LOCAL ONTOLOGIES FOR SEMANTIC INTEROPERABILITY

IN SUPPLY CHAIN NETWORKS

Milan Zdravković and Miroslav Trajanović

Laboratory for Intelligent Production Systems, University of Niš, Faculty of Mechanical Engineering

Aleksandra Medvedeva 14, Niš, Serbia

Hervé Panetto

Research Centre for Automatic Control (CRAN – UMR 7039), Nancy-Université, CNRS, Nancy, France

Keywords: Ontology, Enterprise interoperability, Supply chain management, SCOR.

Abstract: Most of the issues of current supply chain management practices are related to the challenges of

interoperability of relevant enterprise information systems (EIS). In this paper, we present the ontological

framework for semantic interoperability of EISs in supply chain networks, based on Supply Chain

Operations Reference (SCOR) model, its semantic enrichment and mappings with relevant enterprise

conceptualizations. In order to introduce the realities of the enterprises into this framework, namely their

models, we define and implement the approach to generation of local ontologies, based on the databases of

their EISs. Also, we discuss on the translation between semantic and SQL queries, a process in which

implicit semantics of the EIS’s databases and explicit semantics of the local ontologies become inter-related.

1 INTRODUCTION

Despite the advances in the relevant research areas

and rapidly growing demand for flexible,

customized production, manufacturing supply chains

are still primarily focused on a cost reduction as a

key aspect of collaboration. The fact that supplier

relationship management contributes largely to the

overall costs of the supply chains’ final products has

great impact to their configuration-related decisions.

For example, manufacturers tend to reduce the

number of suppliers. Moreover, relationships are

dyadic – rarely expanded to include vendors’

vendors and customers’ customers. Also, high level

of integration is required in order to reduce costs –

manufacturers tend to view their suppliers as

extensions of themselves. Traditional approach to

supply chains’ configuration may have negative

impact to their performance. First, high-speed, low-

cost supply chains are often unable to respond

efficiently to unexpected structural changes in

(customized) demand or supply. Second, high level

of integration reduces flexibility of small and

medium enterprises, main constituents of the lower

levels of supply chains. Third, investments in

technical framework for enterprise integration,

which could maximize the efficiency and

productivity, cannot be returned in a short term.

Furthermore, starting collaboration in such

traditional settings is reactive and not proactive

decision. Namely, relationship establishment or

development is motivated by the internal, rather than

external factors: complexity and volume of supply

relationships, potential for cost reduction (Lamber et

al., 2006), high frequency of transactions between

parties (Jespersen and Larse, 2006), degree of asset

specificity (Williamson, 1985), etc.

In a response to the issues of static and integrated

architecture of the supply chain, a notion of virtual

enterprise has been introduced and widely discussed

in academic community. Virtual enterprise is a

temporary network of independent enterprises, who

join together quickly to exploit fast-changing

opportunities and then dissolve (Browne and Zhang,

1999). It is characterized by a short-living

appearance of a supply chain, capable to produce

low volume of high variety of products, by drawing

from the loosely-coupled, heterogeneous

environment of available competences, capabilities

Zdravkovi

c M., Trajanovi

c M. and Panetto H..

LOCAL ONTOLOGIES FOR SEMANTIC INTEROPERABILITY IN SUPPLY CHAIN NETWORKS.

DOI: 10.5220/0003416500220031

In Proceedings of the 13th International Conference on Enterprise Information Systems (ICEIS-2011), pages 22-31

ISBN: 978-989-8425-53-9

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

and resources, sometimes referred to as Virtual

Breeding Environment (Sánchez et al., 2005).

Paradigms of virtual enterprises and their breeding

environment are based on the capability of an

enterprise to configure or reconfigure quickly,

according to the circumstances of the market, often

not known in advance, or even in the moment of

configuration. Hence, efficiency and effectiveness of

this joint endeavour depends on the interoperability

of enterprises, rather than their integration. The main

prerequisite for achievement of interoperability of

the loosely coupled systems is to maximize the

amount of semantics which can be utilized and make

it increasingly explicit (Obrst, 2003), and

consequently, to make the systems semantically

interoperable.

In this paper, we discuss on the notion of

semantic interoperability in supply chain networks,

namely the overall architecture of the enterprise

information systems environment and corresponding

ontological framework. One of the greatest

challenges in building this framework is related to

the implicitness of semantics of the enterprises’

realities. In our approach to face this challenge, we

assume that: 1) these realities are represented by the

corresponding enterprise information systems (EIS),

and 2) enterprise message models (crucial for

flexible economic integration) are based on EISs’

data models, represented implicitly in their

databases. The proposed approach aims at making

this representation - explicit.

