Complexity and Adaptive Enterprise Architecture
Wissal Daoudi, Karim Doumi and Laila Kjiri
AlQualsadi Team, ENSIAS Mohammed V University in Rabat, Rabat, Morocco
Keywords: Adaptation, Complexity, Dynamic Environments, Enterprise Architecture.
Abstract: In the current VUCA (Volatility, uncertainty, complexity and ambiguity) environment, enterprises are
facing constant threats and opportunities due to internal and external factors. Those factors can impact
various parts of the enterprise in the form of changes. Thus, Adaptive Enterprise Architecture (EA) is
leveraged to assist the continuous adaptation to the evolving transformation. On the other hand, the
complexity has been identified as one of the major challenges of the discipline of Enterprise Architecture.
Moreover, one of the criteria of Adaptive EA is the ability to monitor and control the complexity of changes.
Consequently, in this paper, we suggest a conceptualization of EA complexity measurement drilled down
into factors and indicators. First, we begin with a brief summary of the criteria that we consider compulsory
for Adaptive Enterprise Architecture and we give an overview of the model that we worked on in previous
work. Then we investigate related work about complexity in a broader view. Finally, we describe our
approach of assessment of complexity based on the proposed indicators.
1 INTRODUCTION
In current turbulent environment, enterprises are
often required to adapt in the form of disruptive
changes that impact its various parts. Thus, they
need to become adaptive and agile by facing unique
challenges that they encounter with the specificities
of each of them (cycles, recurrence, frequency, etc.).
They are required to recognize the impact of change,
detect obstacles and facilitate decision-making.
Also, they need to consider the uncertainty and the
diversity of change and respond effectively to it.
In order to support the evolving requirements,
Enterprise Architecture can be leveraged. As the
watcher of changes and facilitator of adaptation, EA
should focus on the methods and tools needed to
move from an initial, detailed, complex,
documentation-centred and prescriptive EA to an
EA that focuses on principles of adaptation to
expected changes and unforeseen ones. More
importantly, EA should provide continuous
improvement to proactively address development
needs with the right level of complexity. In this
regard, we introduced Adaptive Enterprise
Architecture model (Daoudi et al. 2020a).
Adaptive EA takes into account the uncertainty
of change and its diversity. It allows the proactive
detection of change and responds to it efficiently. It
also permits the management to make the adequate
trade-offs between the components involved and that
are sometimes competing. Most importantly it leads
dynamic transitions, in the form of projects, from an
“as is” to a “to be” by ensuring the right level of
complexity.
On the other hand, the Cambridge Dictionary
defines complexity as “the quality of having many
connected parts and being difficult to understand”.
In the literature, the notion of complexity can be
found in different domains and there is a lot of
definitions and a lack of consensus on it or on how
to measure complexity (Padalkar et al., 2016). In
regards with EA, this concept can have many
interpretations as there are many stakeholders
involved in an EA and each one have a different
perception of it.
In this paper, we explore the state of the art
related to complexity and we define some factors
that we consider as drivers of complexity in
Enterprise architecture. Then, we introduce EA
DDC (Degree of Dynamic Complexity) as a
measure. In Section 2, we summarize the results of
our previous work in regards with Adaptive
Enterprise Architecture. Section 3 focuses on related
work of complexity of change management. In
Section 4, we define the factors and the metrics of
Complexity in an elementary EA transition. Finally,
Daoudi, W., Doumi, K. and Kjiri, L.
Complexity and Adaptive Enterprise Architecture.
DOI: 10.5220/0010475707590767
In Proceedings of the 23rd International Conference on Enterprise Information Systems (ICEIS 2021) - Volume 2, pages 759-767
ISBN: 978-989-758-509-8
Copyright
c
 2021 by SCITEPRESS – Science and Technology Publications, Lda. All rights reser ved
759

in the last part, we conclude our work and present
our perspectives.
2 ADAPTIVE ENTERPRISE
ARCHITECTURE MODEL
In order to put into context our paper, we present in
this part the main results than we achieved and that
were published in previous work. We proposed a
definition for Adaptation: Adaptation ensures that
the EA is consistent with the changes, to maintain its
normal functioning. It is a process of adjustment and
of continuous improvement to reach an EA in
harmony with its environment. Then, we defined
some criteria that we consider compulsory
ingredients for Adaptive Enterprise Architecture
(Daoudi et al., 2020b).
