Actors Based Competences Supporting Enterprise Modeling Changes
Marwen Jabloun, Yemna Sayeb and Henda Ben Ghezala
Riadi–GDL Laboratory, Manouba University, Manouba, Tunis, Tunisia
Khaled Gaaloul
Public Research Centre Henri Tudor, Luxembourg
Keywords: Enterprise Architecture, Actor, Competency, Performance, Ontology.
Abstract: The competitive environment of companies triggers changes having impacts on the enterprise information
system. In this context, Enterprise Architecture (EA) offers instruments to steer enterprises changes and to
ensure cohesion and alignment between the different aspects of an enterprise. Despite considerable numbers
of EA frameworks, there are not many giving importance to actors competencies supporting EA changes
when modeling. In this paper, we present an approach based on competencies to support enterprises’
changes. In doing so, we introduce a performance assessment method from the process modeling to the
competence selection using a competence-oriented ontology. Moreover, we develop a prototype as an
evaluation part of our work illustrated by a case study supporting EA decision making.
1 INTRODUCTION
The growing importance of alignment, compliance
and manageability issues have increased the
attention to EA (Braun et al., 2005). EA promises to
better align technical projects with business needs,
where business values could best be realized by a
holistic approach to systems architecture that
explicitly looked at every important issue from every
important perspective (Zachman, 1987).
EA presents methodology for guiding changes.
Those changes can impact different levels of the
enterprise. This needs an approach steering EA
changes. More specifically, the organizational aspect
is more and more considered where actors, roles and
competencies have a fundamental contribution in the
success or the failure of desired changes (Gagnon,
2008). Enterprise performance depends on the
alignment of the corporate strategy and business
process. The performance of business process
depends on the involved actors. These actors have
skills to achieve their tasks using relevant IT
systems (CIGREF, 2008). Hence, any EA change
should take into account actor’s competencies to
assess its performance. The importance of actor and
competency concept is clearly shown in enterprise
models while ignoring change issues (Le Boulaire et
al., 2008).
In this paper, we aim to focus on competencies
management supporting EA change. Our approach is
based, firstly, on a BPEL (business process
execution language) process that optimizes EA
change scenarios. The BPEL process ensures the
orchestration of services using business rules from
our performance assessment method. The
performance assessment integrates, then, a
competence-oriented ontology to ensure capturing
relevant information for actors’ performance (e.g.,
skills, training, etc.). Finally, we develop a prototype
as an evaluation part of our work illustrated by a
case study supporting EA decision making.
The remainder of this paper is structured as
follows. Section 2 introduces the research context
and problem statements. The approach is presented
in section 3, where we explain our methodology to
monitor process execution based a competence-
oriented ontology. Section 4 evaluates our work
illustrated by a real world scenario. Section 5
concludes and discusses future works.
148
Jabloun M., Sayed Y., Ghezala H. and Gaaloul K.
Actors Based Competences Supporting Enterprise Modeling Changes.
DOI: 10.5220/0005425301480152
In Proceedings of the Fourth International Symposium on Business Modeling and Software Design (BMSD 2014), pages 148-152
ISBN: 978-989-758-032-1
Copyright
c
 2014 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

2 BACKGROUND
2.1 Related Work
Traditionally, improvement initiatives are currently
positioned at two types of performance
improvement: continuous improvement and radical
improvement. EA is a continuous improvement
approach. The improvement is one of the basics of
quality managed by a set of standard such as the ISO
9000 that describes which requirements a system
needs for quality management. The ISO 9001
standard emphasizes the continuous improvement of
the overall performance of the company. This
principle means that management must measure
performance based on quality (Hachicha, 2012).
Competence management aims to improve the
enterprise performance through the effective
deployment of resources allocated to the business
process. Competence management involves the
following steps: (i) Analysis of existing data; (ii)
Analysis of decision change, and (iii) The design of
the new organization.
