Limited Offer Get 25% off — use code BESTW25
No AI No Plagiarism On-Time Delivery Free Revisions
Claim Now
Prices start at $5 per page · plagiarism-free papers · free revisions for 14 days 24/7 support: support@bestwriters.com
Uncategorized

Data & Knowledge Engineering

  • Posted
  • Authoradmin
  • Length10,985 words
  • Reading time48 min
Write My Assignment FREE

Empirical Insights into the Development of a Service-oriented Enterprise
Architecture
Ayed Alwadain, Erwin Fielt, Axel Korthaus, Michael Rosemann

PII:
DOI:
Reference:
S0169-023X(15)00082-8
doi: 10.1016/j.datak.2015.09.004
DATAK 1544

To appear in: Data & Knowledge Engineering

Received date:
Revised date:
Accepted date:
7 March 2015
14 September 2015
15 September 2015

Please cite this article as: Ayed Alwadain, Erwin Fielt, Axel Korthaus, Michael Rosemann, Empirical Insights into the Development of a Service-oriented Enterprise Architecture, Data & Knowledge Engineering (2015), doi: 10.1016/j.datak.2015.09.004
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
1
Paper Title: Empirical Insights into the Development of a Service-oriented Enterprise
Architecture
Ayed Alwadain
Computer Science Department
King Saud University
Riyadh, Saudi Arabia
aalwadain@ksu.edu.sa
Erwin Fielt
School of Information Systems
Queensland University of Technology
Brisbane, Australia
e.fielt@qut.edu.au
Axel Korthaus
College of Business, Information Systems Discipline
Victoria University
Melbourne, Australia
axel.korthaus@vu.edu.au
Michael Rosemann
School of Information Systems
Queensland University of Technology
Brisbane, Australia
m.rosemann@qut.edu.au
Abstract
Organisations use Enterprise Architecture (EA) to reduce organisational complexity, improve communication, align
business and information technology (IT), and drive organisational change. Due to the dynamic nature of
environmental and organisational factors, EA descriptions need to change over time to keep providing value for its
stakeholders. Emerging business and IT trends, such as Service-Oriented Architecture (SOA), may impact EA
frameworks, methodologies, governance and tools. However, the phenomenon of EA evolution is still poorly
understood. Using Archer’s morphogenetic theory as a foundation, this research conceptualises three analytical
phases of EA evolution in organisations, namely conditioning, interaction and elaboration. Based on a case study
with a government agency, this paper provides new empirically and theoretically grounded insights into EA
evolution, in particular in relation to the introduction of SOA, and describes relevant generative mechanisms
affecting EA evolution. By doing so, it builds a foundation to further examine the impact of other IT trends such as
mobile or cloud-based solutions on EA evolution. At a practical level, the research delivers a model that can be used
to guide professionals to manage EA and continually evolve it.
Keywords
Enterprise Architecture, EA, Service-oriented Architecture, SOA, critical realism, morphogenetic theory
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
2
Empirical Insights into the Development of a Serviceoriented Enterprise Architecture
Abstract
Organisations use Enterprise Architecture (EA) to reduce organisational complexity, improve communication, align
business and information technology (IT), and drive organisational change. Due to the dynamic nature of
environmental and organisational factors, EA descriptions need to change over time to keep providing value for its
stakeholders. Emerging business and IT trends, such as Service-Oriented Architecture (SOA), may impact EA
frameworks, methodologies, governance and tools. However, the phenomenon of EA evolution is still poorly
understood. Using Archer’s morphogenetic theory as a foundation, this research conceptualises three analytical
phases of EA evolution in organisations, namely conditioning, interaction and elaboration. Based on a case study
with a government agency, this paper provides new empirically and theoretically grounded insights into EA
evolution, in particular in relation to the introduction of SOA, and describes relevant generative mechanisms
affecting EA evolution. By doing so, it builds a foundation to further examine the impact of other IT trends such as
mobile or cloud-based solutions on EA evolution. At a practical level, the research delivers a model that can be used
to guide professionals to manage EA and continually evolve it.
Keywords
Enterprise Architecture, EA, Service-oriented Architecture, SOA, critical realism, morphogenetic theory
1 Introduction
Enterprise Architecture (EA) has been successfully employed to reduce organisational complexity, improve
communication, align business and information technology (IT) and drive organisational change (Lankhorst, 2005;
Schekkerman, 2005). The importance EA has gained is reflected by a recent Gartner survey, which stated that EA
practitioners strongly influence organisations’ IT budgets and globally are either the final decision maker or greatly
influence more than $1.1 trillion in enterprise IT spending (Gartner, 2012a). There are various perspectives of what
is meant by the term EA, and as yet no common definition has emerged. As a working definition, we follow
Lankhorst (2005, p. 3) who defines EA as: “… a coherent whole of principles, methods, and models that are used in
the design and realization of an enterprise’s organisational structure, business processes, information systems, and
infrastructure” (Lankhorst, 2005, p.3). Given the dynamic nature of organisations and the continual emergence of
new technological and business paradigms, however, it is essential that EA must not be static. As a prerequisite to
successful benefits realisation, it is crucial that EA is continually developed, maintained and kept up to date.
Frameworks, methodologies and terminology used in developing EA need to take into account that organisations are
dynamic, adaptive systems of systems (Sampaio, 2010; Sousa, Lima, Sampaio, & Pereira, 2009) and it is, therefore,
essential to understand and plan EA evolution (MacLennan & Belle, 2014; McKendrick, 2010; Shah & Golder,
2011).
EA evolution needs to be planned on both the level of architectural descriptions (the EA frameworks, meta-models,
concepts etc.) as well as on the level of architectural representations (Martin, Purao, & Robertson, 2009). The level
of architectural representations relates to the models that represent the architecture of the analysed organisation
from a business and IT perspective and thus form the central deliverable of the EA practice. When EA evolution is
not managed and aligned at this level, misrepresentation and occasionally even failures result (Martin, et al., 2009).
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
3
Related architecture models must be accurately and traceably linked to their implementation to manage the
complexity, development and maintenance of evolving systems. Any changes to the implementation have to be
reflected back in the architecture to keep EA evolving and correctly describing the organisation (Mens, Magee, &
Rumpe, 2010). To align EA models with the corresponding real world, enterprise architects have to be aware of
changes affecting the enterprise and its EA (Roth, Hauder, Farwick, Breu, & Matthes, 2013).
Architectural descriptions, on the other hand, are the “vehicle” for building architectural representations (Martin, et
al., 2009). On this level EA as a discipline faces the fundamental challenge of re-thinking EA frameworks, metamodels and concepts in response to emerging business and IT paradigms and capabilities, such as, for example,
Service-Oriented Architecture (SOA) and cloud computing (McKendrick, 2010), and embedding them adequately in
existing EA frameworks (Jung, 2009; Mens, et al., 2010; Roth, et al., 2013). This may require changes to
architectural elements and the relationships that are used as part of the EA practice. The architectural elements are
the elements that enclose and describe an organisation’s business, people and technology (more specifically, its
strategies, business principles, stakeholders, locations, functions, activities, processes, products, information,
applications, systems, infrastructure, etc.) (Schekkerman, 2004, p. 22). When new paradigms, such as SOA, emerge
and change the enterprise, new architectural elements and relationships may need to be considered (Banerjee &
Aziz, 2007; Infosys, 2009; Postina, Trefke, & Steffens, 2010; Varnus & Panaich, 2009).
The phenomenon of EA evolution is still poorly understood. While EA evolution presents significant challenges to
organisations (Land, Proper, Waage, Cloo, & Steghuis, 2009; Short, 2013), research on this phenomenon so far is
very limited and there is a lack of understanding of how EA actually evolves, in particular on the level of
architectural descriptions. More specifically, there are no empirical studies that describe or explain how EA in
organisations evolves due to emergent business and IT trends such as SOA (Knippel & Skytte, 2007; McKendrick,
2010; Postina, et al., 2010; Saat, Aier, & Gleichauf, 2009; Shah & Golder, 2011). A limited number of studies have
either focused on the representational changes of EA, such as changes to applications (e.g. see Buckl, Ernst,
Matthes, & Schweda, 2009; Farwick, Schweda, Breu, Voges, & Hanschke, 2012), or provided examples of EA
evolution on the architectural descriptions level without considering the underlying process of evolution or what
may impact the evolution (e.g. see Banerjee & Aziz, 2007; Correia & Silva, 2007; Postina, et al., 2010;
Shankararaman & Kazmi, 2011; Sharma, 2013 ). The existing approaches do not clearly consider external influences
on the EA planning process or changing conditions in an organisation’s environment (Saat, et al., 2009).
The emphasis of this research lies on changes to architectural descriptions that result from the emergence of new
business or IT paradigms, which introduce new concepts, architectural elements and new ways of thinking about
organisations. We specifically study the introduction of SOA as this is one of the major, relatively recent new
architectural paradigms with the potential to trigger substantial EA evolution. SOA is defined as “an architectural
style that supports service orientation, and service orientation is a way of thinking in terms of services and servicebased development and the outcomes of services” (The Open Group, 2010). SOA is among the top issues that
organisations are trying to address using EA (Varnus & Panaich, 2009) and SOA integration into EA needs more
attention (Dico, 2012). In fact, many studies explicitly argue that EA needs to evolve to address and integrate SOA
(Khoshnevis, Aliee, & Jamshidi, 2009; MacLennan & Belle, 2014; Postina, et al., 2010; Sanders, Hamilton, &
MacDonald, 2008; Sharma, 2013 ; Viering, Legner, & Ahlemann, 2009). Kistasamy, Van der Merwe, and De La
Harpe (2012) argue that, although both EA and SOA have matured, there is a lack of understanding of the
relationship between them, which has resulted in a marginal realisation of their combined benefits (Kistasamy, et al.,
2012).
The objective of this study is to describe the EA evolution process and explain observed EA evolution outcomes at
the level of architectural descriptions. It specifically focuses on the introduction and implementation of SOA in an
organisation as one exemplary trigger for an EA evolution process. From a conceptual point of view, this research
views EA evolution as an interaction between the existing EA and the action of introducing a new IT phenomenon,
which may result in EA evolution outcomes using Archer’s (1995) morphogenetic theory as a foundation. We have
chosen a case study-based approach to be able to derive rich empirical observations from an organisation that has
introduced SOA, which had an impact on the evolution of the organisation’s EA.
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
4
We proceed as follows. The next section will lay the theoretical foundation for the study by introducing the
morphogenetic theory as the analytical lens to examine the EA evolution phenomenon and developing the study’s
research model. After that, the research methodology is outlined and background information about the case
