A New Approach for SBPM based on Competencies Management
Wafa Triaa
1
, Lilia Gzara
1
and Herve Verjus
2
1
GSCOP, Technology Institute of Grenoble, Grenoble France
2
LISTIC, Universite Savoie Mont Blanc, Annecy, France
Keywords:
BPM, Agile-BPM, SBPM, Collaboration, Competency Management.
Abstract:
In such continuous changing business work environment, traditional BPM has two principal issues: firstly
the model-reality-divide, the typical separation between process’s design and execution. Secondly, the loss
of innovation associated to the lack of internal performers’ implication. To overcome these issues and to
stress continuous adaptation and rapid innovation, BPM has to be agile. Otherwise, an agile enterprise is
basically an enterprise of knowledge and skills. Human dimension the key element of an agile enterprise was
and stills not taken into consideration within BPM. One of the recent solutions to support BPM agility is the
integration of Social Software (SS) principles within BPM leading to the emergence of Social BPM (SBPM).
Although the importance and the innovative ideas of the proposed approaches, they are not able to address all
the identified issues of traditional BPM and to support all the phases of its lifecycle. Thus, in our approach,
we integrate competency management to answer how stakeholders can find the right performers at the right
time for the right type of contribution. It is mainly based on three phases: 1) identification of the required
competencies to fulfil a specific need. Based on a semantic analysis, the system will be able to identify the
required competencies and automatically extract the possible candidates. 2) Then the identified candidates
will be evaluated against our defined criteria (related to time dimension, human dimension, cost dimension,
etc.) to select the relevant ones. 3) Finally, after selecting the relevant performers, the process model will be
adjusted based on the identified competencies. In this paper, we will typically present the first phase.
1 INTRODUCTION
With the advancement of technology over the past
decade and the increase of competition in the in-
dustry, the need for effective management of orga-
nization’s business processes has become more im-
portant than ever before (Ryan., 2009). Priority of
every organization or company is to increase oper-
ational efficiency, reduce costs, improve quality of
their products or services and better manage opera-
tional knowledge. Many organizations are using busi-
ness process management (BPM) as a key component
in automating their processes, increasing standardiza-
tion and improving performance and customer satis-
faction. BPM typically consists of series of activ-
ities for the ongoing improvement of business pro-
cesses that are carried out within an iterative life cy-
cle (Weske, 2012). In addition, business processes
are classified into two main categories depending on
their nature: the first concerns well structured, highly
repetitive processes. These processes are subject to
little change over time and they are often supported
by traditional BPM. The second category concerns
loosely structured processes known as knowledge-
intensive processes which cannot be supported by tra-
ditional BPM (Gottanka, 2012). As it was affirmed by
(Gilbert, 2010) in an example of a large bank, more
of 60% of the processes are knowledge-intensive pro-
cesses known also as ad-hoc processes, not covered
by classical BPM methods. While just 2.5% of them
are highly complex repetitive and allow a substan-
tial automation. Such ad-hoc processes can be seen
as to what (G., 2011) called the accelerated ”pace of
changes” as well as the ”spreading of context infor-
mation and the demand for quickly created process
solutions” of BPM. Furthermore, research in the field
of BPM pays more attention to reduce its incapacity
in order to support ad-hoc processes. The evolution of
BPM over the years suggests that there are still lim-
itations within the different stages of BPM’s lifecy-
cle. One of the main limitations of traditional BPM is
the separation between the process design and execu-
tion, which is often referred to as model-reality divide
(Palmer, 2011). Thus, during process execution, ideas
suggested by internal performers may remain unused
during process design and cannot be shared within
Triaa, W., Gzara, L. and Verjus, H.
A New Approach for SBPM based on Competencies Management.
