This article was downloaded by: [148.241.128.

12]
On: 08 November 2013, At: 07:10
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Enterprise Information Systems
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/teis20
An approach for formalising the supply
chain operations
Milan Zdravkovi
a
, Herv Panetto
b
, Miroslav Trajanovi
a
&
Alexis Aubry
b
a
Faculty of Mechanical Engineering , Laboratory for Intelligent
Production Systems, University of Nis , ul. Aleksandra Medvedeva
14, 18000, Nis Serbia
b
Research Centre for Automatic Control (CRAN UMR 7039) ,
Nancy-Universit , CNRS, Nancy, France
Published online: 04 Jul 2011.
To cite this article: Milan Zdravkovi , Herv Panetto , Miroslav Trajanovi & Alexis Aubry (2011) An
approach for formalising the supply chain operations, Enterprise Information Systems, 5:4, 401-421,
DOI: 10.1080/17517575.2011.593104
To link to this article: http://dx.doi.org/10.1080/17517575.2011.593104
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
Content) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

An approach for formalising the supply chain operations
Milan Zdravkovic
a
*, Herve Panetto
b
, Miroslav Trajanovic
a
and Alexis Aubry
b
a
Faculty of Mechanical Engineering, Laboratory for Intelligent Production Systems, University
of Nis, ul. Aleksandra Medvedeva 14, 18000 Nis Serbia;
b
Research Centre for Automatic
Control (CRAN UMR 7039), Nancy-Universite, CNRS, Nancy, France
(Received 13 January 2011; nal version received 29 May 2011)
Reference models play an important role in the knowledge management of the
various complex collaboration domains (such as supply chain networks).
However, they often show a lack of semantic precision and, they are
sometimes incomplete. In this article, we present an approach to overcome
semantic inconsistencies and incompleteness of the Supply Chain Operations
Reference (SCOR) model and hence improve its usefulness and expand the
application domain. First, we describe a literal web ontology language (OWL)
specication of SCOR concepts (and related tools) built with the intention to
preserve the original approach in the classication of process reference model
entities, and hence enable the eectiveness of usage in original contexts. Next,
we demonstrate the system for its exploitation, in specic tools for SCOR
framework browsing and rapid supply chain process conguration. Then, we
describe the SCOR-Full ontology, its relations with relevant domain
ontology and show how it can be exploited for improvement of SCOR
ontological framework competence. Finally, we elaborate the potential impact
of the presented approach, to interoperability of systems in supply chain
networks.
Keywords: enterprise information systems (EIS); supply chain; ontology;
inter-enterprise interoperation; inter-organisational enterprise systems
1. Introduction
In order to gain real benets from supply chain management (SCM), relevant
systems must span full horizontal organisation of enterprises and beyond its
customers and suppliers. For dealing with the complexity of such an environment,
reference models play an important role. Supply chain operations reference (SCOR)
(Stewart 1997) is a standard approach for analysis, design and implementation of ve
core processes in supply chains: plan, source, make, deliver and return. SCOR
denes a framework, which aims at integrating business processes, metrics, best
practices and technologies with the objective to improve collaboration between
partners. In this article, we present an approach to overcome semantic incon-
sistencies and incompleteness of the SCOR model, by using ontologies and thus
enabling eective knowledge management in supply chain network, facilitating
*Corresponding author. Email: [email protected]
Enterprise Information Systems
Vol. 5, No. 4, November 2011, 401421
ISSN 1751-7575 print/ISSN 1751-7583 online
2011 Taylor & Francis
DOI: 10.1080/17517575.2011.593104
http://www.informaworld.com
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

semantic interoperability of its participants and contributing to a further
improvement of the reference model.
Semantics analysis can be useful at dierent levels of supply chain networks.
First, the semantic representation of queries and information may improve the
relevance of the results and thus improve the quality of partners selection process. It
can be used instead of or in addition to usual requests representation. Second,
semantics can be used to represent participants, or groups of them, leading
participants to better know each other. Such information can be useful for routing
the requests to other participants in order to obtain the relevant answers within a
short time and with a low trac load. Third, this information can also be used to
organise the network so as to improve eciency. This is very important for the
open settings of the supply chain networks, where the traditional approaches to
business process management, which attempt to capture processes as monolithic
ows, have proven to be inadequate, resulting to moving research focus from
process to interaction modelling (Desai et al. 2006). All these research directions
have been followed by the researchers community, but more work is needed on
the interaction between all these elements and their impact on the eciency of the
global system. The use of domain ontology is already proven as benecial for
SCM, in the development of self-integrating SCM systems (Jones et al. 2001), or
facilitating collaboration of inter-enterprise design teams (Lin and Harding 2007),
simulation of supply chain network (Fayez et al. 2005), or online negotiations
(Pathak et al. 2000), development of approaches to semantic integration of
industrial information systems (Izza 2009), etc. There are also inuential eorts to
provide the exhaustive ontology-based semantic models for SCM (Ye et al. 2008),
organised in a modular way to support the reusability and maintainability of the
involved micro-theories. However, it is still very hard to integrate all these eorts
in a single formal theory which would enable a multiplication of achieved
impacts. Ontologies are formal models of collective knowledge. The consensus on
their structure is extremely hard to reach (Hepp 2007), particularly for a very
expressive (or richly axiomatised) ontology with large number of concepts. In
response to this problem, the approach of collaborative conceptualisation is
proposed and applied on the case of electronic product catalogues integration
(Guo 2009).
The development of reference models in dierent domains is a community
response to interoperability problems. They aim at the standardisation of domain
collaboration by providing categorisation schemes or taxonomies knowledge
structures, interpreted in organised way to be used as guidelines in the collaboration
of humans and systems.
Like most of the other reference models, SCOR is a form of knowledge
organisation system. The key feature of these models is subjectivity, or context-
dependent determination (Zdravkovic and Trajanovic 2009). They are not developed
with the intent to be semantically rich or precise, but to provide human-
understandable knowledge on the specic domain. However, their implicitness is
considered as an obstacle for a machine-based interpretation. SCOR lacks semantic
precision. SCORs Input/Output entity entails all resources exchanged between
process elements and actors physical or non-physical, states, events, documents, etc.
System entity includes information systems, modules, capabilities, approaches or
volume of use, integration levels, etc. Sometimes, reference models do not provide
enough expressivity for a complete formal model. In the case of SCOR, this is evident
402 M. Zdravkovic et al.
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

