Strengthening Competitiveness

through Production Networks

A perspective from European ICT research projects in
the field of ‘Enterprise Networking’















2005 Edition




LEGAL NOTICE
By the Commission of the European Communities, Information Society and Media Directorate-General.
Neither the European Commission nor any person acting on its behalf is responsible for the use which might
be made of the information contained in the present publication. The European commission is not responsible
for the external web sites referred to in the present publication.
The views expressed in this publication are those of the authors and do not necessarily reflect the official
European Commission’s view on the subject.
Europe Direct is a service to help you find answers
to your questions about the European Union
Freephone number (*):
00 800 6 7 8 9 10 11
(*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls
may be billed.
A great deal of additional information on the European Union is available on the Internet.
It can be accessed through the Europa server (http://europa.eu.int).
Cataloguing data can be found at the end of this publication.
Luxembourg: Office for Official Publications of the European Communities, 2005
ISBN 92-894-9924-9
© European Communities, 2005
Reproduction is authorised provided the source is acknowledged.
Printed in Belgium
PRINTED ON WHITE CHLORINE-FREE PAPER

Foreword
Research is still too fragmented in Europe, including in the ICT for
enterprise networking domain. Therefore, the European Commission
is promoting a structural approach aimed at integrating collaborative
relevant RTD at a pan-European level, thus contributing to the
creation of a genuine and powerful European Research Area (ERA).
One way to achieve this has been the clustering of research projects,
intended to promote exchange between projects working in related
topics, allowing their individual partners to come together and to share
their knowledge and experiences for mutual benefit. Project clusters
also provide a common basis for dissemination among and interaction
with people interested in the domain.
This report is an initiative of the “Ambient Intelligence Technologies for the Product Lifecycle”
(AITPL) cluster. It is an instrument to widely disseminate recent results obtained by projects from this
cluster and by other projects in the ICT for Enterprise Networking domain. All these projects share in
common the aim – and the passion – to tackle the challenge of pinpointing strategies for further
research in new forms of dynamic networked co-operative processes, and of keeping Europe’s
manufacturing industry not only alive, but fully competitive and in a strategically leading position.
The papers presented here outline some of the major challenges which face the community of
European researchers in the e-business area. Although these papers are almost devoted to a single
debate – is there such a thing as a new way to develop production networks in the knowledge-based
economy? – they exemplify a number of innovations deemed to support integrated enterprise
modelling for supply business processes, Build-to-Order strategies for vehicle design in the
automotive industry, co-operative SME networks, the interoperability of enterprise systems and
applications, standardised business processes for effective collaboration across the full supply chain,
cross industry and end-user involvement in RFID standardisation, Web-based services for cost
engineering, and ICT integration and the simultaneous collaboration amongst the companies in the
entire supply chain to achieve far-reaching goals (e.g. the “5-day car”).

The AITPL cluster starts from the conviction that future products and services will be designed to
offer customers more value and enable manufacturers to respond faster and in a flexible manner to
changing market demands. Customers are demanding ever more advanced and sophisticated products,
greater choice and shorter delivery times. To satisfy this demand for differentiated and customised
products at competitive prices, companies with different expertise must collaborate. But they need to
do so in ways which ensure that the value chain remains flexible, so as to realise the full benefits of
rapid product innovation and open competition. In addition, manufacturers are looking to make their
products “smarter” by designing in added-value services as part of the customer offering. This
“extended product” approach combines a product with services and enhancements that improve
marketability. The customer proposition may subsist more in the benefits of the value-added elements
than in the physical product itself. Enhancements can incorporate tangible features that make the
product more intelligent, customised or user-friendly, including embedded features like maintenance.
Other aspects, such as services, engineering or software, are intangible and make the offering more
information- or knowledge-intensive.
Secondly, the cluster believes that the strength of the European economy is increasingly based on
relationships among many enterprises, which together form agile networks, able to react to market
demands in shortest time. These networks (sometimes developed as a virtual enterprise for a specific
product) are still competing successfully on a global scale with enterprises from distant countries,
where labour costs are just a fraction of those in Europe. This success can be retained, only, if the
networks establish and maintain smooth interactions, which cover the complete life cycle of the
product.
3