DOI: 10.5220/0006704706730681
In Proceedings of the 20th International Conference on Enterprise Information Systems (ICEIS 2018), pages 673-681
ISBN: 978-989-758-298-1
Copyright
c
2019 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
673
the organization (Schmidt, 2009). Owing to the fre-
quently stated fast changing business world and un-
predictability of processes, several works in academia
and industry propose concepts to enable the continu-
ous and rapid adaptation of processes to change. This
capability known as agility considered as inescapable
feature of today’s forward-looking corporates. Thus,
BPM must be agile in order to be able to react quickly
and adequately to internal and external events. One
of the recent solutions to support BPM agility is the
integration of Social Software (SS) principles and
techniques within BPM leading to the emergence of
Social BPM (SBPM). In literature, combination of
BPM and social software is discussed under the terms
subject-oriented BPM (Felishman, 2010), social BPM
(Schmidt, 2009) and BPM 2.0 (Kern, 2015). These
terms refer to the improvement of business processes
that seeks to break down silos by encouraging a more
collaborative and transparent approach. In our pa-
per, we use the term Social BPM to describe the inte-
gration of social software principles within BPM. In
such context, BPM paradigm changes from closed to
open and social. Rather than centrally defining pro-
cesses by the managers and deploying them for exe-
cution by internal performers, business processes can
be reached to a broader class of stakeholders. The in-
tegration of social software with BPM depends on the
companies’ needs. Some of them will only use so-
cial software functionalities for communication, oth-
ers will use it to reduce their time to market, and
yet others will use it for transformation. Actually, in
such continuously changing and turbulent work envi-
ronment, using social software principles to enhance
process adaptation and transformation seems to be
strongly important. While most previous research was
focused on improving the collaboration between the
model creators and internal performers during process
design phase, minor loosely coupled social features
within BPMS are suggested. This is far from enabling
full exploitation of the SBPM benefits and the princi-
ples of social software which have been identified a
long time ago but not properly implemented. As well
as traditional BPM, SBPM allows efficiently man-
aging and coordinating business processes indepen-
dently from human resources. They are designed to
provide a support to the stakeholders involved in these
processes to answer the questions: what needs to be
done? When should it be done? These stakeholders
are assigned to perform tasks in specified sequences
without taking into consideration their acquired com-
petencies. Indeed, competencies management, which
has been suggested as a way to more effectively uti-
lize employee skills, stills not taken into consideration
while managing business processes. Thus, in this pa-
per we aim to answer how stakeholders can find the
right performers at the right time for the right type
of contribution. Our work is mainly based on three
phases: 1) identifying the required competencies to
fulfill a specific need. Based on a semantic analy-
sis, the system will be able to identify the required
competencies and automatically extract the possible
performers. 2) Then the identified performers will be
evaluated against our defined criteria (availability, co-
hesion, flexibility...) to select the relevant ones. 3)
Finally, after selecting the relevant performers, the
process model will be defined based on the identi-
fied skills. Therefore, processes could be designed,
modified and adjusted dynamically during execution
to include unplanned participants. In this paper, we
present our SBPM approach to support collaboration
during execution and designing processes on the fly.
So the remainder of this paper is structured as follows.
Section 2 gives a depth review of BPM and the main
existing issues the roots of SBPM’s emergence. Then
Section 3 presents the suggested approach for effec-
tive and efficient SBPM improvement. And finally,
section 4 concludes the paper.
2 BACKGROUND AND RELATED
WORK
2.1 BPM
Business Process Management (BPM) is the dis-
cipline that combines knowledge from information
technology and knowledge from management sci-
ences and applies this to operational business pro-
cesses (Weske, 2012). The goal of BPM is to achieve
the organization’s objectives by aligning them with
business processes and to continuously improve these
processes. It includes concepts, methods, and tech-
niques to support the design, administration, configu-
ration, enactment and analysis of business processes
(Van der Aalst, 2007). BPM provides a platform to
manage business processes through their lifecycle as
it is depicted in Figure 1, which represents one of
the simplest proposed models of BPM’s lifecycle in
the literature. The technical tool to manage busi-
ness processes and support the entire lifecycle is Busi-
ness Process Management System (BPMS). Owing
to the complexity and turbulence of work environ-
ment, BPMS support the execution of stable and sim-
ple business processes. The need to support quick ad-
justment of business processes in order to meet chang-
ing environmental conditions was behind the devel-
opment of several approaches and suggestions in the
ICEIS 2018 - 20th International Conference on Enterprise Information Systems
674
Figure 1: Van der Aalst’s BPM lifecycle.
literature. One from the best and newest way of sur-
vival and success of organizations is agility. Over the
last two decades, corporates have focused on improv-
ing the agility of their business processes. Several
solutions emerged today, to support processes agility
heavily influenced by context and application domain.
One of the recent solutions is Social BPM (SBPM)
mostly defined as the integration of Social Software
principles and techniques within BPM. The motiva-
tion for including social software and BPM contains
many facets: fostering collaboration, sharing knowl-
edge, support process models adaptation and others.
In fact, the aim of SBPM is to solve the principles is-
sues of traditional BPM, which will be revealed in the
next section.
