fr´eres Aissou, Ben Aknoun, Algiers, Algeria
2
Houari Boumediene University of Sciences and Technology, USTHB, Algiers, Algeria
Keywords:
Complexity, Enterprise Information System, Agility.
Abstract:
The term complexity is present everywhere. Complexity surrounds information, information system, orga-
nization as well as computing systems. In this work we present some challenges that deal with complexity.
Why can enterprise organization be seen as complex? From what, complexity can arise within enterprise in-
formation system? What is relationship between complexity and agility? What are factors that may evolve
complexity within enterprise information system? How computing system led information system towards
complexity? To response to all this challenges, several dimensions dealing with complexity are displayed such
as dynamics in enterprise information system components, while interacting, that can be tolerated by compo-
nents autonomy; intelligibility that causes knowledge evolving; and inter-connectivity that is necessary with
distribution. All these factors influence each others, within a three dimensional system. UML modeling of
enterprise information system that includes complexity parameters is given.
1 INTRODUCTION
Technological advances means of information pro-
cessing and dissemination encourage new uses emer-
gence putting users at the center of the computing sys-
tem. This leads us to more complex systems, whose
evolution is guided by the use, which is in its turn
modified. We can cite in this context, the evolution
of the Internet and Web for example or Web1.0 to
Web2.0. Future computing systems and their use are
facing an increasing complexity due to the following
changes:
Emergence of new computing environments such
as large-scale as well as mobile ad hoc networks
Emergence of new uses and needs from this evo-
lution, requiring sophisticated applications, using
large complex data volumes
A strong need for user-centered applications,
where Information Technology (IT) system user
is considered as a major component or as an im-
portant integral part entity of the system. The de-
velopment of these applications and their evolu-
tion are driven by user-system interactions. We
can consider the system as a set of IT components
interacting with each other and with the environ-
ment
The advent of service technology, the web, large-
scale networks, etc., making the computing envi-
ronments more open and distributed and compo-
nents are no longer under the control of a single
organization
Computing environment is heterogeneous, thus,
becoming difficult to configure, maintain and
troubleshoot
In this context, thinking about IT system engineer-
ing requires the integration of its complexity as an
essential characteristic. IT system complexity arises
from its opening to its environment, itself complex,
distributed and dynamic (Hassas, 2006), (Benatallah
et al., 2003), (Hassas, 2003). These IT developments
and their features have a strong impact when it comes
to be deployed in environments like that of the en-
terprise. The enterprise, taking advantages of these
evolutions, finds itself constraint to change its orga-
nization and its information system. These devel-
opments are also dictated by the enterprise business
environment evolution, beyond its computer system,
and rely on informational and structural entries made
by the Enterprise Information System (EIS). The lat-
ter belongs to complex systems category because it
interacts with other EIS, computer systems, resources
and users. These information systems are distributed,
74
Mellah, H. and Drias, H.
Enterprise Information System, Agility and Complexity - What is the Relationship?.
In Proceedings of the 1st International Conference on Complex Information Systems (COMPLEXIS 2016), pages 74-80
ISBN: 978-989-758-181-6
Copyright
c
2016 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
heterogeneous, decentralized, interdependent, and of-
ten operate in a dynamic and unpredictable environ-
ment (Mordinyi and Kuhn, 2011). EIS represents
the mean used by the enterprise to protect its data,
to have a memory, retrieve, store, process, dissemi-
nate, exchange information that is the foundation of
business intelligibility and knowledge. According to
(Peaucelle, 1997) it embodies the language used to
represent the organization reliably and economically
aspects, it is seen as as the organization. However, in-
formation is considered as one of the main resources
of the enterprise’s business, that is being evolved
through interaction and communication.
Figure 1: Information system, Organization, information
flow: A complex framework.