organisation and the data collection and analysis is provided. The following section presents the case study findings
in detail, structured along the components of the research model presented in the second section. Finally, a
discussion of the findings and a report on the insights gained from the study conclude the paper.
2 Theoretical foundations
To add a unique contribution to the body of EA knowledge, we position this research from two perspectives: its
focus (EA evolution at the level of architectural descriptions) and its theoretical basis (i.e., theory building or theory
testing). With respect to the latter, EA is still a young domain that presents many challenges for researchers and EA
studies often lack sound theoretical foundations (Schmidt & Buxmann, 2011; Simon, Fischbach, & Schoder, 2013).
Therefore, this research aims at building theory that describes and explains how EAs evolve as a complex,
organisational level phenomenon and an important aspect of EA management (Winter, Buckl, Matthes, & Schweda,
2010). Lynham (2000) argues for theory building (1) to advance professionalism in and maturity of a given field and
(2) to help bridge the gap between research and practice. For the purposes of this research, we use Lynham’s (2000,
p. 222) definition of theory as “a coherent description, explanation, and representation of observed or experienced
phenomena”.
To theorise about EA evolution, this research recognises both the inherent complexity and the temporality of EA
evolution and the need for an analytical lens for understanding its evolution. The study investigates EA evolution
using the morphogenetic theory (Archer 1995), a critical realism-based theory, to comprehend the evolution process
triggered by SOA introduction. Archer’s morphogenetic theory is adopted because it considers an explicit temporal
dimension to study change, which is fitting for an investigation of EA evolution (see Figure 1). It provides a useful
conceptualisation for the examination of organisational changes, particularly those involving technology (Mutch,
2010).
According to Archer (1995), every morphogenetic cycle (see Figure 1) differentiates between three broad analytical
phases which comprise: “(a) a given structure (a complex set of relations between parts), which conditions but does
not determine (b), social interaction”. The social interaction is also influenced by agents’ orientations and “in turn
leads to (c), structural elaboration” (Archer, 1995, p. 91). Archer (1995) argues that the analysis must start at time
(T1) to include the structural conditioning that is formed by previous actions, not at the time of (T2) when the social
interaction takes place. Cuellar (2010, p. 41) states that “previous cycles have created a particular set of existing
structures and distributions of resources as the result of prior cycles which condition the actions of existing
agencies”. The interaction phase is enabled or constrained by pre-existing conditions. It is also influenced by agents’
orientations, interests, and interpretations (Archer, 1995; Danermark, Ekström, Jakobsen, & Karlsson, 2002). The
third analytical phase is the elaboration phase, which can be either the reproduction or transformation of pre-existing
structure: Structures may stay as they are with no changes (reproduction) or may change (transformation) (Archer,
1995). The three phases: conditioning, interaction, and elaboration are continuous through time but, as an analytical
tool, dualism is employed to let the researcher cut into reality and project cycles forwards and backwards.
According to Archer (1995), the point of examining any morphogenetic cycle is to provide an analytical perspective
on the emergence of outcomes under investigation (EA evolution outcomes in this case). The interplay between the
architectural conditioning phase (T1) and the architectural interaction phase (T2-T3) occurs in a morphogenetic
cycle (SOA integration into EA), which defines how the architectural elaboration (EA evolution outcomes) occur
(Archer, 1995; Cuellar, 2010). Building on the morphogenetic theory, the elaboration (EA evolution outcomes) is
the result of the interplay between the action-formation mechanisms of the interaction (SOA introduction) and the
conditional generative mechanisms of the conditioning phase. The bottom half of Figure 1 shows the appropriation
of the theory in the context of this study to conceptualise the EA evolution process by distinguishing three phases:
(1) architectural conditioning (due to an organisation’s pre-existing EA), (2) architectural interaction (e.g., SOA
introduction) and (3) architectural elaboration (outcomes of EA evolution). Next, we developed the model into more
detail (Figure 2) based on a literature review and explorative interviews as described in previous work (Alwadain,
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
5
Fielt, Korthaus, & Rosemann, 2014). For the architectural elaboration (T4), five possible levels of EA evolution
outcomes were identified: business architecture, information systems architecture, technology architecture, EA
governance, and EA methods and tools. The first three outcome levels are commonly addressed in EA frameworks
as perspectives or layers. The latter two were added to the research model as a result of the explorative interview
phase which was reported in Alwadain, et al. (2014).
Structural conditioning
T2 Social interaction T3
T1
Structural elaboration T4
T1 Architectural conditioning
T2 Architectural interaction T3
Architectural elaboration T4
Figure 1. Mapping between morphogenetic theory and EA evolution
Of particular interest for this study are the generative mechanisms related to the architectural conditioning (T1) and
architectural interaction (T2-T3) phases that can explain EA evolution outcomes. Generative mechanisms (causal
powers) are one of the main components of critical realist studies. A central aspect of mechanisms in the critical
realism tradition is that they present a source of explanatory power (Archer, 1995; Bhaskar, 1998). Archer (1995)
states that structures have properties that enable them to influence the world around them (Archer, 1995; Cuellar,
2010). The architectural conditioning phase includes mechanisms that are situational (conditional) and the
architectural interaction phase includes action-formation mechanisms (Volkoff & Strong, 2013). We identified three
conditional generative mechanisms related to architectural conditioning (T1): EA framework, EA objectives and EA
maturity, and six action-formation generative mechanisms related to architecture interaction (T2-T3): View of SOA,
SOA perceived benefits, SOA scope, SOA governance, SOA design and Business-IT collaboration.
T1 Architectural Conditioning
T2 Architectural Interaction T3
Architectural Elaboration T4
 Business architecture
 IS architecture
 Technology architecture
 EA governance
 EA methods and tools
 View of SOA
 SOA perceived benefits
 SOA scope
 SOA governance
 SOA design
 Business-IT collaboration
 EA framework
 EA objectives
 EA maturity
Figure 2. The study’s theoretical model
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
6
In this study, the phases of the morphogenetic cycle of SOA’s integration into EA were determined by using the
stability-change-stability approach (Njihia, 2008; Trosper, 2005), which suggests that the researcher tentatively
adopt a set of periods, distinguished by times of stability and times of change, to identify the analytical
morphogenetic cycle boundaries (Njihia, 2008; Trosper, 2005), where each morphogenetic cycle signifies a
substantial change in the structure (Njihia 2008). The scope of change under investigation here is limited to SOA
introduction as a trigger of EA evolution. Any other aspects that potentially cause changes to EA (e.g, new corporate
strategy, emerging technologies such as mobile or social) are outside the scope of this study.
3 Research Methodology
Given the complexity of EA evolution as a phenomenon to be investigated and the nature of the research, a
qualitative approach involving explorative interviews, reported in Alwadain, et al. (2014), followed by a case study
was undertaken to collect and analyse the empirical data. A qualitative approach is generally considered appropriate
for conducting critical realism-based studies (Danermark, et al., 2002; Sayer, 1992). In this paper, we will present
the findings of a case study conducted at a public sector agency in the United Arab Emirates. The case study method
was chosen to contextualise the developed research model of EA evolution, reported in Alwadain, Fielt, Korthaus,
and Rosemann (2013b) and Alwadain, et al. (2014), as suggested by (Danermark et al. 2002), and to further
understand EA evolution in real world settings.
The case study method is one of the most extensively used qualitative research methods in information systems
research (Benbasat, Goldstein, & Mead, 1987). Yin (2003, p. 13) defines the case study as “an empirical inquiry that
investigates a contemporary phenomenon within its real-life context, especially when the boundaries between
phenomenon and context are not clearly defined”. It is an appropriate method to investigate emerging phenomena
about which few previous studies have been conducted. Indeed, Eisenhardt (1989) highlights that the case study
method is “especially appropriate in new topic areas” or in areas in which few studies have been conducted
(Benbasat et al., 1987). Furthermore, the case study method is well suited to conduct critical realist research when
studying contemporary socio-technical phenomena (such as EA evolution) to uncover the causal mechanisms that
generate evolution outcomes (Easton, 2010; Mingers, 2004; Wynn & Williams, 2012; Wynn Jr & Williams, 2008).
Another aspect related to the case study design is identifying the unit of analysis. The unit of analysis is a
fundamental aspect in qualitative research studies: it defines what the case is (Yin, 2003). Easton (1998) proposes a
form of embedded unit of analysis based on time; that is, the case must have a longitudinal component and should be
seen as a series of cases (embedded) depending on the period of time being investigated and described. As such, in
this thesis, the overall unit of analysis is EA evolution due to SOA’s introduction, which includes embedded units of
analysis. These embedded units of analysis are the investigation of EA prior to SOA’s introduction, a detailed
examination of SOA’s introduction, and the outcomes of SOA’s integration into EA. The aggregation of analyses of
the sub-units composed the analysis of EA evolution (here: SOA’s integration into EA).
3.1 Background of the case
The case study organisation being reported on in this paper is a government agency in the United Arab Emirates. It
has approximately 3,000 employees. The agency is a leader in delivering innovative services supporting national
objectives and improving customer experiences. According to an internal document of the agency ([D-11], see Table
2), contemporary systems, easy practices and cooperative and chivalrous service are and will be maintained as
keystones of the agency practices. For a long time, the agency has contributed to the wide country development,
creating a far-reaching sphere of operations, and facilitated the country’s economical power.
The agency adopted an Enterprise Architecture program in 2006 to facilitate its strategy, business and technology
alignment. Its EA program was part of a larger transformation initiative and EA was used to support fast decisionmaking and the dynamic business needs of the organisation. Moreover, the agency was one of the first leading
public agencies in the United Arab Emirates to undergo an e-government transformation and to progress toward the
broad electronic delivery of services. In particular, the agency introduced SOA in 2008 to support the delivery of
services.
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
7
3.2 Data collection and analysis
During June 2012, eight interviews were conducted with eight senior executives at the agency premises. Table 1
presents the participants’ information. Interviews were conducted following a case study protocol. Each interview
lasted between 40 and 90 minutes and was recorded and transcribed.
Table 1.Participants’ demographics

