ONT

OLOGY SUPPORT FOR PRODUCT DEVELOPMENT

Successful Application of Ontologies in Product Development

Reiner Anderl, Kai Mecke, Andre Sprenger and Oliver Weitzmann

Department of Computer Integrated Design, Darmstadt Technical University, Darmstadt, Germany

Keywords:

Ontology for Industry, Company Data Management, Product Development.

Abstract:

Today knowledge management is a very important factor for the success of enterprises. To support knowledge

management ontologies have been established. In the product development a lot of knowledge, methods and

workﬂows are used, which have to be protected, managed and distributed. The challenge to manage the diver-

sity of product development knowledge is to combine knowledge management methods and requirements of

product development. For this purpose we present functions to support access rights for ontologies, integration

of new information, release and storage of information (workﬂow). Another function to support the efﬁcient

work with knowledge is to use templates in order to reduce time and mistakes while integrating information

and knowledge in the system. These functions are integrated into the ontology-based product development

system (OPDS); the system fulﬁlls the requirements of the product development and used knowledge repre-

sentation forms.

1 INTRODUCTION

Enterprises are discovering the importance of knowl-

edge and identiﬁed knowledge as an important part of

the success factor for enterprises (Staab and Studer,

2004). To become a more successful enterprise they

have to store, distribute and use their knowledge.

Knowledge management becomes signiﬁcant and en-

terprises are willing to invest into knowledge manage-

ment.

The problem of the acquisition of knowledge is

that knowledge is bound to humans. If an employee

leaves the enterprise, the knowledge will be lost. Es-

pecially in the current demographic situation, a lot of

the employees knowledge will be lost, when they are

going into retirement (Schuett et al., 2007). Thus en-

terprises try to keep the knowledge in the enterprise

and this leads to the necessity of using knowledge

management.

According to Probst (Probst et al., 2003) knowl-

edge management is deﬁned as a knowledge process.

We can subdivide the knowledge process into two

categories. The ﬁrst category containing identiﬁca-

tion, acquisition and development, relies on the hu-

man ability to generate and extend knowledge. Com-

puter abilities in this ﬁeld stayed limited, but become

useful in the second category. In the second category

it is possible, that the elements of the knowledge pro-

cess can be supported by computer and new technolo-

gies. Knowledge distribution, Knowledge utilization

and Knowledge storage belong to this category. The

commitment of computers and information technol-

ogy (IT) contributes that the vast amount of infor-

mation will be controllable and guarantees the dis-

tributed use for everyone, who has access to the in-

formation.

A descriptive representation of knowledge in IT is

the ontology. Ontologies are abstract models of some

aspects of the real world (Gruber, 1993).These mod-

els consists of a collection of knowledge concepts,

properties and relationship between the knowledge

concepts (Kashyap et al., 2008).

Ontologies have been established as important

tools to represent knowledge for human and machines

in the knowledge management (Staab and Studer,

2004), (Tempich et al., 2006).

Because of the high ﬂexibility of ontologies it is

possible to use them in various thinkable contexts

to represent a part of real circumstances. Especially

complex structures like products and the product de-

velopment can be represented and supported by on-

tologies.

For creating ontologies, special-editors are used.

These editors are applications, which are developed

177

Anderl R., Mecke K., Sprenger A. and Weitzmann O. (2009).

ONTOLOGY SUPPORT FOR PRODUCT DEVELOPMENT - Successful Application of Ontologies in Product Development.

In Proceedings of the International Conference on Knowledge Engineering and Ontology Development, pages 177-182

DOI: 10.5220/0002298201770182

 SciTePress

for designing and manipulating ontologies. They vi-

sualize the complex structures of ontologies in clearly

presented graphs.

The question is, how enterprises could realize

knowledge management and what functions an enter-

prise would need to do it well. Ontologies are a pow-

erful method to implement knowledge management.

Therefore this paper will focus on enterprises in the

producing industry and especially in product devel-

opment.

There are only a few approaches that described

conceptual solutions of applications, which expand

ontologies with further functions to improve their

functionality. These approaches are our starting po-

sition and are presented in following paragraphs.

Montoni et al. (Montoni et al., 2004) present an

ontology-based Enterprise-Oriented Software Devel-

opment Environment tool. With the help of Taba

Workstation, a software development environment

(Montoni et al., 2004), the different software pro-

cesses will be controlled.

Wu and Chen (Wu et al., 2006) developed an

ontology-based Role-based access control model. In

the ontology they integrated all information and their

relationship, which are necessary for security policy.

