CONCEPTUALISATION APPROACH FOR COOPERATIVE

INFORMATION SYSTEMS INTEROPERABILITY

Mario Lezoche, Hervé Panetto and Alexis Aubry

Research Centre for Automatic Control (CRAN), Nancy-University, CNRS, Campus Scientifique

Faculté des Sciences et Technologies, BP 70239, 54506 Vandoeuvre-lès-Nancy Cedex, France

Keywords: Conceptual modelling, Cooperative information systems, Semantic interoperability, Data model

conceptualisation.

Abstract: In order to increase enterprise performance, economics paradigms focus, now more than ever, on how to

better manage information. The modern architecture of information systems is based on distributed

networks with a grand challenge representing and sharing knowledge managed by those ISs. One of the

main issues in making such heterogeneous Cooperative Information Systems (CIS) working together is to

remove semantics interoperability barriers. This paper firstly analyses interoperability issues between CISs

and then proposes patterns for data models conceptualisation for knowledge explicitation, based on expert

knowledge injection rules and a fact-oriented approach. A case study is proposed related to a work order

process in Sage X3, an Enterprise Resource Planning application.

1 INTRODUCTION

The actual archetype for the Information Systems

(ISs) involves large number of ISs distributed over

large, complex computer/ communication networks.

Such cooperative information systems (CIS) have

access to large amount of information and have to

interoperate to achieve their purpose. The

cooperative information systems architects and

developers have to face a hard problem:

interoperability.

Interoperability can be defined as the ability for

two or more systems to share, to understand and to

consume information (IEEE, 1990). Some work

(Chen et al., 2006) in the INTEROP NoE project has

identified three different levels of barriers for

interoperability: technical, conceptual and

organisational. Organisational barriers are still an

important issue but out of scope of this paper. The

technological barriers are strongly studied by

researchers in computer science and are generally

based on models transformation (Frankel, 2003).

Our research focuses on the conceptual level of

interoperability that is the ability to understand the

exchanged information. A concept is a cognition

unit of meaning (Vyvyan, 2006), an abstract idea, a

mental symbol. It is created through the action of

conceptualisation, that is, a general and abstract

mental representation of an object. During the

history of human effort to model knowledge,

different conceptualisation approaches regarding

different application domains were developed

(Aspray, 1985).

This paper is dealing with a first step from a

more general work focusing on the study of the

semantic loss during the exchange of information

representing business concepts. Quantifying the

semantic gap between interoperating ISs implies

enacting their semantics through their normalized

conceptual models. Indeed, in this context, the

starting point for semantics interoperability is related

to models conceptualisation.

We will present a conceptualisation approach to

make explicit the finest-grained semantics embedded

into conceptual models for finally enabling two

different information systems seamlessly

interoperating.

Next section presents the general context of our

work. Then, the following section details the

fundamental pillars of our conceptualisation process.

Then, we will propose a knowledge explicitation

process starting from an implemented relational

model to a fact-oriented conceptual one. This

process allows us emphasizing the finest-grained

101

Lezoche M., Panetto H. and Aubry A..

CONCEPTUALISATION APPROACH FOR COOPERATIVE INFORMATION SYSTEMS INTEROPERABILITY.

DOI: 10.5220/0003508401010110

In Proceedings of the 13th International Conference on Enterprise Information Systems (ICEIS-2011), pages 101-110

ISBN: 978-989-8425-53-9

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

semantics that must be enacted to study semantics

interoperability between collaborating ISs.

Finally, to validate our proposal, a practical

case study is presented based on an Enterprise

Resource Planning application involved in a B2M

(Business to Manufacturing) interoperation process.

2 COOPERATIVE

INFORMATION SYSTEMS

Information Systems are systems whose activities

are devoted to capture and to store data, to process

them and produce knowledge, used by any

stakeholders within an enterprise or among different

networked enterprises. It is commonly agreed that

Cooperative Information Systems provide a

backbone for the Integrated Information

Infrastructure (Sheth, 1998). Fully understanding

and exploiting the advances in computing is the only

way to encompass the complexity of constructing

and maintaining such systems.