2.2 BPM Principles Issues
The current state of traditional BPM has two princi-
pal issues identified by (Granitzer M. and Schmidt R.,
2010):”model reality divide” and ”lost innovations”.
Model reality divide: it represents the gap be-
tween what the process actually is and what hap-
pens in real life. In fact, during the design stage of
BPM lifecycle, process models are created using
modeling languages like BPMN, Petri-nets and
others. Since these process models are often an
abstraction from the real world, exceptions are of-
ten not covered by them as well as tasks that are
difficult to be modeled.
Innovation’s loss: during business process lifecy-
cle, the lack of model users’ implication leads to
lose some important ideas and information for in-
novation. BPM normally follows a top-down ap-
proach, where processes are designed by a group
of individuals and passed on to internal perform-
ers to follow (Sinur, 2011). Working under a strict
top down manner, employees are not motivated to
share ideas for process improvement and innova-
tion. Besides, their knowledge is either lost en-
tirely, or applied on the local scale of individual
process instances.
These properties of the standard BPM cycle make
it unsuitable for so-called knowledge-intensive pro-
cesses. These problems were behind the emergence
of SBPM.
2.3 Social BPM
Research in SBPM formally started in 2008 (Schmidt,
2009) and it has evolved ever since. Several defini-
tions have been proposed to understand what SBPM
is and how it operates. Although there is not a sin-
gle understanding of the concept of SBPM (Houy C.
and Loos P., 2011), depending on the specified needs,
SBPM could be adapted and its functionalities could
be integrated to satisfy these needs. Within BPM, so-
cial software principles can be used to support the
different lifecycle steps of a business process or to
support an individual lifecycle phase. In the litera-
ture, various works were carried out to well integrate
BPM and social software and to answer the research
question of how to overcome model-reality divide and
lost innovation principal issues of BPM. The first cat-
egory in SBPM is to support collaborative model-
ing of business process. Technical and non-technical
people have to participate in the discovery, modeling
and design of business processes in order to ensure
its acceptance. The first framework which is called
BPM4PEOPLE was developed by (Fraternali, 2012).
A social extension of BPMN known as BPMN 2.0
based on the use of design patterns is the principal fea-
ture of this framework. The extension made by (Fra-
ternali, 2012) aims to support collaboration among
stakeholders and to reduce model-reality-divide issue
using means of communication that will enable em-
ployees to exchange, talk about, integrate and lever-
age existing knowledge from different sources. In
(Hauder, 2014), they developed a solution based upon
hybrid wiki that empowers users to collaboratively
design and adapt information structures. Using hybrid
wiki, no special syntax or modeling concepts are re-
quired to utilize the structured information elements.
Another tool was presented by (Houy C. and Loos P.,
2011) called CoMoMod. This tool supports several
aspects of collaborative process modeling, such as si-
multaneous work of spatially distributed modelers on
one process model diagram. Socially support busi-
ness process during their execution phase is seldom
taken into account. Few approaches were developed
since the emergence of SBPM. One of these propo-
sitions is a framework called AGILIPO (AR., 2010).
AGILIPO follows the principles of agile software de-
velopment (Rosemann M., 2001) and of organiza-
tional design and engineering. An incomplete process
definition is specified by a set of activities that de-
scribe part, but not all, of the process instances behav-
A New Approach for SBPM based on Competencies Management
675
ior (AR., 2010). Another proposition to improve the
SBPM functionalities is developed by (Fettke, 2013).