First, we highlighted multi-level of dynamics
factor as some types of change occur at different
layers and impact the relations inter-layers and intra-
layers. Then, we explained the sensing of change
part which is the ability to detect continuously the
need for change proactively at internal and external
levels. We also underlined the process of
adaptation which is the core of the adaptive
enterprise architecture. We pointed out the
complexity of change management that is related
to the degree of complexity of the different
components and relationships in an EA. It is for
example related to business diversification,
geographic diversification or network
interconnectedness. Moreover, a complex document-
oriented framework will certainly fail to handle
abrupt changes that happen at high pace. Then, we
defined the ability of handling unforeseen changes
which is the proactive definition of unexpected
change specifities location, severity, probability and
kinds of adaptations needed. Another criterion
specified was related to the explicit management of
adaptability trade-offs. It allows the archiving,
tracking and knowledge sharing of trade-offs
necessary when deciding of an architecture. Finally,
we underlined the importance of evaluation of
adaptation which allows the assessment of the
improvements made through the adaptation process.
Then driven by those criteria, we tried to propose
an Adaptive Enterprise Architecture approach based
on agile methodologies (Daoudi et al., 2020a). The
Figure 1 is a simplified diagram that shows the main
elements of our Adaptive Enterprise Architecture
Approach.
Figure 1: Simplified diagram of the proposed model.
Our approach allows having a dynamic architecture
that is continuously evolving through time. Thus, in
order to analyse the components of the EA we take a
static snapshot at a certain time (EAi). We consider
that during an enterprise lifecycle we move from an
EAi (i∈N*) to EAi+1 (i∈N*) (Elementary
transition). So as to ensure those continuous
transitions, every elementary transition is a project
with the main objective to close the gap between the
“As-Is” and “To-Be”.
3 RELATED WORK
In the current VUCA (Volatility, uncertainty,
complexity and ambiguity) environment,
management approaches need to adapt to the new
requirements and to manage complexity. In the
literature, multiple papers and researches have
shown the importance of complexity management
as it impacts various project phases during its
lifecycle, it hinders the identification of goals and
objectives and it can affect different project
outcomes in terms of time, cost and quality
(Baccarini, 1996) and (Parsons-Hann et al., 2005).
Also, the larger and the more complex a project is,
the riskier it is. In fact, this type of projects face
significant, unpredictable change, and are difficult or
impossible to forecast (Taleb et al., 2009). If we
focus only on the IT (Information System),
complexity has been attributed as one of the causes
of high failure rates in IT projects. So as to give
some statistics, One in six IT projects is expected to
be a black swan, with a cost overrun of 200% on
average (Flyvbjerg et al., 2011). In general,
complexity is taken as having negative impact on
project performance (Bjorvatn et al., 2018). But in
order to maximize the effectiveness of an
ICEIS 2021 - 23rd International Conference on Enterprise Information Systems
760

architecture, some new concepts evolved recently
like “requisite complexity”. It shows that is
important to find the right balance between
complexity excess and deficit, that is, to find an
optimal level of complexity (Schmidt, 2015).
On the other hand, limited research has been
conducted on metrics and measuring IT complex
projects and less in defining methods for managing
them. Most research concludes that metrics and tools
are required but not available or not reliable
(Morcov et al., 2020). Specifically talking about
complexity and Enterprise Architecture, complexity
has been identified as one of the major challenges
faced by the discipline of enterprise architecture
(Lucke et al., 2010). But little research on
complexity management in other areas is applicable
to the field of enterprise architecture (Lee et al.,
2014).
In addition to that, systems are increasingly
exposed to hazards of disruptive events (Zio, 2016).
e.g., new business requirements, unexpected system
failures, climate change and natural disasters,
terrorist attacks. Risk assessment is, then, applied to
inform risk management on how to protect from the
potential losses. It is a mature discipline that allows
analysts to identify possible hazards/threats,
understand and analyze them, describe them
quantitatively and with a proper representation of
uncertainties (Zio, 2018). Its principles are based on
assessment of risk as a scientific activity depending
on the available knowledge and the uncertainty
inherent in risk, and decision making based on risk
is regarded as a political activity. According to Qazi
et al. (2016), in the current literature, some
researchers are supporters of the existence of a
relationship between complexity and risk. They
argue that the adoption of a disintegrated approach
of evaluating complexity and risks in silos raises the
possibility of selecting sub-optimal risk mitigation
strategies While others are detractors of this link and
suggest that these two concepts are distinct.