Although the progress made in information

technology considerably improved the efficiency of

applications development, its drawbacks and

limitations are obvious and serious. In fact, the

application models involved in a single application

are numerous and different, each coping only with

particular and partial aspects of the overall task.

Moreover, the components technologies are

heterogeneous, platform- and machine-dependant.

The above-mentioned limitations and barriers

measurably hinder the development and the

maintenance process.

There is a growing demand to integrate such

systems tightly with organizational work so that

these information systems can be directly and

immediately used by the business activity.

Here, the need of interoperation clearly appears.

In fact, to achieve the purpose of the cooperation

between the different Information Systems,

information must be physically exchanged (technical

interoperability), must be understood (conceptual

interoperability) and must be used for the purpose

that they have been produced (conceptual and

organisational interoperability). When trying to

assess the understanding of an expression coming

from a system to another system, there are several

possible levels of interoperability (Euzenat, 2001):

• encoding: being able to segment the

representation in characters;

• lexical: being able to segment the representation

in words (or symbols);

• syntactic: being able to structure the

representation in structured sentences (or

formulas or assertions);

• semantic: being able to construct the

propositional meaning of the representation;

• semiotic: being able to construct the pragmatic

meaning of the representation (or its meaning in

context).

This tiered structure is arguable in general; it is

not as strict as it seems. It makes sense because each

level cannot be achieved if the previous levels have

not been completed (Euzenat, 2001).

The encoding, lexical and syntactic levels are the

most effective solutions for removing technical

barriers for interoperability, but not sufficient, to

achieve a practical interoperability between

computerised systems. Dealing with trying to enable

a seamless data and model exchange at the semantic

level is still a big issue that needs conceptual

representation of the intended exchanged

information and the definition of the pragmatic

meaning of that exchanged information in the

context of the source and destination applications.

Different cooperation types have been

investigated in ISO 14528 (ISO, 1999). In fact, this

standard considers that models could be related in

three ways:

(1) integration when there exists a standard or

pivotal format to represent these models;

(2) unification when there exists a common

meta-level structure establishing semantic

equivalence between these models; and

(3) federation when each model exists per se, but

mapping between concepts could be done at

an ontology level to formalise the

interoperability semantics.

Integration is generally considered to go beyond

mere interoperability to involve some degree of

functional dependence (Panetto, 2007). Classifying

interoperability problems (Tursi, et al. 2009) may

help in understanding the degree of development

needed to solve, at least partially, these problems but

conceptualisation and semantics extraction is still an

important issue because of the various contextual

understanding of tacit knowledge embedded into

those applications. The main prerequisite for

achievement of interoperability of information

systems is to maximise the amount of semantics

which can be used and make it increasingly explicit

(Obrst, 2003), and consequently, to make the

systems semantically interoperable. To highlight this

issue, the paper is based on a referenced scenario

involving enterprise systems applications.

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

102

Most of reverse engineering approaches

(Fonkam, 1992) (Chiang, 1994) return the

information structure but present a model with tacit

semantics. The ADM (Architecture-Driven

Modernization) initiative (OMG, 2003) from OMG

(Bézivin et al., 2005) is tackling this problem by

promoting a common Knowledge Discovery Meta-

model to facilitate discovering tacit knowledge

embedded inside existing software. In our scenario,

those applications are still implemented and running

using databases. We can extract, from them, by

using reverse engineering approaches, some

knowledge in a form of a conceptual model. We

have then to enrich that model with enterprise

applications best practices (knowledge coming from

users). Finally, we make explicit all disclosed

knowledge hidden in the resulting model.

3 OUR APPROACH FOR

SEMANTICS ENACTMENT IN

CONCEPTUAL MODELS

In order to cooperate, two (or more) Information

Systems have to interoperate. As previously

discussed, we focus our interest on the conceptual

level of interoperability letting different information

systems to share and use knowledge models that

they represent. Our principal issues are, therefore,

first to understand the conceptual relationships

between those models in the context of their use and

secondly how, through conceptualisation, to unhide

the tacit knowledge buried inside them. A usual

approach for making explicit the tacit knowledge,

concealed in attributes and classes, is the

relationships-oriented perspective composed of a set

of transformation rules. In that transformation

method, an attribute a

of type T

pertaining to class

is modelled as a relationship between the class C

and a standard type T

. This approach does not

resolve entirely the semantics elicitation problem

because it focuses its point of view on the values

instead of on the concepts. The attribute semantics is

somewhat yet hidden in the relationship just created.