from the lack of relationships between metrics and systems, which could point out to
the source of information needed for performance measurement.
Industrial reference models are not formal models. They are descriptive
languages. They were created with an objective to aggregate entities for some
purpose, rather than to describe the nature of the entities. Hence, they are very hard
to maintain and to evolve in a consistent way. Dynamics and volatility of concepts
are much easier to manage if they are represented by a set of meaningful statements
or expressions, rather than by narrative descriptions. Also, higher levels of
expressivity and axiomatisation extend the opportunities for automated support.
However, industry acceptance of the reference models shows that practical benets
are more likely to be achieved when they are focused on highly contextualised
approaches where formalising domain knowledge is involved. Domain knowledge
evolves at highest rate at lower levels of abstraction, in domain community
interaction, where consensus is more likely to be reached.
Enterprise formal models usually stand at the opposite end of the abstraction
ladder. Although there are many enterprise modelling frameworks used in an
industry, this is not the case with enterprise ontologies. In this article, we present
three existing enterprise ontologies with dierent levels of expressivity and show
how one of these can be used for improving the competence of the semantically
enriched SCOR model. Although these ontologies had signicant impact to
enterprise modelling scientic community and some briey reported applications,
there are no strong evidences of their industry acceptance with their intended
purpose. While obvious lack of practical implications can be justied by the
technology-related implementation diculties, it is a fact that many of the
existing eorts in development of the common enterprise model are based on an
inspirational approach to enterprise modelling, implemented by the groups of
experts, not by community (Grubic and Fan 2010). Moreover, enterprise
ontologies are usually created from scratch. As a result, the development and,
especially, validation processes take a very long time to complete (Yan 2007), due
to a typically large amount of work needed for analysis and synthesis of the
domain knowledge, as well as achievement of the consensus on developed
conceptualisations within the relevant community. However, they do provide a
formal basis for semantic interoperability of systems. Thus, we consider the
coherence between creation, evolution and use of specic, highly contextualised
knowledge and development of formal expressive models as a very important
factor for the usefulness of domain ontology.
In this article, we present the approach to semantic enrichment of the SCOR
reference model, for the benet of enhanced interoperability of resulting ontologies,
on multiple layers of expressivity and abstraction. This approach extends already
veried model and maps its original elements to their semantics. It secures the
integrity of existing standard by using multiple levels of models expressivity.
Sections 2.1 and 2.2 of this article describe the ontology development issues, relevant
for systems interoperability, and some development decisions made in the
conceptualisation of the enterprise modelling domain and SCOR model formalisa-
tion. In Section 2.3, we describe how our approach layers the implicit knowledge,
application and domain representation models. This approach is demonstrated by
developing semantically aligned models (or model layers) of: (a) implicit knowledge
on the Supply Chain operations, namely, SCOR reference (SCOR Knowledge
Organisation System SCOR-KOS); (b) specic application (problem domain)
Enterprise Information Systems 403
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

ontology (SCOR-Cfg for process conguration) and (c) micro theory for supply
chain operations (SCOR-Full), which semantically enriches the SCOR reference
model. These models are described in detail in Section 3. The main intention behind the
decisions made in the implementation of this formalisation approach is to facilitate the
semantic interoperability of systems, relevant for supply chain networks management.
Hence, in Section 4, we elaborate on the notion of semantic interoperability of systems
and describe the potential impact to the architecture of the relevant enterprise
information systems (EIS) and some expected practical benets.
2. Theoretical background
2.1. Approaches to ontology development and interoperability
One of the main issues of ecient use of information systems is interoperability between
multiple representations of reality stored within the systems and/or between these
representations and systems users, namely, their perceptions of reality (Hepp 2008).
This issue is expected to be resolved by use of ontologies logical theories for formal,
explicit, and partial specication of conceptualisation (Guarino and Giaretta 1995).
Interoperability of information systems depends on the quality and mutual
consistency of the underlying ontologies. Dierences in conceptualisations (or
paradigmatic stands) to which ontologies are committed can cause semantic
mismatch, and hence have a negative eect on interoperability. Namely, in ontology
development, knowledge workers or domain experts can choose descriptive or
prescriptive approach, temporal or static representations, objective or subjective
stand, etc. Negative eects of the inconsistent conceptualisations can be reduced by
employing additional eorts in mapping, alignment, translation, transformation or
merging the corresponding ontologies (Noy 2004).
Using dierent levels of granularity is a common approach to engineering of an
ontological framework. It is applied in building upper ontologies, combining
continuant, enduring perspectives of reality and concepts extended through time
(Grenon and Smith 2004). Four-dimensional perspective on reality within a single
framework can be granularised further to strategic, operational and tactical sub-
perspectives. A variety of granularity levels in an ontological framework extends the
scope of inference. Also, it contributes to achievement of the semantic interoper-
ability between corresponding systems.
2.2. Enterprise and supply chain ontology
TOronto Virtual Enterprise (TOVE) project ontology (Fox et al. 1996) applied a
formal approach to enterprise modelling, representing activities, states and time
(top-level ontology), organisation, resources, products, costs and activity-based cost
management. The primary focus of the TOVE enterprise model has been in linking
the structure and behaviour through the concept of empowerment the right of an
organisation agent to perform status changing actions. TOVE aims at providing
sophisticated support to decision making by enabling the inference, not only on basis
of what is explicitly stated in the model, but also on the basis of what is implied by
the model. It introduces the notion of an ontology competence and corresponding
competency questions the ontology benchmarks, in the sense that the ontology is
necessary and sucient to represent the tasks specied by the competency questions
and their solution. Thus, TOVE organisational ontology denes three sets of
404 M. Zdravkovic et al.
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

competency questions: (1) structure; (2) behaviour and (3) authority, empowerment
and commitment competency.
The Enterprise Ontology (Uschold et al. 1998) is a result of the work in
development of a method and a computer tool set for enterprise modelling, based on a
formal model. Firstly, it aims at ensuring that all parties involved in the enterprise have
a shared understanding of the relevant aspects. Its role is to act as a communication
medium (primarily, but not exclusively between humans). Secondly, it is intended to
assist acquisition, representation and manipulation of enterprise knowledge. Thirdly, it
is intended to enable the interoperability, by using the ontology as an interchange
format for terms related to business enterprise. The main criteria for selection of the
terms were common usage (consensus on the meaning) and avoid of ambiguity. The
building blocks on the Enterprise Ontology are notions of an entity, a relationship, a
state of aairs and a role. These are the primitives that are used for expressing the
denitions in ontology and they are referred to as concepts of meta-ontology. They
are specialised to the concepts of four sections: (1) activity, plan, capability and
resource; (2) organisation; (3) strategy and (4) marketing.
IDEON
TM
ontology (Madni et al. 2001) is one of the candidate ontologies for
modelling collaborative distributed enterprises. It employs four complementary
perspectives to capture the key concepts and relationships of the enterprise. First, the
enterprise context view represents the interaction between an enterprise and its
external environment (partners, customers, competitors, etc.). It introduced the
interesting concept of sensors, employed for observing the environment, enabling
the enterprise to act upon the assessment of the observation, with a goal to achieve a
specic eect to its environment. Second, the enterprise organisational view captures
its inner structure, assigns goals and selects strategies for their achievement and acts
upon them, by employing relevant processes. Third, the process view represents
planningexecutioncontrol cycle. Fourth, the resource view elaborates on the
various types of resources that might be needed to execute a process.
Table 1 shows the comparative overview of the above enterprise conceptualisa-
tions, which illustrates used development approaches (purpose, modularisation
decisions) and implementation features (expressivity, notations and applications).
To the best of our knowledge, there were only a few attempts to ontologise
SCOR model. SCOR (Productivity appex website 2011 http://www.productivity
apex.com/products/scor.asp) is directed towards overcoming the limitations of the
basic SCOR model through an ontology-based tool. This tool will enable an
automated and comprehensive denition of the supply chain at four of its distinctive
levels: supply chain level, the enterprise level, the elements level and the interaction
level. It enables generation of generic explicit views and models that represent the
four levels. Unfortunately, SCOR is a proprietary product, and details on the
formalisation approach are not accessible. Lin (2008) extended the SCOR model by
generalising existing elements to 3A concepts (activity, artefact, actor-role), dened
in general process ontology (GPO). Also, she used the model for development of the
goal ontology, by modelling SCOR performance attributes as general soft goals and
deriving domain-specic goals from attributes metrics. Vegetti et al. (2005) used
SCOR to develop the SCOntology. They extended SCOR with the notions of an
enterprise model, with the aim to provide the foundations for the specication of
information logistics processes in extended supply chains associated to process
industries. Considering that enterprises are collaborating for a certain purpose,
supply chain may be considered as a kind of system-of-networked enterprises. Lu
Enterprise Information Systems 405
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