Our approach to semantic interoperability in

supply networks assumes fragmentation of the

problem into three inter-related areas: a) formal

model of supply chain, b) enterprise semantics (body

of knowledge), and c) local semantics.

Formal model of supply chain (Zdravković et al.,

2010) builds upon a widely adopted supply chain

process reference model – SCOR (Stewart, 1997). It

is represented on two layers of abstraction. First

layer models implicit semantics of SCOR elements

and stores actual knowledge on supply chain

operations. Second layer represents SCOR’s

semantic enrichment - it identifies common

enterprise notions, maps those to SCOR entities and

classifies them into more general inter-related

concepts. Both layers are then represented by OWL

models – SCOR-KOS and SCOR-FULL. This

approach is shortly described in section 2. In this

section, we also discuss on the semantic

interoperability in supply networks and on how we

can use its formal definition to evaluate it. Finally,

we describe the role of local ontologies in

semantically interoperable systems.

SCOR-FULL ontology identifies and classifies

common enterprise notions. However, their

semantics is defined externally. We strongly believe

that enterprise semantics is well described in many

efforts of conceptualizing its architecture, functions

and processes and that additional effort in this

direction would be redundant. Thus, different

enterprise formalizations, contexts and views of

existing architectures and other conceptualizations

need to be used as sources of specifications of

enterprise semantics, and mapped accordingly to the

enterprise notions in SCOR-FULL ontology.

Currently, SCOR-FULL ontology is mapped to

TOVE (Fox et al., 1996) organizational and

foundational ontology (in fact, to its OWL

representation).

While formal model of supply chain and

enterprise semantics provide a theoretical context for

semantic interoperability in supply networks, local

ontologies introduce actual enterprises contexts,

namely, the language which they are going to use to

communicate, in a collaboration environment. We

believe that enterprises’ capability to efficiently

collaborate between each other depends on the

correspondence between their local semantics and

the general context above. Main focus of the work,

presented in this paper, is on the analysis of the

source of this local semantics, namely relational

database systems and, consequently, its

explicitation.

The research addresses some of the identified

weaknesses of the existing approaches (see section

3.1) to database to ontology mapping and aims at

using the OWL expressivity to enrich the implicit

semantics of ER (Entity-Relationship) models. It

delivers a method and corresponding software tool

which: 1) imports the database structure and

classifies ER entities; 2) classifies (infers) OWL

types and properties; 3) enables lexical refinement

and 4) generates local ontology. The concepts of the

local ontology are mapped backwards to the

corresponding concepts of the intermediary models,

in order to enable transformation of semantic to SQL

queries. The method and the software tool are

described in section 3.2, and are applied in the case

of OpenERP database. Some of the experiences

gained during implementation of this case are

described in section 3.3. The method for execution

of semantic queries on the local ontology, namely,

instantiation of its concepts according to the content

of the relevant database, is described in section 3.4.

It is important to emphasize that the scope of the

presented approach is limited only to selected ER

patterns which are associated to semantics,

LOCAL ONTOLOGIES FOR SEMANTIC INTEROPERABILITY IN SUPPLY CHAIN NETWORKS

expressed by the OWL constructs. Although the

process overcomes some of the gaps, identified in

the current state-of-the-art in database to ontology

mapping, its end result typically requires

considerable amount of customization. Since direct

mapping is unlikely to produce a useful ontology,

the result of this analysis may be considered as

intermediary. Thus, it is necessary to put additional

work in enactment of this intermediary ontology,

which aims at facilitating the final stage of semantic

mapping of local ontology to relevant domain

ontologies.

2 SEMANTICS OF SCOR

AND SEMANTIC

INTEROPERABILITY

The concepts and tools presented in this paper are

using the formal framework of supply chain

operations presented at Fig. 1. It is developed with

goals to enable the semantic interoperability

between SCOR-based systems and other relevant

enterprise information systems, and to improve the

expressivity of SCOR-based models.

In the remainder of this sub-section, we briefly

describe the elements of this framework. Their

detailed elaboration can be found in Zdravković et

al., 2010.

SCOR-FULL ontology is developed by semantic

analysis of SCOR Input/Output elements,

identification of core terms and their generalization.

It extends what we call the SCOR-SYSTEM

ontology, which formalizes the SCOR System

element.

It is then extended by the SCOR-GOAL

ontology, which semantically maps its concepts to

SCOR Performance Metrics element. SCOR-FULL

is exploited by different application models, which

formalize specific design goals. For example,

SCOR-CFG OWL model is used to develop a

semantic web application for supply chain process

configuration (Zdravković et al., 2010).