In literature, (Meersman, 2003) presented the

definition of two different objects types, a lexical

object (LOT), a term, is an object in a certain reality

that can be written down. LOTs always consist of

letters, numbers, symbols or other characters. They

can be used as names for or references to other

objects. A non-lexical object (NOLOT), a concept, is

an object in a certain reality that cannot be written

down. Non-lexical objects must be named by lexical

objects or referred to by means of lexical objects.

Applying these definitions, we can flatten the nested

knowledge embedded in a model to simplify

semantic enactment resulting from a set of

modelling transformations. Our contribution is to

have at our disposal an approach letting us to

fragment knowledge through the transformation of

attributes into entities and relationships, and thus to

emphasize some fine-grained knowledge atoms. In

the proposed approach, that is the first part (Figure

1) of our general methodology, the starting point can

be various: an application, a data model, a logical

view, a model. We have already mentioned that

there are several reverse engineering methods, such

as in (Fonkam, 1992) and in (Chiang, 1994), through

which a model from the application or schema level

can be derived (Step 1). Then, the resulted initial

model is enriched and corrected through an Expert

Knowledge Injection step (Step 2). In fact, the

model is examined with the help of a domain expert

or an end-user, who are the most qualified persons to

describe the context of the peculiar domain and to

put in evidence the contextual knowledge.

Figure 1: Conceptualisation approach.

CONCEPTUALISATION APPROACH FOR COOPERATIVE INFORMATION SYSTEMS INTEROPERABILITY

103

According to the enterprise best practices and its

data, they would clean and better organise the

knowledge represented in the derived model.

However, the obtained initial conceptual model, in

the form of a UML class diagram, has yet a major

limit. In fact, its semantics is in a tacit form because

all the attributes are buried inside single classes and

it is then difficult to make their semantics explicit.

Thus, the next step of our approach (Step 3) is a

Fact-Oriented Transformation (Halpin, 1991)

through the application of a set of patterns rules for

transforming the enriched conceptual model to a

fact-oriented model (FOM) with its semantics

completely displayed. The consequence is that all

the classes and their attributes are transformed into

respectively LOTs and NOLOTs objects. The

resulting fact-oriented model, displaying the finest-

grained semantic atoms, is then used as an input for

the second part of our methodology for semantic

loss evaluation (not presented in this paper).

In the following sub-sections, we will discuss, in

detail, the proposed 3 steps.

3.1 Step 1: Reverse Engineering

Conceptualisation is a decision process (Guarino,

1998), a view, in which studied part of reality

knowledge, usually in an implicit and complex form,

is reorganised in different aggregates usually simpler

to be represented.

According to (Engelbart, 1962), developing

conceptual models means specifying the essential

objects or components of the system to be studied,

the relationships of the objects that are recognised

and what kinds of changes in the objects or their

relationships affect the functioning of the system and

in which ways.

Conceptual models range in type from the more

precise, such as the mental image of a familiar

physical object, to the abstractness of mathematical

models that do not appear to the mind as an image.

Conceptual models also range in terms of the scope

of the subject matter that they are taken to represent.

The variety and scope of conceptual models is due to

the variety of purposes that people had while using

them.

Conceptualisation approaches are numerous and

have been developed in different knowledge

domains (LaOnsgri, 2009).

Our scenario assumes that we start from

enterprise application database. So, the first studied

approach is the Reverse engineering. It is, in

database (DB) community, an approach to extract

the domain semantics from the existing database

structures. Typically, it concerns making the reverse

transformation from logical to conceptual schema. In

(Fonkam, 1992), the authors propose a general

algorithm based on several old attempts to make

explicit the logical structure buried into DB

schemas, application programs and in the minds of

designers and developers. (Chiang, 1994) presents a

methodology for extracting an extended Entity-

Relationship model from a relational database,

through a combination of data schema and data

instance analysis. In our study we will consider at

profit the reverse engineering experiences developed

in the past. These methods are, by now, acquired by

the software industry that produces countless tools.