Several meanings and definitions of competence
exist, but no common definition seems to satisfy all
viewpoints (Pepiot, 2012). In fact, there is an
evolution in the considerations of the competence
concept in enterprise modeling languages. The
Unified Enterprise Modeling Language (UEML)
proposes a consensus in the scientific community
both at the terminology and conceptual (Petit et al.,
2002). The actor concept in UEML is considered as
a resource which is a specialization of the enterprise
object. Resources can be specialized into three
categories: applications, human resources and
machines. UEML project has open research
perspective. There are investigations of
complementary language for competencies modeling
such us the Unified Enterprise Competence
Modeling Language (UECML). UECML is an
enterprise modeling approach based on
competences. UECML is based on a set of core
constructs and a set of additional constructs. These
additional constructs are specialized constructs
required by the competence and resource based on
enterprise modeling needs (Pepiot, 2012). The actor
here is considered as an enterprise object. It is a
subclass of enterprise object class. Therefore, it
inherits all the properties of objects organization.
The actor is assigned to a role and is characterized
by a set of competence. We notice three granularity
levels related to competence: (1) Unit competence
which is considered as the basic level and entity in
the model is tightly linked to an activity; it is defined
as the ability to mobilize effectively material and
non-material resources in order to respond to an
activity; (2) Individual competence is the set of unit
competences and resources developed/required by
an actor within the framework of assigned activities;
(3) Collective competences is considered as the
highest competence level and linked to processes
and group of actors.
Another aspect is about the link between
competence and performance. In literature, many
researches develop the competence integration in
performance assessment. In (Bennour et al., 2006)
authors proposed to model and assess the
competence based on knowledge, rather than an
overall assessment of competence. In (Hlaoittinun et
al., 2009), they quantify the impact of both
individual core competencies (technical, decision
autonomy) and collective (management, relational)
to assess the modulation rate of the nominal
performance to execute an activity. Finally, the
approach proposed in (Gruatet al., 2006) introduced
the level of competency where a performance is
calculated based on the operator’s productivity such
as in production line.
2.2 Research Motivations
Despite the competence extension, UECML is still
lacking of a systematic approach evaluating
competences based on resources. In a context where
actors vary from the manager who makes strategic
decision to the worker who performs operational
activities, the issue is to ensure consistency with
regards to actor’s understandings and viewpoints.
We think that leveraging competence features in EA
can ensure this consistency. To that end, we propose
a performance method that predicts impacts based
on actor’s competence. Prediction will allow
architects to deal with situations such as the lack of
resources when an actor is absent or has no skills to
execute an activity with the process. Based on our
approach, we can offer alternatives as anticipatory
actions such as training, outsourcing to solve such
problems.
3 THE APPROACH
The performance assessment based on competence
is realized in three main steps: (1) Identification of
KPIs and metrics derived from EA strategy in order
to deal with the normative dimension of
performance assessment; (2) Collecting relevant data
relevant for competence computation where we
Actors Based Competences Supporting Enterprise Modeling Changes
149

propose an ontology to deal with cognitive
dimension; (3) Development of business rules in
order to orchestrate services invocations from the
BPEL process.
The normative dimension is based on a set of
KPI an d corresponding metrics. It provides metrics
for business rules to assess performance. From the
literature (Rezaei et al., 2011), we have identified
the different services: developing a performance
measurement system needs the conversion of
strategic objectives into key goals, the establishment
of metrics to compare the desired performance with
the actual achieved standards, the gaps identification
to allow us to understand performance and finally
the initiation of improvement actions.
The trend in performance assessment is geared
towards intangible success factors such as
competencies, customer satisfaction, motivation and
staff satisfaction (Rezaei et al., 2011). In our
approach, we are based on competence concept
because it is interrelated within EA layers, and so,
provides a broad vision that facilitates decision
making. Moreover, KPIs depend on the competence
family. In fact, according to (CGIREF, 2009)
competencies related to design, operation,
infrastructure and technical architecture tend to
move away from the enterprise core and even
to move geographically away.