In the following, we explore the broader state-of-
the-art related to the definition of complexity with a
focus on research papers related to IT, business and
project management as, in our model, the elementary
transition from EAi to an EAi+1 is a project. We
also, identify the main contributions that discussed
complexity measurement in Enterprise Architecture.
3.1 Definition of Complexity
The notion of complexity can be found in STEM
(Science, Technology, Engineering, and
Mathematics), social, economic and management
disciplines. The main challenge is that there is a lack
of consensus on the definition of complexity of a
project (Padalkar et al., 2016). In the Table 1, we
summarize the main definitions of complexity.
Table 1: Definitions of Complexity.
In our paper, we consider that complexity of an
Adaptive Enterprise Architecture involves many
unknowns and many interrelated factors as
explained by the previous criteria. In fact,
complexity is related to the different parts of an
enterprise with their specificities, to the interrelation
between layers and to the environment. Moreover,
we also have the dynamic aspect between EAi and
EAi+1. This means that the complexity is not
applied to a static approach but has a dynamic part.
Thus, our reasoning tends towards the definitions
given by Vidal et al. (2008), Schütz et al. (2013),
and Trinh et al. (2020).
3.2 Complexity Measurement
As shown in the previous part, complexity can have
many interpretations sometimes even in the same
field. In the following, we focus on papers that
discussed the measurement of complexity.
According to San Cristóbal et al. (2018), in order
to comprehend project complexity concept can be
drilled down into factors and characteristics. They
Complexity and Adaptive Enterprise Architecture
761

identified the main factors that are considered in the
literature: Size, Interdependence and Interrelations,
Goals and Objectives, Stakeholders, Management
Practices, Division Labor, Technology, Conccurent
engineering, Globalization and context dependence,
Diversity, ambiguity, Flux.
Also, with a focus on IT projects, Morcov et al.
(2020) identified the below characteristics of
complexity : Multiplicity, ambiguity, uncertainty,
Details (Structural), Dynamics, Disorder, Instability,
Emergence, Non-Linearity, recursiveness,
irregularity, randomness, Dynamic complexity,
uncertainty of objectives and methods, varied
stakeholder and competing views, changing
objectives, adaptive evolving, explanation states of
stability-instability, Size, Variety, interdependence,
context, innovation, difficult to understand, Difficult
to foresee and difficult to control.
Lagerström et al. (2013) applied Design
Structure Matrices. They classify applications based
on their dependencies into core, control, shared and
periphery applications and calculate the propagation
costs.
In Schneider et al. (2014), the authors identified
eight aspects frequently examined in complexity
science literature and proposed a conceptual
framework that aims to unify views on complexity
through four dimensions : Objective vs Subjective /
Structural vs Dynamics/ Quantitative vs Qualitative/
Ordered vs Disordered.
Kahane’s approach to complexity used a process
called the U-process. Basically, the project managers
try to sense the current reality of the project, then
analyse it and propose action items, and finally they
implement those actions (Kahane, 2004).
Cynefin Decision-Making Framework originated
from Snowden‟s work in knowledge management. It
is a sense-making framework that sorts systems into
five domains that require different actions based on
cause and effect relationships: simple, complicated,
complex, chaotic and disorder (Kurtz et al., 2003).
In relation with Enterprise Architecture, Iacob et
al. (2018) worked on the conceptualization of EA
complexity measurement, including the variables
and the metrics to measure them. Through an
analysis of the state-of-the-art, they proposed a
measurement model that integrated existing
complexity metrics and introduced new metrics.
Janssen et al. (2006) considered enterprises as
complex adaptive systems and attributed to them
properties like emergence and self-organization. In
addition, they provided concrete architectural
guidelines.
Mocker (2009) provided one of the first
empirical evaluations of complexity measures
including interdependencies of applications,
diversity of technologies, deviation from standard
technologies and redundancy.
Kandjani et al. (2013) presented a co-evolution
path model, which is based on the idea of Ashby’s
law of requisite variety. The model shows that each
time the complexity of an enterprise’s environment
changes, the enterprise itself has to adjust its
complexity.
According to the IEEE Standard 1471-2000 in
IEEE Architecture Working Group (2000) and
Schütz et al. (2013), we can consider EA as a
system, consisting of its components and its
relations to each other. Zio (2016) stated that
systems are increasingly exposed to hazards of
disruptive events. Thus, risk assessment is applied to
act proactively to those events and prevent eventual
losses.