We choose MEGA Suite (http://www.mega.com), a

modelling management environment to transform

relational models into conceptual ones.

3.2 Step 2: Expert Knowledge Injection

After the reverse engineering process has created a

conceptual model, the current step concerns

enriching it by injecting the enterprise knowledge,

expressed by users’ best practices or experts. These

stakeholders know the domain peculiarities and they

are capable to embed specific constraints into the

new conceptual model. The first stage is the

renaming process. Usually the database tables, and

the derived concepts, have not standard names. The

renaming process is essential to bring coherence and

semantics in concepts that otherwise would be of

very difficult comprehension. The following stage is

the redefinition of the attributes and of the

associations’ roles multiplicities according to the

enterprise users’ best practices. This step is

fundamental to define the real constraints that are

not always made explicit into the implementation

model. As an example, considering a particular

attribute a

, two cases can be considered:

1) a

is a non-mandatory attribute in the conceptual

model but, as users are requested to always fill it

with a specific value, the enriched model must

formalise that this attribute a

is to be treated as

mandatory;

2) a

is defined as mandatory in the conceptual

model but, by practice, the users never care about its

value and fill it with some dummy one. In such case,

the enriched model may formalise that this attribute

is not mandatory.

Note that the same cases may happen also to the

roles of associations.

The last stage concerns of making explicit some

implicit associations. Those implicit associations

relate some concepts but they are defined only by

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

104

enterprise practices even if they are not expressed in

the model itself.

At this time, the enriched conceptual model

formalises the whole application semantics (both the

explicit one and the users’ implicit one).

3.3 Step 3: Fact-Oriented

Transformation

The quality of a conceptual model is often

influenced by the conceptual language used for its

specification. Most conceptual languages for data

modelling are based on a version of Entity-

Relationship modelling (E-R) (Barke, 1990) (Czejdo

et al., 1990) (Hohenstein et al., 1991). However,

these modelling languages are making a distinction

between entities, attributes and relationships. On the

contrary, in order to normalise the way that

knowledge is represented, NIAM (Natural-language

Information Analysis Method) (Nijssen & Halpin

1989) proposed to model the world in term of facts

(either presenting terms (real things), or representing

characteristics (attributes) of these real things), and

relationships between facts. NIAM is attribute-free,

it does not use explicitly the notion of attribute,

treating all elementary facts as relationships. Some

authors have extended the concepts and notations

developed by NIAM with object orientation. It is the

case of ORM (Object Role Modelling) (Halpin,

1998). Our purpose is to adapt this fact-oriented

modelling approach to enriched conceptual models

represented using the UML (OMG, 2004) class

notation. Thus, we developed a set of transformation

modelling rules, to be applied to selected UML

patterns (Table 1).

Let us refer to the definitions of LOT and

NOLOT facts given in the beginning of section 3.

Transforming a particular conceptual model in a

fact-oriented model must follow these rules:

1. all classes are transformed into LOT facts.

Using UML Class notation, a LOT fact is

represented by a UML Class.

2. all attributes are transformed into NOLOT

facts. Using the UML Class notation, a

NOLOT fact is represented as a UML Class.

3. for each attribute a belonging to a UML Class

C, an association is created between the

corresponding LOT a and the corresponding

NOLOT C, created by the two previous

rules.

4. the multiplicity associated to each attribute a

is copied as the multiplicity of the role of

the previous (rule 3) association attached to

the NOLOT a. The opposite role of the

same association must have a constraint

multiplicity equal to one.

5. all “simple” associations between classes are

transformed into “simple” associations

between NOLOTs.

6. all generalisation relationships between

classes are transformed into “simple”

associations with a constraint multiplicity

equal to one on the role attached to

generalised NOLOT and a non constraint

multiplicity equal to * on the opposite role.

In order to trace the fact that this association

was coming from a generalisation, we

annotate semantically the new

corresponding association with a logical rule

using OCL (Object Contraint Language)

notation.