The framework is based on a premise that

domain knowledge evolves at highest rate at lower

levels of abstraction, in domain community

interaction. Consensus on the specific notions is

more likely to be reached than agreement on the

generalizations and abstractions. However, this level

is often characterized by the implicit semantics of

the standards, reference models, database structures,

etc. Thus, we consider coherence between creation,

evolution and use of specific, highly contextualized

knowledge and development of formal expressive

models as a very important factor for usability of the

models.

In the process of development of a formal

framework for supply chain operations, we start with

modelling the implicit semantics of SCOR model

and representing it by using OWL language (SCOR-

KOS OWL). OWL (OWL 2 Web Ontology

Language) is a family of knowledge representation

languages, which provides the syntax for authoring

and exchanging the ontologies among relevant tools

and applications.

Figure 1: Formal framework of supply chain operations

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

Thus, this common syntax (OWL) enables us to

reuse the resulting model with existing tools such as

ARIS EasySCOR by IDS or e-SCOR by Gensym.

Next, the semantics of the SCOR elements is made

explicit through its formalisation with ontology rules

embedded into the SCOR-FULL ontology. These

rules are used for mapping those concepts to SCOR-

KOS OWL concepts. We call this set of rules,

SCOR-MAP. SCOR-FULL may then be considered

as a micro theory which formalizes knowledge about

supply chain operations, by identifying and

aggregating common enterprise notions. It is using

those concepts to define the semantics of chosen

generalizations, namely, the notions of Course,

Setting, Quality, Function and Resource.

2.1 Semantic Interoperability

in Supply Chain Networks

One of the crucial competitiveness factors of the

enterprises, especially SME’s, is their ability to

ensure the quality performance, when

simultaneously participating in more than one

supply chain, with the same or different products.

This ability is mainly driven by the enterprises’

capacity to conform to different collaboration

(processes) requirements, issued by the customers,

where these requirements are described by the

different standards or models (SCOR, RosettaNet,

ISO/TS 16949 quality specification for automotive

industry supply chain and others) and/or methods

(CPFR, Vendor-Managed Inventory, etc.).

One of the main challenges is to define the

references between the standards, make the

corresponding models compatible or complementary

and thus, ensure the interoperability of the relevant

systems, driven by those models. This challenge can

be addressed by formalizing collaboration standards

and methods, identifying common enterprise notions

(and relating those to the standards’ elements) and,

mapping them to a general body of knowledge,

namely, enterprise models. Hence, the relevant

systems, based on those models, will become fully

or partially – semantically interoperable.

ISO/IEC 2382 defines interoperability as the

“capability to communicate, execute programs, or

transfer data among various functional units in a

manner that requires the user to have little or no

knowledge of the unique characteristics of those

units”. Semantic interoperability builds upon this

notion and it means ensuring that the precise

meaning of exchanged information is uniquely

interpreted by any system not initially developed for

the purpose of its interpretation. It enables systems

to combine and consequently process received

information with other information resources and

thus, to improve the expressivity of the underlying

ontologies. In our research, we adopt the formal

definition of John Sowa (Sowa, 2000; SUO, 2001),

because we can use it to evaluate semantic

interoperability of enterprise systems:

“A sender's system S is semantically operable

with a receiver's system R if and only if the

following condition holds for any data p that is

transmitted from S to R:

For every statement q that is implied by p on the

system S, there is a statement q' on the system R

that: (1) is implied by p on the system R, and (2) is

logically equivalent to q. The receiver must at least

be able to derive a logically equivalent implication

for every implication of the sender's system.”

We represent this definition in controlled natural

language, as asymmetric logical function

semantically-interoperable(S,R):

data(p) ∧ system(S) ∧ system(R) ∧

semantically-interoperable(S,R) ⇒

∀p (

(transmitted-from(p,S) ∧

transmitted-to(p,R)) ∧

∀q(statement-of(q,S) ∧ p⇒q)

∃q’(statement-of(q’,R) ∧ p⇒q’ ∧

q’⇔q)

)

Figure 2 illustrates the following assumption of

semantic interoperability of systems, represented by

the local ontologies: when two different application

ontologies of two partners in the supply chain (or

two departments or contexts of the same enterprise)

are mapped to the same domain ontology, relevant

information systems whose knowledge they

represent will become fully or partially semantically

interoperable in specific direction, depending on the

mappings.