et al. (2010) extended the ONTOlogy for Product Data Management (ONTO-PDM)
Product Ontology developed by Tursi et al. (2009) with the SCOR model. The
resulting ONTO-SCOR ontology then denes a product-centric supply chain ontology
for facilitating the interoperation between all enterprise applications involved in an
extended supply chain. On basis of the analysis of the contribution of the SCOR model
to the alignment of business processes and information systems, Millet et al. (2009)
proposed the extended reference model, including the structure of information
exchanged between processes. This model is proposed in response to the identied
weaknesses of the current SCOR model and lack of important process dependencies.
In addition, there are many relevant papers with reported work on other
reference models formalisation, addressing the semantics of RosettaNet (Haller
et al. 2007), The United Nations Standard Products and Services Code (UNSPSC)
(Hepp 2006a, 2006b), The Automotive Industry Action Group (AIAG) and
Standards for Technology in Automotive Retail (STAR) (Anicic et al. 2005),
electronic data interchange (EDI) (Foxvog and Bussler 2006), etc. Presented results,
methods, tools and gained experiences were extremely useful in setting up and
implementing this approach.
2.3. Description of our approach
Our approach builds upon three of the ve general approaches to ontology design:
inspiration, induction, deduction, synthesis and collaboration (Holsapple and Joshi
2002). Induction is used in the phase of semantic enrichment of a reference model, by
improving the semantic precision of the categorisations. It is combined with
inspirational approach which is characterised by an individual (authors) viewpoint
about the used abstractions. Inspiration is also used for formalising problem solving
models application models, based on the design goals. Finally, synthesis is
Table 1. Overview of the enterprise ontologies.
The enterprise
ontology TOVE IDEON
Purpose
(motivation)
Enterprise modelling Enterprise modelling Collaborative enterprise
modelling
Domains
(modularisation
approach)
Activity, plan,
resource
Structure Interaction
Organisation
Strategy
Behaviour authority,
empowerment and
commitment
Organisation Process
Marketing Resource
Expressiveness Low High Low
Existing (known)
notations
Ontolingua Web ontology
language (OWL)
Unied modeling
language (UML),
knowledge interchange
format (KIF)
Known
applications
The enterprise toolset
(business process
models, agent-
based architecture
for integration)
Integrated supply
chain
management,
process integration
in enterprise
engineering
Process-centric crisis
action planning and
execution, integrated
product-process
development (IPPD)
406 M. Zdravkovic et al.
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

employed in mapping of a semantically enriched model with relevant
ontologies, with the aim to enable semantic interoperability and/or to extend an
inference scope.
In order to formalise the reference model, we propose the approach of using the
semantically aligned layers of a literal specication of a reference model, its semantic
enrichments and resulting domain ontology and application models, developed on
basis of the dierent design goals. Approach reects the practices from AI domain of
using the dierent granularities of domain knowledge in solving engineering
problems of dierent abstraction (Falkenhainer and Forbus 1991). The approach is
demonstrated, in the next section, on the example of SCOR model and its
application for generating process thread diagrams.
Figure 1 shows the framework for semantic enrichment of reference models, and
includes source data (reference models and existing domain ontologies), various
tools and resulting models: literal web ontology language (OWL) specication,
semantically enriched model and application ontologies.
As reference models are stored in number of dierent formats and representa-
tions, the use of import facilities in support to initial development and continuous
update of the OWL model is recommended. Some of the examples of the import
tools are EulerGUI
1
, a lightweight Integrated Development Environment (IDE) that
translates UML XMI format and XML schemas into N3 triples and Anzo for
Excel
2
, which extracts RDF data from Excel spreadsheets. Update of the models or
instantiation of relying concepts can be automated when the import tools and
respective application programming interfaces (API) are used for alignment of OWL
models, and native data formats of the applications which are using the reference
models. In case of SCOR, some of the examples of such applications are ARIS
EasySCOR by IDS
3
or e-SCOR by Gensym
4
, used for the benet of SCOR
implementation process.
Layering of application and domain representation models reects the paradigm
of separation of domain and task-solving knowledge (Gangemi 2005) and assume
their mutual independence (Guarino 1997). Thus, arbitrary design goals can be
dened, formalised to set of competency questions and used for development of a
task-solving application ontology. Although problem domain is restricted to a
Figure 1. Framework for semantic enrichment of reference models.
Enterprise Information Systems 407
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

supply chain context, level of its variety can be extended in process of synthesis,
namely, mapping of semantically enriched model with other relevant ontologies,
such as enterprise ontology, TOVE ontologies and others.
3. Semantic enrichment of SCOR model
For building the fully expressive SCM semantics, we start from the obvious point of
community consensus reference models, in specic SCOR. Because of the SCORs
weak semantics, in the rst step, we model it as a KOS, but we use semantic tools to
represent this model in a computerisable language web ontology language (OWL).
OWL (OWL 2 Web Ontology Language 2009) is a family of knowledge representation
languages, endorsed by the World Wide Web Consortium, which provides the syntax
for authoring and exchanging the ontologies among relevant tools and applications.
Figure 2 shows entities of SCOR-KOS OWL model and relationships between them.
The aim of the literal OWL specication is to preserve the classication approach
and to represent models concepts and properties, thus enabling the use of a resulting
SCOR-KOS model for the original purpose. This purpose can be formalised by the
competency questions, used for the validation of resulting model. Competency of a
SCOR-KOS OWL model is validated by using following questions: which process
elements constitute one SCOR process and in which order? What are the input and
output resources for the selected process element? What are the metrics and best
practices for the selected process element? Which systems can facilitate the
improvement of the selected process element and/or process category?
The actual order of process elements is determined by executing SPARQL
Protocol and RDF Query Language (SPARQL) queries against asserted precedes
(meaning direct precedence) triples. The denition of concurrency in a SCOR-KOS
OWL model is used only for the determination of ows branching, and hence it is
not semantically correct. Concurrency is inferred on basis of isConcurrentWith
relation and modelled by property chain axioms, on basis of asserted precedes and
inferred (inverse) succeeds property:
Figure 2. Entities of SCOR-KOS OWL model.
408 M. Zdravkovic et al.
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