Moreover, the inheritance feature of the

generalisation association is mapped as new

associations between LOTs representing the

attributes of the generalised NOLOT, and all

the specialised NOLOTs (sub-classes).

7. composition and aggregation relationships are

transformed into simple association (rule 3)

that keep unchanged the existing roles’

multiplicities but trace their specific

semantics through an attached semantic

annotation formalised with an OCL logical

rule.

8. association classes are transformed into a

LOT fact with two associations linked to the

corresponding initial LOT facts. The

multiplicities of the roles of these two

associations are determined inverting the

ones initially formalised on the roles of the

previous association.

9. any other specific constraints (generally

modelled using OCL logical rules) are kept

during the transformation process.

10. we did not take into account special cases of

constraints in generalisations because they

are not usually used in data conceptual

modelling.

One of the conceptual modelling requirements is

that a conceptual model must have formal

foundations, which allow comparing that model with

other conceptual models in a formal and exact way.

3.4 Patterns Represented in FOL

(Berardi et al, 2005) and (Tursi, 2009) formalise

UML class constructs semantics in First Order

Language (FOL) axioms. We propose to adapt these

works to formalise the fact-oriented model patterns

CONCEPTUALISATION APPROACH FOR COOPERATIVE INFORMATION SYSTEMS INTEROPERABILITY

105

Table 1: Fact-Oriented modelling patterns using UML notation.

UML FOM (UML) UML FOM (UML)

Class and Attributes Composition

Association Class Generalisation

Aggregation

(presented previously) in FOL axioms.

Due to the lack of space we will present only one

pattern rule formalisation in FOL: the “Class and

Attributes” as reported in

Table 1

A class in UML designates a set of object with

common features. Formally a class C corresponds to

a FOL unary predicate C.

An attribute a of type T for a class C associates

to each instance of C a set of instance of T, its

multiplicity [i..j] specifies that a associates to each

instance of C at least i and at most j instances of T.

Formally, an attribute a of type T for class C

corresponds to a binary predicate.

An association in UML is a relation between the

instances of two or more classes. The multiplicity

[m..n] attached to the role of a binary association

specifies that each instance of the class C can

participate at least m times and at most n times to the

related association. An association A between two

classes can be formalised as a binary predicate. In

the studied pattern, we formalise a class C

containing two attributes A

and A

with respectively

a multiplicity of 1 and [0..1], and with associated

types respectively, A

Type and A

Type.

Its formalisation in FOL assertions is the

following:

∀,,









∧





,



⊃











∀,,









∧





,



⊃











∀, 









⊃













,



∀, 









⊃



01











,



Applying the transformation rule, presented in

3.3, to the class C

, and to the two attributes A

and

, we will obtain the Fact-Oriented Model (FOM)

in UML notation as shown in

Table 1

“Class and

Attributes”.

Its formalisation in FOL assertions is the

following:

∀



,



, 









,





⊃











∧











∀



,











⊃

















,



∀



,











⊃

















,



∀



,



, 









,





⊃











∧











∀



,











⊃



01















,



∀



,











⊃

















,



Using a FOL engine such as the Haskel engine

(http://www.cs.yale.edu/homes/cc392/node1.html),

based on Russel and Norvig algorithms (Russel and

al, 1995), we are able to demonstrate that the

semantics formalised in the initial conceptual model

is equivalent or included into the one transformed in

FOM.

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

106

Figure 2: Sage X3 work order enriched process model.

4 CASE STUDY

Interoperability between organisational and man-

ufacturing activities is crucial in manufacturing

enterprises. Production services have to produce,

quickly and efficiently, the good product at the right

moment. For this reason, they need at time

information coming from others services, which

need in return precise and update data on production.

We propose here to study and present the first

part of such a B2M interoperability issue by

considering a particular IS implemented in a real

manufacturing environment: Sage X3 as an

Enterprise Resource Planning (ERP) application.

4.1 Specific Analysed Enterprise

Information System: Sage X3 ERP

An Enterprise Resource Planning (ERP) is an

integrated computer-based system used to manage

internal and external resources including tangible

assets, financial resources, materials, and human

resources (Bidgol, 1997). Its purpose is to facilitate

the flow of information between all business

functions inside the boundaries of the organization

and manage the connections to outside stakeholders.

