Enterprise Architecture

An Approach to Promote Traceability and Synchronization of Computational Models

José Rogério Poggio Moreira and Rita Suzana Pitangueira Maciel

Department of Computing Science, Federal university of Bahia, Av. Adhemar de Barros, S/N, Ondina, Salvador, Brazil

Keywords: Enterprise Architecture, Model Synchronization, Traceability of Models, Alignment between Business and

IT, Alignment between Strategy and Business Processes, Strategic Alignment, Business Model.

Abstract: In the context of Enterprise Architecture (EA) modeling, the lack of alignment between the computational

models from the strategic level to the operational level of Information Technology (IT), focusing on

information systems, constitutes an organizational problem. The root cause of this problem is the low

capacity of traceability and lack of synchronization between the computational models present in the EA.

Among the negative impacts related to this problem are, for example, the obsolescence of models, the

difficulty of carrying out impact analyzes and making decisions, in scenarios of organizational changes.

This paper aims to propose an approach to enable the traceability and synchronization between the

computational models of enterprise architecture. It is hoped to provide a greater alignment, understanding

and adaptation of the organizational environment, from the strategic level to the operational level of IT.

1 INTRODUCTION

An enterprise architecture is the organizing logic of

strategy, business processes and IT infrastructure,

including information systems, which reflects the

integration and standardization requirements of an

institution's strategic-operational model, such as the

Zachman and TOGAF models that have driven and

consolidated, respectively, the works in this area

(Zachman, 1987; Ross, 2008; Josey et al., 2011). An

enterprise architecture is formed by levels that are

related: (i) strategic; (ii) business processes

(tactical); (iii) services and; (iv) IT infrastructure

(operational) (Ross, 2008). In the modeling of

enterprise architecture, the relationship between

strategy, business processes and information systems

can be represented by the alignment of the

computational models elaborated within the

organization (Engelsman et al., 2011). The models

represent instruments to reach and graphically

represent the different levels of abstraction of

enterprise architecture and to contribute to the

achievement and maintenance of the strategic

alignment between business and IT, in addition to

their unions providing an integrated visualization of

an organization's EA (Ross, 2008). These models,

then, represent the achievement of: (i) corporate and

IT strategic planning; (ii) the business tactic to

achieve the strategic objectives defined in the

strategic planning and; (iii) the IT operation,

focusing on information systems. However, the

isolated construction of models such as the

construction of a requirements model, without

having a process model and a strategic model

previously defined, can result in the elaboration of

semantically fragile and non-aligned models with

the business. If there is no established strategy and

the processes are not defined and are unstructured,

without maturity and inefficient, there is no reason

for an institution to start developing information

systems.

In this scenario, to allow a better understanding

of this research, this work defines the term

alignment as the ability to trace and synchronize the

strategic, tactical and operational structures present

in the computational models of an organization.

The lack of traceability or its delivery in an

inadequate way constitutes a problem and makes it

difficult to see and understand how a set of models

and their structures are related, which contributes

negatively to the analysis of impacts in scenarios of

organizational changes (Bleistein, 2006).

The absence of synchronization between

different models (such as business process model

and system requirements model) is another problem

and results in the obsolescence of these models,

which are now outdated and inconsistent, since the

460

Moreira, J. and Maciel, R.

Enterprise Architecture - An Approach to Promote Traceability and Synchronization of Computational Models.

DOI: 10.5220/0006374304600465

In Proceedings of the 19th International Conference on Enterprise Information Systems (ICEIS 2017) - Volume 3, pages 460-465

ISBN: 978-989-758-249-3

change in some element of the model, does not

guarantee the updating of the other associated

structures. This makes it difficult to maintain

business strategy or at least generates a significant

impact analysis effort in a scenario of change. In

addition, there is still an equally significant

probability of occurrence of failures, since the

activity of checking the points that can be affected

by a possible change, is manual (Salgado et al.,

2013).