precedes o succeeds 4 isConcurrentWith
, or by using RDQL query:
9precedes.(2 succeeds)
Flows of input and output resources are determined by SPARQL queries, which
return instances of SCOR_InputOutput concept from domain of asserted triples of
hasInput and hasOutput properties. The source of these properties is determined
from the domain of fedBy property, which is used to assert connections between
process elements from dierent process categories. Inference of systems which can
facilitate improvement of selected process elements (categories) is achieved by
implementing properties:
implements (SCOR_System, SCOR_BestPractice), and:
facilitates (SCOR_BestPractice, SCOR_ProcessElement),
as inverse to implementedBy and hasBestPractice, used for the assertion of
relationships between process elements, best practices and systems. The properties
above are dened as sub-properties of transitive property enable, hence, enabling
reasoning of relationships between SCOR_System and SCOR_ProcessElement.
By dening inverse property enabledBy, the inference on relationships between
systems and process elements (categories) becomes possible in the opposite direction.
Thus, it is possible to identify systems which can improve the performance of a
selected process element and/or category.
SCOR-KOS OWL is used for the development of the web application for
browsing and visualisation of the SCOR framework. Main features of the web
application are: display of the selected process category map, display of the input/
output resources (including sources/destinations), best practices and metrics for
selected process element and customisation of the display. Figure 3 shows the web
applications work area, with displayed output resources, best practices and metrics
for P4.04. Establish delivery plans process element of P4. Plan Deliver process
category.
3.1. SCOR-Cfg application ontology
For demonstrating the approach, a design goal has been set generation and
representation of a SCOR thread diagram. A thread diagram is a standard tool used
in implementation of a SCOR model. In this case, it is inferred as conguration of
source, make and deliver processes, on basis of asserted product topology,
participants and production strategies for each component. Dierent process
patterns (and roles) are inferred as a result of SPARQL queries executed against
SCOR-Cfg model in each of the three possible manufacturing strategies: made-to-
stock, made-to-order or engineered-to-order.
The approach is demonstrated on a simplied case of snow making facility
product engineering, which involves purchase of fan guns (from stock), hydraulic
equipment (by order) and sourcing engineering, and production of a pump house.
Figure 4 shows the basic interface for the denition of snow making facility product
topology and generated SCOR thread diagram.
Product conguration is asserted to application ontology: SCOR-Cfg
OWL model, consisting of the following concepts: SC_project, SC_product,
Enterprise Information Systems 409
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

SC_production_type, SC_process (with child concepts, corresponding to dierent
process types) and SC_participant. The generation of a SCOR thread diagram is
done by selecting (and rendering) participants of supply chain project, its products
(components) and, nally, processes, in exact order.
Main features of the web application are as follows: development of complex
thread diagrams, generation of process models and workows and generation of
implementation roadmap.
First, the above example shows only interactions and collaborations between
customer and its rst-tier suppliers. The number of visualised levels depends on the
submitted product topology: if the detailed product topology is entered, the full
supply chain would be represented, with the number of tiers corresponding to the
depth of a product topology. Also, the horizontal organisation of individual supply
chain actors can be represented in more detail, by inferring additional participants
for dierent manufacturing strategies: warehouses (D, S), plants (M) and head-
quarters (P).
Second, a SCOR thread diagram is not a process map. In fact, it is just a
representation of a supply chain conguration. The full process model can be
generated by adding new rules for conguration of the SCOR PLAN activities and
by exploiting alignment relations between the SCOR-KOS and SCOR-Cfg OWL
models.
Third, alignment relations between the SCOR-KOS and SCOR-Cfg OWL
models also provide opportunities for the generation of a detailed implementation
roadmap, consisting of proposed best practices, relevant systems (or their modules,
capabilities, intended use, etc.) for their execution, resource tracking (SCOR Inputs
and Outputs) and environment for measuring the performance of a supply chain, by
using the SCOR metrics.
Figure 3. P4. Plan Deliver process category.
410 M. Zdravkovic et al.
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

3.2. SCOR-Full ontology
SCOR-Full is a domain ontology a micro theory for representation and
management of knowledge of the supply chain operations. It formalises core
concepts of supply chain operations, embedded in SCOR model denitions, and is
developed by semantic analysis of SCOR Input/Output elements, identication of
core terms and their categorisation. It extends SCOR-SYSTEM ontology, which
formalises the SCOR System element. It is extended by SCOR-GOAL ontology,
which semantically maps its concepts to SCOR Performance Metrics element. These
ontologies are not in the scope of this article and will not be elaborated.
SCOR-Full ontology does not aim at formalising the supply chain, but only to
resolve semantic inconsistencies of a SCOR reference model. Thus, its scope is
strictly limited to using the common enterprise notions for expressing the existing
elements of SCOR model. Central notion of the SCOR-Full ontology (as it is the
case for SCOR model) is a generalisation of process, in the sense that it acts as the
main context for semantic denition of other concepts in the ontology.
Main concepts of the SCOR-Full ontology are: agent, course, resource item (and
its sub-concepts: information item, physical item, congured item and communicable
item), function, quality and setting. Figure 5 shows the main concepts of SCOR-Full
ontology and relationships between them.
Agent is the concept which describes an executive role and entails all entities
which perform individual or set of tasks within the supply network, classied with
the concepts of equipment, organisation, supply chain, supply chain network, facility
Figure 4. Web interface for denition of a product topology and generated SCOR thread
diagram.
Enterprise Information Systems 411
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

and information system. Although semantically described as roles, agents do not
have explicit denition of functions. Functionality is dened as a property of a
course, performed by an agent. Hence, agents are functional in a context of a course
they execute.
Course classies prescriptions of ordered sets of tasks: Activity, Process, Method,
Procedure, Strategy or Plan, at the same level of abstraction. The notion of course
generalises doable things with common properties of environment (enabling and
resulting states, constraints, requirements, etc.), quality (cost, duration, capacity,
performance, etc.) and organisation (agent and business function).
Setting concept provides the description of environment of a course. It aggregates
semantically dened features of the context in which course take place its
motivation, drivers and constraints. Thus, it classies rules, metrics, requirements,
constraints, objectives, goals or assumptions of a prescribed set of actions.
Quality is the general attribute of a course, agent or function which can be
perceived or measured, e.g. capability, capacity, availability, performance, cost or
time/location data.
Function concept entails elements of the horizontal business organisation, such
as stocking, shipping, control, sales, replenishment, return, delivery, disposition,
maintenance, production, etc.
Instead of representing process ows, SCOR-Full is used to model enabling and
caused states of the relevant activities. These states are represented by the concept of
congured item (Conf-Item), the range of the has-postcondition and has-precondi-
tion properties of Course and its sub-concept Activity.
A resource item is a general term which encloses communicated (Comm-Item,
e.g. Notication, Response, Request) and congured (Conf-Item, with dened state)
information items (Inf-Item), such as Order, Forecast, Report, Budget, etc. and
physical items (Phy-Item). Where information items are the attributes of a Quality
Figure 5. Main concepts of SCOR-full ontology and relationships between them.
412 M. Zdravkovic et al.
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