Participant Position Years in org. Years of experience Background
P-1 Head of IT planning and
enterprise architecture
5 7 Business and
management
P-2 IT strategist 6 15 Strategy/planning
P-3 IT strategist 4 0 IT/strategy
P-4 Senior business architect 6 20 Business
P-5 Senior business architect 6 25 Business
P-6 Senior business architect 5 5 Business
P-7 Senior tech architect 5 10 Technology
P-8 Senior tech architect 3 10 Technology

To achieve triangulation, besides the conducted interviews, sixteen documents, see Table 2, related to both EA and
SOA were obtained from the department (internally) or online from the agency’s website or other websites. These
documents are cited as D-1, D-2, and so on throughout this paper. The transcribed interviews and the obtained
documents were imported to NVivo 9 to prepare them for analysis. Analysis of the collected data was informed by a
thematic analysis procedure and, more specifically, a deductive approach using the theoretical model, shown in
Figure 2, as a lens (Fereday & Muir-Cochrane, 2006).
Table 2. Collected evidence (documents)

ID Source Description
D-1 Online A presentation about the implementation of EA and its benefits (2007).
D-2 Online A presentation about the use of EA as a governance practice (2007).
D-3 Internal A suitability report represents the agency efforts and practices to achieve sustainability (2011).
D-4 Online Report of COBIT implementation: the government efforts to assess governance levels at
different agencies.
D-5 Online A white paper from IBM about the benefits of using EA to align business and IT. The agency
was presented as one of the examples that achieved benefits from its EA implementation.
D-6 Internal This document presents an e-services delivery excellence model for the electronic provision
and improvement of government services. It outlines guiding principles for services
enablement evaluation.
D-7 Internal A presentation on the use of EA for knowledge management and knowledge sharing tool.
D-8 Internal An internal document for services identification and classification.
D-9 Internal Classification of the agency’s domains.
D-10 Internal The agency’s EA meta-model.
D-11 Internal A document with information related to the agency’s SOA program.
D-12 Internal A document with information related to the SOA program (guide of SOA program).
D-13 Internal A presentation about the agency’s EA implementation and benefits.
D-14 Internal A report from IBM describing the implementation of EA in the agency.
D-15 Online A report about the launch of the agency’s SOA-based electronic system.
D-16 Internal Information about some technical aspects of SOA implementation.