The International Risk Governance Council
(2012) defined risk as an uncertain (generally
adverse) consequence of an event or activity with
respect to something that human beings value. As
for the risk description, the focus is on the accident
scenarios, their possible consequences and
likelihoods, and the uncertainties therein (Bjerga et
al., 2016). The post-accident recovery process, is not
considered. As the accuracy of scenarios and of
estimations are evaluated against the available
knowledge which is limited, risk needs to take into
account the uncertainties associated to the risk
assessment (Kaplan, 1981). Aven et al. (2010)
integrated knowledge as an explicit component in
the definition of risk. The challenge is to have under
analysis all the knowledge from experts observations
and model prediction about rare but potentially
disastrous accident events (Zio, 2018). The
relatively recent discussions on the concept of risk,
have clearly stated the outcomes of risk assessment
are conditioned on the knowledge available on the
system and/or process under analysis (Aven, 2016).
This means that there is inevitable existence of a
residual risk related to the unknowns in the system,
and/or process characteristics and behaviors.
For a specific project, the identification and the
tracking requirement are not sufficient as they are
based on unconstrained plans. Thus, Perera et al.
(2005) proposed to integrate 7 pillars of risk
management (Schedule, people, technical,
configuration management, Safety, Environment,
and cost/Budget) with the three major areas of
emphasis of project management which are project
ICEIS 2021 - 23rd International Conference on Enterprise Information Systems
762

control, systems engineering, and safety and mission
assurance.
Recognizing the common framework used to
describe the uncertainties in the assessment stands
on probability theory, and particularly on the
subjectivist (Bayesian) theory of probability, as the
adequate framework within which expert opinions
can be combined with statistical data to provide
quantitative measures of risk (Kelly and al., 2011).
According to Perera et al. (2005), NASA’s
(National Aeronautics and Space Administration)
risk management strategy is a continuous and
iterative process performed to reduce the probability
of adverse threats. It includes also an approach of
knowledge archiving and sharing as a basis for
future mitigation activities. It focuses on the
following activities. First the identification of
potential problems. Then the analysis of those
threats by understanding the nature of the risks,
cleaning by merging elements and eliminating
duplicates, classifying and prioritizing them which
help with the creation of the mitigation plans. The
next step is risk planning (action plan). After that is
the tracking part. Finally, risk control which is the
decision making in relation with each risk and the
actual action plan. One Other contribution of this
paper is the measure of effectiveness of the risk
management process though four dimensions: Input
(documentation hinders), Speed (time to get from
source to right destination), Fidelity (risk input
changes) and Synthesis (view of correlated input
from different sources).
In addition, Dynamic Risk Assessment (DRA) is
defined as a risk assessment that updates the
estimation of the risk of a deteriorating system
according to the states of its components, as
knowledge on them is acquired in time (Yadav et al.,
2017). Most existing DRA methods, only use
statistical data that require the occurrence of the
accidents or near misses (Zio, 2018).
4 COMPLEXITY IN ADAPTIVE
ENTERPRISE ARCHITECTURE
In this paper and in relation with our approach, we
propose the assessment of one of the criteria of
“Adaptive Enterprise Architecture” that were
proposed in Daoudi et al. (2020b): the complexity of
change management.
Before tackling the core of this part we consider
EA as a system, consisting of its components and its
relations to each other. This consideration is aligned
with IEEE Standard 1471-2000 in IEEE
Architecture Working Group (2000) and Schütz et
al. (2013) work. In regards with the tools of
modelisation, we suggest the use of Archimate
notation. ArchiMate is a modeling language that
provides a uniform representation of diagrams
describing enterprise architectures. This provides an
integrated architectural approach that describes the
different domains of architecture, their components
and their relationships and dependencies. As such,
we suggest considering the complexity of change
management or the complexity of moving from an
EAi to an EAi+1 as a function of time that has
multiple factors that we will define later. We named
this metric: EA Degree of Dynamic Complexity
(DDC).
DDCi,i+1(t) =
∑
(𝑓𝑗(𝑡)


+ 𝑓𝑗) where n∈N
Where i the indicator of the EA version, f
j
(t) the
values of dynamic factors and f
j
the values static
factors.