In other words, if there exist two isolated

enterprise information systems S

and S

and

corresponding application ontologies O

and O

and if there are mappings M

L1D1

and M

L2D1

established between the concepts of O

, O

and

domain ontology O

, respectively, then there exist

mappings M

L1L2

which can be inferred as logical

functions of M

L1D1

and M

L2D1

. Each of the local

ontologies may (not necessarily) represent one of the

contexts of the enterprise (C

-C

In our research, we aim at confirming this

assumption by inferring the mappings between

two contexts of the enterprise, represented by

LOCAL ONTOLOGIES FOR SEMANTIC INTEROPERABILITY IN SUPPLY CHAIN NETWORKS

Figure 2: Semantic interoperability of systems.

formalizations of an ERP system (S

-O

) and

SCOR-based system (S

-O

) for managing the

supply chain operations of the enterprise. Mappings

are inferred as logical functions of mappings

between corresponding formalizations and SCOR-

FULL ontology (O

). Also, by exploiting the

mappings between SCOR-FULL (O

) and TOVE

organizational and foundational ontologies (O

), we

aim at showing how the expressivity of the overall

ontological environment can be increased, for the

benefit of improved semantic interoperability and

increased competence. Finally, we evaluate the

semantic interoperability of the systems by using the

definition above.

In this paper, we focus only on the local

ontologies, namely, the formalizations of enterprise

information systems’ data models and we

demonstrate the approach to automated (or semi-

automated) generation of local ontologies on basis of

the relational database structure.

3 INTEROPERABLE LOCAL

ONTOLOGIES

One of the major challenges in the efficient use of

computer systems is interoperability between

multiple representations of reality (data, processes,

etc.) stored inside the systems, or actual

representations and reality itself – systems’ users

and their perception of reality (Hepp, 2007). Where

latter can be formalized by the domain ontologies, as

shared specifications of the conceptualizations,

former relies upon the local ontologies – wrappers

for heterogeneous sources of information, business

logic and presentation rules.

In our work, the range of semantic

interoperability is clearly set to enterprise

information systems. The semantic interoperability

of the enterprises is considered as more complex

problem and is not addressed in this paper. The

conceptualization of their information systems is

made on basis of the business logic, which is hidden

in the actual code, in most cases, and data model,

represented by the corresponding relational database

structure. We consider EIS’s databases as legitimate

starting point for building a relevant local ontology.

Obviously, business logic which is encapsulated in

the EIS’ will remain hidden – only underlying data

model is exposed by ontology. The exceptions are

database’s triggers, which can be considered as

business rules, if they are not implemented only to

enforce referential integrity of the database.

In the remainder of this section, some of the

reported work in database to ontology mapping is

presented. Then, our approach to local ontology

generation is described and demonstrated on the case

of ER model (database) of OpenERP enterprise

software.

3.1 State of the Art in Database to

Ontology Mapping

Review of the relevant literature reveals several

approaches which address database to ontology

mapping. In this section, we present the main

features of four distinctive frameworks, made with

different objectives, and we identify gaps, in terms

of the usability and coverage of the frameworks.

Work on DB2OWL mapping facility is a part of

development of a general interoperability

architecture (Ghawi and Cullot, 2007) that uses

ontologies for explicit description of the semantics

of information sources, and web services to facilitate

the communication between the different

components of the architecture. DB2OWL (Cullot et

al., 2007) looks for some particular cases of database

tables to determine which ontology component has

to be created from which database component.

According to these cases, conversion process is

performed (table -> class, column-> property,

constraint -> relation) where the set of

correspondences between database and ontology

components is conserved, thus enabling the

translation of ontological to SQL queries and

retrieval of corresponding entities. However, it

remains unclear how this translation will be

implemented. More important, the semantics of

existential constraints of the columns and cardinality

of relations is not taken into account.

Relational.OWL (de Laborda and Conrad, 2005)

is a candidate for data and schema representation

format, relevant for database to ontology mapping. It

provides a meta-model, which describes the

components of the relational database. Hence, it can

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

be used as an intermediary in the process of database

to ontology mapping, instead of a document with

correspondences, used by DB2OWL. Unfortunately,

it suffers from the same problems as DB2OWL -

multiplicity of the foreign keys is not in the model.

Thus, it is not possible to use it to assign source and

destination cardinality to OWL properties.

Moreover, source multiplicity determines important

aspect of the semantics of the underlying concept or

database table. Namely, where source multiplicity of

the foreign key is 1, the corresponding OWL relation

shall be necessary condition for instantiation of the

concept in its domain. This is important semantic

feature, because it enables intensional

conceptualization of the entity.

Where DB2OWL is used to create new ontology

from existing schema, D2OMapper (Xu et al., 2006)

is a tool for automatic or semi-automatic creation of

the mappings between database schema and existing

ontology. D2OMapper follows a set of predefined

heuristic rules, based on the conceptual

correspondences between the schema and ontology.

This work is based on the authors’ experience in

developing ER2WO (Xu et al., 2004) tool for

translating ER schema into OWL ontology.

Vis-A-Vis tool (Konstantinou et al., 2006) uses

the Protégé libraries for graphically representing an

ontology, a database model (MySQL or

PostgreSQL) and the mappings between them. The

plug-in allows queries to be asked to the ontology

and returns results from the database. The key

motivation of the authors was to keep the instances

stored in a database and maintain a link to the

dataset. Thus, ontologies become smaller.