This approach enables stakeholders to communicate
with each other, to create groups for discussions or
to ask question to the audience of a specific business
process. The given answers are not restricted to one
person rather they are visible to all members of the
created group. (Karakostas B., 2013) developed an-
other proposition that follows this idea. Their ap-
proach discusses how social tagging can be used in
the context of social business process management to
assist and support the execution of business processes
in a social environment. Even there are several ap-
proaches proposed in the literature to integrate social
software techniques within BPM, some problems re-
main unsolved. In many situations, a performer has
to improvise or to find the suited competencies to get
the work done. There are situations, where changes
on an existing process model are necessary, when ex-
ternal context of the process is changing. Although
the importance of competencies management none of
the proposed SBPM approaches supports this con-
cept. Offering BPM as social software is a promising
approach that fosters communication and collection
of knowledge by allowing multiple users to work on
the design, operation and improvement of a business
process simultaneously and without access control re-
strictions typical of traditional BPM (Karakostas B.,
2013). Enterprises have found it difficult to utilize
social software in such manner to 1) achieve its ob-
jectives, 2) add value and make it attractive to the
members and 3) avoid unintended consequences. Al-
though the importance of organizational integration
none of the proposed SBPM approaches support this
concept. Considering agility as the most important
aspect of processes management (Wafa Triaa, 2017),
SBPM needs to provide contextually useful informa-
tion, customized services connecting specific users to
each other in order to respond to subsequent process
exceptions. Lack of customization, community cre-
ation and expert’s retrieval process need to be studied
and implemented to better support and exploit SBPM
benefits. The accession of the required customized
information in real time is a crucial concept for a ro-
bust execution of business processes. Within actual
SBPM and even traditional BPM, information is not
classified, structured and organized; each actor can
express his or her opinion freely and give further sug-
gestions that need analysis before select the most rele-
vant ones. Instead of working alone, employees have
to establish and maintain relationships with one an-
other and to perform several interactions. We talk
about weak ties which are of special importance in
this context as they form the long tail of knowledge as
it is affirmed by (Schmidt, 2009). So leveraging the
collective intelligence of a business community can
only be accomplished if all relevant participants are
actually included and their needs considered. How
stakeholders can find the right actors at the right time
for the right type of contribution. To overcome this is-
sue, we think to support customized information dis-
tribution during run time. Actually, people cannot lo-
cate knowledgeable colleagues because they are not
provided with the proper means to do so. Based on
competencies retrieval process, employees can iden-
tify whom to talk and capture their knowledge effi-
ciently and easily. Our focus is to extend the reach
of SBPM for good inter organizational involvement
of employees during business process execution by
answering the following: how, when, and which ex-
ternal and internal actors should be included to per-
form processes activities? We also want to deal with
process actor autonomy as processes are often pre-
scriptive as the reality is mostly empirical and non-
determinist.
3 PROPOSED SBPM
FRAMEWORK
3.1 Functional Requirements
Our goal aims to analyze and construct the social net-
work between processes’ actors, to provide answers
to two important questions of ”who owns what?” and
”who needs what?” Owing to competencies retrieval
process and data mining techniques, unplanned par-
ticipations could be rapidly and dynamically taken
into account. To move forward the following areas
need to be investigated and addressed:
Elimination of distinctions between designers and
executors in the SBPM framework. Executors
transmit the required information to complete the
process model. This is in order to avoid the typical
issue of model-reality divide as faced by current
approaches in BPM as well as SBPM.
Merging of the design and execution phase in the
BPM model. So that there is no explicit or ordered
distinction between the two phases. Such ap-
proach will support agility of business processes.
Elimination of pre-defined tasks in the processes,
in order to support knowledge intensive processes
characterized by activities that cannot be planned
easily. These processes may change on the fly
and are often driven by the contextual scenario in
which they are embedded in. The scenario dic-
tates who should be involved and who the right
ICEIS 2018 - 20th International Conference on Enterprise Information Systems
676
person to execute a particular task is. The set of
users involved may be not formally defined and be
discovered as the process scenario unfolds.
Identifying and exploiting competencies to op-
timize the choice of individual (or groups) for
emergent tasks. Relationships between the pro-
cess participants are mapped and measured within
social network analysis. One of its aims is
supporting collaboration in workplaces and team
building.
The above needs can be achieved through incorpo-
rating an approach where both design and execution
are blended, competency management is integrated
and social systems behavior is involved. Actors will
be identified due to their competencies to support pull
and push service. Therefore, we’ll support the cre-
ation of an organizational environment that enables
and fosters continuous customized contributions of all
stakeholders. The scenario dictates who should be
involved and who the right person to execute a par-
ticular task is. Therefore, the framework will sup-
port the design of a process during its execution (de-
sign by doing). Actually, one of the main basic pro-
cesses of competency management (Giuseppe Berio,
2006), (LeBoterf, 2004) is: competency utilization.
The aim of this process is to optimize the competency
allocation based on predefined human resources’ pro-
files. To better support the required-owned compe-
tency corresponding, performer’s profile should be
modeled. Without a clear sense of identity, there can
be no foundation for trust or reputation. Thus, when
using sufficient information about each performer’s
participation, each performer will have a specified
profile defined as a set of his owned competencies
facilitating the expert’s retrieval process. To do so,
we need to develop a competency referential to facil-
itate the identification of the required competencies
and the allocation of the acquired competencies. Now,
we will give some definition of the competency con-
cept and its main characteristics in the next section.