Built on a centralized database, ERP systems

centralise all business operations into a uniform

CONCEPTUALISATION APPROACH FOR COOPERATIVE INFORMATION SYSTEMS INTEROPERABILITY

107

system environment. Sage X3 presents different enterprise management functions: finance,

commercial, industrial and services.

The focus for this case study is (i) to analyse

how the work order process inside the Sage X3

application is modelled, (ii) to use the proposed

modelling process to externalise the implicit

knowledge in the model structure.

The model depicted in figure 2 is already the

result from the two first steps of our approach. This

means that we have already passed the “Reverse

Engineering” and the “Expert knowledge injection”

stages. The “Manufacturing Order Heading” concept

is the management function of production orders and

planned activities. It allows the generation of a

production order by variation of one or more

classifications and a single production line. For each

production order, the achievement of the material

benefits and sequencing operations is possible. This

block captures general information about the work

order, such as, planning facility and facility of

production, status of the order (manufacturing order

product). It allows entering general information

about the production order. The availability of

components is checked through the information

given by the bill of material related with the

launched products.

Once that initial information is determined, the

system updates the list of materials and operations of

the created or modified orders.

Step 1: Reverse Engineering

All these information are coded in the Sage

application database. The first step of our method is

the reverse engineering to extract the initial

conceptual model.

Step 2: Expert Knowledge Injection

Currently the model depicted in Figure 2 is the result

of the reverse engineering step enriched by a domain

expert because the architecture of the Sage X3 ERP

is built with all the database relationships

implemented directly into the application layer and

not in the database. The reverse engineering result,

as shown in the lower part of the Figure 3, creates a

model containing unlinked classes with coded

names.

The expert work was about cleaning this conceptual

model according to the best practices in the

enterprise, modifying the attributes multiplicity,

adding explicit names to the concepts, the attributes

and the associations and others operations to fit the

conceptual model to the “real” use of the Enterprise

Information System. A usual case that requests the

domain expert attention is about the mandatory

properties in forms’ attributes.

Figure 3: Sage X3 architecture and expert knowledge

injection.

Step 3: Fact-Oriented Transformation

Applying the pattern transformation rules, presented

in the previous section, class attributes are

transformed into NOLOTs to increase the atomic

representation of the knowledge embedded into the

model. These rules have been coded using a

programming language and then automatically

executed inside MEGA Suite.

Figure 4 shows an extract of the resulting FOM

after applying our approach to the Sage X3 work

order process. The resulting full FOM is composed

of 23 NOLOTs, 56 LOTs and 46 associations.

It seems then that the resulting model is much

more complex than the initial one, which it is true in

a visual point of view but it is false in term of

expressiveness of its semantics. Indeed, the fine-

grained atoms of semantics are now made explicit,

which helps any automatic computing. An important

result is that such semantically detailed model will

help automating the next part of our methodology

for semantic gap evaluation, as explained in section

5 CONCLUSIONS

In this article, a conceptualisation approach for

enacting implicit semantics from Enterprise

Information Systems is proposed. Our approach if

divided into 3 steps from the traditional reverse

engineering process, through a knowledge elicitation

and model enrichment by domain experts, till

making use of fact-oriented modelling patterns to

externalise tacit knowledge. These patterns have

been formalised in FOL axioms to verify their

semantic coherence. Our contribution can be

assimilated to a reverse engineering methodology.

Expert

work

Automatic

work

Sage X3

Relationships,

Primary keys,

trigger, i nde x…

Software interface

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

108

However, the main objective is to formalize the whole semantics of such models in order to help

Figure 4: Sage X3 work order process model part transformed with fact-oriented approach.

automatic knowledge computing. An industrial case

study, related to an enterprise information system

implemented into an ERP system demonstrates the

applicability of our approach.