3.2 Our Approach to Database to

Ontology Mapping

Mapping is a process in which implicit semantics of

a database schema is mapped to the explicit and

formal knowledge structure of the ontology. In our

approach, we use the database schema to generate

this formal structure, while preserving the logical

mappings between ER meta-model and generated

local ontology. These mappings will enable the

translation of semantic to database queries.

Generation process consists of 4 phases: a) data

import and classification of ER entities; b)

classification (inference) of OWL types and

properties; c) lexical refinement; d) generation of

local ontology; and is illustrated on Figure 3, below.

The process is supported by a web application,

developed by using RAP API (Oldakowski et al.,

2005), a PHP-based package for parsing, querying,

manipulating, serializing and serving RDF models.

Web application consists of modules for data

import/assertion of ER meta-model instances, lexical

refinement and transformation of classified OWL

types and properties to a local ontology.

First, database schema is investigated and OWL

representation of the ER-model is constructed. This

is realized by developed application, which connects

to the database, uses introspection queries to

discover its structure and asserts the relations

between the artifacts by using proposed ER

formalization (er.owl).

Following assertions are made for each field

of the corresponding table: has Attribute (entity,

attribute), hasType (attribute, type) and

Figure 3:

Approach to database-to-ontology mapping.

LOCAL ONTOLOGIES FOR SEMANTIC INTEROPERABILITY IN SUPPLY CHAIN NETWORKS

hasConstraint(attribute,’not-null’) and/or

hasConstraint(attribute,’unique’) (if applicable).

Following assertions are made for each relation:

hasDestinationAttribute (relation, attribute),

hasSourceAttribute(relation, attribute).

Second, resulting (serialized) OWL

representation of the database ER-model is imported

into meta-model (s-er.owl), which classifies future

OWL concepts (Ax1) and domains and ranges of the

object and data properties, according to defined

axioms (Ax2, Ax4). Although specification of object

and data properties may impose the unnecessary

restrictions on the resulting ontology, we consider

those as important for improving the efficiency of

mapping or alignment process, which is critical for

the semantic interoperability. Another reason for the

assertion of object properties in OWL representation

of database ER-model is that object properties of the

resulting local ontology will be annotated with the

URI’s of the respective relations, in order to enable

the correspondence between the ontology and

database representation, for the benefit of query

transformation. On the other hand, existential

constraints from the ER-model are associated to an

explicit semantics of the resulting ontology, namely,

necessary conditions for entailment of the

corresponding concepts. According to these

constraints, axioms for intensional conceptualization

(necessary conditions, or inherited anonymous

classes) for particular entity are identified by

inferring ranges of hasDefiningProperty(concept,

concept) and hasDefiningDataProperty(concept,

data-concept) relations (Ax2.2 and Ax4.2). Finally,

the approach takes into account the functionality of

the properties (owl:FunctionalProperty). Functional

property is property that can have only one (unique)

value y for each instance x. They are classified when

relation one-to-one is identified between two

concepts (Ax2.3).

Classification of future OWL concepts is inferred

by exploiting following axioms:

Ax1. Concepts are all entities of the ER model’s

OWL representation, except the entities whose all

attributes are relation sources (corresponding to

intermediary tables, connecting two tables with

many-to-many relationship).

er:entity(x) ∧ not (er:hasAttribute

only (er:attribute ∧

(er:isSourceAttributeOf some

er:relation))) ⇒ s-er:concept(x)

Ax2.1. Domains and ranges of the object

properties are inferred by using the rule below.

er:entity(x) ∧ er:entity(y) ∧

er:relation(r) ∧ er:hasAttribute(x,

a1) ∧ er:hasAttribute(y, a2) ∧

er:isDestinationAttributeOf(a2, r) ∧

er:isSourceAttributeOf(a1, r) ⇒ s-

er:hasObjectProperty(x, y)

Ax2.2. Domains and ranges of the defining

properties (necessary conditions of the concept) are

inferred by using the rule below. Defining property

is a sub-property (rdfs:subPropertyOf) of the object

property (hence, simplified representation of the rule

below).

s-er:hasObjectProperty(x, y) ∧

er:hasConstraint(a1,'not-null') ⇒ s-

er:hasDefiningProperty(x, y)

Ax2.3. Domains and ranges of the functional

properties are inferred by using the rule below.