What is noticed is that the activity in the enter-
prise generates information flows (figure 1), and for
that some authors such as Hugues Angot (Angot,
2006) addressed the organization in information flows
terms, and what is emphasized are structured rela-
tions between the different information sources com-
ponents. This what explains the JJ Lambin (LAM-
BIN, 1990) vision, for the EIS, as structured relation-
ships complex network, where are involved humans,
machines and procedures, which aims to generate rel-
evant information ordered flows from the enterprise
internal and external sources, intended to serve as a
basis for decisions and solutions presented to the user.
EIS can be viewed through the dynamic organiza-
tion (and even self-organization) of its flow. Informa-
tion flow generated during the execution of an activity
or task is susceptible to disturbances that may block
its routing, thus making the system obsolete. In or-
der to overcome the disturbance, one solution would
be through other network topologies, thus generating
new communication links, which did not exist a priori
between different interacting entities. This explains
our vision of flow management problem, as a self-
organization problem. And consequently a solution to
the problem mentioned above would be the induction
of a self-organizing mechanism that is a key feature of
complex systems. With the large scale processors net-
works evolution like Internet, considered as the most
used mean for information exchange between enter-
prises, new practices are highlighted and new needs
have emerged towards the computing environment. It
is therefore necessary to think about computing sys-
tems developmentable to work intelligently with their
environment systems. Tackle EIS in a holistic view
considers it as a whole or ”holism” without separating
the user from the EIS contents and the environment in
which this system interacts. This is especially true
in a dynamic environment context where interactions
between different parts of the computing system are
constantly changing, requiring challenges of its struc-
ture, organization, interactions and its dynamics. Cur-
rently the information flow increasing systems’ com-
plexity is a characteristic of enterprises and manufac-
turers regarding products, services to offer, and all
information exchange. This complexity also comes
from changing environments that are disturbance sur-
rounded, in which EIS operate. Difficulties arise and
the system has to solve unforeseen problems on the
basis of available information that is generally incom-
plete and imprecise.
A complex system is a system, characterized with self
organization, adaptation, evolution, etc., it consists of
plenty of constituents which interact in a hierarchi-
cal frame to function as a whole. Emergence, adapta-
tion, evolution are the elementary properties of com-
plex systems, where emergence is the most essential
feature (Zundong et al., 2010).
2 COMPUTING SYSTEMS AND
ENTERPRISE INFORMATION
SYSTEMS
Computing systems evolution, evolved by fact and in
the same direction EIS. Computing systems evolution
and the emergence of new technologies, reveals evo-
lution needs in the same direction of information sys-
tems, which is seen at the same time as an opportu-
nity for EIS to benefit from these evolution. Further,
EISs as complex systems are systems whose compo-
nents interact to accomplish goals. The developers
of such systems must, in addition to the application
logic expressing the reaching goal, to include a man-
agement and supervision logic in order to overcome
the environmental problems for which the organiza-
tional goal cannot be reached. The management logic
is important, for example, in case of EIS components
non functioning (Mordinyi and Kuhn, 2011).
2.1 Enterprise Organization
Evolution in enterprises organizations was clearly no-
ticed in the last decade (Mintzberg, 1994). It took
Enterprise Information System, Agility and Complexity - What is the Relationship?
75
into account the networks, knowledge and skills man-
agement, cooperation, enterprise organizational struc-
ture, for the following reasons:
The economic transformation that requires re-
sponsiveness
The competitive nature that is a factor of organi-
zational requirement
The costumers/users growing power pushing en-
terprises to reshape their organization to better re-
spond
The facility offered by the IT for information ex-
change internally and externally
The reasons cited above showed that the organiza-
tional implementation of strategies is a key perfor-
mance factor and the organization became a part of
the enterprises competitiveness. These organizational
changes result in two heavy and common trends to
many enterprises that are exploding the organization
boundaries and the transversal management. The
basic elements constituting an organization (Mellah
et al., 2007) are connected by a variety of complex
flows that are all important and explain how an or-
ganization operates (considering all of these flows).