3.3 Application of the theoretical framework
The agency’s morphogenetic cycle comprising SOA integration into EA is shown in Figure 3. It was determined by
using the stability-change-stability approach as discussed in section 2. EA was established in 2006 and was
completed by the end of 2007. The period between the establishment of EA (end of 2007) until SOA introduction in
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
8
2008 is considered the architectural conditioning (T1) of the morphogenetic cycle (SOA integration into EA). The
change period (architectural interaction) began when SOA was introduced in mid of 2008 and finished in 2010. This
study was conducted two years later, in June 2012, after SOA introduction was complete.
Figure 3. The morphogenetic cycle of SOA integration into EA
Thus, to understand the outcomes of the integration of SOA into EA, the event (SOA introduction) and the status of
EA prior to the interaction were studied retrospectively (see Figure 4). The retrospective analysis was achieved
through intensive interviews with executives involved in EA and SOA, and was supported by the analysis of
obtained relevant documentations. In an effort to address the possible limitations of exploring a time-consuming
phenomenon through retrospective interviews, multiple participants with different backgrounds and hierarchical
levels were interviewed, and internal and online documents were examined to provide multiple triangularly
perspectives as mentioned above in section 3.2.
SOA integration within
EA outcomes
Retrospective Analysis of both EA and SOA
In 2006, EA was
established
In 2008, SOA was
introduced
In 2010, SOA was
finished
In 2012, EA evolution
was examined
Figure 4. Retrospective analysis
4 Case Study Findings
This section presents the case study findings using the three analytical phases of the theoretical model of this study.
The development division, which is one of five main divisions, is the custodian of the agency’s enterprise
architecture. The development division ensures constant modernisation and development of the agency’s
procedures, operations, projects, and services to keep pace with the most recent international practices and
techniques. Participant [P-2] described the division’s role as being in charge of the organisation’s development and
changes:
“Any changes, any development within the organisation is actually done within the [agency’s] development
division. So, we deal with people, processes, technology and information.”
The agency has a business and IT perspective of EA where EA encompasses both domains and provide the needed
support for decision making and changing business needs. For example, as reported in [D-13] EA “articulates and
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
9
connects organisational information to support fast decision making and to have a single point of truth containing
information to support the dynamic business needs….. Enterprise architecture is the broker between Business and
IT. It provides the benefit of knowing why we need to build, what to build, when to build it, and how to build it” [D-
13].
4.1 Architectural conditioning T1
This section addresses the EA-related conditions prior to the architectural interaction (SOA introduction). The basic
argument for considering the architectural conditioning phase that precedes the action (here: SOA’s introduction) is
that EA evolution cannot be fully explained without reference to antecedent architectural conditioning (Archer,
1995). This section briefly presents an overview of the EA implementation and then organises the findings
according to the following three conditional generative mechanisms: EA framework, EA objectives and EA
maturity. These conditional generative mechanisms condition (enable or constrain) the subsequent architectural
interaction phase, which leads to the architectural elaboration (Archer, 1995).
The agency has adopted EA to align its strategy, business and technology. A Senior Business Architect [P-4]
reported that the EA program was part of a large transformation initiative. It was adopted to make sure that the
transformation objectives were realised. He stated: “When the transformation initiatives started… the organisation
wanted a mature practice to be followed to ensure that the realisation of the objectives happens”.
The program was implemented in three phases. The EA program was launched in late 2006 and the first two phases
were completed by June 2007. By November 2007, most EA artefacts such as strategies, business processes,
activities, technical artefacts and their relationships were documented and stored into IBM System Architect (a
software tool) [D-1]. The EA program was successfully implemented across the organisation [D-1].
When the EA program started in 2006, efforts focused on the processes and technical levels of the organisation. The
organisation identified the value chain, business groups, business processes and functions at the business level.
Then, the business architecture was mapped to the technical architecture by identifying how business processes are
realised at the technology level to answer questions such as what applications support what processes, and what
applications run on what infrastructure [P-4 and P-5]. According to the participants’ statements, the organisation’s
EA approach, which focused on business processes, was based on an internal perspective of the organisation itself.
This approach was called “inside-out” by an IT strategist [P-2].
4.1.1 EA Framework
The agency built their own customised EA framework following the principles of the Zachman (Sowa & Zachman,
1992) and TOGAF (The Open Group, 2009) frameworks. The organisation needed a way to help decision-makers
make the right decisions by providing all the information needed about the organisation from different angles. The
organisation divided EA into multiple layers: strategy, resources, process, information and technology. The first
layer, strategy, encompasses the organisation’s business vision, objectives, enablers and performance measures. This
layer holds strategy-related elements such as directions, guidance, objectives, the means of delivering these
objectives and performance KPIs [P-3]. The second layer, resources, holds elements such as people, assets,
organisation and locations. The third layer, process, holds business processes, business process definitions and
metrics. The information layer includes information models and information flow. The technology layer includes
applications, data models, technical reference models, hardware and network.
According to a senior business architect [P-5], the whole organisation was decomposed to understand its current (asis) state. It became a reference point for identifying the gaps and the changes that might be required to move to a
future state. These as-is business architecture models were used to discuss project proposals and demands. EA
documentation outcomes were stored in a repository using the IBM System Architect tool.
To improve access to the information stored in their IBM System Architect, the organisation internally developed
the enterprise connected view (ECV), an interactive interface used for navigating and querying stored information.
The executive director of the development division stated: “we use System Architect as the basis for our enterprise
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
10
connected view (ECV), which enables us to manage enterprise information, run impact analysis, and make decisions
more effectively” [D-5].
4.1.2 EA Objectives.
Since the early adoption of EA at the agency, a strategic long-term vision was held. Reported EA objectives, in this
case study, are classified into strategic, operational, IT and governance-oriented objectives (see Table 3) as
suggested in EA literature and the previous explorative interview findings (Alwadain, Fielt, Korthaus, & Rosemann,
2013a).
EA holds organisational strategies for the corporation, departments and divisions. It holds business and technical
information and stores them in one repository. The agency uses EA to align its strategy with that of the government.
EA enables effective governance of both business and technical architecture and facilitates fast responses to changes
in business and IT requirements [D-2]. According to the Head of IT planning and EA [P-1], EA focuses
fundamentally on documenting all informational assets in the organisation to manage and govern the organisation on
multiple levels: strategy, business, information, and technology. EA and ECV were introduced to generate blueprints
of the organisation and to be the single source of official information. They were used for planning, governance,
decision-making, and impact-analysis purposes.
Participant [P-4] commented on the reasons for establishing EA at the agency. He mentioned that EA was adopted to
improve the decision-making process, prioritise work, develop new capabilities, improve technology
implementation and guide the future of the organisation. The organisation has used EA to achieve benefits in areas
such as business and IT alignment, impact analysis and strategic decision-making assistance [D-7].
Table 3. EA objectives at the agency

EA objectives
Strategic Business and IT alignment, strategic decision-making assistance, change management, knowledge
management and gaps identification
Governance Holds strategies, holds business and technical requirements, and effective governance of both
business and IT architectures
Operational Documentation of all enterprise components, reuse of components, impact analysis, discovery of
duplications and standardisation
IT Provide solutions requirements, monitor their development, reduce IT duplications and manage IT
complexity

4.1.3 EA Maturity
As mentioned before, the agency adopted an organisation-wide EA program in late 2006. By the end of 2007, most
of the architectural artefacts were captured and stored in the EA repository. The EA maturity assessment was based
on: (1) the obtained documentations, which describe the early stages of EA prior to SOA introduction, (2)
participants’ responses and (3) an EA maturity assessment survey questionnaire based on the NASCIO maturity
model (NASCIO, 2003), which was handed to the participants. Four completed forms were received.
Table 4. EA maturity dimensions of the agency

EA maturity dimensions Description of the agency’s EA maturity level as reported
Documentation  EA documentation was comprehensive
 Strategy, business, IS & infrastructure information were captured and stored in a
repository
 An interface tool was built to browse, query and navigate the repository and its
content by business and IT personnel
Planning  EA program was well defined and had a structured framework and timeline for
developing the EA
 EA planning was well integrated with major strategic initiatives to help the
organisation to achieve these initiatives objectives

ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
11

 EA was involved in building roadmaps for organisational improvements
 EA enabled assessment of the current situation, identification of gaps and
development of roadmaps and action plans
Governance  EA governance standards, processes and procedures were established and
employed. EA roles were defined and review committee was established.
Evaluation  EA and its products were evaluated on two sides. First, EA methodology & meta
model were reviewed and assessed periodically every two years. Second, EA was
reviewed and changed when needed (e.g., new trends or requirements)
Team and resources  EA team was defined. Tools, frameworks and resources were available for EA
team to support their activities
Business Support