Our current work concerns applying this approach

for evaluating the (non)-interoperation through the

measurement of the semantic gap occurring between

CISs interoperability (

Yahia, 2011

REFERENCES

Aspray, W. F. 1985. The Scientific Conceptualization of

Information: A Survey. Annals of the History of

Computing, 7: 117-40. IEEE

Badia, A., 2002. Conceptual Modeling for Semistructured

Data. In Proceedings of the 3rd International

Conference on Web Information Systems Engineering

Workshops (WISE 2002 Workshops), p. 170-177.

Singapore.

Barker, R. 1990, CASE* Method: Entity Relationship

Modelling, Addison-Wesley, Wokingham, England.

Berardi, D., Calvanese, D., De Giacomo, G. (2005).

Reasoning on UML class diagrams, Artificial

Intelligence 168 (1–2) 70–118.

Bézivin, J., Kurtev, I. 2005. Model-based Technology

Integration with the Technical Space Concept.

Proceedings of the Metainformatics Symposium,

Esbjerg, Denmark, November 8-11, 2005. Springer-

Verlag

Bidgol, H., 2004. The Internet Encyclopedia, Volume 1,

John Wiley & Sons, Inc. p. 707.

Carney, D., Fisher, D., Morris, E., Place P., 2005. Some

current Approaches to Interoperability. In technical

note CMU/SEI-2005-TN-033.

Chen, D., Dassisti, M., Elvesaeter, B., Panetto, H., et al.,

2006. In DI.2: Enterprise Interoperability Framework

and knowledge corpus, Interoperability Research for

Networked Enterprises Applications and Software

Network of Excellence, n° IST 508-011.

Chiang Roger, H. L., Barron, T. M., Storey Veda, C.,

1994. Reverse engineering of relational databases:

Extraction of an EER model from a relational

database. In Data & Knowledge Engineering. Vol. 12,

Issue 2, 107-142.

Czejdo, B., Elmasri, R., Rusinkiewicz, M. & Embley,

D.W. 1990, ‘A graphical data manipulation language

CONCEPTUALISATION APPROACH FOR COOPERATIVE INFORMATION SYSTEMS INTEROPERABILITY

109

for an extended entity-relationship model’, IEEE

Computer, March 1990, pp. 26-37.

Engelbart, D.C., 1962. Augmenting human intellect: a

conceptual framework. In Menlo Park, CA: Stanford

Research Institute.

Euzenat, J., 2001. Towards a principled approach to

semantic interoperability. In CEUR Proceedings of the

IJCAI-01 Workshop on Ontologies and Information

Sharing, Seattle, USA, August 4-5, , ISSN 1613-0073,

Vol. 47., 19-25.

Fonkam, M.M., Gray, W.A., 1992. An Approach to

Eliciting the Semantics of Relational Databases. In

CAiSE 1992, Manchester, UK, May 12-15, 1992.

Lecture Notes in Computer Science 593 Springer,

ISBN 3-540-55481-5. 463-480 Manchester, UK.

Frankel D. S. 2003. Model Driven Architecture: Applying

MDA to Enterprise Computing. John Wiley & Sons.

Guarino, N., 1998. Formal Ontology in Information

Systems (Ed.) IOS Press.

Halpin, T.A. 1991, ‘A fact-oriented approach to schema

transformation’, Proc. MFDBS-91, Springer Lecture

Notes in Computer Science, no. 495, Rostock.

Halpin, T., 1998. Handbook on Architectures of

Information Systems Chapter 4, eds P. Bernus, K.

Mertins & G. Schmidt, Springer-Verlag, Berlin.

Henrard, L., Hainaut, J.-L., 2001. Data Dependency

Elicitation in Database Reverse Engineering Software

Maintenance and Reengineering. In Fifth European

Conference on Software Maintenance and

Reengineering, pp. 11

Hohenstein, U. & Engels, G. 1991, ‘Formal semantics of

an entity-relationship-based query language’, Entity-

Relationship Approach: the core of conceptual

modelling (Proc. 9th ER conf.), ed. H. Kangassalo,

Elsevier Science Pub., Amsterdam

IEEE: Standard Computer Dictionary, 1990. A

Compilation of IEEE Standard Computer Glossaries.

In NY. 610-1990. ISBN: 1559370793.

International Organization for Standardization, 1999. ISO

14528: Industrial Automation Systems – Concepts and

rules for Enterprise Models, TC 184/SC5/WG1,

Geneva, Switzerland.