Based on Schneider et al. (2014) and as shown in
the formula, we proposed a first dimension of
classification of our factors. Thus, we have
“Dynamic” one who can have many values
overtime. Those factors can allow us to study their
trends and to assess their evolution during the
elementary transition. On the other hand, we have
“Static” ones that have the same value overtime
during the elementary transition. Those factors can
be picked by the management in collaboration with
the Architecture owner. In our proposition, we won’t
consider any static factor.
The second classification dimension is
Objectivity. Thus, we consider that we have factors
that are assessed based on expert judgment and
available knowledge. Those are “Subjective”
Factors. In opposite, we define “Objective” factors
that can be calculated using mathematic formulas
based on the characteristics of the components of the
architecture.
Trinh et al. (2020) considered that the attributes
of project complexity are parts of the following
groups: organizational complexity, technical
complexity and environmental complexity. Also,
Schütz et al. (2013) introduced a system theoretic
conceptualization of complexity in enterprise
architectures. Similarly and in application to EA, we
also propose a third dimension of classification that
is based on the below EA sub-systems. The first one
is “Architecture”. It encompasses the factors that are
drivers of complexity of the whole project of
transitioning from an EA
i
to an EA
i+1
. It contains
Complexity and Adaptive Enterprise Architecture
763

also factors that are related to multiple layers of the
EA. Then, we have “Strategy”, “Business”,
“Organisation” and “Information System”. The
factors in these categories translates the specificities
of complexity at respectively each level. We added
“External” category, it is not a sub-system of EA but
it is worth mentioning as some environmental
requirements may have an impact on the complexity
studied. The Table 2 shows the factors that we
consider as drivers of the complexity of each
increment or elementary transition (project) in our
proposed approach. The list is not exhaustive.
Table 2: Proposed complexity factors in EA transition
project.
The selection of the factors was mainly based on
the interrelations between layers and the
heterogeneity of elements (Schütz et al., 2013). In
addition, it is also related to the characteristics of the
dynamic aspect in our approach: elementary
transition EAi to EAi+1 and also sprints and weekly
vertical alignment in each transition (Daoudi et al.,
2020a).
In the following we will define each complexity
factor. For objective factors we used quantitative
metrics. As for subjective factors, we adopted a
scoring methodology to translate the expert
judgment in numbers Perera et al. (2005).
First at architectural level for subjective factors,
we proposed “context awareness” which expresses
the ability to catch internal changes and to adapt the
project details accordingly at architectural level.
Then we have “ambiguity” that shows to which
extent the decisions and the communication are
traceable and clear to all the stakeholders. There is
also “uncertainty” that shows the level of
uncertainty in project estimations according to the
owners (Architecture, Business and IT) and the
assumptions taken for the unconstrained plans.
Another factor is “Security”. It describes the degree
of complexity of security requirements in the whole
project. Finally, we have “Risk assessment” which
shows the assessment of risks in the elementary
architectural transition. Then, we have objective
factors at architectural level. We considered in those
the “Interdependencies between different layers”
which is the different relationships interlayers. We
also identified some factors related to the project of
moving from EA
i
to EA
i+1
: “Number of deliverables
estimated of the project”, “effort estimated of the
project”, “Cost/budget of the project” and
“Duration of the project estimated”.
At strategy level, we identified two subjective
factors. “Context awareness” is the first one it and it
expresses the degree of integration of strategic
priorities and the level of support from management.
Then, we have “Competing soft goals”. As we
proposed the use of goal modelling (Doumi, 2013),
the assessment of soft goals and the identification of
the competing ones gives an outlook over the trade-
offs that will be needed. The other objective factor is
related to interdependencies and the number of
interrelated had goals.
Regarding organisation, we suggest three
objective factors. “Variety of Stakeholders and
competing views” allows the calculation of the
concentration of the business units, the geographic
dispersion, the division labor, the competing
stakeholders views and the implicated contingent
companies. Then, “team size” which is the number
of collaborators implicated in the project. Finally,
the “Variety of skills” that shows the distribution of
skills that are needed in the project.
At business layer, we propose the calculation of
interdependencies that are impacted by the
elementary transition. We will deep dive into the
method of calculation of this objective factor in the
next parts. The other factor is the existence of
“Business KPIs” to monitor the transition or the
ICEIS 2021 - 23rd International Conference on Enterprise Information Systems
764

necessity of creation of new ones. This one is
subjective and assessed by expert judgment.