In this way, it is proposed an approach that has

the objective of enabling the synchronization and

traceability between the computational models of an

EA. The proposed approach consists of: (i) a meta-

model set comprising the strategic, tactical and

operational levels of the EA; (ii) an independent

traceability model that supports configuration and

change management through the use of COBIT

(ITGI, 2012) and ITIL (OGC, 2007) best practices

and; (iii) the application of Model-Driven

Development (MDD), through the construction of a

model transformation process, as a way to

synchronize elements between different models.

The remainder of this paper is organized as

follows: Section 2 presents the related works.

Section 3 describes the meta-model set. Section 4

presents de traceability model. Section 5 describes

model transformation process. Finally section 6

presents the conclusion of this work.

2 RELATED WORKS

This work sought to gather relevant contributions,

focused on the theme of strategic alignment between

business and IT, focusing on the alignment between

computational models of enterprise architecture. For

this, a bibliographic survey was performed using

systematic review techniques. Were considered the

1,500 works that were most relevant to the theme

and that were returned by the search. Posteriorly, 96

works were identified from the association of their

titles, with the respective theme of the research and

the application of filters related to authors relevant

to the theme. Of these, only 44 articles were

selected, based on the analysis of inclusion and

exclusion criteria defined for the research. The study

covered the work that was identified during the

research and which was published in the last 13

years (2004 to 2016) and was in English and

Portuguese. The following databases were searched:

(i) Compendex; (ii) Web of Science; (iii) IEEE

Xplorer; (iv) Springer e; (v) Elsevier, in addition to

the data sources: EJIS, SCIENCEDIRECT,

WORLDSCIENTIFIC and AISEL. Also included in

the bibliographic collection, from this work, sources

of classic information (Zachman, 1987) and

reference (Josey et al, 2011). From the analysis of

the works it was possible to answer the following

research questions:

 Q1. Are there techniques that align IT and

business through traceability and synchronization

between computing models at all levels of an

enterprise architecture?

A1. There was no identification of studies that

performed alignment (traceability and

synchronization) between computational models

at all levels of EA. However, research was

carried out that deals in a theoretical and partial

way with the alignment between computational

models (Babar et al, 2011) (Giannoulis et al,

2010; Fritscher et al., 2011; Jonkers et al., 2004;

Engelsman et al., 2011).

 Q2. Is there a graphical language pattern to

represent the elements present in the models of

the different levels of enterprise architecture?

A2. A graphical language pattern has not been

identified for the modeling techniques that are

used at the levels of an EA, since each level of

the EA uses a different modeling technique and

even within an EA level, such as the tactical,

represented by business processes, exist different

languages that can be used to modeling business

process (BPMN, EPC, UML) (De Souza et al.,

2014; Kurz, 2016; OMG, 2015).

 Q3. Are there standard concepts that are used to

define the components used in the models of

enterprise architecture?

 A3. It was possible to identify a conceptual

standard to define the components used in the

models of enterprise architecture. At the process

level, for example, there are process and activity

concepts (De Sousa et al, 2014). At the strategic

and operational levels of IT, it was also possible

to identify a conceptual standard (Babar et al.,

2011; Engelsman et al., 2011).

The analysis performed in the correlated work

identified, showed that information technology is

relevant to organizational performance and

contributes directly to business operations. However,

the lack of alignment between IT and business still

prevails. For the analysis of the related works, tree

technical criteria were defined. Each criterion

represents the alignment between computational

models at a certain level of the business architecture.

The criterions defined were:

 Criterion I - Alignment between strategic

models: that reflects the alignment between

Enterprise Architecture - An Approach to Promote Traceability and Synchronization of Computational Models

461

strategic corporate and IT maps;

 Criterion II - Alignment between strategic and

tactical models: that reflects the alignment

between strategic maps and business process

models;

 Criterion III - Alignment between tactical and

operational IT models: this reflects the

alignment between business process models and

IT requirements models, necessary for the

development of information systems.

In order to improve the analysis of the works, tree

attributes were defined for each criterion. Each

attribute represents a characteristic that can be

treated by an identified criterion. The attributes

identified were:

 Theoretical: it should be used when the work

theoretically addresses subjects related to the

criterion;

 Traceability: should be used when the work

deals with traceability between models;

 Synchronization: should be used when the work

deals with synchronization between models.