(figure2 )shows the environment in which EIS evolves
considering different dimensions that make the sys-
tem complex. The characteristics of this flow are:
The massive character. Given the important need
of customers/users that is increasing and the flows
variety and their complexity. These flows connect
all the basic elements of an organization and may
involve authorities, materials, or simply commu-
nications and all these parameters need control to
assure their connectivity and survive. Informa-
tion uses increase its intelligibility that is a key
for knowledge management.
Dynamic and autonomy. Tolerated by the advent
of communication IT (eg intranet), in the sense
that IT break formal organization barriers to form
working groups that are non formal organizations.
ITs endow EIS of autonomy.
Distribution and inter-connectivity. ITs provide a
growing organization, the means to be split into
different parts distributed on the network that can
communicate later with each other. Maintaining
this distribution, inside or outside the enterprise,
must be assured by means like self organizing
protocols (Mellah et al., 2007). Interaction net-
works link organizational positions to form orga-
nizational structures. In the context of cooperative
distributed problem solving (CDPS), an organiza-
tional structure can be seen as a way to specify the
coordination strategy (pa S. Young and Edmund,
1993).
Figure 2: Complexity parameters surrounding information.
2.2 System-environment Coupling
The vision adopted by (Hassas, 2003) considers the
computing system as a set of components in inter-
relational and retroactive interactions, evolving in a
shared, dynamic and uncertain environment that cor-
respond to that of a complex system, is inspired from
autopoetism/Enactivism characteristics. But we must
not be confused by autopoetism abstract conceptual-
ization and by complexity theory which treats sys-
tems as entities that we see outside ourselves. We
are part of the problem and of the solution (Cachon,
1999). The challenge of this type of system is to
find an effective organization of its components (by
self-organization), in organizational structures imple-
menting or highlighting a coherent behavior towards
the environmentcomplexity. However, living systems
have managed such a challenge for millions of years.
As supported by the Enactivism (Hassas, 2006), we
believethat this ability to living systems is due to their
coupling with the environment. Natural phenomena
are an inspiration source in different computer sys-
tems areas (Bonabeau et al., 1999)(Drias et al., 2005).
This phenomenon is due to the fact that living sys-
tems have properties that allow them to adapt to a
complex environment. Systems such as cell organ-
isms, insect colonies or societies of individuals ex-
hibit complex global behavior that self organizes their
components into robust and flexible structures. These
systems are characterized by a high reactivity, robust-
ness to individual failures, and a great capacity to
adapt to environmental changes. Several computing
areas have referred (Hassas, 2006) to bio-inspired ap-
proaches. Rats, spiders, ants, fish, birds, bacteria, etc.,
are all natural living systems that attracted researchers
attention, for differentiation, information dissemina-
tion, aggregation, optimization,...etc. All these works
mainly refer to the Stigmergy mechanism (Grasse,
1959) which was discovered by P.P Grasse, and is
COMPLEXIS 2016 - 1st International Conference on Complex Information Systems
76
used by insect colonies to coordinate their behavior.
Coordination is assured through the persistent effects
left in the environment by previous actions on the fu-
ture actions of agents. The environment is considered
as a space fit these traces of actions and interactions.
Structures and processes emerge in a autopoetic vi-
sion (ie the permanent co- evolution of the process
and structure as a result and process support (Hassas,
2003).
To overcome the deadlock situations that may en-
counter a distributed system and that will disturb its
inter-connectivity, IT designers (or sellers) recognize
that it is necessary, even essential to think about de-
signing systems that assume themselves, their man-
agement. This is recognized in academic and indus-
trial settings by Oriented Autonomic Computing, rec-
ognized in 2001 by ”Paul Horn” (senior vice presi-
dent of IBM Research) as analogous to the human’s
nervous system (Markus and Julie, 2008). The Dis-
tributed Autonomic Computing or DAC is accom-
plished mainly through the implementation of self *
properties which can be classified according to a tax-
onomy of criteria and in this taxonomy we find the
resource allocation, which is characterized by the fact
that a system needs to allocate limited services or allo-
cate tasks to resources,etc. (Frei and Giovanna, 2011).