3.2 Competency Definition and
Characteristics
In industrial engineering competency is integrated as
an essential point of view for enterprise modeling. In
the literature, several definitions of competency are
available. For example in the HR-XML Consortium
Competencies Schema ((HR-XML, ), a competency
is defined as: A specific, identifiable, definable, and
measurable knowledge, skill, ability and/or other de-
ployment related characteristic (e.g. attitude, behav-
ior, physical ability) which a human resource may
possess and which is necessary for the performance
of an activity within a specific business context. There
are two main categories of competency derived from
the available literature, (SJanas, 2008), (Giuseppe Be-
rio, 2006):
Hard competency: which refers to two types: a)
Know: It concerns everything that can be learned
from educational/formative systems. It represents
the theoretical understanding of something such
as a new or updated method or procedure, etcand
b) Know-how. It is related to personal experiences
and working conditions. It is learned by doing,
by practice and by experience. It is the practical
knowledge consisting in ”how to get something
done”.
Soft competency: which consists on relational
know-whom, cognitive know-whom and behav-
ior. It is referred to individual characters, talents,
human traits, or qualities that drive someone to
act or react in a certain way under certain circum-
stances.
During our competency retrieval process, the aim is
firstly to identify what type of hard competency is
needed to reach/satisfy the goal. Thus, in our work,
we focus mainly on the first category of competency:
Hard competency. Many studies evaluate different
methods of competency modeling (Vernadat, 1999),
(Zarifian, 2001), (SJanas, 2008) show that ontology
have a greater expressiveness and meet better the re-
quirements and the goals of the competency model-
ing. Ontology are collections of concepts, instances
of concepts and relations among them that are ex-
pressed at the desired level of formality and that are
deemed to be important in characterizing the knowl-
edge domain under consideration at the desired level
of detail (Prilla, 2008). In our work, we have cho-
sen an ontological approach to define the function-
alities covered by the entreprise. In our ontology as
depicted in Figure 2, the central concept is ”Compe-
tency” which is linked to the five identified concepts:
Situation, Know-how, Know, ResourceType and De-
liverable. We propose that a competency manifests as
knowing how to act. It results in one or more knowl-
edge in action (Know how) in a given situation by
mobilizing knowledge (theories, procedures) and re-
sources (technologies, materials ) to provide a result.
The objective and the requirements of the correspond-
ing situation define the composition of the competen-
cies and specify the corresponding criteria for their
implementation. Possible relationships might exist
between two or more hard competencies. In some
cases, the execution of one competency requires the
presence of other one(s). Indeed, if a stakeholder
defines his goal in the form of a request, semantic
matching, which will be used to enable the identifi-
A New Approach for SBPM based on Competencies Management
677
Figure 2: Competency ontology.
cation of the required competency, will not be suffi-
cient if the relationships between them are not taken
into consideration. In this case, competency defini-
tion and modeling is supported by the relationship:
C
i
Requires C
j
where C
i
and C
j
are two hard compe-
tencies. Adding to that, symantic relations have been
defined to avoid the no recovery situation. We consid-
ered two types of ontological relationships: concep-
tual relations and semantic relations. A conceptual
relation is a relation between two concepts (ie ”pro-
duces” is a conceptual relation between the compe-
tency concept and the deliverable concept), where a
semantic relation is a relation between two names of
the same concept. (ie synonymy relation between two
names of the same competency: program in java ”is
a” develop in java). If we found different names for
the same concept, so we have to define a father con-
cept and the list of son concepts. We have defined the
relationships between the concepts of ontology as il-
lustrated in the Figure 3. The semantic relations exist
between the father concept and its son concept which
has the same meaning. Conceptual relations exist
only between two father concepts. When express-
Figure 3: Identified relations.
ing the need for the requestor through his request, the
extracted concept(s) can be directly aligned with the
competencies of our ontology or with other concept(s)
that are conceptually related to competency or seman-
tically related to another concept that is conceptually
related to one or more competencies. Each concept of
the stakeholder’s request will be randomly integrated
in the competency’s ontology and using the Dice mea-
sure as defined below, the corresponding concept will
be localized in the ontology.
Dice(C
r
, C
o
) =
2 Card(C
r
C
o
)
Card(C
r
) +Card(C
o
)
right} (1)
The similarity values are real numbers between 0 and
1 where:
Case 0.0: zero recovery.