Based on the identified criterion and attributes, the

analysis of the studies found that the vast majority of

research focuses on the theme related to criterion III.

However, the study revealed that even in the theme

of concentration (criterion III), the surveys do not

cover the alignment in its completeness, because this

does not contemplate the attribute of

synchronization between models. The topics related

to criterion I, as well as criterion II, have been very

incipient, containing proposals that discuss their

concepts and focus on the extraction of knowledge

from strategic and tactical models (strategic maps

and business process models) to identify

requirements of systems.

The evaluation of the analyzed works indicated

that the contributions of (Bleistein, 2006) and

(Engelsman et al., 2011) that originated the

languages BSC-P and ARMOR can be adapted to

allow the closure of the gaps found in the themes

associated to criteria I and II, since these works deal

theorically traceability between models that are

associated with criterion III. The B-SCP, also stands

out as a contribution, due to the validation and

verification of business requirements models, in

terms of business process and strategy, which may

allow an adaptation to the reality of the I criteria and

II. (Babar et al, 2011), together with (Giannoulis et

al., 2010) also presented works relevant to the topic

of this article, when discussing concepts related to

strategic maps, and can be used to support research

in the first and second criterion.

The architecture model proposed by (Fritscher et

al., 2011) also provides a relevant contribution and

can be evaluated with a view to adapting to a more

strategic business view. Considering the above, the

evaluation validated the need to study the strategic

alignment theme between business and IT, focusing

on the computational models of an enterprise

architecture, since it can be verified, the absence of

work for some aspects (criterion I and II) and lack of

completeness in others (criterion III).

Considering the above, the evaluation validated

the need to study the strategic alignment theme

between business and IT, focusing on the

computational models of an enterprise architecture,

since it can be seen, the absence of work for some

aspects (criterion I and II) and lack of completeness

in others (criterion III).

3 META-MODEL

The proposed set of meta-models aims to represent

the semantics of EA, proposed by Zachman and

TOGAF, and enable the tracking and

synchronization of the models present in their

different levels of abstraction, keeping the EA

always aligned and without obsolescence. For this

purpose, meta-models were constructed through the

Meta Object Facility (MOF) (OMG, 2016), with the

purpose of representing the main concepts,

characteristics and relationships present in the

strategic, tactical and operational levels of an EA.

Each meta-model is composed of: (i) concepts;

(ii) aspects and; (iii) relationships. The concepts

represent relevant elements that exist within the

levels of the EA and are represented by meta-

classes. The aspects represent the important

characteristics that are related to the identified

concepts and are represented by meta-attributes.

Already the relationships represent the semantic

associations that exist between the concepts present

in the meta-models, being represented by

associations. The meta-models were also designed to

be independent of the technology used to construct

the models (strategic, tactical and operational) that

are associated with the respective meta-models. The

concepts and aspects present in the set of meta-

models were adopted from the perspective of

techniques and good practices related to strategic

planning, process management and software

engineering, which, in turn, are associated with the

strategic, tactical and operational levels of EA.

A relevant issue in the conception of this set of

meta-models is the integrated and holistic approach

to modeling (strategic, tactical and IT operational),

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

462

viewing them as a single organizational model and

reinforcing the institution's strategic thinking

development (Objectives, strategies, strategic

initiatives, processes and system requirements).

Figure 1 presents the proposal of the meta-model

of strategic IT alignment, with a focus on

information systems. In it, it is possible to observe

that there are three types of modeling: (i) strategic;

(ii) tactic and; (iii) IT operational. These types of

modeling are represented, respectively, in the set of

meta-models by the packages: (i) MMEstrategic; (ii)

MMTactical and; (iii) MMOperational.

4 TRACEABILITY MODEL

This research works with the concept of horizontal

and vertical traceability (forward and backward),

and may also be associated with low and high

granularity. The traceability model proposed in this

work is based on several traceability models such as

those proposed by (Sayão et al., 2006; Oglio, 2010;

Ramesh, 2001).

Figure 4.1 shows the traceability model proposed

in this work. The model is divided into three parts.