Information delivered by an EIS is captured in a com-
plex way. It can be framed in a three-dimensional
system with content, use and structure dimensions
(Mellah et al., 2008), as it is considered that infor-
mation content must be related to other content, and
is nothing without external factors (environment), to
this information, without the value given by the user
and without context raised pertinence (Mellah et al.,
2013b).
3 COMPLEXITY AND AGILITY
The first changes, in perspective, known by orga-
nizational theory, is the transition from a closed to
the perspectives of an opened system (Mellah et al.,
2008) taking into account organization’s external
factors. To evolve, an EIS must be opened to its
environment. We find this openness and evolution
in the characteristics of agile information systems.
Agility is defined by Gartner as ”the ability of an
organization to sense environmental change and
responds efficiently to that change”(Mellah et al.,
2013b). Agile information system is a system able to
adapt to the functional and technical developments,
it is often more open to absorb management process
with third parties, and allow organization evolution
without challenging business applications. In prac-
tice, agility is materialized by ”services” orientation
[wikipedia].
Definition. We define an agile information system
as a self organizing system or a system endowed with
a self organizing mechanism. We consider the agility
of an EIS as a new class in the taxonomy of self *
properties criteria.
This definition underlines that an agile information
system as it is characterized by self organizing prop-
erty is a complex system. The computing resources
offered by an agile information system are expressed
in terms of services. They have the following fea-
tures(Mellah et al., 2013a):
May extend beyond the enterprise boundaries
through transit interfaces such as Internet.
Are requested by the mean of interfaces used to
interact with a variety of enterprises (other orga-
nizations).
Conversely an organization providing any service is
often called upon to interact with a set of service re-
questers (Benatallah et al., 2003) the service can have
the following characteristics:
changing and evolving over time(Thompson,
1967).
appearing in a new version instead of the old one.
new version of services with advanced features
can be offered by organizations integrating inter-
action process.
”Computer services” create the technological
foundations and management required to support en-
terprise agility (McCoy, 2007) and complexity. To
ensure this agility and enabling services use in or-
der to meet the business process and the user, it is
important to have a software service-oriented archi-
tecture (SOA) (McCoy, 2007)(Mecella and Pernici,
2006). System-environment coupling, that charac-
terizes agile information system, is among the ar-
guments justifying the consideration of the above
issues in terms of EIS modeling through informa-
tion flows, their inter-dependencies, their dynamics.
Within MAS paradigm, this coupling can be assured
by a self organizing protocol (Mellah et al., 2007)
and can be projected within SOC (Service Oriented
Computing) as an SOA solution for agility (Mellah
et al., 2013a) that leads systematically an SOA to an
ASOA (Agile SOA). Consequently an ASOA is de-
fined as being a SOA whose components self orga-
nize while encountering disturbance within environ-
ment. Agility as it is matched to complexity consid-
ers EIS in two levels: (i) a core centered data, users,
and connectivity between them, (ii) a shell evolving in
Enterprise Information System, Agility and Complexity - What is the Relationship?
77
Figure 3: Agility parameters within EIS.
three dimensions, which axis are supported by intel-
ligibility, dynamics and inter-connectivity. The three
parameters are involved and represent the support of
knowledge, autonomy and distribution intra and extra
enterprise (figure3).
4 ENTERPRISE INFORMATION
SYSTEM MODELING
Organization theory deals with complexity as a struc-
tural variable that characterizes organizations and
their environments (Zaho et al., 2007). Draft (Syn-
tec, 2007) matches complexity to the number of ac-
tivities or subsystems, which basically composes the
organization. This implies that a complex system is a
system of systems. More the system has subsystems
and more it is complex. Complexity can be decom-
posed into three dimensions: vertical, horizontal and
spatial, corresponding respectively to the number of
levels in the organizational hierarchy, the number of
posts or departments within an organization, and the
number of geographical locations. We can identify
the following understandings (Zaho et al., 2007):
Activities/subsystems. Each subsystem can be
composed by other subsystems
Activities/subsystems have a geographical loca-
tion (networks node) that may change well.