 EA and its activities were supported by top management.
 EA was engaged with business and IT
The documentation of the whole organisation was comprehensive and it was stored in the EA repository. EA had a
well-defined methodology. EA governance practices were established and integrated with organisational
governance. EA was involved in demand and project governance to ensure that projects were aligned with both
strategy and architectural (business and IT) standards. EA’s content and its meta-model were kept up-to date. EA
had diverse and skilled team members. There were about twelve strategists, business architects and technology
architects. EA was supported by top management.
Based on the findings of the interviews, the obtained documents and the answers to the accompanying survey
regarding EA maturity (see Table 4), it is evident that the agency’s EA practices were quite mature (between level 3:
well-defined program and level 4: managed program out of 5) before SOA introduction when considering the fact
that EA practices internationally are still emerging and in the early stages of maturity (Gartner, 2012b).
In summary, the findings of the architectural conditioning phase built the foundation of the analysis to understand
how EA evolved due to SOA introduction. Following the architectural conditioning phase, the architectural
interaction (SOA introduction) began in 2008 and lasted for two years. The findings of that phase are presented in
the next section.
4.2 Architectural interaction T2-T3
This section presents the findings relevant to the second analytical dimension of the theoretical model. It first
provides an overview of SOA introduction at the agency and then presents the six action-formation generative
mechanisms of the research that influence SOA introduction.
While other potentially change-inducing activities may have been in process during the architectural interaction
phase, the scope of our study covers only SOA-related activities and the outcomes of its integration into EA.
However, participants did not mention another large-scale event other than SOA introduction that could have
significantly affected the observed evolution outcomes, nor were any identified in the obtained documents.
The data showed that the agency took initiatives to improve its service delivery. In particular, a service-oriented
initiative in 2008 to deliver eServices to citizens, called eServices Delivery Excellence Model [D-6], accompanied
by internal thoughts to embrace what an IT strategist [P-2] called an “outside-in strategy”, had led to changes in the
organisation and its EA. The outside-in strategy takes an external view of an organisation (customers’ needs) and
targets its restructuring based on that view. For example, some participants described the move to serviceorientation:
“But later we noticed, we were lacking in this area and the global trend is purely from service-orientation
perspective.” [P-4]
“The challenge at the IT level implementation is until you define your services at the business level,
customers’ level, you can’t do them right.” [P-2]
The service-oriented initiative started in 2008 and was officially completed in 2010. It was well-described in a report
(Oracle., n.d.):
“[The agency] required a new operating model, based upon a service-oriented approach, to deliver the
flexibility and scalability needed to accommodate existing and projected levels of trade. Additionally, to
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
12
increase efficiency and reduce unneeded use of internal resources, [the agency] aimed to implement a
Web-based, self-service system for its customers.”
The SOA introduction (SOA implementation) initiative involved a web-based, scalable and feature-rich business-togovernment suite for the agency’s customers and partners. It has been large: a great deal of funds, effort and time
has been put towards the success of its implementation [D-12]. At the beginning of the project, during the design
phase, the agency contracted external vendors to plan and develop its requirements. It was developed by a competent
internal work force with the help of external expertise when needed. In a nutshell, the SOA suite was developed
using a combination of in-house development and products from vendors such as Oracle, IBM and Microsoft [D-
12]. Further, to ensure successful SOA implementation, employees were trained, the organisational restructuring
was completed and the assessment and training section was established. This initiative was managed and supervised
by experts to facilitate the transition to the new system. A selection of clients representing all trade sectors were
involved in development of the SOA, primarily in the design of the system. Multiple focus group meetings were
held with clients to announce the new procedures. The project was successful and led to paperless operations, cost
reductions and improved customer service.
The previous paragraphs have provided an overview of the agency’s SOA introduction and the following paragraph
examines SOA introduction-related action-formation generative mechanisms.
Based on the developed theoretical model (Figure 2), there are six action-formation generative mechanisms that
influenced the introduction of SOA in this case study. First, SOA introduction was driven by both business and IT
perspectives of SOA. It was undertaken to redesign business processes and improve services delivery. Second, SOA
implementation was associated with perceived benefits at the strategy, process and IT levels, and most of these
benefits were reported as achieved in the SOA implementation documentation and by some participants. Third, SOA
introduction encompassed the whole organisation and lasted for two years. Fourth, SOA introduction was governed
on multiple levels. It was governed by the eServices delivery excellence model and organisational wider governance
practices (COBIT). Fifth, SOA was introduced based on a long-term roadmap that employed a defined SOA
reference architecture. Services were identified using a top-down approach and were classified into business and
technical services. Also, a service repository (IBM System Architect) was employed to track these services in
relation to other architectural elements. Finally, SOA introduction was a large project that involved key business and
IT stakeholders during design and implementation. It was supported and driven by the organisation’s top
management as part of a transformation initiative to improve services delivery. External and internal (business and
IT) highly skilled team members were involved in SOA implementation. Table 5 summarises the SOA introductionrelated generative mechanism in this case study.
Table 5. Summary of generative mechanisms relevant to SOA introduction

Generative
mechanisms
Description as reported in the case
View of SOA
SOA perceived
benefits
 SOA implementation was associated with perceived benefits at the strategy, process
and IT levels
 Most of these benefits were reported as achieved in SOA implementation
documentation and by some participants
SOA scope
SOA governance
SOA design  SOA was implemented using a long term roadmap
 It employed a defined SOA reference architecture. Services were identified using a
top-down approach

 SOA introduction was driven on both business and IT levels
 It was undertaken to redesign business processes and improve services delivery.
 SOA implementation was organisation-wide
 It lasted for two years
 SOA implementation was governed on multiple levels
 It was governed by the eServices delivery excellence model
 It was governed by the adopted IBM SOA reference architecture
 It was governed by internal governance practices (COBIT) and EA governance

ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
13

 Services were classified into business and technical services
 A service repository (IBM System Architect) was adopted to track these services in
relation to other architectural elements
Business and IT
collaboration
 Business and IT stakeholders were involved during design and implementation
 It was supported and driven by the top management of the organisation as part of a
transformation initiative to improve services delivery
 Diverse skilled teams were involved in SOA implementation
 External vendors and consultants were involved in SOA’s implementation

4.3 Architectural elaboration T4: reproduction or transformation
Using the last phase of the theoretical model, this section discusses architectural elaboration (EA evolution
outcomes). Possible outcomes are that the pre-existing architectural settings are either reproduced or transformed on
five levels: business architecture, Information Systems architecture, technology architecture, EA governance and EA
methods and tools. In this case study, the five levels were transformed as EA evolved to accommodate SOA on all
the five levels. Following the data analysis, the observed architectural elaboration (evolution outcomes) are
summarised in Table 6.
Table 6. The observed architectural evolution

Architectural
transformation level
Description
Business architecture
(transformed)
 Process layer became “business layer” to incorporate business services besides
the other elements of the business architecture
 Re-design of the organisation in terms of domains, and each domain has its
provided services
 Design of business architecture in terms of services
 New SOA-related elements were added to business architecture, such as
business services, their descriptions, supported channels, client groups, service
scenarios and owners
 Business services were mapped to other business architecture elements
 Business services viewpoints were added
IS architecture
(transformed)
 Applications were designed and documented in terms of technical services that
support business processes and services
 A technical service was represented, which had a schema, used a service
operation, and had a service realisation diagram
 Technical services were aligned and used by business processes and services
on the business architecture
 Granularity of technical services was considered at the design level to ensure
proper reuse
 Services were used to integrate internal systems and external systems such as
external payment services
 Use of SOAP protocols, WSDL for services description and XSD for services
schema definitions
 Technical services were mapped to business processes and supporting
infrastructure
Technology architecture
(transformed)
 SOA infrastructure such as BPEL engine, web services manager and ESB
documented using technology environment, instance, interface, interface
messaging and message structure
 Use of services-related communication protocols such as SOAP and services
security protocols such as WS-security to document used SOA protocols
 Service repository (integrated into IBM System Architect) that hosts the meta
data of services and related information
 Services/infrastructure mapping to show the infrastructure that supports

ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
14

services
 Services SLAs were configured and monitored at the application and the
infrastructure layer to make sure that the SLAs were met
EA governance
(transformed)
 EA covered governance aspects regarding demands management and
alignment with strategy and architectural standards
 SOA (and its projects) had its own governance frameworks that were aligned
with the overarching EA governance
 EA governed service documentation, service identification and service delivery
 Services were monitored using the orchestration engine
 SOA demands were also governed by EA, similar to any other demands,
against the architectural standards and strategy
 Every service was governed by technical and business SLAs
 Every business service had an owner
EA methods and tools
(transformed)
 EA was integrated with demands/projects, which include SOA projects New
SOA-related elements and new relationships were created in the used EA tools.
 New views were created in used EA tools to support services and associated
elements
 Service identification methods and services were identified using EA products
(repository)