The first part (in blue) represents the structure that

allows to trace the elements of the models, present in

the different levels of the EA, of vertical or

horizontal form, forwards or backwards and in the

granularities high and low.

The first part is organized so that a component

represents the abstraction of an element or a

traceability link between two elements. An element

may represent a concept or an aspect, and every

traceability link is formed by two elements, one of

origin and one of destination. In this way it is

possible to create traceability links between

concepts, between aspects and between concepts and

aspects, besides enabling, as previously mentioned,

traceability of low and high granularity, vertical and

horizontal.

The second part (in gray) is formed by the

configuration management structure. This structure

is necessary to identify, control and provide the

necessary information about an Organization

Configuration Item (CI) (OGC, 2007). A CI, in the

context of this work, is represented by the Class

“Model” that represents the strategic, tactical and

operational computational models of enterprise

architecture, along with its versions, types and

components.

The third part (blank) is formed by the change

management structure that allows changes to be

recorded, prioritized, evaluated and authorized by

the Change Advisory Board (CAB), tested and

implemented, allowing control over the changes.

Which occur on the models and their components,

reducing the risk of incidents and consequently of

damages to the organization.

Figure 1: Meta-model set.

Also according to the elaborated model, a change in

a component can generate positive and / or negative

impacts for the organization, besides having a

requester, a responsible and the affected parties. A

change may be requested for the purpose of

correcting any problem, for the purpose of

functional improvement or inclusion of a new

component. In addition, a change always has a

reason, which is why the change must be carried out,

together with a strategy backout, which is necessary

to restore a component to the situation immediately

prior to the change.

Enterprise Architecture - An Approach to Promote Traceability and Synchronization of Computational Models

463

Figure 2: Traceability Framework.

5 TRANSFORMATION PROCESS

In order to synchronize the elements of a set of

models, preliminary theoretical studies were

performed, demonstrating that the use of the Model-

Driven Development (MDD) (Lucrédio, 2009)

approach would allow the synchronization of the

concepts and aspects present in the models of an EA,

allowing, therefore, that the elements of the models

are always updated.

With MDD it is possible to transform more

abstract models into more specific models. Among

these models, there is a set of transformations rules

that are applied to achieve the expected result.

The transformations are relevant and many

works have been carried out in order to improve

them and even automate those (Dias et al, 2006).

However, most of these works deal only with the

transformations between the Platform Independent

Model (PIM) and the Platform Specific Model

(PSM), leaving behind the transformations that

involve the Computation Independent Model (CIM)

(Lucrédio, 2009). For (Dias et al, 2006), the

transformations involving the CIM models are

relevant, since it is through this model that an

adequate understanding of the business is obtained

that will give rise to the requirements that the

information system must attend. Therefore, from the

business specification represented by the CIM, the

requirements of the information systems that make

up the PIM must be determined. For this work, the

CIM are represented by the strategic and tactical

models of an EA. The operational model is a

representation of the PIM.

The process aims to carry out transformations of

models (strategic, tactical and operational),

originated by the set of meta-models proposed in

Section 3. The process, according to Figure 3,

consists of five activities: (i) define transformation

models; (ii) define transformation strategies; (iii)

design transformations; (iv) implement

transformations and; (v) execute transformations.

Figure 3: Model Transformation Process.

6 CONCLUSIONS

This article presents the proposal that aims to enable

the synchronization and traceability of the

computational models present in enterprise

architecture. The development of this work was

carried out based on bibliographic research on the

subject of enterprise architecture and under from the

perspective of computational models and the

strategic alignment of business and IT. At this stage

of the work, no studies were found to perform the

alignment (traceability and synchronization) of

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

464

models in their completeness. The possible benefits

expected for this research are:

 Creation of the integrated modeling (strategic,

tactical and IT operational) approach in a

traceable and synchronized way, viewing the

models as a single organizational asset.

 Decreased obsolescence of organizational

models and, consequently, of the organizational

holistic view;

 Improve impact analysis in scenarios of

organizational changes, allowing the institution

to be more flexible;

 Facilitate the construction and maintenance of

information systems.

 Contribute to the achievement of strategic

alignment of IT, through by aligning the models

present in A.E.