Therefore, they do not have a persistent physical
location
Activities/subsystems can be decomposed to
highlight features that are invisible at first sight.
That is an important characteristic of complexity.
The decomposition can be done based on relative
uses that characterize subsystems. While they are
decomposed, a structure or a pattern is generated
(figure 4).
Figure 4: EIS classes diagram in relation with entities im-
plied in information flow.
These points highlight the key features of agile in-
formation systems, well supported by a SOA (McK-
elvey, 1999). The key features appear to be compati-
ble with Drafts proposal. In order to tackle informa-
tion system complexity, it is required to harmonize
the quality of service offered to users internally and
externally. A new version for the design of complex
systems is called organic computing (Mellah et al.,
2013a). The latter satisfies conventional requirements
for trustworthy systems, which autonomously adapt
to dynamic changes of the environment, and have
self-x (self-organizing, self-healing, etc.) properties
(McKelvey, 1999) as postulated for autonomic com-
puting (Markus and Julie, 2008). From our perspec-
tive, we recognize the information complexity, issued
from a set of interacting EIS or from a SOA, in a
three-dimensional system, handling service, context
and service discovery structure (Draft, 1992).
Each EIS represents one or more organizations,
one or more subsystem. We can note a mapping be-
tween our work and that of Draft, interpreted in the
following points:
Each level of an organizational hierarchy corre-
sponds to the existence of services, and the num-
ber of departments in an organizational level may
correspond to the number of contexts that we as-
cribe to services.
The geographical locations constitute the service
discovery structures generated by the various ser-
vice discoveries, to meet the needs of the user and
achieve the desired goals.
Considering these technicalities, we confirm that
complexity theory is easily supported by a SOA, in
the sense that SOA offers issues to integrate self-
organization within a SOA as it is a key feature of
complex systems(figure 5). Self organization can be
handled on the basis of layered architecture(Mellah
et al., 2013b) and a Self Organizing (SO) protocol
(Mellah et al., 2009; Mellah et al., 2010). The SO
protocol controls interaction between nodes support-
ing EIS contents, by the mean of checking and routing
COMPLEXIS 2016 - 1st International Conference on Complex Information Systems
78
Figure 5: Illustrative schema on agility, complexity and system-environment coupling.
process to maintain connectivity of the hole system
(Mellah et al., 2009; Mellah et al., 2010).
5 CONCLUSION
In this work we have presented some challenging sit-
uations that deal with complexity and that turn around
information, EIS, computing system. The complexity
parameters that surround information, and that make
complex EIS, are presented. A self organizing view
to agility (that is matched to complexity) is given.
While discovering contents, offered by a EIS or its
subsystems, a structure is generated. Without consid-
ering the use context which can be either relative use
characterizing the content itself (can varied), or global
use, characterizing the system in its hole, the structure
represents itself a complexity support(Mellah et al.,
2013b), because it is this structure that gives robust-
ness to the system, by self organization, recognized as
a feature of complex systems and of agile information
systems.
REFERENCES
Angot, H. (2006). Systeme d’information de l’entreprise:
Analyse theorique des flux d’information et cas pra-
tiques. Editions de Boeck universit. Bruxelles.
Benatallah, B., Muas, M., Fauvet, M.-C., and Rabhi, A.
(2003). Towards pattern of web services composition.
In Patterns and skeletons for parallel and distributed
computing.
Bonabeau, E., Dorigo, M., and Theraulaz, G. (1999).
Swarm Intelligence: From Natural to Artificial Sys-
tems. Oxford University Press.USA.
Cachon, G. (1999). Complexity theory and organization
science. organization science.
Draft, R. L. (1992). Organization theory and design. Saint
Paul Minnesota: 4th west publishing company.