The challenge facing supply chain management is easy to define but less easy to run! It is to provide
the right products in the right quantity at the right place and time. As the papers in this publication
demonstrate, any supply chain strategy should address key themes, such as the growing gap between
the financial performance of companies with disciplined supply chain management and the financial
performance of those with ‘ad hoc’ supply chain management, the development of end-to-end
processes (rather than local/functional processes) and their integration beyond organisational
boundaries to realise the benefits of vertical integration without integrating, the management of
uncertainty through various techniques of forecasting (‘range forecasting’) and risk pooling (to build
flexibility), and the capability of supply chains to evolve as business conditions change.
Meanwhile, as companies are shifting from integrated, vertical business models to horizontal models
that rely heavily on outside vendors and suppliers, the entire supply chain needs to be strategically
aligned. While managing the four ‘clusters’ of the ICT for Enterprise Networking unit, we are well
aware that getting everyone working in concert (concertation) is very difficult – inside the four walls
of a single organisation and, even more, when it is other organisations that are performing critical
activities that a particular organisation used to handle itself. How to ensure that everyone has the same
priorities and, more fundamentally, that all stakeholders within a collaborative business network speak
the same language? The answer lies in large part in the strategic use of performance metrics, i.e.
identifying the key supply chain indicators that drive the overall success of business and developing
effective means of tracking these indicators.
Work in the AITPL cluster has a high economic and political relevance. Because the stakes grow
higher each year, we must change course now. Improving the knowledge of the product and
production processes will in the longer term imply that the required engineering knowledge is
maintained, making Europe indeed more competitive on a global scale and enhancing the prospects for
employment in the networked economy
1
. The European Council report of March 2004 identified de-
industrialisation as a risk and requested concrete steps from the European Commission to improve the
competitiveness of the European industry
2
. In addition, the Kok report (November 2004) stated that in
the period 1996-2003 ICT investments contributed to half of Europe’s productivity gains. Therefore,
an ICT-focus on innovation in the product lifecycle seems a good measure to reduce the “productivity
divide” between the EU and the most dynamic regions in the world, to combat
the risk of de-industrialisation, and finally to get close to achieving the Lisbon
goal of making the EU the most competitive knowledge-based economy in the
world.
Overall, without question we face immense challenges in Europe, as we work
to keep our union and its member states at the forefront of the hyper
competitive global knowledge-based economy. Despite the enormity of these
challenges, we in the Commission are optimistic that we can build a better and
stronger Europe thanks to the leadership, the skills and devotion of so many
outstanding professionals in academic and industrial research teams.
The views expressed in the articles are those of the authors and do not necessarily reflect those of the
European Commission. We would like to express particular thanks to the VERITAS project for co-
ordinating the contributions and editing of the book.

Gérald Santucci Florent Frederix
Head of Unit Cluster Officer


1
Although manufacturing creates less then 30% of the European jobs directly, it indirectly creates more then
70% of the European jobs if the full product lifecycle is considered (Manufuture 2003 conclusions).
2
Presidency Conclusions – Brussels, 25/26 March 2004, comm. SN 100/04, paragraph 17.

4

Table of Contents




Foreword.................................................................................................................................................................
1 1. . I In nt tr ro od du uc ct ti io on n: : N Ne et tw wo or rk ke ed d o or rg ga an ni is sa at ti io on ns s a as s a a c co or re e d dr ri iv ve er r t to o c co om mp pe et ti it ti iv ve e a ad dv va an nt ta ag ge e i in n t th he e 2 21 1s st t c ce en nt tu ur ry y....
2 2. . O Op pe er ra at ti io on n o of f t th he e s su up pp pl ly y c ch ha ai in n.................................................................................................................
2.1 Analysis and comparison of supply chain business processes in European SMEs.................................
2.2 Making Build to Order a reality: The 5-Day Car Initiative.....................................................................
3 3. . E Es st ta ab bl li is sh hi in ng g a an nd d m ma ai in nt ta ai in ni in ng g S SM ME E n ne et tw wo or rk ks s .........................................................................................
3.1 Co-operative SME networks in manufacturing .....................................................................................
3.2 Issues in the Management of COllaborative DEmand and Supply NETworks.......................................
4 4. . I In nt te er ro op pe er ra ab bi il li it ty y a an nd d s st ta an nd da ar rd di is sa at ti io on n........................................................................................................
4.1. Interoperability of Enterprise Systems and Applications.......................................................................
4.2. E-Business Standardisation in the Automotive Sector – The situation of SMEs....................................
4.3 Challenges in the adoption of RFID standards .......................................................................................
5 5. . T To oo ol ls s a an nd d t te ec ch hn no ol lo og gi ie es s s st tr re en ng gt th he en ni in ng g c co om mp pe et ti it ti iv ve en ne es ss s ..........................................................................
5.1 Virtual Cost Engineering Studio (V-CES): a new service for cost engineering professionals ...............
5.2 ICT as an Enabler to the 5-Day Car: A Central Challenge to the ILIPT Project ..................................
About the Authors...............................................................................................................................................
Index of Keywords (pages) .................................................................................................................................