Functional property is a sub-property

(rdfs:subPropertyOf) of the defining property

(hence, simplified representation of the rule below).

s-er:hasObjectProperty(x, y) ∧

er:hasConstraint(a1,'not-null') ⇒ s-

er:hasDefiningProperty(x, y)

Ax3. Data concepts are all attributes of the ER

model’s OWL representation which are not at the

source of any relation.

er:attribute and not

(er:isSourceAttributeOf some

er:relation) ⇒ s-er:data-concept

Ax4.1. Domains and ranges of the data

properties are inferred by using the rule below.

Ranges of the data properties are data types,

corresponding to the simple types from XML

schema.

er:type(x) ⇒ s-er:data-type(x)

s-er:concept(c) ∧ er:attribute(a) ∧

er:type(t) ∧ er:hasAttribute(c, a) ∧

er:hasType(a, t) ⇒ s-

er:hasDataProperty(c, t)

Ax4.2. Domains and ranges of the defining data

properties are inferred by using the rule below.

Defining data property is a sub-property

(rdfs:subPropertyOf) of the data property (hence,

simplified representation of the rule below).

s-er:hasDataProperty(c, t) ∧

er:hasConstraint(a,'not-null') ∧

er:hasConstraint(a,'unique') ⇒ s-

er:hasDefiningDataProperty(c, t)

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

Rules above classify instances of the OWL

representation of the database ER model (er.owl)

into a meta-model (s-er.owl). Inferred triples can be

edited in a simple web application, which also

launches the process of local ontology generation. In

this process, meta-model entities are transformed

into corresponding OWL, RDF and RDFS constructs

– a resulting local ontology. Concepts of the

generated local ontology are annotated with URI’s

of the corresponding ER entities from er.owl model.

Thus, translation of semantic to SQL queries

becomes possible.

3.3 Case Implementation

The approach above is implemented on the case of

OpenERP enterprise information system. OpenERP

is an open source suite of business applications

including sales, CRM, project management,

warehouse management, manufacturing, accounting

and human resources. It uses PostgreSQL relational

database for data storage and application server for

enterprise logic.

With all modules installed, OpenERP database

counts 238 tables. In the first step of database import

into er.owl model, namely, instantiation of the OWL

representation of ER model, 3806 individuals are

created (2633 attributes, 238 entities, 934 relations)

and 7999 object property assertions are made. In the

second step of classification of OWL concepts and

properties, 696 of individuals’ entailments (193

concepts and 493 data-concepts) and 2779 properties

are inferred, on the basis of axioms, presented in

section 3.2. All inferences are stored in a separate

OWL file in order to reduce the processing

requirements for the final step. In the final step of

local ontology generation, application transforms

classified instances of the meta-model of the

openERP database to the corresponding OWL

concepts and properties (see Fig. 4).

Figure 4: OpenERP local ontology in Protege.

One of the benefits of the semantically

interoperable systems (see Fig. 2) is the possibility

to use the single criterion (or criteria) to infer the

statements that hold true in all these systems, despite

their heterogeneous structure. Namely, specific

semantic query executed against the local ontology

would normally infer triples of information from

the database of S

. However, if mappings (or logical

function of mappings) between O

and O

exist,

inferred triples will also include information from

the database of S

. For example, in supply chain

networks, a single semantic query can be used to

find out the availability of specific resource or

competence, of all - owned and used by the

enterprises from the Virtual Breeding Environment

(for the benefit of virtual enterprise formation

process).

3.4 Reasoning with Local Ontologies

and Translation of Semantic

to SQL Queries

In this section, we describe the method for instance

assertions to local ontology on basis of the semantic

query results. Method consists of the following

steps: 1) decomposition and analysis of the semantic

query; 2) data extraction and instance assertions; 3)

reasoning.

Semantic query can be considered as a pair (O,

C), where O is a set of concepts which need to be

inferred and C - a set of restrictions to be applied on

their properties, namely value (owl:hasValue and

qualified cardinality restrictions, owl:allValuesFrom,

owl:someValuesFrom) and cardinality constraints

(owl:cardinality, owl:minCardinality,

owl:maxCardinality). This consideration

corresponds to a simplified representation of a SQL

query which includes tables (and fields) and

comparison predicate, namely restrictions posed on

the rows returned by a query. In addition, different

types of property restrictions correspond to different

cases (or patterns, where complex semantic query is

mapped) of SQL queries.

Where relevant entailments can be reasoned only

by property domain and range inferences, a set C

may be considered as necessary and sufficient for

representation of the semantic query. For example,

in openERP ontology (see Fig. 4), a DL query

“hasAccountAccountType some (hasCode value 3)”

returns all instances of account_account concept

whose type’s code is exactly 3. This kind of query

representation (only by using properties restrictions)

may produce unpredictable and misleading results

where the restrictions are posed on the common

LOCAL ONTOLOGIES FOR SEMANTIC INTEROPERABILITY IN SUPPLY CHAIN NETWORKS

lexical notions of different concepts, such as

“name”, “type”, “id”, etc. Ambiguity of the

corresponding properties is reflected on the relevant

ontology in the sense that their domains are typically

defined as union of large number of concepts. For

example, in openERP ontology, domain of the

“hasName” data property is union of 170 concepts.