Especially the expansion of semantic information to

the model is an innovative approach.

Using ontologies in industry requires, that an

ontology-based application is supported by ontology-

lifecycle model. Tran et al. (Tran et al., 2007) devel-

oped an information-model for lifecycle management

of ontologies. In a second step they presented in (Tran

et al., 2007) a generic OIS Architecture with lifecycle

supports.

In this contribution will be discussed which ele-

ments are necessary in regard of developing knowl-

edge management system, which applies all industry

requests. In the second chapter the requests will be

deﬁned. In chapter three the concept regarding the

deﬁned requests will be proposed.

2 INDUSTRY REQUESTS

For a successful introduction of semantic technolo-

gies in the industry, a number of requirements have to

be met. Existing workﬂows of product development

have to be supported and additional requests regard-

ing security, collaboration and social networks con-

sidered.

Figure 1: Product development and Product Life (Birkhofer

et al., 2009).

2.1 Product Development

Product development is the process, which is neces-

sary to get a product from an idea to a salable solu-

tion. This process is subdivided in several succeeding

processes. The single processes are shown in ﬁgure

1. To support the product development process the

VDI-guideline 2222 has deﬁned tasks for these pro-

cesses (VDI, 1997). The product development pro-

cess is also documented by several authors (Spur and

Krause, 1997) and (Pahl et al., 2003).

The product development is part of the Product

Lifecycle. This process describes the whole life of

products from planning to disposal. It is divided in

the major categories of a product life: plan, design,

build, support, and dispose. (Grieves, 2006)

The key to success of enterprises is the reduc-

tion of time consumption of the phases of Product

Lifecycle plan, design and build. For this purpose

the methods Simultaneous Engineering, parallelizing

processes to decrease time and Concurrent Design,

subdividing construction space and working simulta-

neous at each space are developed.

Product Data Management (PDM). Product data

management is part of the product data technology

and it is used for the processing of product data

(Grabowski et al., 1998). In the product lifecycle

product data management (PDM) supports the man-

agement of product data and controlling of product

data ﬂows.

2.2 Practical Requests

Distributed Product Development. The develop-

ment of a product at multiple research centers around

the globe is a special challenge for every company

dealing with complex products. In case of the inte-

gration of a subcontractor or supplier and the con-

sideration of variants of the product the challenge in-

KEOD 2009 - International Conference on Knowledge Engineering and Ontology Development

178

creases signiﬁcantly. Several concepts, like ”‘follow

the sun”’, ”‘Function sharing”’ and Module sharing

have been developed.

All approaches rely on common distributed data,

which is presented to the participating developers in

an intuitive manner. Ontologies have the basic pos-

sibility to make everyone understand the relation be-

tween concepts. As a web technology, the distribu-

tion of information in an intranet, analog to the World

Wide Web, is easy.

Styling, Design and Management. Styling deals

with shapes, impressions and appearances of a prod-

uct in early phases of the product development pro-

cess. Contrary to this, the design process from the en-

gineering side considers technical requirements, func-

tions and construction space, an abstract approach

considering a high granularity of the product (Klug,

2007).

A common knowledge base is a necessary step for

interdisciplinary product development. An important

support of this step is graphical, intuitive presenta-

tion of relevant information and relations which can

be provided by ontologies. The development process,

supported by graphical mapping techniques (”Design-

mapping”) is described in (Malzacher et al., 2008).

Intellectual Property Protection. Impressive occur-

rences of plagiarism have sensitized the industry to

protect intellectual property as a key factor to stay one

step ahead of competitors (Nitterhouse, 2003).

Industry espionage has increased over the last

decades. Therefore access control to information is

needed, ensuring that it is available at the right place,

to the right person at the right time (iViP Association,

2008).

There are two basic strategies for intellectual

property protection information hiding and informa-

tion ﬁltering.

The idea of securing electronic business docu-

ments is an increasing necessity nowadays. (Alkassar

et al., 2007).

2.3 Need for Action

From the compact overview of industry requirements

the need for action can be derived. Even though on-

tologies can easily satisfy some of these requirements,

such as being a web technology, some requests lead to

unsolved problems.

The storage of distributed information is not a

challenge for a web technology. The unique and uni-

ﬁed identiﬁcation of ontology elements in the web

and the vast possibilities of interconnection of dis-

tributed ontologies can be a strong support for global

collaboration in product development.

The emerging structures in interdisciplinary prod-

uct development can be reduced to graph structures.

With additional rules of an appropriate expressivity

level, these structures can be represented by ontolo-

gies as a uniform datamodel.