Performance assessment may have some
semantic issues when dealing with metrics
interoperability due to cultural differences between
actors, e.g., business people and information
technology professionals (Frank, 2002). It is a
critical problem because it may leads to data
collection aberrations.
Ontologies have been considered as solution to
ensure the metrics interoperability in process
reengineering (Charlet et al., 2002). We define an
ontology entitled Ontology of Enterprise Oriented
Competence (OnEOC) (see figure 1). The OnEOC
ontology is composed of the following components:
- Thing, an abstract object representing the class
of all things.
- Urbanization view refers to the different EA
views: strategy view, the business view, the
organizational view and the technical view.
- Work unit is a component that refers to a
decision Unit it is linked with the performance
estimation and the choice of changes reasons.
- The actor concept is a key concept. Actors have
different competencies and roles. The actor in
OnEOC can be considered through different
views depending on his role and his capabilities.
- The competence concept identifies the unitary
competence, personal competence and
collective competence.
- The change concept is also defined in the
OnEOC where it ensures the evolution of an
existing state (AS-IS) to a desired target state
(TO-BE).
The OnEOC ontology allows different types of
research via ontology queries. For instance, actors
executing tasks with a given KPI value such as the
number of resolved complaints in an hour. The
ontology allows a semantic interpretation of KPI,
where technical actors interprets the KPI query
execution time by the computer system, while actors
at the strategic level interprets this KPI according to
customers satisfaction percentage. Then, after
identifying KPI and collecting relevant data; the
performance assessment needs to develop business
rules. Every business rules depends on the identified
KPI and the extracted data from the ontology. These
business rules will be executed via a process
implemented by the business process execution
language (BPEL). The idea is to allow combining
a set of services for specifying business rules
where performances are evaluated by orchestrating
these rules.
BPEL is an XML language designed to allow the
composition and orchestration of web services.
BPEL inputs translate the different possible changes
scenarios and the output is a value of the expected
performance after change. BPEL inputs define the
subject (reason) of the desired change, e.g., the
stakeholder management to modify existing business
processes. Changes criteria are either immediate or
continuous. Change’s nature can be related to
specific actions such as add, delete or modify
involving actors, systems, and processes. Then
depending on the input, an invocation of a web
service is executed based on business rules. The
principle of performance assessment depends on the
identification of the gap between existing
competences and required competences allowing the
adaptation to a given change. So the output of t this
process represents the value of performance after
change. Note that depending on the importance of
the impacted actor and the importance of the
impacted business domain, a decision can be taken
in order to plan appropriate actions to deal with any
performance lowering.
Fourth International Symposium on Business Modeling and Software Design
150

Figure 1: Ontology of Enterprise Oriented Competence (OnEOC).
4 EXPERIMENTATION
To illustrate and validate the proposed approach, we
present here a case study of a Telecom operator. We
have implemented the liaison between the ontology,
the BPEL process and its business rules. The
synchronized services by BPEL are deployed as web
services. They are defined as requests using
SPARQL queries (SPARQL Protocol and RDF
Query Language) (Perez et al., 2009). These
SPARQL queries are used to express the business
rules implemented in Java classes.
Figure 2: Prototype supporting process change.
Figure 2 depicts the overall architecture of the
implemented prototype where the main components
are presented by rectangles. The headed-arrow
connections define the interaction between
components and lozenges represent web services.
The Implemented BPEL process defines the
execution of invoked processes by the web
application. It is composed of many portion
separated by an else/if condition. This condition
depends on change’s subject. Each process’s portion
calls the following operations: Receive to get the
input, Assign to map the process input data and the
Web service input data, Invoke to calls a web
service, Assign to map the process output data and
the Web service output data, and Reply to return
results.
The proposed prototype is composed of a web
application supporting a web interface that allows
user to monitor change. The user can define the type,
the subject and the nature of the change scenario via
a web form. The web application offers also an
interpretation to show the returned results.