However, this ambiguity may be considered as an

advantage in some cases. Value restrictions on

ambiguous data properties may produce relevant

inferences and thus, facilitate semantic querying

without a need to have extensive knowledge on the

underlying ontology structure. This kind of query is

mapped to a SQL UNION query which combines

SELECT sub-queries made on the each element of

the property domain, with the WHERE statement

corresponding to the relevant rows restrictions. For

example, in a mapping process, DL query “hasName

value ‘Derek Porter’” is first used to infer all 170

possible entailments (property domains), which are,

then, used to assemble qualified (O,C) pairs, e.g.

“res_users and hasName value ‘Derek Porter’”.

When corresponding element of the UNION query is

assembled, a static field with appropriate label (a

reference to the concept) is added to each of the

elements, so as to become possible to decide on the

entailments. In other words, we need this to

determine which sub-query actually returned the

results.

In the first step of the method, decomposition

and semantic analysis of the input query is

performed. The 4-tuplets in forms of (subject

predicate some|only|min n|max m|exactly o bNode)

and (subject predicate value {type}) are extracted

from the input query. In case of the DL query which

returns all concepts which are related to a company

whose primary currency is EURO

(“hasResCompany some (hasResCurrency some

(hasName value "EUR"))”), following 4-tuplets are

identified:

X hasResCompany some bNode1

bNode1 hasResCurrency some bNode2

bNode2 hasName value "EUR"

Next, a database connection is established and

SQL query is constructed and executed for each 4-

tuplet, in reverse order, as a result of analysis

described above. Each query returns data which is

used to generate OWL statements which are asserted

to a temporary model. Each set of the OWL

statements corresponds to a sub-graph whose focal

individual is an instance of the concept, inferred on

basis of the 4-tuplet’s property domain or returned

result (label). Other individuals or values correspond

to defining properties of this concept (inherited

anonymous classes). In case of ambiguity, resulting

blank nodes are represented as the sets, which are

filtered as a result of range inference of the parent 4-

tuplet, in a final stage of the method.

4 CONCLUSIONS AND FUTURE

WORK

Work, presented in this paper is a part of the

research of semantic interoperability in supply chain

networks. This research is based on formalization of

widely adopted supply chain process reference

model and includes development of its OWL

representation, semantically enriched model,

specification of some of its entities (namely, system

and goal) and correspondences with other models. It

transforms implicit semantics of the reference model

to the explicit specification which uses common

enterprise notions, assumingly defined in other

domain ontologies and/or conceptualizations of

relevant enterprise models, architectures and

frameworks. Used approach is characterized by the

multiple, cross-referenced levels of abstraction,

represented by the OWL models of different

expressivity. Modular design contributes to the

usability of the ontology framework, by avoiding

performance related problems in reasoning, as well

as by providing increased potential for ontology

matching. Thus, it is expected to facilitate the

semantic interoperability in supply chain networks.

In this paper, we focus on introducing the partial

realities of the enterprises, namely data

representations of their information systems, into

heterogeneous environment of a supply chain

network. In presented approach, enterprise data

models are used to generate local ontologies, by

applying a set of rules for interpreting the semantics

of an ER model, namely database schema. Although

“database to ontology mapping” is not a novel

concept, we show that existing approaches are

characterized by weaknesses, most of which are

related to lack of completeness of properties’

semantics. Our approach and corresponding tools

aim at overcoming those, thus enabling the complete

(from the aspect of OWL expressivity) interpretation

(explicitation) of the implicit semantics of the ER

model, as well as full correspondence between

semantic and database queries.

In context of the semantic interoperability in

supply chain networks, resulting local ontologies

may be considered as enterprise message models. As

such, they aim at enabling the semantic

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

interoperability of corresponding enterprise

information systems, not the enterprises themselves.

Still, significant research efforts are needed for

representation and exposition of the enterprise

business logic, which is hard-coded in the systems,

as well as the semantics of the instances, namely

information which is stored in the database (for

example, occurrence patterns). Another line of

research in the future will aim at enactment of the

generated ontologies, as they are considered only as

intermediary models. We consider those research

directions as important for increasing collaboration

in a supply chain network, as its fulfilment will

enable logic driven, automatic and transparent

decision making, thus, facilitating a transition from

traditional supply chains to virtual enterprise and

related paradigms.