(of Function, Agent or a Course), their congurations are realisations of the rules,
metrics, requirements, constraints, goals or assumptions of a course.
Congured items model state semantics of the resource physical or information
item, the notions which are used to aggregate the atomic, exchangeable objects in
enterprise environment. Examples of information items are Order, Forecast, Budget,
Contract, Report, Proposal, Bill-Of-Material, etc. Their structure is not addressed by
SCOR-Full ontology from this perspective, these are the atomic concepts which
can be semantically dened when mapped to other enterprise ontologies. Physical
items are Product (MRO-Product, Defective-Product and Part) and Scrap.
Congured items are characterised by one or multiple states of information or a
physical item:
(Phy-Item(?x) _ Inf-Item(?x)) ^ has-state(?x,state(?y)) ) Conf-Item(?x)
Available states are identied in the analysis of SCOR model and include 25
possible attributes of the congured item, which can be associated to dierent
information and physical items. Some of the examples of the states are: adjusted,
approved, authorised, completed, delivered, installed, loaded, planned, released,
returned, updated, validated, etc.
Where Inf-Item denes the semantics of the relevant resource, Conf-Item
describes its dynamics. Note that SCOR-Full asserts the semantic relation (realises
(Agent, Conf-Item)) which can be used to infer which Agent is responsible for a
particular state of the resource, although this specic information cannot be
extracted from the original SCOR model. SCOR-Full will rely on the external
enterprise knowledge to ll this and other gaps.
For the expressive process model, it is crucial to dene how resources are
communicated among activities and their corresponding actors. This knowledge is
embedded (explicitly or implicitly) in original SCOR model (in natural language) and
is used by SCOR-Full ontology to formalise abstract communicated item (Comm-
Item) which aggregates specic concepts of Notice (or its child concept - Signal),
Request, Response and Receipt. SCOR model does not provide explicit information
about who communicates congured items, but this can be inferred by using external
knowledge when property chain of
performs(Agent, Course)o issue(Course, Comm-Item)
is exploited, where former relation is inferred on basis of the mappings with external
ontologies and latter from SCOR-KOS OWL. Necessary conditions for a Comm-
Item are that it is issued (requested, responded, notied or received) by a course and
that it communicates a congured item:
Course(?x)^Conf-Item(?y)^issue(?x,?z)^communicates(?z,?y) ) Comm-Item(?z)
More specic axioms are set for the sub-concepts of Comm-Item, by using the
sub-properties of issue (Course, Comm-Item) property, namely: issue-request
(Course, Request), issue-response (Course, Response), issue-notice (Course, Notice)
and issue-receipt (Course, Receipt).
Currently, SCOR-Full ontology has 212 concepts and 32 properties and is
semantically mapped to the SCOR Input/Output elements. In order to increase the
Enterprise Information Systems 413
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

exibility of semantics framework, semantic web rule language (SWRL) rules are
used for mapping the SCOR-Full concepts to SCOR-KOS OWL instances. For
example, all instances of the business-rule class from SCOR-Full ontology are the
same as SCOR Input/Output concept Business_Rules_For_Return_Processes, if
there exists a return process in SCOR-Full ontology which has a business rule from
above, as a setting:
business-rule(?x) ^ return-process(?y) ^ has-rule(?y, ?x) ) SameAs(?x, Busi-
ness_Rules_For_Return_Processes)
Semantic mappings between SCOR-Full and SCOR-KOS enable characterisa-
tion of supply chain operations managed by using SCOR-Full ontology, in context
of SCOR reference model. For example, based on the above SWRL implication, it
can be inferred that a business rule, which is asserted in SCOR-Full ontology as a
setting for an instance of the return process, is an output of the SCOR process
element ER.01 Manage Business Rules for Return Processes. In the opposite
direction, relevant inferences of SCOR-KOS OWL model can result with a formal
semantics of the selected SCOR element.
SCOR-Full ontology is expected to support knowledge management in supply
chain operations. It classies concepts and relevant data objects, to be used in
collaborative systems, such as Semantic information pool for manufacturing supply
networks (SIP4SUP) (Zdravkovic et al. 2010), which is currently in development. It
enables lookup of data objects, required for consistent and complete denition of
supply chain operations concepts. It provides a roadmap for implementation of
SCOR reference model. It does not improve the expressivity of SCOR, because it
only uses common enterprise notions and proposed generalisations to formalise core
concepts of supply chain operations, embedded in SCOR model denitions.
However, these generalisations enable alignment of SCOR-Full model with relevant
enterprise models, such as TOVE ontology, and thus exploitation of its knowledge
for improving the competence of SCOR. Last, and most important, SCOR-Full
ontology is expected to facilitate the semantic interoperability of systems, which is
relevant for supply chain networks management.
Where SCOR-KOS provides implicit semantics of the supply chain operations by
using a semantics representation language, SCOR-Full (and its corresponding
mappings with the domain ontologies and SCOR-KOS itself) makes this semantics
explicit. Objective conceptualisation and corresponding explicit representation of
domain knowledge is considered as a main condition for making the relevant systems
semantically interoperable. In the next section, we discuss on the semantic
interoperability in supply chain networks. We show how the presented approach
can be exploited to improve the expressivity of the underlying ontological framework
and describe some practical impact of the semantically interoperable systems to
SCM.
4. Ontologies and semantics issues in supply chain networks interoperability
Supply chain networks may be considered, to a certain extent, as systems-of-
networked systems. The System-of-Systems (SoS) paradigm is widely recognised
and has become quite studied since a decade (Sage and Cuppan 2001), as it has
potentially practical applicability in systems engineering. SoS-organised systems,
414 M. Zdravkovic et al.
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