In figure 3, we can change the actor assigned to
task 3. This change can be permanent, temporary or
immediate. The change may include additional
actions (e.g., add, delete, modify).
Actors Based Competences Supporting Enterprise Modeling Changes
151

5 C
O
The ma
i
enterpri
s
concept.
assessm
e
making
compete
n
compete
n
defining
We
a
method
help bu
s
domain
vocabul
a
describe
Semanti
c
Rules
(
Langua
g
rationali
z
REF
E
Braun, C
.
A
rch
i
a M
e
Man
a
Zachman
Syste
m
26.
Gagnon,
Y
de l
a
natio
n
CIGREF
Un c
a
l’ent
r
Le Bou
l
comp
:quel
Instit
u
Charpon,
méth
o
du s
y
Figure 3: C
h
O
NCLUS
I
i
n contributi
o
s
e changes
This conce
p
e
nt to monito
r
in EA. Sup
p
nce is sup
p
o
r
nce addres
s
metrics for a
c
a
im to impro
v
for modeling
s
iness experts
in a natur
a
a
ry that supp
the overal
l
c
s of Busin
e
(
SBVR) an
d
g
e (SWRL)
z
e actors’ per
E
RENCES
., Winter, R.,
2
i
tecture Metam
o
e
tamodeling P
l
a
gement Unive
r
, J.A., 1987.
m
s Architectur
e
Y-C., 2008. T
h
a
réussite du
c
n
ale d’adminis
t
livre blanc,
2
a
dre global d
e
r
eprise.
l
aire, M., R
e
étences, straté
g
est le rôle
d
u
t d’Administr
a
J., 2006. Thè
s
o
de et aide à l
a
y
stème d’info
r
h
ange process i
n
I
ONS
o
n of this pa
p
based on
p
t contribute
s
r
changes and
p
orting chan
g
r
ted by the O
n
s
ing semant
i
c
tors’ perfor
m
v
e this approa
c
business rul
e
to model th
e
a
l language
ly the neces
s
l
framework
.
e
ss Vocabul
a
d
the Sema
will be o
u
formance in
E
2
005.
A
Compr
e
o
del and Its Im
p
l
atform. Instit
u
r
sity of St. Gall
e
A
Framewo
r
e
. IBM Syste
m
h
èse:
L
es trois
c
hangement te
c
t
ration publiqu
e
2
008.
L
’Archit
e
e
coopération
p
e
tou
r
, D., 2
0
g
ie et perform
a
d
e la fonction
a
tion des Entre
p
s
e:
L
'urbanism
a
décision pou
r
r
mation de l’
e
n
terface.
p
er is to su
p
the compet
e
s
in perform
a
support deci
s
g
e process
u
n
EOC ontolo
g
i
c issues
w
m
ance.
c
h by proposi
e
s. The idea
i
e
desi
r
ed busi
supported b
s
ary semanti
c
.
Standards
a
ry and Busi
n
tic Web
R
u
r next ste
p
E
A framewor
k
e
hensive Enter
p
p
lementation
U
u
te for Inform
a
e
n, Switzerlan
d
r
k for Inform
a
m
s Journal, Vo
l
leviers stratég
i
c
hnologique.
É
e
du Québec.
e
cture d’Entre
p
p
our les acteu
r
0
08. Gestion
a
nce de l’entre
p
RH ?. CERA
G
p
rises de Gren
o
e
organisation
n
r
piloter l’évol
u
e
ntreprise. L'
E
p
port
e
nce
a
nce
s
ion-
u
sing
g
y of
w
hen
i
ng a
i
s to
i
ness
b
y a
c
s to
like
i
ness
R
ule
p
to
k
s.
p
rise
U
sing
ation
d
.
a
tion
l
ume
i
ques
É
cole
p
rise
r
s de
des
p
rise
G et
o
ble.