The lack of a suitable speciﬁcation of semantics in

enterprise models leads to inconsistent interpretations

and uses of knowledge. Reasoned by (Grueninger

et al., 2000) it is necessary to establish a uniform data

model.

The protection of intellectual property is a central

challenge for the use of ontologies in product devel-

opment. Basically knowledge in ontologies is acces-

sible to every user. Data, which are referenced by

the ontology, is protected by access control mecha-

nism of an additional datasource or data system. The

basic idea of information for everyone is useful for

the application in the World Wide Web, but not in a

company environment of research and development,

where data leakages can be fatal for the product suc-

cess.

Contemporary ontologies-editors are not suitable

for the use of knowledge-management in enterprises.

Therefore, an enhanced ontology-editor/access sys-

tem has to be developed, which increases the usability

for enterprises.

3 CONCEPT

We deﬁned a system called Ontology-based Product

Development System (OPDS). This sketched system

is able to support every phase of the product develop-

ment process. It allows retrieving information about

the design, the decisions and the know-how of a prod-

uct. Especially know-how from manufacturing, uti-

lization, recycling etc. like uncertainties (e.g. dimen-

sion tolerance, production accuracy etc.) can be rel-

evant factors of designing of high-quality products.

So the information from later phases of the product

lifecycle can be stored in the system, design engi-

neers get the information and consider this by the

design process. This system is distinguished from

PDM-Systems, because the use of ontologies for rep-

resenting knowledge and especially the representing

of knowledge is not possible for PDM-Systems today.

To manage these requirements it is necessary to

integrate new functions in the ontology-editor, in or-

der to become an Ontology-based Product Develop-

ment System.

In the following sections the different functions

and templates will be explained.

ONTOLOGY SUPPORT FOR PRODUCT DEVELOPMENT - Successful Application of Ontologies in Product

Development

179

3.1 Functions of OPDS

The requirements of enterprises towards an applica-

tion to manage their knowledge are to save and to

protect their knowledge, so that the application has to

control the access and also to control and to support

the enterprise workﬂows.

Based on the requirements the application must

have different functions, which are

• the elementmanagement,

• the rightsmanagement,

• the workﬂowmanagement and

• the datamanagement.

The functions will be explained in the following sec-

tions.

3.1.1 Elementmanagement

Ontologies’ main function is to store and to manage

knowledge and information. To reuse and retrieve

knowledge and information it is necessary to build a

uniform described element model. The component of

this element model is attributes to identify and clas-

sify the elements. This information is stored in meta-

data.

Identiﬁcation attributes have to be interpretable by

application and user. An attribute consists of an iden-

tiﬁcation number and of a human interpretable name.

This name is subject to a naming convention, which

is given by the enterprise restrictions.

Through the classiﬁcation of elements the belong-

ing is given, the element is extended by codiﬁcation,

which includes the organizational structure of the en-

terprise. The codiﬁcation is a part of the classiﬁcation

attributes.

The elementmanagement supports the organisa-

tion of the knowledge management and the users to

retrieve data.

3.1.2 Rightsmanagement

Standard ontologies provide information to all users.

Everyone has access to get this information. There-

fore a restriction of access is necessary in form of a

rightsmanagement, so that only authorized users have

to access to the information. The rightsmanagement

includes user management, aggregation to groups,

controlling the access to information amd controlling

the access to functions.

To control the rightsmanagement a system module

called role-based access control (RBAC) will be inte-

grated. A concept to implement an ontology-based

RBAC is mentioned by (Wu et al., 2006).

In the following paragraphs the elements of the

rightsmanagement will be described in detail.

User Management, Groups, Roles and Rights The

access to the system is only for authorized users. The

authorization process requires a restriction of the user.

In conjunction with the account, the user belongs to

special groups and roles. This helps to organize the

user into groups and roles and which different rights

to access the different information they have.

A user will be a part of a special ontology, which

includes all information about users, their groups,

their roles and further information about the person.

The Access Control Module uses the user-ontology to

check, if the user is allowed to access the application.

The module distinguishes between the two basic

roles of the system: user and administrator.

Controlling the Access to Information. The pieces

of information of the ontology are the central el-

ements and have to be protected by the system.

Through the protection code, which is part of the

metadata, the information about the kind of access

rights a user has got is stored.

Controlling the Access to Functions. Analog to the

access to information it is also necessary to control

the access to functions. To support the product devel-

opment process and to organize the system the OPDS

have to provide functions, which supports e.g. several

workﬂows.

Not every user should have the possibility to use

every function, because he has not got the competence

to decide the circumstances. The restriction of access

to function is a protection against misuse.