such as supply chain networks, could make ecient use of resources from a variety
of domains. One of the main characteristics of SoS is what authors are calling the
connectivity property (Maier 1998). Connectivity refers to interoperability of the
many suppliers taking part in supply chain network.
Interoperability is typically dened as the ability of two or more systems or
components to exchange and use information (IEEE 1990). Integration is
generally considered to go beyond mere interoperability to involve some degree
of functional dependence (Panetto 2007). Integration is desirable within the
horizontal organisation of the single enterprise or, in some cases, between focal
partner and rst-tier (strategic) suppliers (for example, with third-party logistics
partners). However, in lower levels of supply chain, the tight couplings can
produce unacceptable outcomes, mostly related to decrease of exibility. The
main prerequisite for achievement of interoperability of the loosely coupled
systems is to maximise the amount of semantics which can be utilised and make
it increasingly explicit (Obrst 2003), and consequently, to make the systems
semantically interoperable.
While database interoperability has been widely studied by the research
community, it takes into account technological concerns. Interoperability in supply
chain is mainly related to semantics issues, where the semantics is discussed in
multiple levels of abstraction, as well as in multiple contexts. It needs to align strategic
views, expressed in enterprise architecture, with business process organisation and
control, and consequently, with IT architecture and infrastructure (Vernadat 2007).
Many researches are trying to demonstrate that semantic interoperability can be
enabled through setting up an Ontology. The use of ontology is required as it acts as a
conceptual model representing enterprise consensus semantics (Obrst et al. 2003). It
aims at reducing the semantics loss among heterogeneous information systems that
are sharing mostly common concepts from the same area of knowledge. Furthermore,
ontology provides a common understanding and a formal model of the basic entities,
properties and relationships for a given domain that are essential to overcome
semantic heterogeneity. Semantic interoperability ensures that the meaning of the
information that is exchanged is automatically interpreted by the receiver of a
message. In centralised systems, this property improves the relevance of query
answers. In distributed heterogeneous systems, it is compulsory to enable
autonomous heterogeneous sources understand each other to obtain relevant results
(Panetto and Molina 2008).
Many works rely on the assumption that a single ontology is shared by all the
participants of the system. However, in SoS with autonomous sub-systems, this
assumption is not realistic anymore. On the contrary, one has to consider that the
participants create their ontologies independently of each other. Thus, most often the
ontologies dier. Still, the distinctness of ontologies does not prejudice logical
inconsistency of their terms, especially if they focus on dierent contexts of the same
concepts. Namely, ontology is not a tool for checking correctness of reality, but for its
subjective or objective representation. To tackle this problem, research on ontology
matching proposes several techniques to dene correspondences between entities of
two ontologies. So, in some way, ontology matching highlights the shared parts of
two ontologies. Thus, it provides the basis for interoperability between heterogeneous
systems and by transitivity in the whole system (Panetto 2010). Typically,
correspondences between two interacting ontologies are expressed by logical
equivalences, subsumption or sameness relations, assertions of constraints, based
Enterprise Information Systems 415
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

on the object properties or identication of rules, with the form of logical implication
between the antecedent and consequent statements.
Also, meanings from ontologies, developed in isolation, can be reconstructed or
re-created by using contextualisation or logical theories, such as ontology of
descriptions and situations (DnS) (Gangemi et al. 2002), which enable the rst-order
manipulation of micro-theories and models, independently from an upper ontology.
4.1. Interoperability and expressivity of the formal model
SCOR shows lack of expressivity for a complete formal model. One of the evidences
is the lack of relationship between metrics and systems, which could point out to the
source of information needed for performance measurement. This is obvious
limitation of the reference model and it cannot be addressed in the process of
semantic enrichment, as this relationship does not exist.
However, semantically enriched model enables us to establish the references
between formalised systems, system capabilities, intended uses, etc., and goals,
mapped to the metrics of the SCOR model, by using the external knowledge,
formalised in various domain ontologies. Namely, if there exist systems S1 and S2,
driven by the ontologies O
1
and O
2
(external knowledge), and if there exist alignment
between these ontologies O
1
:O
2
, the competence of O
1
will be improved and S
1
will
be enabled to make more qualied conclusions about its domain of interest.
For example, in TOVE organisation ontology, the concept of Communication-
Link (cl) captures the notion of benevolent communication in which organisation
agents voluntarily provide information that they believe are relevant to other agents.
TOVE organisation ontology can be extended with a property chain axiom of the
new information-provided-by(inf,oa) relationship, established between the concepts
of Organisation-Agent (oa) and Information (i):
information-provided-by(inf,oa) : inverse(inverse(has-sending-agent(cl,oa))
o will-volunteer(cl,i))
Assertions of the above TOVE relationships can be exploited for inference of the
sources of information relevant for measuring the performance of the process
elements if the following assumptions hold true: (1) Organisation agent is an
abstraction of an information system concept; (2) the correspondences between
TOVE Information and SCOR-Full Inf-Item instances are established or inferred
and (3) SCOR-Full Inf-Item are congured (Conf-Item) and these congurations are
mapped to the corresponding goal concepts.
Alignment of SCOR-Full ontology with other relevant ontologies make all
the research eorts based on these ontologies, complementary with this one, thus,
improving the competence of the SCOR-Full ontology. For example, mapping of
Location instances to geographic information systems (GIS) ontologies can
provide routing services for the shipment companies. Mapping of Product
instances and corresponding identiers to UNSPSC or eClass ontologies can
enable customers to identify the suppliers of the substitutable or alternative parts
or assemblies. Mapping of Process elements to Partner Interface Process
instances in RosettaNet ontology can enable the collaboration between two
companies using dierent standards for modelling and tracking their supply chain
processes.
416 M. Zdravkovic et al.
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

In order to improve the expressivity of the SCOR-Full ontology, it is mapped to
OWL representations of TOVE ontologies (resource, organisation and underlying
activity-state-time ontologies). TOVE Resource ontology sets semantic relations
(and constraints - axioms) between the notions of resource and activity. These
relations enable the inference on the commitment of the resources to specic activity,
their consumption and availability at given time. Thus, we can exploit the above
mappings to improve the competence of SCOR-Full ontology and ask additional
questions about SCOR activities, such as: Which resources are committed
(or available for commitment) to a process element at given time? Is there an
alternative to an unavailable resource, to be used by a process element at a given
time? Or, more specic: Can the unplanned order for manufacturing of the 10
hydraulic pumps, to be delivered until September 2010, be accepted (in context of
available resources)?
In addition, mappings with TOVE Organisation ontology enable us to improve
structural and behavioural (in context of organisational goals) competence of the
SCOR-Full model. For example, answers to the following questions may become
available: Whose permission (if any) is needed in order to perform the specic task of
selected process element (activity)? Who has authority to verify the receipt of the
sourced part? Which communication link can be used to acquire specic
information? etc.
4.2. Semantic interoperability and EIS
The stack of the semantic technologies, consisting of informal or formal ontologies,
their representations, inference engines and semantic applications, provides the
means for development and implementation of a new layer of the enterprise systems
architecture. The main role of this layer is to make the implicit semantics of the
dierent existing enterprise systems (or underlying reference models) explicit, and
consequently, mutually correspondent. Thus, the layer will enable the semantic
interoperability of these systems and facilitate better integration of the hetero-
geneous environments, such as supply chain networks.
In this scenario, EIS will be represented in the semantic layer by local ontologies
semantically weak representations (OWL models) of the implicit knowledge related to
the enterprise, and typically stored in relational databases of the relevant systems and
in other data sources. Semantic matching techniques and tools can facilitate the
contextualisation and explicitation of the individual representations, by helping to
establish the correspondences between these representations and relevant formal
micro-theories, such as SCOR-Full. Consequently, semantic mappings between
SCOR-Full notions and other domain and problem ontologies can be exploited for
applying an integrated approach to solving some of the supply chain networks issues.
For example, the partner selection problem can be associated with the denition
of the individual semantic query which expresses the sucient and necessary
conditions, regarding the capability, capacity, cost, availability, etc. (SCOR-Full
notion of quality) of a specic resource or an agent (among resources and agents of
the whole network). Mappings between these notions, used in a query and dening
correspondences between concepts in the dierent local ontologies, expressed as
logical functions, ensure that the single query is interpreted correctly in each of the
network partners systems and corresponding data storage facilities. Thus, it
becomes possible to use a single query (expressed in a formal semantics) to explore
Enterprise Information Systems 417
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