Case 1.0: Total recovery.
To match the concept of the request with the corre-
sponding one in the ontology, two situations are iden-
tified. On the first hand, the stakeholder could men-
tion the name of the required competency in his re-
quest, in such case the ontology is used to extend
the list of the required competencies with the similar
ones. On the other hand, the stakeholder may don’t
know what the required competency is. In this case
the ontology is used to (1) find the required compe-
tency and (2) extract the similar competencies if there
exist. Thus, a set of required competencies will be
identified which will be used to identify the relevant
performers.
3.3 Performer’s Profile
The profile of each performer is represented as a vec-
tor combing each acquired competency with the cor-
responding mastery level. Only father concepts are
used to define each performer’s profile. Where C
i
is
the index of competency i and n
mi
is the level of mas-
tery of competency i by the performer m. To evaluate
the acquired competency, we defined five levels ac-
cording to the French repository ROME (it provides a
simple description of competencies to be easily iden-
tified): (0: Absence), (0.25: Sensitization), (0.5: Ca-
pacity to put into practice), (0.75: Mastery) and (1:
Expertise). We have to note that the mastery level
is dynamic. It depends on two factors: learning (it
increases) and degradation (it decreases). According
to the available literature, the degradation of compe-
tency (respectively learning) depends on the number
of execution of the corresponding competency and the
period of interruption (respectively depends on the
number of execution and the learning ability of the
corresponding performer). As part of our approach,
the level of mastery is considered static but it is possi-
ble to take into account its dynamics in order to update
the repository of competency. To provide an efficient
access to relevant performer, we decide to present all
ICEIS 2018 - 20th International Conference on Enterprise Information Systems
678
the identified profiles under the Galois approach. It
provides a formal and efficient classification process
to discover and represent hierarchies of concepts. It is
the basis of what is also called ”formal concept anal-
ysis”, mainly used in data analysis and knowledge ex-
traction. We chose to group all the profiles in the same
repository. In such case, the framework will sup-
port presenting performers with their owned assessed
competencies. The various defined performer’s pro-
file, presented in the form of competency’s vector,
will be aggregated. And due to an incremental struc-
turing algorithm, 1) competencies trellis will be con-
tinuously updated, and 2) possible candidates (per-
formers) to execute the set of required competencies
will be extracted. The result of using the Galois lat-
tice is a list of qualified competencies corresponding
to the required competency as shown in the following
picture: The reasons that prompt us towards the use
Figure 4: Result of using Galois Lattice.
of Galois Lattice are:
It decomposes the context to characterize into
concepts (a set of performers with the same mas-
tered competencies). Indeed, every concept exists
in the lattice is a response to a query (search of
competencies). It gathers all actors mastering the
same competencies in a concept of Galois. It is
then easier to find for given competencies all cor-
responding performers.
It is a dynamic method for concepts classification
where new concepts can be added to the lattice
or existing relationships between objects and their
properties can be modified. The profile of each ac-
tor is dynamic on two factors learning and degra-
dation. Indeed, the mastery level of the compe-
tence C
j
by the actor a
i
may increase or decrease
or actor a
i
can master a new competency.
The relation in the Galois lattice classifies the
identified concepts (set of actors with set of
competences) in a decreasing or increasing way
(generalization or specialization). Indeed, while
browsing the lattice from bottom to top (inversely
from top to bottom), the number of the shared
competencies decreases and the number of actors
increases (conversely the number of competencies
increases and the number of actors decreases).
3.3.1 Basics of Galois Lattice
In this subsection, we briefly describe Galois lattices
(or concept lattices) and we give a basic overview of
the algebraic notions needed for our purposes. Ga-
lois lattices allow users to obtain every subset of in-
stances distinguishable according to the chosen at-
tributes. Further information, results and proofs may
be found in (M. Montjardet, 1970), (Birkhoff., 1973).
As input, we consider a non-binary relation between
a set of individuals and a set of attributes, in our case
a set of actors A and a set of keywords W of their ac-
quired competencies. A context is a triplet (A, C, R),
where R A x C. R (a, c) means that the actor ”a”
has the competency c with a mastery level equals to R.
Using the relation R; we can derive each performer’s
profile (a set of keywords referring to all his/her ac-
quired competencies). Similarly, the set of perform-
ers mastering the same competency can be identified.