In this way, this work intends not only to improve

the way organizations think about performing their

computational modeling at the levels of enterprise

architecture, but also to improve the way the current

tools work with their models.

REFERENCES

Babar, Abdul., Wong, B., Gill, A. Q., (2011), An

Evaluation of the Goal-Oriented Approaches for

Modelling Strategic Alignment Concept. Research

Challenges in Information Science (RCIS), 1-8. Ieee.

Bleistein, S. J., Cox, K., Verner, J., & Phalp, K. T., (2006),

B-SCP: A requirements analysis framework for

validating strategic alignment of organizational IT

based on strategy, context, and process. In:

Information and Software Technology.

De Sousa, Rafael T.; et al, (2014), Business process

modelling: A study case within the Brazilian Ministry

of Planning, Budgeting and Management. In: CISTI.

Dias, Felipe.; Morgado, Gisele.; Oscar, Pedro.; Silveira,

Denis.; Alencar , Antonio Juarez.; Lima, Priscila.

Schmitz, Eber. (2006). Uma Abordagem para a

Transformação Automática do Modelo de Negócio em

Modelo de Requisitos.

Engelsman, W., Quartel, D., Jonkers, H., & Sinderen, M.

V, 2011, Extending enterprise architecture modelling

with business goals and requirements. Enterprise

Information System.

Fritscher, B., Pigneur, Y., (2011), Business IT Alignment

from Business Model to Enterprise Architecture.

Workshop on Business/IT Alignment.

Giannoulis, C., Petit, M., Zdravkovic, J., (2011), Modeling

Business Strategy: A metamodel of Strategy Maps and

Balanced Scorecards. Research Challenges in

Information Science.

ITGI. (2012) COBIT 5 - A Business Framework for the

Governance and Management of Enterprise IT. In:

<http://www.isaca.org/cobit/Documents/COBIT-5-

Introduction.pdf >. Accessed in June 2016.

Jonkers, H., Lankhorst, M., Buuren, R. V., 2004, Concepts

for modeling enterprise architectures. International

Journal of Cooperative Information Systems.

Josey, Andrew et al. (2011). TOGAF Version 9.1. In:

http://pubs.opengroup.org/architecture/togaf9-

doc/arch/. Accessed in June 2016.

Letelier, P. (2002). A framework for requirements

traceability in UML–based projects. In International

Workshop on Traceability In Emerging Forms of

Software Engineering.

Lucrédio, Daniel. (2009). Uma Abordagem Orientada a

Modelos para Reutilização de Software. USP – São

Carlos.

Kurz, Matthias., (2016)., BPMN Model Interchange: The

Quest for Interoperability., 8th International

Conference on Subject-oriented Business Process

Management.

Oglio, Pablo Dall. Silva, João Pablo Silva da. Crespo,

Sergio. (2010). Um Modelo de Rastreabilidade com

suporte ao Gerenciamento de Mudanças e Análise de

Impacto.

OGC. The Official Introduction to the ITIL Service

Lifecycle. UK: TSO, 2007.

OMG., 2015, UML 2.5 Specification. In:

http://www.omg.org/spec/UML/2.5/PDF/.

OMG., (2016), MOF 2.5.1 Specification. In:

http://www.omg.org/spec/MOF/2.5.1/PDF. Accessed

in July of the 2016.

Ramesh, B., & Jarke, M. (2001). Toward reference models

for requirements traceability. IEEE Transactions on

Software Engineering.

Ross, J. W., Weill, P., Robertson D. C., (2008),

Arquitetura de TI como Estratégia Empresarial.

Salgado, Carlos., Machado, Ricardo., Suzana, Rita.,

(2013), Using Process-level Use Case Diagrams to

Infer the Business Motivation Model with a RUP-

based Approach.

Sayão, Miriam. Leite, Julio Cesar. Rastreabilidade de

Requisitos. (2006). Revista de Informática Teórica e

Aplicada.

Zachman, J. a. (1987). A framework for information

systems architecture. IBM Systems Journal, 26(3),

276-292.

Enterprise Architecture - An Approach to Promote Traceability and Synchronization of Computational Models

465