the whole supply chain network, despite the heterogeneity of used systems and their
data sources (syntax, data modelling patterns, etc.).
Collaborative process management can be facilitated by monitoring the state
(congurations) of the resource items in the semantic layer (by using a software
agent), and triggering appropriate actions (e.g. initiating SCOR process elements, or
equivalently, launching the process activities) when desired congurations are
established. Hence, desired congurations of the resource items, whose parameters
are stored as semantic annotations of the process models (generated by the process
modelling tool, such as the one described in Section 3.1), are continuously compared
with the specic entities of the relevant local ontologies, and logically related with
those items. Once they become logically same, the software agent would assert a new
individual of the Activity type, assign an agent to this individual and set other
necessary properties. This change will also be propagated backwards, by assertion of
the logically equivalent concepts of corresponding local ontologies, and conse-
quently, update of the relevant database(s). Thus, appropriate enterprise informa-
tion system(s) will be aected by automatic insertion of the work order, web service
invocation, issuing of the request for approval (authorisation) or similar action.
5. Conclusions and future work
Although ontology matching, contextualisation and other semantic web techniques
provide a basis for interoperability, the challenge is still to dene a whole semantic
infrastructure in which supply chain participants search for information is both
relevant and ecient. In response to this challenge, this approach proposes the use of
sets of semantically aligned models, on dierent levels of expressivity, consisting of
(SCOR-KOS), helper contextual models (SCOR-Cfg) and domain ontology (SCOR-
Full), and thus potentially improving the relevance of ontology matching and
facilitating contextualisation.
The proposed approach relies on and builds upon widely accepted industry
practice, represented in its native format (SCOR-KOS). The native representation is
expected to attract attention and gain understanding of SCM experts community
and, hence, to facilitate the transition towards using more sophisticated, knowledge-
based tools in the domain. Its mapping and alignment with higher-level ontologies
(such as SCOR-Full, TOVE, and others) make the implicit, human-understandable
knowledge of the supply chain operations correspondent with the explicit, machine-
based representations of the relevant models and, thus, corresponding systems. It
enables or will enable a structured support in other relevant processes, such as
resource management, partners selection, performance tracking, exceptions hand-
ling, etc. Also, it is expected to contribute to further development and/or renement
of the SCOR reference model.
In the development of a micro-theory for supply chain operations, a bottom-up
approach is applied, where existing, implicit knowledge is semantically analysed in
order to identify the specic concepts. Ontological decisions are made in the
categorisation of these concepts, when common general properties are recognised
and used to aggregate the concepts into general notions, such as agent, course,
setting, quality, function and resource items. This type of approach signicantly
reduces the development time, as it builds upon the already existing consensus of the
domain experts, transposed into the SCOR reference model. Thus, the problem of
validation of the resulting micro-theory (SCOR-Full) can be actually reduced to
418 M. Zdravkovic et al.
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

checking the correctness of the logical relations made between this theorys concepts
and axioms and the implicit SCORs knowledge representation (SCOR-KOS).
The main potential benet of the SCOR-Full and associated models, including
SCOR-KOS, application ontologies and the set of mappings with relevant enterprise
ontologies is the facilitation of the semantic interoperability of systems in the supply
chain networks, in a way described in Section 4 of this article. Multiple levels of
abstraction and modular approach to the formalisation of dierent relevant sub-
domains and problems, applied in the development of the models are expected to
improve the relevance of semantic matching results. However, there are still no
evidences for this assumption. In ongoing and future eorts, we are committed to
introducing the individual realities of the actual enterprises, namely corresponding
EIS, into the existing formal framework. The objective then is to show that this
framework represents common reality, in context of enterprises participation in a
joint endeavour, such as the supply chain network. These eorts include, but are not
restricted to: generating local ontologies from the relational databases of the
enterprise systems; their semantic enactment; developing methods and measures for
evaluation/assessment of the semantic interoperability and development of a test bed
for scenarios, briey described in Section 4.2.
Notes
1. http://eulergui.sourceforge.net
2. http://www.cambridgesemantics.com/semanticexchange
3. http://www.ids-scheer.com/ru/ARIS/ARIS_Reference_Models/ARIS_EasySCOR/
115741.html
4. http://www.gensym.com
References
Anicic , N., Ivezic , N., and Jones, A., 2005. An architecture for semantic enterprise application
integration standards. In: Proceedings of the rst international conference on the
interoperability of enterprise software and applications (I-ESA05), 2325 February,
Geneva, Switzerland.
Desai, N., et al., 2006. OWL-P: A methodology for business process development. Lecture
Notes in Computer Science, 3529, 7994.
Falkenhainer, B. and Forbus, K.D., 1991. Compositional modeling: nding the right model
for the job. Articial Intelligence, 51 (13), 95143.
Fayez, M., Rabelo, L., and Mollaghasemi, M., 2005. Ontologies for supply chain simulation
modelling. In: Proceedings of the 37th winter simulation conference, 47 December 2005.
Orlando, USA.
Fox, M.S., Barbuceanu, M., and Gruninger, M., 1996. An organization ontology for
enterprise modelling: preliminary concepts for linking structure and behaviour. Computers
in Industry, 29 (12), 123134.
Foxvog, D. and Bussler, C., 2006. Ontologizing EDI semantics. In: Proceedings of the
workshop on ontologising industrial standards, Tucson, AZ, USA. New York: Springer,
301311.
Gangemi, A., 2005. Ontology design patterns for semantic web content. Lecture Notes in
Computer Science, 3729, 262276.
Gangemi, A., et al., 2002. Sweetening ontologies with DOLCE. Knowledge engineering and
knowledge management: ontologies and the semantic web. Lecture Notes in Computer
Science, 2473, 223233.
Grenon, P. and Smith, B., 2004. SNAP and SPAN: towards dynamic spatial ontology. Spatial
Cognition and Computation, 4 (1), 69103.
Grubic, T. and Fan, I.-S., 2010. Supply chain ontology: review, analysis and synthesis.
Computers in Industry, 61 (8), 776786.
Enterprise Information Systems 419
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