The basic idea of using Galois lattices is the partition
of data in a set of basic classes which are clusters of
instances sharing the same basic type (in our work
competency). The identification of the actors is done
using the following formula:
g(C) =
{
a A s.a c C, R(a, c) α
}
(2)
Where α is the required mastery level to practice com-
petency C. This level as defined in the previous sec-
tion is between [0, 1]. Only actors, with acquired
mastery level above α, are selected. Actors, with
a mastery level below α, are considered as under-
competent and are not selected. To present the pro-
file of each actor, the set of his acquired competencies
with the corresponding mastery levels, we use the fol-
lowing expression:
f (A) =
{
c C s.a a A, α = minR(a, c)
}
(3)
4 FURTHER SUGGESTIONS FOR
OUR APPROACH
Selecting the best performer depends on various at-
tributes from several dimensions: social, temporal,
business etc To socialize our framework and to opti-
mize its behavior with a high degree of positively af-
fecting stakeholders’ satisfaction, we add some func-
tionality. When looking at the time dimension, re-
garding the fuzzy nature of the only considered at-
tribute ”availability” (of performer and resources), to
trace it a need of a simple concept and easy to grasp
is recommended. A simple technique already used
with social applications (Facebook, twitter, Tumblr,
Google Plus) is adapted. Presence is about being able
A New Approach for SBPM based on Competencies Management
679
to share one’s availability status (online or offline)
so everyone is aware of it and can act accordingly.
Traceability of the status makes day-to-day allocation
easier. The social dimension typically focuses on the
cohesion of the group and the closeness between the
stakeholder and the performer(s). While selecting the
relevant performer(s) two cases should be taken into
consideration: 1) all owned competencies from one
performer correspond to the required ones, in this case
this candidate is adequate to satisfy the goal. Another
case, one performer has less than the required com-
petencies. Thus, team could be constructed based on
”cohesion” criteria and in this case soft competency is
taken into consideration. Overall, closeness between
stakeholder and selected performer(s) offered the best
combination of speed and efficiency; in general, the
more closed the network, the more misbehavior will
be detected and a significant benefit on problem solv-
ing will be offered. Regarding the cost dimension,
the costs of executing a competency may be fixed
or depend on the type of used resources. In our re-
search, we are focused on cost of used resources to
practice the required competencies. Costs related to
human resources are out of the scope of our research.
To reduce the level of uncertainty prevalent in this
type of decision-making (selection of the relevant per-
former(s)) a multi-criteria approach will be adopted
for the purpose of this step. Subsequent to the given
brief description of the identified attributes, to rapidly
and dynamically take into consideration unplanned
participations the following requirements need to be
investigated:
R1: Identification and assessment of performer
owned competencies
R2: Presentation of each performer with his
owned assessed competencies
R3: Allowing goal and required competencies
correspondence
R4: Allowing required and owned competencies
match
R5: Selection of the relevant performer according
to context dependent criteria to make a personal-
ized service.
Each defined goal will provide the intention behind
the needed competencies to be performed. Once the
goal is defined and analyzed the corresponding re-
quired competencies will be identified. All these
identified requirements are classified into three main
steps as shown in Figure 6: Analysis, Identification
and then Selection. As a last step of our framework,
the process’s model will be adjusted according to the
identified competencies. As defined in our ontology,
Figure 5: Proposed framework.
the concept competency exhibits an intentionality for-
mulated by the goal (which characterizes the situation
of using the competency). Using each competency
goal and corresponding relations, processes could be
designed, modified and adjusted dynamically during
execution in order to include unplanned participa-
tions.
5 CONCLUSION
In this article we discussed the problem of traditional
BPM approaches. These approaches suffered from
several issues and a lot of works have been made
in order to support BPM and improve their agility.
Lastly, social software has been used to support the
different lifecycle steps leading to the emergence of
SBPM. Current SBPM approaches present their own
challenges and problems that first need to be over-
come. Our study shows that these approaches espe-
cially support the design phase of business processes
while the execution phase is seldom taken into ac-
count. In that way, knowledge intensive processes
still not supported. In our work, we aim to support
running processes which can be adapted during exe-
cution to include unplanned participants and complete
the work more effectively. We presented a framework
for SBPM using competencies management and so-
cial network features to support emergent processes
and find rapidly the relevant performers. Our frame-
work is mainly based on competency (required and
owned competency) and the context in which they can
be practiced. Effective and automated competency
management creates a real time and predictive inven-
tory of the capability of any workforce. It is more
than ever a primordial factor for many companies to
better assess their human capital, to plan the execution
of emerged tasks, to tackle highly innovative projects.