5 Discussion
Using Archer’s (1995) morphogenetic theory as a lens, the architectural elaboration (the evolution outcomes in
terms of integrating SOA into EA) can be explained by looking retrospectively at both the architectural interaction
(SOA’s introduction) and the conditioning (pre-existing EA) phases. The architectural conditioning phase shapes the
architectural interaction (T2-T3) that generates the evolution outcomes. Figure 5 presents a high-level overview of
the case study findings using the theoretical model of this study. Next we will discuss the most important findings
(propositions) from the case study related to EA settings and its conditioning role, the intervention (here: the
implemented SOA and its characteristics), and the resulted EA evolution outcomes.
T1: Architectural Conditioning T2: Architectural Interaction :T3 Architectural Elaboration: T4
 Business and IT view of SOA
 Strategic, Process and IT benefits
 Organisation-wide scope
 Governance framework and a reference
architecture used
 Top-down approach, services are
classified, a long term roadmap
 High level of collaboration, business and
IT and very skilled team
 Mature EA
 Business Architecture
(Transformed)
 IS Architecture
(Transformed)
 Technology Architecture
(Transformed)
 EA methods and tools
(Transformed)
 EA Governance (Transformed)
 Strategic, operational, IT and
governance oriented EA
 In-house developed using
TOGAF and Zachamn,
Organisation-wide, Welldefined EAF, Well-established
and managed EA repository
Time
Prior to SOA introduction in 2008 SOA introduction between 2008-2010 EA changes after SOA introduction (examined in
2012)
Figure 5. The case study’s morphogenetic cycle of SOA’s integration into EA
Proposition 1: Extensive SOA integration into EA can benefit from pre-existing EA as enabling context when the EA
framework is advanced, EA objectives are comprehensive and EA practices are mature.
The context is described in terms of three conditional generative mechanisms that conditionally influence EA
evolution. In our case study, the three conditional mechanisms created an enabling context for EA evolution and the
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
15
outcome was a transformation of the EA due to the introduction of SOA. The first conditional generative
mechanism is EA framework. In this case, EA framework had an organisation-wide scope, a well-defined metamodel, a well-defined methodology and well-maintained deliverables. Such practices create an enabling context for
EA related activities in general (Bui, 2012) and EA evolution specifically. The second generative mechanism is EA
objectives. In this case, EA objectives were strategic, business and IT oriented. The agency realised EA benefits and
appreciated them. Thus, it seems that the realisation and maintenance of comprehensive objectives of EA improves
the likelihood of EA engagement with organisational activities that may require EA’s evolution. This finding is
supported by another argument by Haki, Legner, and Ahlemann (2012) who state that organisations follow different
EA development approaches based on the architecture objectives, which, in turn, may affect EA activities in later
stages. The third conditional generative mechanism is the EA maturity. In this case, the agency’s EA practices were
mature on several dimensions. For example, well-established architectural governance facilitated the EA team’s
engagement with the SOA introduction and, thus, SOA introduction’s compliance with EA. The well-established
and populated EA repository contributed to the SOA introduction process through the use of existing EA models and
information to design and implement SOA. The three conditional mechanisms collectively facilitated SOA
integration into EA.
Archer (1995) calls this type of context, marked by the three EA-related conditional mechanisms in this study, an
“enabling context” whereas it could be a “constraining” one in other cases when opposite conditions are present. In
other words, it is expected that EA evolution might be impeded when EA framework is not well-developed, EA
objectives are not clearly defined and when EA is not mature enough. While not explicitly studied in terms of
architectural elaboration, this is aligned with related insights as discussed below.
Enterprise architecture needs to be mature in order to have an impact on the enterprise (Lagerstrom, Sommestad,
Buschle, & Ekstedt, 2011). Kotusev, Singh, and Storey (2015) reported that ambiguous EA vocabulary, unclear
objectives of EA, and incompetence to promote EA practices are key factors of a poor adoption of EA and isolation
of EA practices. Deficient governance practices, inadequate support for the EA development from key stakeholders,
as well as insufficient resources and skills hinder the realization of the strategic alignment potential of EA (Löhe &
Legner, 2014). Companies that did not consider mechanisms to update their EA artifacts, such as through afterproject changes, end up with outdated and low quality EA repositories leading to a low utilization of EA artifacts
(Löhe & Legner, 2014).
Further, an example that echoes Archer’s (1995) argument of the constraining impact of the architectural
conditioning phase is provided by Carvalho and Sousa (2014). They address how incomplete and immature EA
(architectural conditioning as reported in this study) influences new IS implementation. They found that three
domains of EA: data, application, and technology architectures were poorly documented resulting in EA’s lack of
support for the new IS implementation. The efforts needed to implement the new tool were increased to improve and
generate EA artefacts in order to compensate for the immature state of EA prior to the new IS implementation. The
EA in this case was needed to achieve organizational agility through the implementation of the new IS tool, but due
to the EA’s immature state, EA needed to be re-developed and improved during the change situation (Carvalho &
Sousa, 2014).
Nevertheless, treating these conditional generative mechanisms as the primary factors in the EA evolution process is
not suitable according to the findings and Archer’s (1995) argument that states neither the structure (EA) nor the
action (SOA introduction) alone determines the outcomes. Thus, the next paragraph discusses the SOA introduction
phase and its impact on SOA integration into EA.
Proposition 2: The extent of SOA integration into EA results from the actions related to the introduction of SOA that
are influenced by the agents’ view of SOA, perceived SOA benefits, SOA scope, SOA design, governance and
business/IT collaboration.
Archer (1995) recognises the influence the agent’s orientation, interests and resources can have on the action (T2-
T3). In this study, SOA introduction is influenced by the six identified action-formation generative mechanisms
(view of SOA, perceived SOA benefits, SOA scope, SOA design, governance and business/IT collaboration)
summarised in Figure 5. The findings support the theoretical arguments of Archer (1995), Hedström and Ylikoski
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
16
(2010) and Cuellar (2010) about the influence that a combination of interests, orientations and resources can have on
the action (SOA introduction). The action-formation generative mechanisms collectively (acknowledging that one
generative mechanism may counterbalance others) shape the way SOA is introduced. This conclusion is supported
by Mutch’s (2010) argument that IS implementations could be configured in different ways based on different
factors to produce very different outcomes for organisational arrangements. Thus, when introducing SOA,
organisations need to consider the implications of the combination of these generative mechanisms on (1) SOA
implementation and (2) the organisation and its enterprise architecture in general.
Proposition 3: Extensive SOA integration into EA involves transformation at all five possible levels of EA evolution
outcomes.
The evolution outcomes (architectural elaboration) are classified into five levels. In other words, EA could be
transformed (integrated with SOA) or not (reproduced) on one or more of these levels. In this case study, EA was
transformed on all five architectural levels due to the introduction of SOA (in the form described above) in an
enabling context (architectural conditioning phase). In other words, both the enabling context and the introduction of
SOA in the way described above (the actualisation of the generative mechanisms as shown in Figure 5) resulted in
SOA’s integration into EA on all the identified five architectural levels. Such separation of the five levels facilitates
the examination of whether an emerging capability requires EA evolution and, if so, what levels of EA need to be
evolved.
Proposition 4: EA evolution in organisations can be described and explained by three analytical phases:
architectural conditioning, architectural interaction and architectural elaboration as part of a morphogenetic cycle.
Building on the morphogenetic theory (Archer 1995), the introduction of SOA (T2-T3) in a given architectural
conditioning (T1) results in an architectural elaboration (evolution outcomes) (T4) based on the interplay between
the action-formation generative mechanisms and the conditional ones. In this case, looking at one EA alone or SOA
introduction in isolation of pre-existing EA would not provide a comprehensive picture of EA evolution process and
outcomes. While a pre-existing EA can be an enabling context if EA framework is advanced, EA objectives are
comprehensive and EA practices are mature, this will not guarantee SOA integration in EA. Specific attention to
SOA introduction is required. On the other side, while specific attention to SOA introduction is important, it may be
difficult to achieve SOA integration in EA when the pre-existing EA does not provide an enabling context.
6 Conclusion
Due to the dynamic nature of environmental and internal factors, EA development in an organisation is not a one-off
activity that leads to static descriptions of the business and IT artefacts of that organisation. Rather, it is a process
that parallels the evolution of the organisation and its strategy (Shah & Golder, 2011). EA needs to change over time
to provide value for its stakeholders. However, the phenomenon of EA evolution is still poorly understood, in
particular at the level of architectural descriptions. Using Archer’s morphogenetic theory as a foundation, this paper
conceptualises three analytical phases of EA evolution in organisations, namely architectural conditioning,
architectural interaction and architectural elaboration. Moreover, motivated by the lack of empirical studies that
investigate EA evolution, a case study was conducted with a government agency to contextualise the developed
theoretical model and to enrich the understanding of how an organisation evolves its EA.
The case study findings provide first theoretically and empirically grounded insights into EA evolution, in particular
in relation to SOA. The potential changes in EA are the outcomes of a process that is conditioned by the pre-existing
EA, and influenced by the way in which SOA is introduced. The findings provided deeper insights into how the
three identified conditional mechanisms related to the pre-existing EA: EA framework, EA objectives, and EA
maturity, which enable or constrain EA evolution based on their actualisations in a given context. The study also
showed that the six identified action-formation mechanisms related to the SOA introduction: view of SOA,
perceived SOA benefits, SOA scope, SOA governance, SOA design and business and IT collaboration, influence
SOA introduction. These mechanisms reflect the actors’ orientations, beliefs, interests and resources (Archer, 1995;
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
17
Hedström & Ylikoski, 2010),. Thus, it is crucial to explicitly pinpoint these mechanisms prior to introducing SOA
not only to ensure successful SOA implementation but also to understand their influence on SOA’s integration into
EA. The process eventually results in an architectural elaboration (evolution outcomes) based on the interplay
between the described conditional generative mechanisms and the action-formation generative mechanisms. The
interplay and the actualisations of these generative mechanisms are expected to be dissimilar in different contexts,
which plausibly explain varied EA evolution outcomes. That is, EA is evolved (transformed) or reproduced on one
or many of the five identified levels (business architecture, IS architecture, technology architecture, EA governance
and EA methods and tools). More generally, the findings show the significance of the impact of the new emerging
business and IT trends on EA frameworks, methodologies, governance and tools. Thus, it is essential for