nel :
u
tion
E
cole
Ha
c
Pet
i
Pép
Be
n
p
Hla
Gr
u
p
Re
z
CI
G
Fra
n
Ch
a
p
Per
e
Nationale Sup
é
l'Université Je
a
c
hicha, M.,
M
odélisation
e
sous Contrai
n
UNIVERSITE
i
t, M., Doumei
n
A
rt in Enterp
r
Enterprise Mo
d
io, G., Chei
k
Glardon, R.,
Competence
M
Fédérale de
L
Management
a
n
nou
r
, M. and
C
in performan
c
p
rocess. Uni
d’Informatiqu
e
de Montpellie
r
o
ittinun, O.,
construction
d
s
tructuration
d
Université Fra
n
u
at, F.A., Bott
a
M
odélisation
d
p
rise en comp
t
Maroc.
z
aei, A.R., Çeli
k
measurement
Department
Mediterranean
G
REF, 2008.
compétences
l
Management
d
n
k, U., 2002.
M
(MEMO) -
C
Languages, P
r
Conference on
a
rlet, J., Bachi
m
p
our le Web s
é
Institut Nati
o
Compiègne.
e
z, J., Arenas,
Complexity o
f
ACM, New Y
o
é
rieure des Mi
n
a
n Monnet.
2012. Thèse
e
t Résolution d
u
n
tes de Com
p
PARIS 8 Vinc
n
gts, G., 2002.
R
r
ise Modeling,
d
eling Langua
g
k
hrouhou, N.
,
2006. UEC
M
M
odeling Langu
a
L
ausanne, La
b
n
d Processes,
L
C
restani, D., 2
0
c
e estimation:
v
ersité Mont
p
e
, de Robotiqu
e
.
2009. Thès
e
d
’équipes de
d
u projet et le
p
n
che-Comté, F
r
a
-Genoulaz, V.
d
’un problème
t
e des compéte
n
k
, T., Baaloush
a
in a qualit
y
of Civil
University.
Livre blan
c
l
es Systèmes
d
d
es Systèmes d’
M
ulti-Perspect
C
onceptual Fr
a
r
oceedings of
t
System Scien
c
m
ont, B., Tron
c
é
mantique. Mi
s
o
nal Univers
i
M
., Gutierrez,
SPARQL. AC
M
o
rk, USA.
n
es de Saint Et
i
e
: Contributi
o
u
Problème d’
A
p
étences et P
r
c
ennes-Saint-D
e
R
eport on the
S
,
Projet UEM
L
g
e.
,
Furbringer,
M
L: Unified
E
age. Ecole Pol
y
b
oratory for
P
L
ausanne, Swit
z
0
06. Using co
m
From the acti
v
t
pellie
r
2,
L
e
et de Microé
l
e
: Contributi
o
conception c
o
p
ilotage des co
m
r
ance.
.
, Campagne,
J
d’ordonnanc
e
n
ces. MOSIM’
h
a, Y., 2011. P
e
y
manageme
n
Engineering,
c
sur l’évol
u
d
’Information,
Information d
e
t
ive Enterprise
a
mework and
the Hawaii In
t
c
es.
n
cy, R., 2002.
O
s
sion de recher
c
i
té Technolo
g
C., 2009. Sem
M
Trans. Dat
a
i
enne et
d
e
o
n à la
A
ffectation
r
éférences.
e
nis.
S
tate of the
L
: Unified
M., and
E
nterprise
y
technique
P
roduction
z
erland.
m
petencies
v
ity to the
L
aboratoire
ectronique
o
n à la
o
uplant la
m
pétences.
J
.P., 2006.
ment avec
06, Rabat,
e
rformance
n
t system.
Eastern
u
tion des
École de
e
Grenoble
Modeling
Modeling
t
ernational
O
ntologies
c
he STIM,
g
ique de
a
ntics and
a
base Syst.
Fourth International Symposium on Business Modeling and Software Design
152