3.1.3 Workﬂowmanagement

In the product development there are a lot of work-

ﬂows like releases, change, versioning etc., which are

supported by the OPDS to get consistency. Even in-

formation and knowledge are subjected to changes,

releases and versioning.

Through the distributed work, which is required

by Simultaneous Engineering and Concurrent Design,

the OPDS is suited to control the exchange workﬂow

and to start workﬂows automatically(Schuschel and

Weske, 2003).

A general function of workﬂowmanagement is the

management of information, which has been stored

in OPDS. Further functions for workﬂowmanagement

are constrained by the requirements of the enterprise.

Besides the automated control of workﬂows, the

knowledge about these workﬂows is also represented

in workﬂow ontology. Doing so, the staff understand

the ﬂow between the processes.

KEOD 2009 - International Conference on Knowledge Engineering and Ontology Development

180

In following paragraphs the three basic functions

for the management of knowledge and information

will be explained.

Inspection Procedures. To guaranty the consistency

of the ontologies, inspection procedures like change-

management and releasemanagement are necessary.

These inspection procedures consist of a set of rules.

The modiﬁed structures are checked against these

rules, before they get the release to be published for

usage.

State Change. The knowledge is subjected to per-

sistent change, new experiences lead to new knowl-

edge, which has to be stored in the ontology. Before

new knowledge will be used, it has to be checked by

authorized staff, which has the competence to decide

(Tempich et al., 2006).

Versioning. A permanent Knowledge acquisition is

a factor of success. One part of Knowledge acquisi-

tion can be the modiﬁcation of information. By the

modiﬁcation of information a new version of infor-

mation with a version number to distinguish will be

stored in the ontology. The versioning supports the

integration of the new information in the existing on-

tology, therefore there is a versioning assistant nec-

essary, which tracks the relations of the old version

and give the choice to integrate the new version with

existing relations.

3.1.4 Datamanagement

The use of Ontologies leads to a lot of data and infor-

mation. To store this data and information a database

is preferred. Because databases enable a structured

storage and scalability, fast access and save transac-

tions.

This character of databases permits a central man-

agement of data. Another reason for using database

is the factor of safety; through the encapsulation it is

not possible to access the database without the mid-

dleware.

3.2 Templates

A Template is a standard pattern, which includes

abstract information about repeating data constructs.

The beneﬁt of templates is that they are implemented

once and they can be used for several similar cases.

Templates will support and improve the imple-

mentation, the uniﬁcation and the recognition of sim-

ilar structures. The users only have to complete the

data content and have not build up a new structure.

With template features are being integrated in

OPDS, it is possible to generate templates for spec-

iﬁed problems. The templates are independent from

application versions.

To support, for example, projectmanagement the

basics structures of a project will be represent in a

template, so that a user only have to complete the

speciﬁed project data. This reduces the time to ini-

tialize new projects and minimizes the probability to

make mistakes.

3.3 Model of OPDS

Based on the representation of knowledge in form of

ontologies a model for a system to control knowledge

management in enterprises was developed.

Through the presentation in form of a graphical

user interface the user/administrator can access the

OPDS.

Before he can enter the ontology, he has to log in

and the RBAC checks if he is authorized. The RBAC

belongs to the workﬂowmanagement, which as well

controls the workﬂows Versioning, Change Manage-

ment and Inspection Procedures.

Authorized user have the possibility to create tem-

plates, therefore they can use extracted structures

from the ontology and for further knowledge acqui-

sition.

4 CONCLUSIONS

In this paper we discussed, that knowledge manage-

ment is an important factor for enterprises to become

successful in developing new concepts. The integra-

tion of knowledge management is a challenge for en-

terprises.

To support knowledge management, ontologies

have been established as form of representation. Es-

pecially in the product development the complexity

of knowledge, information, methods and workﬂows,

which are necessary to be protected and to be man-

aged, requires the commitment of IT and knowledge

representation forms.

To fulﬁll these tasks and requirements we devel-

oped an Ontology-based Product Development Sys-

tem (OPDS). The OPDS has several functions to con-

trol permission to ontologies (RBAC) and manipula-

tion of the ontologies. Furthermore OPDS supports

the enterprise workﬂows to assure the technical and

the content consistency.

To make work with ontologies easier and to guar-

antee the consistent description, templates will use

master to provide the integration of information.

The commitment of OPDS in enterprises leads to

faster access to information and knowledge; therefore

ONTOLOGY SUPPORT FOR PRODUCT DEVELOPMENT - Successful Application of Ontologies in Product

Development

181

Figure 2: Model overview of complete system.

products can be developed faster.