At IT layer, we proposed the assessment of
“Infrastructure and material resource availability”
so as to identify the needed acquisitions, leasing and
partnerships. Then, we identified the “variety of
systems and applications” that shows the
heterogeneity of applications and systems and the
number of their types. Another important factor is
the “Quality-of-Service” required in the IT systems
and the network. We also have the
interdependencies between the impacted
components in IT layer.
Moreover, we added the external perspective
which is an outlook over the environment of the EA.
It is mainly focused on the analysis of external
limitations, compliances and regulations.
Regarding the factor “interdependence and
interrelations” in different sub systems and between
them, we suggest the use of matrix notation based on
ArchiMate 2.0.
This representative matrix XY
n,m
n,m ∈ N
2
will
be constituted of n rows representing one layer and
m columns representing the other layer. Also, X and
Y belongs to {S,B,A,I}. We propose then six
representative matrices: SS, SB, BB, BA, AA, AI.
SS has soft goals in rows and Hard goals in columns,
SB has hard goals in rows and Business process in
columns, BB represents the intra-relations inside the
business layer, BA has business processes in the
rows and applications in the columns, AA represents
the intra-relations in the application layer and AI has
applications at rows and infrastructure components
at columns. The elements of the representative
matrices are couples (a
ij
,w
ij
) ∈ {0,1}xN i, j xN
2
where a
ij
represents the existence of relationship
between rows and columns and w
ij
represents the
weight of this relationship.
Based on this definition, we can automatically
find impacted entities in the business layer,
application layer and in the infrastructure layer
through dependency chain. We use for this purpose
the following operator “x”:
We suppose : i, j, l, n, m ∈ N
5
A ={(a
ij
,w
ij
)}
n,m
where(a
ij
,w
ij
) ∈ {0,1}xN
B={(b
ij
,p
ij
) }
m,l
where (b
ij
,p
ij
) ∈ {0,1}xN
R = {(r
ij
,k
ij
) }
n,l
where (rij,kij) ∈ {0,1}xN
The resulting matrix is then
R= A x B = {(
⋃
𝑎𝑖𝑘 ∗ 𝑏𝑘𝑗


,
⋃
𝑤𝑖𝑘 ∗ 𝑝𝑘𝑗


)}
n,l
Where U is the OR operator.
The number of interdependences is then the sum of
the first part of elements impacted. Based on the
resulting matrix, we select the set of elements
impacted and sum its elements. The value is couple
represented by the number of relations and the
weight of the relations.
Regarding the variety of applications and systems
and the variety of stakeholders we suggest the use of
Entropy.
The term entropy was introduced in 1865 by Rudolf
Clausius. He developed the concept based on the
formulation of the second law of thermodynamics.
The entropy of a system is determined by the
number of states accessible to the system, and the
probability of occurrence of each of those states.
Its formula is :
S= -
∑
𝑝𝑖 ln(𝑝𝑖)


Where S is the entropy and pi the probability of
each state of the studied system.
According to (Martínez-Berumen, 2014), we can
consider the organisational aspect as a system and
thus apply Entropy to it. We will use his definition
of organisational entropy. For the variety of
applications and systems, we will also use entropy
so as to assess the heterogeneity of the landscape.
Regarding risk assessment, we suggest the use of
a matrix that contains the risks based on expert
knowledge, to assess them and characterize their
impact using a scoring notation. The risk of the
project then can be categorized from 1 to 3 (High
risk, medium risk and low risk).
For other factors, apart from the calculation of
the estimations proposed in the Table 1, we suggest
the use of a scoring notation from 1 to 10. This will
allow us to sum all the values gathered and define
classes of complexity of an elementary transition
project.
5 CONCLUSION
In this paper we explored the literature regarding
complexity in general and in relation with Enterprise
Architecture. Then we outlined and formalized our
methodology based on factors and indicators to
monitor complexity when doing an elementary
transition (project) in the Adaptive EA approach.
We also managed to categorize those factors based
on the implication of an expert stakeholder
(Subjective/ Objective) and the perspectives targeted
Complexity and Adaptive Enterprise Architecture
765

(Architecture, Strategy, Business, Information
System and External). The main contributions in
this paper are to provide the project managers
(Architecture owner, Business owner, IT owner and
management stakeholders) with a set of indicators to
monitor complexity and also to stimulate discussion
about complexity in Enterprise Architecture context.