Drias, H., Sadeg, S., and Yahi, S. (2005). Cooperative bees
swarm for solving the maximum weighted satisfiabity
problem. In IWANN.
Frei, R. and Giovanna, D. M. S. (2011). Advances in
complexity engineering. International Journal of Bio-
Inspired Computation.
Grasse, P. (1959). La reconstruction du nid et les coordi-
nations interindividuelles chez bellicositermes natal-
ensis et cubitermes sp. la theorie de la stigmergie: Es-
sai dinterpretation du comportement des termites con-
structeurs. insectes sociaux.
Hassas, S. (2003). Complex systems based on situated
multi-agent systems, in french:systemes complexes a
base de multi-agents situe. hdr. In Liris Lyon.
Hassas, S. (2006). Engineering complex adaptive systems
using situated multi-agents: Some selected works and
contributions. In ESAW 2005.
LAMBIN, J. (1990). La recherche Marketing. Hill, Mc
Graw.
Markus, C. H. and Julie, Q. M. (2008). A survey of au-
tonomic computing-degrees, models and applications.
ACM computing surveys.
McCoy, D. (2007). Defining cultivating and measuring en-
terprise agility. Gartner Research.
McKelvey, B. (1999). Complexity theory in organiza-
tion science: seizing the promise or becoming a fad?
Emergence.
Mecella, P. and Pernici, M. (2006). Cooperative information
systems based on a service oriented approach. Inter-
Enterprise Information System, Agility and Complexity - What is the Relationship?
79
national journal of interoperability in Business Infor-
mation Systems.IJBIS.
Mellah, H., Hassas, S., and Drias, H. (2007). Towards a self
organizing protocol for a multi agents system. In The
sixth International IEEE Conference on Networking
(ICN’07), pages 60–66.
Mellah, H., Hassas, S., and Drias, H. (2008). Modeling in-
formation in a three dimensional system to contribute
to the agility of an enterprise system. System and In-
formation Sciences Notes, CoSIWn.
Mellah, H., Hassas, S., and Drias, H. (2009). Massop :
A self organizing protocol for multi agents systems
inspired by bacteria colonies. International Transac-
tions on Systems Science and Applications.
Mellah, H., Hassas, S., and Drias, H. (2010). A Formal
Approach for a Self Organizing Protocol Inspired by
Bacteria Colonies: Production System Application.
Book Chapter. In Bio-inspired Models of Network, In-
formation, and Computing Systems. LNICST. E. Alt-
man et al. (Eds.).
Mellah, H., Hassas, S., and Drias, H. (2013a). A fault toler-
ant service discovery by self organization: a mas ap-
proach. International journal of Business Information
System.IJBIS.
Mellah, H., Hassas, S., and Drias, H. (2013b). Service com-
puting and agents paradigm for managing distributed
information sources. International journal of Business
Information System.
Mintzberg, H. (1994). Structure et dynamique des organi-
sations. Les editions dorganisation.
Mordinyi, R. and Kuhn, E. (2011). Coordination mecha-
nisms in complex systems. Next generation Data tech-
nologies.
pa S. Young and Edmund, D. H. (1993). An organizational
self design model for organizational change. In Work-
shop on AI and Theories of Groups and Organiza-
tions.
Peaucelle, J. (1997). Systemes dinformation. Encyclopedie
de gestion. Economica.
Syntec, I. (2007). Les architectures orientes services. Livre
Blanc. collection thematique.
Thompson, J. D. (1967). Organizations in Action:Social
Science Bases of Administrative Theory. McGraw-
Hill.
Zaho, J. L., Mohan, T., and Liliang, L. (2007). Ser-
vices computing as the foundation of enterprise
agility:overview of recent advances and introduction
to the special issue. information systems Front.
Zundong, Z., Limin, J., and Yuanyuan, C. (2010). A new
kind methodology for controlling complex systems.
World Academy of Science, Engineering & Technol-
ogy, 68(44):146–159.
COMPLEXIS 2016 - 1st International Conference on Complex Information Systems
80