REFERENCES

Alkassar, A., Husseiki, R., Stble, C., and Hartmann, M.

(2007). ISSE/SECURE 2007 Securing Electronic

Business Processes. Vieweg.

Birkhofer, H., Kloberdanz, H., Mathias, J., and Engelhardt,

R. (2009). Integrated product and process develop-

ment based on robust design methodology. Proceed-

ings of the International conference on engineering

design ICED 2009.

Grabowski, H., Anderl, R., and Polly, A. (1998). Integri-

ertes Produktmodell. Entwicklungen zur Normung von

CIM. Beuth-Verlag.

Grieves, M. (2006). Product Lifecycle Management.

McGraw-Hill.

Gruber, T. (1993). A translation approach to portable ontol-

ogy speciﬁcations. Knowledge Acquisition. Academic

Press Inc., 5(2).

Grueninger, M., Ateﬁ, K., and Fox, M. S. (2000). Ontolo-

gies to support process integration in enterprise engi-

neering. Computational & Mathematical Organiza-

tion Theory, 6(4).

iViP Association, P. (2008). Secure product creation pro-

cesses (sp2). Technical report.

Kashyap, V., Bussler, C., and Moran, M. (2008). The Se-

mantic Web - Semantics for Data and Services on the

Web. Springer-Verlag.

Klug, L. (2007). Methodischer Einsatz von parametrischen

Prototypen in der Produktentwicklung. PhD thesis,

Techn. Univ. Darmstadt.

Malzacher, J., Rambo, J., Geis, C., and Richter, M. (2008).

Designmapping: Ein arbeitsmittel fr die kollabora-

tive produktentwicklung zwischen ingenieuren und

designern. In Hentsch/Kranke/Wlfel:, editor, Indus-

triedesign und Ingenieurwissenschaften. 2. Sympo-

sium Technisches Design Dresden 2008.

Montoni, M., Santos, G., Villela, K., Rodrigo Miranda, A.

R. R., Travassos, G. H., Figueiredo, S., and Mafra,

S. (2004). Knowledge management in an enterprise-

oriented software development environment. PAKM

2004, LNAI 3336, pages 117–128.

Nitterhouse, D. (2003). Plagiarism - not just an ”academic”

problem. Teaching Business Ethics, 7(3):215–227.

Pahl, B., Beitz, W., Feldhusen, J., and Grote, K. (2003).

Konstruktionslehre - Grundlagen erfolgreicher Pro-

duktentwicklung - Methoden und Anwendungen. 5.

Auﬂage. Springer-Verlag, Berlin, Heidelberg.

Probst, G., Raub, S., and Romhardt, K. (2003). Wissen man-

agen: Wie Unternehmen ihre wertvollste Ressource

optimal nutzen. Gabler-Verlag, Wiesbaden.

Schuett, P., Bentele, M., and Weber, M., editors (2007).

Positionspapier: Wichtige Trends im Wissensmanage-

ment 2007 bis 2011, Berlin. BITKOM Bundesverband

Informationswirtschaft, Telekommunikation und neue

Medien e.V.

Schuschel, H. and Weske, M. (2003). Lecture Notes in

Computer Science, chapter Integrated Workﬂow Plan-

ning and Coordination, pages 771–781. Springer-

Verlag.

Spur, G. and Krause, F.-L. (1997). Das virtuelle Produkt:

Management der CAD-Technik. Carl Hanser.

Staab, S. and Studer, R., editors (2004). Handbook on On-

tologies. Springer-Verlag.

Tempich, C., Pinto, H., and Staab, S. (2006). Ontology

engineering revisited: An iterative case study. LNCS,

4011:110–124.

Tran, T., Haase, P., Lewen, H., scar Muoz-Garca, Gmez-

Prez, A., and Studer, R. (2007). Lifecycle-support in

architectures for ontology-based information systems.

ISWC/ASWC 2007, LNCS 4825, pages 508–522.

VDI, editor (1997). VDI-Richtlinie 2222 Blatt 1: Kon-

struktionsmethodik - Methodisches Entwickeln von

Lsungsprinzipien. VDI-Verlag GmbH, Dsseldorf.

Wu, D., Chen, X., Lin, J., and Zhu, M. (2006). Ontology-

based rbac speciﬁcation for interoperation in dis-

tributed environment. Lecture Notes in Computer Sci-

ence, 4185:179–190.

KEOD 2009 - International Conference on Knowledge Engineering and Ontology Development

182