In subsequent work, we aim to propose a
prototype that integrates the complexity factors and
apply it to a case study. We will also deep dive into
the definition of some factors and metrics that can
add more concreteness to the other criteria and also
explore the use of data driven analysis in our model.
REFERENCES
Abankwa D.A., Rowlinson S.,and Adinyira E. (2019).
Conceptualizing team adaptability and project
complexity: A literature review. International Journal
of Innovation, Management and Technology
Aristotle. Metaphysics, Book VIII, 1045a.8–10
Aven, T., and Renn, O. (2010). Risk management. In Risk
Management and Governance, pp. 121-158. Springer,
Berlin, Heidelberg.
Aven T. (2016). Ignoring scenarios in risk assessments:
Understanding the issue and improving current
practice. Reliability Engineering & System Safety. pp.
215-220
Bertelsen S. (2004). Construction management in a
complexity perspective. International SCRI
Symposium
Baccarini D. (1996). The concept of project complexity—a
review. International Journal of Project Management,
vol. 14. pp. 201–204
Bjerga T., et Aven T. (2016). Some perspectives on risk
management: A security case study from the oil and
gas industry. Proceedings of the Institution of
Mechanical Engineers, Part O: Journal of Risk and
Reliability, pp. 512-520
Bjorvatn T. and Wald A. (2018). Project complexity and
team-level absorptive capacity as drivers of project
management performance. International Journal of
Project Management, vol 36, pp. 876–888
Brockmann C. and Girmscheid G. (2007). Complexity of
Megaprojects, in: CIB World Building Congress:
Construction for Development
Cooke-Davies T., Cicmil S., Crawford L., and Richardson
K. (2007). We’re not in Kansas anymore, Toto:
Mapping the strange landscape of complexity theory,
and its relationship to project management. Project
Management Journal, vol. 38, pp. 50-61
Custovic E. (2015). Engineering management: old story,
new demands. IEEE Engineering Management
Review, vol. 43, no. 2, pp. 21–23
Daoudi W., Doumi K., et Kjiri L. (2020a). An Approach
for Adaptive Enterprise Architecture. International
Conference on Enterprise Information Systems ICEIS,
pp. 738-745
Daoudi W., Doumi K., and Kjiri L. (2020b). Adaptive
Enterprise Architecture: Initiatives and Criteria.
International Conference on Control, Decision and
Information Technologies (CoDIT), vol. 1, pp. 557-
562
Doumi K. (2013). Approche de modélisation et
d’évaluation de l’alignement stratégique des systèmes
d’information Application aux systèmes d’information
universitaires. PhD Thesis
Efatmaneshnik M., and Ryan M.J. (2016). A general
framework for measuring system complexity.
Complexity. pp. 533–546
Flyvbjerg B. and Budzier A. (2011). Why Your IT Project
May Be Riskier Than You Think. Harvard Business
Review
Iacob M. E., Monteban J., Van Sinderen M., Hegeman E.,
et Bitaraf K. (2018). Measuring Enterprise
Architecture Complexity. IEEE International
Enterprise Distributed Object Computing Workshop.
pp. 115-124
IEEE Architecture Working Group (2000). IEEE
Recommended Practice for Architectural Description
of Software-Intensive Systems
International Risk Governance Council (2012). An
introduction to the IRGC Risk Governance
Framework. ISBN 978-2-9700772-2-0
Janssen M., and Kuk G. (2006). A Complex Adaptive
System Perspective of Enterprise Architecture in
Electronic Government. Hawaii International
Conference on System Sciences
Kahane A. (2004). Solving Tough Problems: an Open Way
of Talking, Listening, and Creating New Realities.
Berrett-Koehler Publishers
Kandjani H., Bernus P., and Nielsen S. (2013). Enterprise
Architecture Cybernetics and the Edge of Chaos:
Sustaining Enterprises as Complex Systems in
Complex Business Environments. Hawaii International
Conference on System Sciences. pp. 3858–3867
Kaplan S., et Garrick B. J. (1981). On the quantitative
definition of risk. Risk analysis. pp. 11-27
Kelly, D., et Smith, C. (2011). Bayesian inference for
probabilistic risk assessment: A practitioner's
guidebook. Springer Science & Business Media
Kurtz C.F.et Snowden D.J. (2003). The New Dynamics of
Strategy: Sense-making in a Complex and
Complicated World. IBM Systems Journal, vol 42, no.