organisations to explicitly examine whether these emerging trends require EA evolution and, if so, what levels of
EA need to be evolved.
The paper derives its significance and relevance from interrelating important contemporary phenomena based on a
sound theoretical foundation. By providing new empirical and theoretical insights into EA evolution describing
relevant generative mechanisms affecting EA evolution and outcomes in light of SOA, it builds a foundation to
further examine the impact of other IT trends such as mobile or cloud-based solutions on EA evolution. At a
practical level, the paper delivers a model that can be used to guide professionals to manage EA and continually
evolve it in response to emerging business and IT capabilities.
There are a number of limitations in this study. First, the paper only reports on the findings of a single case study
where more cases could provide more insights. Yet, Sayer (2000) argues that one or two cases is enough when using
intensive (qualitative) research methods. This paper intensively examined EA evolution using a case study method
(involved many interviews, several documents, and online materials). The case study was enlightened by preceding
explorative interviews and a comprehensive literature review.
The other limitation is the need for longitudinal data to examine EA evolution prior to the intervention, during the
intervention and post the intervention. In this case, we tried to collect as much data as possible to cover the three
phases but a longitudinal case study is an option for future research.
The results of this study open opportunities for further research in the field of EA evolution. For example, to further
address the impact of poor initial conditions, e.g. incomplete EA framework, fuzzy EA objectives, and low EA
maturity level, on EA evolution, future work is needed to comprehensively understand the influence of these initial
conditions as well as how to overcome them. In addition, further efforts are required to examine EA evolution in
response to other IS interventions such as cloud computing or mobility.
7 Acknowledgments
This research is supported by King Saud University. Author would express their appreciation to the Deanship of
Scientific Research at King Saud University for the provision of funding.
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
18
8 References
Alwadain, A., Fielt, E., Korthaus, A., & Rosemann, M. (2013a). A critical realist perspective of enterprise
architecture evolution: Preliminary findings. In Proceedings of the 24nd Australasian Conference on
Information Systems (ACIS 2013).
Alwadain, A., Fielt, E., Korthaus, A., & Rosemann, M. (2013b). Service-oriented architecture integration within
enterprise architecture: A-priori model. In Proceedings of the 24nd Australasian Conference on
Information Systems (ACIS 2013).
Alwadain, A., Fielt, E., Korthaus, A., & Rosemann, M. (2014). A critical realist perspective of enterprise
architecture evolution: Conditioning and outcomes. Australasian Journal of Information Systems, 18(3).
Archer, M. S. (1995). Realist social theory: The morphogenetic approach. Cambridge, UK: Cambridge university
press.
Banerjee, J., & Aziz, S. (2007). Soa: The missing link between enterprise architecture and solution architecture.
Retrieved April 02, 2010 from http://www.infosys.com/offerings/it-services/soa-services/whitepapers/documents/soa-link-ea-sola.pdf
Benbasat, I., Goldstein, D. K., & Mead, M. (1987). The case research strategy in studies of information systems.
MIS Quarterly, 11(3), 369-386.
Bhaskar, R. (1998). The possibility of naturalism: A philosophical critique of the contemporary human sciences.
London: Taylor & Francis Group.
Buckl, S., Ernst, A. M., Matthes, F., & Schweda, C. M. (2009). Visual roadmaps for managed enterprise architecture
evolution. In Proceedings of the 10th ACIS International Conference onSoftware Engineering, Artificial
Intelligences, Networking and Parallel/Distributed Computing (pp. 352-357).
Bui, Q. N. (2012). Making connections: A typological theory on enterprise architecture features and organizational
outcomes. In Proceedings of the 18th Americas Conference on Information Systems.
Carvalho, J. A., & Sousa, R. D. (2014). Enterprise architecture as enabler of organizational agility: A municipality
case study. In Proceedings of the 20th Americas Conference on Information Systems (AMCIS 2014).
Correia, A., & Silva, M. (2007). Modeling services in information systems architectures. In P. F. Cunha & P. G.
Maropoulos (Eds.), Digital enterprise technology perspectives and future challenges (pp. 157-164). New
York, USA: Springer.
Cuellar, M. J. (2010). Using realist social theory to explain project outcomes. International Journal of Information
Technology Project Management 1(4), 38-52.
Danermark, B., Ekström, M., Jakobsen, L., & Karlsson, J. (2002). Explaining society : Critical realism in the social
sciences. London ; New York: Routledge.
Dico, A. S. (2012). Towards whole-of-government EA with TOGAF and SOA. In P. Saha (Ed.), Enterprise
Architecture for Connected E-Government: Practices and Innovations (pp. 177-204): IGI Global.
Easton, G. (2010). Critical realism in case study research. Industrial Marketing Management, 39(1), 118-128.
Easton, G., & Harrison, D. (1998). Temporally embedded case comparison in industrial network research. In
Proceedings of the 14th Industrial Marketing and Purchasing Group Conference (pp. 241-264).
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
19
Eisenhardt, K. M. (1989). Building theories from case study research. The Academy of Management Review, 14(4),
532-550.
Farwick, M., Schweda, C., Breu, R., Voges, K., & Hanschke, I. (2012). On enterprise architecture change events. In
S. Aier, M. Ekstedt, F. Matthes, E. Proper & J. Sanz (Eds.), Trends in enterprise architecture research and
practice-driven research on enterprise transformation (pp. 129-145): Springer Berlin Heidelberg.
Fereday, J., & Muir-Cochrane, E. (2006). Demonstrating rigor using thematic analysis: A hybrid approach of
inductive and deductive coding and theme development. International Journal of Qualitative Methods,
5(1), 1-11.
Gartner. (2012a). Ea practitioners have significant influence on $1.1 trillion in enterprise it spend. Retrieved Jan 5,
2013 from http://www.gartner.com/id=2286216
Gartner. (2012b). Itscore overview for enterprise architecture. Retrieved November 7, 2012 from
http://www.gartner.com/id=1958816
Haki, M. K., Legner, C., & Ahlemann, F. (2012). Beyond ea frameworks: Towards an understanding of the adoption
of enterprise architecture management. In Proceedings of the European Conference on Information
Systems.
Hedström, P., & Ylikoski, P. (2010). Causal mechanisms in the social sciences. Annual Review of Sociology, 36, 49-
67.
Infosys. (2009). Enterprise architecture expands its role in strategic business transformation, infosys enterprise
architecture survey 2008/2009. Retrieved February 05, 2010 from http://www.infosys.com/offerings/ITservices/architecture-services/ea-survey/Pages/index.aspx
Jung, C. (2009). Actionable enterprise architecture. In Proceedings of the 10th ACIS International Conference on
Software Engineering, Artificial Intelligences, Networking and Parallel/Distributed Computing (pp. 294-
299).
Khoshnevis, S., Aliee, F. S., & Jamshidi, P. (2009). Model driven approach to service oriented enterprise
architecture. In Proceedings of the 2009 IEEE Services Computing Conference (pp. 279-286).
Kistasamy, C., Van der Merwe, A., & De La Harpe, A. (2012). The role of service oriented architecture as an
enabler for enterprise architecture. In Proceedings of the 18th Americas Conference on Information
Systems
Knippel, R., & Skytte, B. (2007). Service oriented enterprise architecture: An opportunity of evolution. Journal of
Enterprise Architecture, 3(3), 17-25.
Kotusev, S., Singh, M., & Storey, I. (2015). Investigating the usage of enterprise architecture artifacts. In
Proceedings of the 23rd European Conference on Information Systems (ECIS 2015).
Lagerstrom, R., Sommestad, T., Buschle, M., & Ekstedt, M. (2011). Enterprise architecture management’s impact on
information technology success. In Proceedings of the 44th Hawaii International Conference on System
Sciences.
Land, M., Proper, E., Waage, M., Cloo, J., & Steghuis, C. (2009). Enterprise architecture: Creating value by
informed governance. Berlin Heidelberg: Springer-Verlag
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
20
Lankhorst, M. (2005). Enterprise architecture at work: Modelling, communication and analysis. Berlin Heidelberg:
Springer-Verlag.
Löhe, J., & Legner, C. (2014). Overcoming implementation challenges in enterprise architecture management: A
design theory for architecture-driven it management (adrima). Information Systems and e-Business
Management, 101-137.
Lynham, S. A. (2000). Theory building in the human resource development profession. Human Resource
Development Quarterly, 11(2), 159-178.
MacLennan, E., & Belle, J. (2014). Factors affecting the organizational adoption of service-oriented architecture
(soa). [Article]. Information Systems & e-Business Management, 12(1), 71-100.
Martin, R., Purao, S., & Robertson, E. (2009). Evolving enterprise architecture. In Proceedings of the Next
Generation Enterprise Architecture Workshop. Bled, Slovenia.
McKendrick, J. (2010). Top 10 enterprise architecture trends, identified by forrester. Retrieved March 02, 2011,
from http://www.zdnet.com/blog/service-oriented/top-10-enterprise-architecture-trends-identified-byforrester/5949
Mens, T., Magee, J., & Rumpe, B. (2010). Evolving software architecture descriptions of critical systems.
Computer, 43(5), 42–48.
Mingers, J. (2004). Real-izing information systems: Critical realism as an underpinning philosophy for information
systems. Information and Organization, 14(2), 87-103.
Mutch, A. (2010). Technology, organization, and structure—a morphogenetic approach. Organization Science,
21(2), 507-520.
NASCIO. (2003). Nascio enterprise architecture maturity model. Retrieved March 20, 2011 from
http://www.nascio.org/publications/documents/NASCIO-EAMM.pdf
Njihia, J. M. (2008). Ict for development reconsidered: A critical realist approach to the strategic context in kenya’s
transition to e-governance. Doctoral Dissertation, University of Warwick.
Oracle. (n.d.). Dubai customs’ web-based, service-oriented architecture streamlines trade transactions. Retrieved
Dec 10, 2012 from http://www.oracle.com/us/corporate/customers/customersearch/dubai-customs-1-soasnapshot-1500953.html
Postina, M., Trefke, J., & Steffens, U. (2010). An ea-approach to develop soa viewpoints. In Proceedings of the 14th
IEEE International Enterprise Distributed Object Computing Conference (pp. 37-46).
Roth, S., Hauder, M., Farwick, M., Breu, R., & Matthes, F. (2013). Enterprise architecture documentation: Current
practices and future directions. In Proceedings of the 11th International Conference on
Wirtschaftsinformatik.
Saat, J., Aier, S., & Gleichauf, B. (2009). Assessing the complexity of dynamics in enterprise architecture planninglessons from chaos theory. In Proceedings of the 15th Americas Conference on Information Systems.
Sampaio, A. (2010). An approach for creating and managing enterprise blueprints. Thesis. Retrieved from
https://dspace.ist.utl.pt/bitstream/2295/767600/1/
Sanders, D. T., Hamilton, J. A., & MacDonald, R. A. (2008). Supporting a service-oriented architecture. In
Proceedings of the the 2008 Spring simulation multiconference (pp. 325-334 ).
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
21
Sayer, A. (1992). Method in social science : A realist approach (2nd ed.). London: Routledge.
Sayer, A. (2000). Realism and social science. London: Sage Publications Limited.
Schekkerman, J. (2004). How to survive in the jungle of enterprise architecture frameworks: Creating or choosing
an enterprise architecture framework. British Columbia: Canada: Trafford Publishing.
Schekkerman, J. (2005). The economic benefits of enterprise architecture. Victoria, BC, Canada: Trafford
Publishing.
Schmidt, C., & Buxmann, P. (2011). Outcomes and success factors of enterprise it architecture management:
Empirical insight from the international financial services industry. European Journal of Information
Systems, 20(2), 168-185.
Shah, H., & Golder, P. (2011). Adappt: Enterprise architecture thinking for information systems development.
International Journal of Computer Science Issues, 8(1).
Shankararaman, V., & Kazmi, P. (2011, 5-7 Sept. 2011). Unifying ea, bpm and soa through a synergestic
framework. In Proceedings of the IEEE 13th Conference on Commerce and Enterprise Computing (pp.
286-293).
Sharma, S. (2013 ). Using enterprise architecture with soa helps in achieving business-it alignment. Retrieved June
15th, 2013 from http://ovum.com/2013/02/19/using-enterprise-architecture-with-soa-helps-in-achievingbusiness-it-alignment/