Guarino, N., 1997. Understanding, building and using ontologies. International Journal of
Human-Computer Interaction, 46 (23), 293310.
Guarino, N. and Giaretta, P., 1995. Ontologies and knowledge bases towards a
terminological clarication. In: N. Mars, ed. Towards very large knowledge bases:
knowledge building and knowledge sharing. Amsterdam: IOS Press, 2532.
Guo, J., 2009. Collaborative conceptualisation: towards a conceptual foundation of
interoperable electronic product catalogue system design. Enterprise Information Systems,
3 (1), 5994.
Haller, A., Gontarczyk, J., and Kotinurmi, P., 2007. Towards a complete SCM ontology the
case of ontologising RosettaNet. In: Proceedings of the 2008 ACM symposium on applied
computing (SAC08), 1620 March 2007. Fortaleza, Ceara, Brazil.
Hepp, M., 2006a. Products and services ontologies: a methodology for deriving OWL
ontologies from industrial categorization standards. International Journal on Semantic
Web and Information Systems, 2 (1), 7299.
Hepp, M., 2006b. The true complexity of product representation in the semantic web. In:
Proceedings of 14th European conference on information system (ECIS 2006), 1214 June
2006. Gothenburg, Sweden.
Hepp, M., 2007. Possible ontologies: how reality constraints the development of relevant
ontologies. IEEE Internet Computing, 11 (7), 9096.
Hepp, M., 2008. Ontologies: state of the art, business potential and grand challenges. In: M.
Hepp, P. De Leenheer, A. De Moor and Y. Sure, eds. Ontology management semantic
web, semantic web services and business applications. Berlin/Heidelberg: Springer Verlag.
Holsapple, C.W. and Joshi, K.D., 2002. A collaborative approach to ontology design.
Communications of the ACM, 45 (2), 4247.
IEEE (Institute of Electrical and Electronics Engineers), 1990. Standard computer dictionary
a compilation of IEEE standard computer glossaries. New York: Inst of Elect & Electronic.
Izza, S., 2009. Integration of industrial information systems: from syntactic to semantic
integration approaches. Enterprise Information Systems, 3 (1), 157.
Jones, A., Ivezic, N., and Gruninger, M., 2001. Toward self-integrating software applications
for supply chain management. Information Systems Frontiers, 3 (4), 403412.
Lin, Y., 2008. Semantic annotation for process models. Thesis (PhD). Norwegian University of
Science and Technology.
Lin, H.K. and Harding, J.A., 2007. A manufacturing system engineering ontology model on the
semantic web for inter-enterprise collaboration. Computers in Industry, 58 (5), 428437.
Lu, Y., Panetto, H., and Gu, X., 2010. Ontology approach for the interoperability of
networked enterprises in supply chain environment. On the move to meaningful internet
systems: OTM 2010 workshops. In: R. Meersman et al., eds. 5th IFAC/IFIP Workshop on
Enterprise Integration, Interoperability and Networking (EI2N2010), 2728 October,
Hernissou, Crete, Greece: Springer Verlag, 229238, ISBN: 978-3-540-88874-1.
Madni, A.M., Lin, W., and Madni, C.C., 2001. IDEONTM: an extensible ontology for
designing, integrating and managing collaborative distributed enterprises. Systems
Engineering, 4 (1), 3548.
Maier, M.W., 1998. Architecting principles for systems-of-system. Systems Engineering, 1 (4),
267284.
Millet, P.A., Schmitt, P., and Botta-Genoulaz, V., 2009. The SCOR model for the alignment
of business processes and information systems. Enterprise Information Systems, 3 (4),
393407.
Noy N.F., 2004. Semantic integration: a survey of ontology-based approaches. ACM
SIGMOD Record Special Section on Semantic Integration, 33(4), 6570.
Obrst, L., 2003. Ontologies for semantically interoperable systems. In: Proceedings of the 12th
international conference on information and knowledge management, 38 November 2003.
New Orleans, USA.
Obrst, L., Liu, H., and Way, R., 2003. Ontologies for corporate web applications. AI
Magazine, Fall, 4962.
OWL 2 Web Ontology Language, 2009. Document overview. W3C recommendation [online].
Available from: http://www.w3.org/TR/owl2-overview/ [Accessed 27 October 2009].
Panetto, H., 2007. Towards a classication framework for interoperability of enterprise
applications. International Journal of CIM, 20 (8), 727740.
420 M. Zdravkovic et al.
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3

Panetto, H., ed., 2010. Special issue on integration and information in networked enterprises,
Editorial. Computers in Industry, 61 (2), 9798.
Panetto, H. and Molina, A., 2008. Enterprise integration and interoperability in manufactur-
ing systems: trends and issues. In: H. Panetto and A. Molina, eds. Special issue on
enterprise integration and interoperability in manufacturing systems, Computers in
Industry, 641646.
Pathak, S.D., Nordstrom, G., and Kurokawa, S., 2000. Modelling of supply chain: a multi-
agent approach. In: Proceedings of the IEEE international conference on systems, man, and
cybernetics. 811 October 2000, Nashville, TN, USA.
Productivity appex website., 2011. Available from: http://www.productivityapex.com/
products/scor.asp [Accessed 11 January 2011].
Sage, A.P. and Cuppan, D.C., 2001. On the systems engineering and management of systems-
of-systems and federations of systems. Information, Knowledge, Systems Management, 2
(4), 325345.
Stewart, G., 1997. Supply-chain operations reference model (SCOR): the rst cross-industry
framework for integrated supply-chain management. Logistics Information Management,
10 (2), 6267.
Tursi, A., et al., 2009. Ontological approach for products-centric information system
interoperability in networked manufacturing enterprises. IFAC Annual Reviews in Control,
33 (2), 238245.
Uschold, M., et al., 1998. The enterprise ontology. Knowledge Engineering Review, 13 (1), 31
89.
Vegetti, M., et al., 2005. Towards a supply chain ontology of information logistics within
process industry environments. In: Proceedings of the fourth Mercosur Congress on
process systems engineering, 1418 August 2005. Rio de Janeiro, Brazil.
Vernadat, F.B., 2007. Interoperable enterprise systems: principles, concepts, and methods.
Annual Reviews in Control, 31 (1), 137145.
Yan, T., 2007. Practical issues in ontology engineering. In: H.R. Arabnia, M.Q. Yang and J.Y.
Yang, eds. Proceedings of the 2007 international conference on articial intelligence, ICAI
2007, Volume II, 2528 June 2007. Las Vegas, Nevada, USA: CSREA Press, 730736.
Ye, Y., et al., 2008. Ontology-based semantic models for supply chain management.
International Journal of Advanced Manufacturing Technology, 37 (1112), 12501260.
Zdravkovic , M. and Trajanovic , M., 2009. Integrated product ontologies for inter-
organizational networks. Computer Science and Information Systems, 6 (2), 2946.
Zdravkovic , M., Panetto, H., and Trajanovic , M., 2010. Concept of semantic information
pool for manufacturing supply networks. International Journal of Total Quality Manage-
ment and Excellence, 37 (3), 6974.
Enterprise Information Systems 421
D
o
w
n
l
o
a
d
e
d

b
y

[
1
4
8
.
2
4
1
.
1
2
8
.
1
2
]

a
t

0
7
:
1
0

0
8

N
o
v
e
m
b
e
r

2
0
1
3