3, pp. 46
Lagerström R., Baldwin C.Y., Maccormack A.D., and
Aier S. (2013). Visualizing and Measuring Enterprise
Application Architecture: An Exploratory Telecom
Case. Harvard Business School Working Paper, pp.
13-103
Lee H., Ramanathan J., Hossain Z., Kumar P., Weirwille
B., and Ramnath R.. (2014). Enterprise architecture
content model applied to 122 complexity management
while delivering IT services. IEEE International
Conference on Services Computing. pp. 408–415
ICEIS 2021 - 23rd International Conference on Enterprise Information Systems
766

Lucke C., Krell S., and Lechner U. (2010). Critical issues
in enterprise architecting: A literature review.
Americas Conference on Information Systems
Martínez-Berumen H. A., López-Torres G. C., et Romo-
Rojas L. (2014). Developing a Method to Evaluate
Entropy in Organizational Systems. CSER, pp. 389-
397
Mocker M. (2009). What is Complex about 273
Applications? Untangling Application Architecture
Complexity in a Case of European Investment
Banking. Hawaii International Conference on System
Sciences.
Morcov, S., Pintelon, L., et Kusters, R. J. (2020).
Definitions, characteristics and measures of IT project
complexity-a systematic literature review.
International Journal of Information Systems and
Project Management
Padalkar M. and Gopinath S. (2016). Are complexity and
uncertainty distinct concepts in project management?
A taxonomical examination from literature.
International Journal of Project Management, vol. 34,
pp. 688–700
Parsons-Hann H. and Liu K.. (2005). Measuring
requirements complexity to increase the probability of
project success. International Conference on
Enterprise Information Systems, pp. 434–438
Perera, J., et Holsomback J. (2005). An integrated risk
management tool and process. IEEE Aerospace
Conference, pp. 129-136
Pich M. T., Loch C. H., and Meyer A. D. (2002). On
uncertainty, ambiguity, and complexity in project
management. Management Science, vol. 48, no. 8, pp.
1008–1023
Qazi, A., Quigley, J., Dickson, A., et Kirytopoulos, K.
(2016). Project Complexity and Risk Management
(ProCRiM): Towards modelling project complexity
driven risk paths in construction projects.
International journal of project management, pp. 1183-
1198
Remington K., Zolin R., and Turner R. (2009). A model of
project complexity: distinguishing dimensions of
complexity from severity. International Research
Network of Project Management Conference
San Cristóbal, J. R. Carral, L., Diaz E., Fraguela, J. A., et
Iglesias, G. (2018). Complexity and project
management: A general overview. Complexity
Schmidt C. (2015). Business architecture quantified: How
to measures business complexity. Business
Architecture Management, pp. 243–268
Schneider A., Zec M., and Matthes F. (2014). Adopting
notions of complexity for enterprise architecture
management. Americas Conference on Information
Systems
Schütz A., Widjaja T., and Kaiser J. (2013). Complexity in
enterprise architectures – Conceptualization and
introduction of a measure from a system theoretic
perspective. European Conference on Information
Systems
Smuts J. C. (1927). Holism and Evolution. 2nd Edition.
Macmillian and Co
Taleb N. N., Goldstein D. G. and Spitznagel M. W.
(2009). The Six Mistakes Executives Make in Risk
Management. Harvard Business Review
Trinh M. T., and Feng, Y. (2020). Impact of project
complexity on construction safety performance:
Moderating role of resilient safety culture. Journal of
Construction Engineering and Management
Vidal L.-A. and Marle F. (2008). Understanding project
complexity: implications on project management.
Kybernetes, vol. 37, no. 8, pp. 1094–1110
Ward S. and Chapman C. (2003). Transforming project
risk management into project uncertainty
management. International Journal of Project
Management, vol. 21, no. 2, pp. 97– 105
Yadav V., Agarwal V., Gribok A.V., and Smith C.L.
(2017). Dynamic PRA with Component Aging and
Degradation Modeled Utilizing Plant Risk Monitoring
Data. International Topical Meeting on Probabilistic
Safety Assessment and Analysis
Zio E. (2018). The Future of Risk Assessment. Reliability
Engineering and System Safety
Zio, E. (2016). Challenges in the vulnerability and risk
analysis of critical infrastructures. Reliability
Engineering et System Safety, pp. 137-150
Complexity and Adaptive Enterprise Architecture
767