ACKNOWLEDGEMENTS

The work presented in this paper was partially

supported by the program for scientific cooperation

between Serbia and France PHC PAVLE SAVIC,

project N

23494VF (2010/2011).

REFERENCES

Browne, J. and Zhang, J. (1999). Extended and virtual

enterprises – similarities and differences. International

Journal of Agile Management Systems, 1(1), 30-36.

Cullot, N., Ghawi, R. and Yetongnon, K. (2007).

DB2OWL: A Tool for Automatic Database-to-

Ontology mapping. In: Proceedings of the 15th Italian

Symposium on Advanced Database Systems. June, 17-

20. 2007, Torre Canne di Fasano (BR), Italy.

de Laborda, C.P. and Conrad, S. (2005). Relational.OWL:

a data and schema representation format based on

OWL. In: Proceedings of the 2nd Asia-Pacific

conference on Conceptual modelling. January 30 -

February 4, 2005. Newcastle, Australia.

Fox, M. S., Barbuceanu, M. and Gruninger, M. (1996). An

organization ontology for enterprise modeling:

preliminary concepts for linking structure and

behavior. Computers in Industry. 29 (1–2) 123–134.

Ghawi, R. and Cullot, N. (2007). Database-to-Ontology

Mapping Generation for Semantic Interoperability. In:

Proceedings of 3rd International Workshop on

Database Interoperability. September, 24. 2007,

Vienna, Austria.

Hepp, M. (2007). Ontologies: State of the art, business

potential and grand challenges. In Hepp, M., De

Leenheer, P., de Moor, A. and Sure, Y. (eds),

Ontology Management – Semantic Web, Semantic

Web Services and Business Applications (pp.3-22),

Berlin/Heidelberg: Springer.

Jespersen, B.D. and Larse, T.S. (2006). Supply Chain

Management - in Theory and Practice.

Copenhagen:Business School Press.

Konstantinou, N., Spanos, D., Chalas, M., Solidakis, E.

and Mitrou, N. (2006). VisAVis: An approach to an

intermediate layer between ontologies and relational

database contents. In: Proceedings of International

Workshop on Web Information Systems Modeling.

June, 6, 2006, Luxembourg.

Lamber, D.M. and Knemeyer, A.M. (2004, December).

We're in This Together. Harvard Business Review on

Supply Chain Management.

Obrst, L. (2003). Ontologies for semantically

interoperable systems. In: Proceedings of the 12th

International Conference on Information and

Knowledge Management. November 3-8, 2003, New

Orleans, USA.

Oldakowski, R., Bizer, C. and Westphal, D. (2005). RAP:

RDF API for PHP. In: Proceedings of the 1st

Workshop on Scripting for the Semantic Web, May

30, 2005. Heraklion, Greece.

OWL 2 Web Ontology Language (2009). Document

Overview. W3C Recommendation 27 October 2009.

http://www.w3.org/TR/owl2-overview/

Sánchez, N.G., Apolinar, D., Zubiaga, G., Atahualpa, J.,

González. I. and Molina, A. (2005). Virtual Breeding

Environment: A First Approach to Understanding

Working and Sharing Principles. In: Proceedings of

the 1st International Conference on interoperability of

Enterprise Software and Applications. February 23-25,

2005, Geneva, Switzerland.

Sowa, J. (2000). Knowledge Representation : Logical,

Philosophical, and Computational Foundations,

CA:Brooks/Cole Publishing Co.

Stewart, G. 1997. Supply-chain operations reference

model (SCOR): the first cross-industry framework for

integrated supply-chain management. Logistics

Information Management. 10 (2) 62-67.

SUO (2001). The IEEE Standard Upper Ontology web

site. http://suo.ieee.org

Williamson, O. (1985). The Economic Institutions of

Capitalism: Firms, Markets, Relational Contracting.

New York:The Free Press.

Xu, Z., Zhang, S. and Dong, Y. (2006). Mapping between

Relational Database Schema and OWL Ontology for

Deep Annotation. In: Proceedings of the 2006

IEEE/WIC/ACM International Conference on Web

Intelligence. 18-22. December, 2006, Hong Kong.

Xu, Z., Cao, X., Dong, Y. and Su, W. (2004) Formal

Approach and Automated Tool for Translating ER

Schemata into OWL Ontologies. Advances in

Knowledge Discovery and Data Mining, Lecture Notes

in Computer Science, Vol. 3056/2004, 464-475.

Zdravković, M., Panetto, H. and Trajanović, M. (2010).

Towards an approach for formalizing the supply chain

operations. In: Proceedings of the 6th International

Conference on Semantic Systems. September, 6-8,

2010, Graz, Austria.

LOCAL ONTOLOGIES FOR SEMANTIC INTEROPERABILITY IN SUPPLY CHAIN NETWORKS