International Organization for Standardization, 2002. ISO

16100: Manufacturing Software Capability Profiling

for interoperability. In Part 1: Framework, TC

184/SC5/WG4, Geneva, Switzerland.

LaOngsri, S. 2009, Semantic Extensions and a Novel

Approach to Conceptual Modelling, Ph.D. Thesis,

School of Computer Science, Engineering and

Mathematics, The Flinders University of South

Australia

Mani, M.: EReX, 2004. A Conceptual Model for XML. In

Proceedings of the Second International XML

Database Symposium (XSym 2004), p. 128-142.

Toronto, Canada.

Manola, F., Miller, E., 2004. RDF Primer. In World Wide

Web Consortium, Recommendation REC-rdf-primer-

20040210.

Meersman, R., Tari, Z., 2003. On The Move to

Meaningful Internet Systems 2003. In OTM 2003

Workshops OTM Confederated International

Workshops, Lecture Notes in Computer Science, Vol.

2889. ISBN: 978-3-540-20494-7, Catania, Italy.

Nijssen, G.M. & Halpin, T.A. 1989, Conceptual Schema

and Relational Database Design, Prentice Hall,

Sydney.

Obrst, L., 2003 Ontologies for semantically interoperable

systems. In Proceedings of the 12th International

Conference on Information and Knowledge

Management. New Orleans, USA

OMG, 2003. Object Management Group. Architecture-

Driven Modernization specification

http://adm.omg.org

OMG, 2004. Object Management Group. UML 2.0

Superstructure Specification http://uml.omg.org

Rumbaugh, J., Blaha, M., Premerlani, W., Eddy, F.,

Lorensen W., 1991. Object Oriented Modeling and

Design. Prentice Hall, Book Distribution Center, New

York, USA.

Russel, S., Norvig, P. Artificial Intelligence, A Modern

Approach'', Prentice-Hall. Inc. 1995

Seeley, R. S., 1997. Manufacturing execution systems in

MED DEVICE DIAGN IND. Vol. 19, no. 11, pp. 64-

68.

Sheth, A., 1998. Changing Focus on Interoperability in

Information Systems: From System, Syntax, Structure

to Semantics. In M. Goodchild, M. Egenhofer, R.

Fegeas, and C. Kottman, editors. In Interoperating

Geographic Information Systems, pp. 5– 30. Kluwer.

Smith, M.K., Welty, Ch., McGuinness, D.L., 2004. OWL

Web Ontology Language Guide. In World Wide Web

Consortium, Recommendation REC-owl-guide-

20040210.

Tolk, A., Diallo, S. Y., Turnitsa, C. D., 2007. Applying the

Levels of Conceptual Interoperability Modelling

Support of Integrability, Interoperability, and

Composability for System-of-Systems Engineering. In

Journal of Systemics, Cybernetics and Informatics,

Volume 5 Number 5, pp. 65-74.

Tursi, A., Panetto, H., Morel, G., Dassisti, M., 2009

Ontological approach for Products-Centric

Information System Interoperability in Networked

Manufacturing Enterprises. In IFAC Annual Reviews

in Control. 33/2, 238-245, Elsevier, ISSN: 1367-5788.

Vyvyan, E., 2006. Lexical Concepts, Cognitive Models

and Meaning-Construction. In Cognitive Linguistics

17 (4): 491-534.

Yahia E., Yang J., Aubry A., Panetto H., 2009. On the use

of Description Logic for Semantic Interoperability of

Enterprise Systems. In On the Move to Meaningful

Internet Systems: OTM 2009 Workshops (Meersman

R., Tari Z., Henerro P. (Eds)), 4th IFAC/IFIP

Workshop on Enterprise Integration, Interoperability

and Networking (EI2N’2009), Vilamoura, Portugal,

Springer Verlag, Lecture Notes in Computer Science,

LNCS 5872, ISBN 978-3-540-88874-1.

Yahia E., Lezoche M., Aubry A., Panetto H. 2011.

Semantics enactment in Enterprise Information

Systems. 18th IFAC World Congress. Milan, Italy

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

110