Int. J. , Vol. x, No. x, xxxx 1
Copyright © 2016 Inderscience Enterprises Ltd.
A reference ontology approach to support global
product-service production
Claire Palmer, Esmond Neil Urwin, and Robert Ian Marr
Young
*
Wolfson School of Mechanical and Manufacturing Engineering,
Loughborough University, UK
Email:
c.palmer3@lboro.ac.uk
Email: e.n.urwin@lboro.ac.uk
Email: r.i.young@lboro.ac.uk
*Corresponding author
Eugenia Marilungo
Università Politecnica delle Marche,
Ancona, Italy
Email:
e.marilungo@pm.univpm.it
Abstract: The need to innovate and compete drives organisations to
constantly seek new approaches to facilitate business and commerce. As
market places become ever more globalised and digital economies grow,
these organisations rely more heavily upon systems to design and deliver
their products and services. Hence, when developing and operating a
global production network the need for systems to interoperate between
different domains and contexts within a global production network
becomes paramount if organisations are to succeed. This paper puts
forwards a reference ontology that has been developed to enable the
interoperation of software tools involved in the global production of new
product-services systems (PSS). It sets out the levels of the reference
ontology, detailing closely the product-service aspects. This has been
developed using a formal logic based approach. An example knowledge
base has been created from industrial end user information with queries
applied to this to provide a set of results showing the ability of the
reference ontology.
Keywords: ontology, reference ontology, product-service system (PSS),
global production network
1 Introduction
To derive commercial and technological advantage in complex global
markets requires companies to innovate and differentiate the way in which
products and services are specified, designed and manufactured. One way
to achieve this which has gained significant industrial interest is to move
from a product-centred perspective to a more service-centred perspective,
with many companies exploring how best to move from selling products
to delivering product-service systems (PSS). The push towards
servitisation (Vandermerwe and Rada, 1988; Goedkoop et al. 1999;
Baines et al., 2007) and the selling of products via services is becoming
ever more popular and more notably, profitable.
A full product-service system is aimed at fulfilling customer needs by
selling the utilisation of a product i.e. the system, instead of providing just
the physical product (Cook et al. 2006). It is also important to note that
there are ranges of possible options that lie between a pure physical
product and a full product-service system, where some levels of service
are provided along with the purchase of a physical product. An example of
this intermediate type of product-service is used in the experimental
section of this paper.
Product-Service Lifecycle Management solutions have recently been
helping the development of such PSS by improving the ability to organise,
share and reuse information within and across organisations (Peruzzini et
al., 2014). However, interoperability has long been recognised as
fundamental to the ability to effectively share information between
software tools (Li et al, 2006) and similarly semantic interoperability,
through the development of formal ontologies, has been recognised for
many years as a key factor in achieving improved interoperability
solutions (Gruber 1993).
Since Gruber’s work a vast amount of ontology research has been
undertaken across a wide range of target areas from very general purpose,
or foundation ontologies, to very specific domain ontologies. There is a
growing recognition of the need for ontologies that lie between these two
extremes, sometimes called reference ontologies (Borgo 2007, Young et al
2009), to which this paper makes a contribution.
The aim of the FLEXINET project, from which this paper originates,
is to create and develop a collaborative heterogeneous decision support
environment to aid the design, configuration and reconfiguration of global
production networks to meet new product ideas, including product-service
A reference ontology approach to support product-service production
ideas. The FLEXINET research project has developed a reference
ontology to support information sharing and thereby aid the decisions
made within the early phases of PSS design.
The development of a reference ontology within the FLEXINET
project has followed a qualitative approach. It has exploited the Noy and
McGuiness (2001) knowledge engineering methodology together with
Yin’s (2013) case study methodology to guide the development of the
reference ontology. The project has worked across three industrial sectors;
white goods, industrial pumps, and food and drinks. They have provided
an important source for derivation of user requirements and provision of
industrial data and information for the research and offered different
perspectives from which to develop and assess the research. While the
reference ontology that has been developed covers a broad range, the
significant elements from the point of view of product service systems and
the production of such systems are the focus of this paper.
The paper is structured as follows. Section 2 provides a review of the
relevant literature and how it reflects upon the FLEXINET approach. The
FLEXINET reference ontology is described in section 3 with section 4
providing experimental results of the software implementation. These
results are then discussed in section 5 and conclusions drawn in section 6.
2 Related work
How to design a product-service system and setup the necessary support
networks to facilitate its services is often complex and not always well
understood (Gebauer et al., 2005; Baines et al., 2009; Spring and Arunjo,
2013). This can be attributed to a number of factors, the first being how
best to integrate and accommodate the more common and accepted
product lifecycles with the newer and progressive service lifecycles. The
second is how to account for and overcome organisational, cultural, social
and geographical differences between suppliers, producers and service
providers within the context of a diverse and divergent Global Production
System. Finally, the third factor underlying all of these issues is that of
interoperability (Li et al, 2006), it becomes paramount to be able to
process and interpret information correctly and succinctly from all parties
between potentially varied and diverse information systems and contexts
to quickly and successfully design, implement and support PSS. The last
of these three is the main focus of this paper.