Short, J. (2013). Agenda overview for enterprise architecture, 2013. Retrieved April 10,
www.gartner.com/id=2292615
2013 from

Simon, D., Fischbach, K., & Schoder, D. (2013). An exploration of enterprise architecture research.
Communications of the Association for Information Systems, 32(1).
Sousa, P., Lima, J., Sampaio, A., & Pereira, C. (2009). An approach for creating and managing enterprise blueprints:
A case for it blueprints. In A. Albani, J. Barjis & J. L. G. Dietz (Eds.), Advances in enterprise engineering
iii (pp. 70-84). Berlin Heidelberg: Springer.
Sowa, J. F., & Zachman, J. A. (1992). Extending and formalizing the framework for information systems
architecture. IBM Systems Journal, 31(3), 590-616.
The Open Group. (2009). The open group architecture framework togaf. LJ Zaltbommel, The Netherlands: Van
Haren Publishing.
The Open Group. (2010). Definition of soa Retrieved January 20, 2010 from
http://www.opengroup.org/projects/soa/doc.tpl?gdid=10632
Trosper, R. L. (2005). Emergence unites ecology and society. Ecology and Society, 10(1), 1-14.
Varnus, J., & Panaich, N. (2009). Togaf 9 survey results. Retrieved January 29, 2010 from
https://www.opengroup.org/conference-live/uploads/40/20081/Mon_-_am_-_3_-_Varnus.pdf
Viering, G., Legner, C., & Ahlemann, F. (2009). The (lacking) business perspective on soa–critical themes in soa
research. In Proceedings of Wirtschaftsinformatik (pp. 45-54).
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
22
Volkoff, O., & Strong, D. M. (2013). Critical realism and affordances: Theorizing it-associated organizational
change processes. MIS Quarterly, 37(1), 819-834.
Winter, K., Buckl, S., Matthes, F., & Schweda, C. M. (2010). Investigating the state-of-the-art in enterprise
architecture management methods in literature and practice. In Proceedings of the Mediterranean
Conference on Information Systems.
Wynn, D., & Williams, C. K. (2012). Principles for conducting critical realist case study research in information
systems. Management Information Systems Quarterly, 36(3), 787-810.
Wynn Jr, D. E., & Williams, C. K. (2008). Critical realm-based explanatory case study research in information
systems. . Paper presented at The International Conference on Information Systems (ICIS).
Yin, R. (2003). Case study research: Design and methods: Thousand Oaks, California: Sage Publications, Inc.
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
23
Ayed Alwadain is an assistant professor at the computer science
department, King Saud University, Riyadh, Kingdom of Saudi Arabia. He
received his PhD degree in Information Systems from Queensland University
of Technology in Australia in 2014. His areas of are Enterprise Architecture,
Service-oriented Architecture and Business Process Management. Ayed has
published and presented his work at many international conferences and
some journals.
Erwin Fielt is a Senior Lecturer in the Information Systems School of the
Science and Engineering Faculty at the Queensland University of
Technology, Brisbane, Australia. In his research and teaching, he focuses on
the intersection between business and IT, where information systems have
to create value for individuals and organizations, in particular in relation to
strategy, business models, innovation, and entrepreneurship. Erwin Fielt has
published and presented his research at numerous international outlets,
such as the Journal of Strategic Information Systems, Electronic Markets,
Communications of the Association for Information Systems, and Info. He
has a PhD from the Delft University of Technology and a MSc from the
University of Twente, both in the Netherlands.
Axel Korthaus is a Senior Lecturer and the Discipline Leader of Information
Systems in the College of Business at Victoria University, Melbourne,
Australia. He received his PhD degree in Information Systems from the
University of Mannheim, Germany, and has gained comprehensive
experience as researcher, lecturer and project manager in areas such as
Service Management and Engineering, Business Process Management,
Component-Based Software Engineering, Crowdsourcing and related fields.
He has been involved in various international collaborations and projects, is
the (co-)author/editor of several books, special issues, journal articles and
conference papers, serves as a track and workshop chair, regular PC
member and speaker at IS- and SE-related conferences and has presented
his work in 11 countries over the last years.
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
24
Michael Rosemann is Professor and Head of the Information Systems School at
Queensland University of Technology. He has a comprehensive publication track record
comprising 7 books and more than 220 refereed papers. His areas of interest are
innovation systems, process design and research management. Michael is a frequent
keynote speaker and provides advice to organizations and their executives across
diverse industries.

The post Data & Knowledge Engineering appeared first on My Assignment Online.

Plagiarism Free Assignment Help

Expert Help With This Assignment — On Your Terms

✓ Native UK, USA & Australia writers ✓ Deadline from 3 hours ✓ 100% Plagiarism-Free — Turnitin included ✓ Unlimited free revisions ✓ Free to submit — compare quotes

admin

Academic writer and editor at BestWriters.

Get your paper off your plate tonight

Post your instructions for free. Prices start at $5 per page.

Place an order