Performer’s selection requires a careful examination
of various attributes. Two aspects evaluate the benefit:
tangible and intangible. Tangible benefit includes en-
gendered cost; we are focused on material resources
ICEIS 2018 - 20th International Conference on Enterprise Information Systems
680
used to practice the required competencies. Intangible
benefits may include social relationship with stake-
holder, availability status of performers and required
resources, the cohesion of the selected performers if
there is not possible performer able to reach the goal.
As a future work, required techniques to support Eval-
uation and Filtering steps of our framework will be
defined and the proposed framework will be proto-
typed and evaluated.
ACKNOWLEDGEMENTS
The authors would like to thank the Region Auvergne
Rhone-Alpes for financial support of this research
work.
REFERENCES
AR., Meziani, R. M. R. M. D. A. A. F. N. S. (2010). Ag-
ilipo: Embedding social software features into busi-
ness process tools. In Proceedings of BPM 2009
International Workshops (pp. 219-230). Germany:
Springer Berlin Heidelberg.
Birkhoff., G. (1973). Lattice theory. In American Mathe-
matical Society Colloquium Publications.
Fettke, P., L. P. V. (2013). Organizational and technologi-
cal options for business process management from the
perspective of web 2.0. In Business and Information
Systems Engineering 2(1) 1528.
Fraternali, p., M. (2012). Combining social web and
bpm for improving enterprise performances. the
bpm4people approach to social bpm. In Proceed-
ings of the 21st international conference companion
on World Wide Web (pp. 223-226). New York: ACM
Publishers.
G., Dengler, F. J. B. K. R. N. S. P. M. S. M. S. R. S. B.
(2011). Key challenges for enabling agile bpm with
social software.
Gilbert (2010). The next decade of bpm.
Giuseppe Berio, Franois B. Vernadat, M. H. (2006). Analy-
sis and modeling of individual competencies: Toward
better management of human resources.
Gottanka, M. (2012). Modelasyougo: Design of sbpm pro-
cess models during execution time.
Granitzer M., Happ S., J. S. J. M. J. P. K. A. N. S. R. D. and
Schmidt R., S. E. (2010). Combining bpm and social
software: Contradiction or chance?
Hauder, M. (2014). Bridging the gap between social soft-
ware and business process management:. In A Re-
search Agenda.
Houy C., F. P. and Loos P., D. (2011). Collaborative busi-
ness process modeling with comomod. In Interna-
tional Workshops on Enabling Technologies.
HR-XML. Consortium competencies schema. In
http://www.hr-xml.org/.
Karakostas B., M. R. (2013). Goal-driven social business
process management. In science and information con-
ference. pages 894901. IEEE.
LeBoterf (2004). construire les comptences individuelles et
collectives : les rponses 90 questions. 3eme edition,
Editions dOrganisation.
M. Montjardet, M. B. (1970). Key challenges for enabling
agile bpm with social software.ordre et classification,
alg‘ebre et combinatoire 2. In Hachette Universite,.
Palmer, N. (2011). The role of trust and reputation in social
bpm.
Prilla, M. (2008). Semantic integration of process mod-
els into knowledge management: A social tagging ap-
proach. In Proceedings of 11th International Confer-
ence on Business Information Systems (pp.130 141).
Austria: Springer Berlin Heidelberg.
Rosemann M., Uthmann C., J. B. (2001). Guidelines of
business process modeling. In vol. 1806 of Lec-
ture Notes in Computer Science, Springer Verlag, pp.
3049.
Ryan., K. S. S. L. E. W. (2009). Business process manage-
ment (bpm) standards: a survey.
Schmidt, N. (2009). Bpm and social software.
Sinur, J. (2011). Social bpm: Design by doing. In Gartner.
SJanas (2008). Choosing the right method to assess and rate
competencies in your organization. In Aug 27th 2008,
Competency Management.
Vernadat, F. (1999). Techniques de modlisation en en-
treprise : application aux processus oprationnels. In
Economica.
Wafa Triaa, Lilia Gzara, H. V. (2017). Exploring the influ-
ence of social software on business process manage-
ment. In IFAC, Toulouse.
Weske (2012). Business process management - concepts,
languages, architectures.
Zarifian, P. (2001). Le modle de la comptence. ditions Li-
aisons, Paris.
A New Approach for SBPM based on Competencies Management
681