2.1 Product Service Systems
There are strategies and methods that currently exist that are
designed to help organisations design for PSS and together with analysis
of those strategies and methods (Spring and Arunjo (2013), Gaiardelli et
al. (2014), Muto et al. (2015), and Vezzoli (2015)). Gaiardelli et al. (2014)
set out a classification model for product service offerings. This consists
of the main types and characteristics relating to product-service oriented
services, use-oriented services, and results oriented services, (as put
forward by Tucker (2004)) together with the product-oriented service
space in an effort to unify and present a multi-dimensional model for the
PSS domain. As such, this can be useful in helping organisations decide
where PSS offerings can be aimed at for the market place. Spring and
Arunjo (2013) put forward a model of service development in
manufacturing focusing on the nature of connection between product and
service. It is an interesting approach because it considers the ability to
reconfigure networks and resources for the making of a product-service.
This is an aspect that is potentially very useful on a global scale.
Additionally, there is work published concerning partnership perspectives
relating to the earlier stages of product-service system lifecycle
management, but, as Lockett et al. (2010) point out that there is little study
of supply networks for PSS, specifically the management and sharing of
information between supply chain partners.
Considering the PSS domain there are a number of European
Framework Programme 7 (FP7) projects that have produced research that
contributes to it. In particular the POP* (Process, Organisation, Product
and others) (Athena, 2006) methodology focused on the domain of
enterprise collaboration and developing methods to capture design and
management concerns within it. The Manufacturing Service EcoSystem
(MSEE) project, focused upon service orientation and collaborative
innovation for Virtual Factory Industrial Models (MSEE, 2014). This
sought to model intangible and tangible assets relating to manufacturing
activities using OWL based formal semantics. The FALCON project
(Falcon, 2016) utilises machine translation and automated term extraction
to construct a network of terms and translations inter-linked to each other
and to source and target documents via URLs. Within the Product Domain
the FP7 LinkedDesign Project has developed an integrated software
platform (LEAP) to support designers, engineers and manufacturers
throughout the product’s lifecycle, from design, through installation and
maintenance to disposal (LinkedDesign, 2014a). The platform contains
tools capable of analysing information knowledge provided by different
A reference ontology approach to support product-service production
data sources present in the LinkedDesign application, such as dimensional
quality control and product lifecycle costs.
2.2 Interoperability and ontologies
To deliver and develop effective Product Service Systems creates a
need to generate, access, process and utilise information seamlessly and
quickly between information systems. Thus, interoperability is of crucial
importance if information generated by different systems is to be
meaningfully applied to develop PSS. To achieve semantic
interoperability enabling machine-understandable data to be shared by
multiple organisations requires an ontological engineering process (Lee et
al, 2009)
Recent ontological approaches within the product domain are those
of Lee et al. (2009) and Panetto et al (2012). Lee et al (2009) suggest a 4-
layered product ontology architecture for an integrated value chain to
achieve interoperability in the product domain. The architecture is based
on the model driven architecture (MDA) with the two upper layers
consisting of generic meta-models, the third layer being the domain
ontology and the fourth layer the ontology instance. The ONTO-PDM
ontology (Panetto et al, 2012) is based on the hypothesis that an
ontological model of a product may be used to facilitate the interoperation
of application software by enabling information sharing during the
physical product lifecycle. ONTO-PDM consists of a common core model
based on the standards ISO 10303 and IEC 62264. ONTO-PDM is based
on First Order Logic and implemented by translation into OWL.
Considering the service domain, ManuHub (Cai et al, 2011) is a
prototype Semantic Web system which enables the use of manufacturing
services within distributed manufacturing environments. ManuHub
achieves manufacturing interoperability by facilitating automatic retrieval
of the required manufacturing services which are derived through a
feature-based semantic matchmaking process. Structural knowledge is
formalized using Ontology Web Language and constraint knowledge
through the use of Semantic Web Rule Language (SWRL).
A formal representation of a PSS ontology is put forward by
Annamalai et al. (2010). This was developed by way of a joint effort with
multiple academic, industrial and governmental organisations contributing
to its development. The ontology focuses on the constituent parts of a PSS
ontology, those being: needs and requirements, stakeholders, business
model perspectives, product-services, their associated lifecycle aspects
and the resultant outcomes of a PSS. It is interesting in that it sets out view
upon higher level domain concepts as constituent parts for an overall