5
3
7
14
14
26
38
38
47
58
58
71
81
91
91
102
113
117


1 1. . I In nt tr ro od du uc ct ti io on n: : N Ne et tw wo or rk ke ed d o or rg ga an ni is sa at ti io on ns s a as s a a c co or re e d dr ri iv ve er r
t to o c co om mp pe et ti it ti iv ve e a ad dv va an nt ta ag ge e i in n t th he e 2 21 1s st t c ce en nt tu ur ry y

Sylvie Feindt, SFC, feindt@sfconsulting.net


Introduction
Today there is significant global overcapacity in most industries. In this environment of scarce
demand customers are less forgiving of poor customer service and more demanding of customised
products or services. As the competition continues to introduce new offerings tailored to the special
needs of different segments of the market, companies have to respond by similar custom made and
highly personalised offerings. The ensuring proliferation of product variety for multiple countries,
customer segments and distribution outlets create headaches in forecasting, inventory management,
production planning, and after sales service support (Lee, 2000). A key issue for many companies is
thus managing the supply base – including sourcing, supplier integration and inbound parts
management. In addition product life cycles are getting ever shorter and technology is evolving fast in
many sectors.

Under the impact of information technology and the resulting globalisation of markets and production,
new methods of combining activities into offerings are generating new opportunities. In such a change
driven environment, a single company rarely provides everything on its own anymore. Rather the most
attractive offerings involve customers and suppliers, allies and business partners in various
combinations. Consequently companies do not really compete with one another anymore. Rather, it is
offerings that compete for the time, attention and money of customers (Norman and Ramirez, 1993).
The networked business can take different shapes ranging from integrated supply changes around a
key player, to collaborative SME networks, strategic alliances, virtual organisations, extended
enterprises, industrial districts, clusters, etc. Principally these enterprise networks can be divided into
more focal, vertical networks with a supply chain topology and more equal co-operative business
networks (Koeszegi, Gruber and Nöster, 2005).

Although this co-operation across company boundaries is seen as a panacea for different types of
organisations their emulation of the existing best practice by all players in the market seems more
difficult than theory suggests. One reason is certainly insufficient interoperability and standardisation.
There is however agreement in the management literature that equating the supply chain with a
technological concept is worst practice.


Aim and structure of the book
Accordingly this book focuses on non-technical issues, without forgetting its background as a cluster
in information society technologies. The book aims to document the results of the first year of research
of the FP6 projects in the area of ICT for enterprise networking. Despite their diverging starting
points, all projects involved – ILIPT, SPIDER-WIN, ATHENA, Co-DESNet, NO-REST, V-CES and
VERITAS are trying to enhance the competitiveness of organisations in the machinery and mechanical
engineering industry with a particular focus on the automobile sector.

The chapters have been divided into four broad categories: supply chain integration, SME networks,
interoperability and tools and technologies to enhance competitiveness within a certain sector or group
of professionals.

The first two chapters look at supply chains.
• The chapter of Rabe and Mussini presents IT solution requirements of companies in
different supply chain positions, based on an assessment applying the integrated enterprise
modelling method to three supply chains addressing the product portfolios motorbikes,
aircraft supply and helicopters.
7

• The second chapter by Stone/Miemczyk and Esser assesses the requirements with respect to
construction, capacity, platform and IT in order to move to a build to order strategy (BTO) in
the automotive industry.

The following two contributions look at rather horizontal SME networks:
• Nöster and Gruber focus on structural settings and success factors for co-operative business
networks.
• Villa/Antonelli and Cassarino present a model for the transformation of traditional industry
clusters, characterised by geographic proximity, into collaborative SME networks.

Three contributions focus on interoperability and standards in technologies supporting inter-
organisational co-operation:
• The contribution of Ruggaber and Berre presents an interoperability framework based on
industry scenarios and pilot case studies in aerospace, automotive, furniture and
telecommunications.
• Gerst and Jakobs look at the influence of SMEs on standardisation in the automobile sector,
by examining the two different approaches: a sector specific portal and the participation in the
open standards-setting process.
• Gerst and Bunduchi assess the development process of Radio Frequency Identification (RFID)
standards, which are determined by the notable absence of cross industry user involvement
and the resulting danger that future standards are shaped by interests of component
manufacturers and large retailers.

A final group of contributions deals with specific tools and technologies to enhance the
competitiveness of companies:
• This includes a contribution of Rios et al who developed a set of virtual services to support
cost estimation (accuracy and speed) at different points in the product life cycle to enhance the
competitiveness of suppliers in the automotive, aerospace and electronic equipment
manufacturing.
• The final contribution of Stone et al discusses the specificities of ICT supporting new logistic
concepts with flexible production to enable the 5-day-car.


Key issues tackled
In the following, the key issues around supply chain integration, interoperability, seller-buyer
relationships, cost reduction and SMEs, which were tackled across various contributions, are outlined
leading to a number of recommendations:

Supply chain integration
In almost all industries, supply chains have become a much more important strategic and competitive
variable. Businesses are increasingly relying on their suppliers to reduce costs, improve quality and
develop new processes and products faster that their rivals can (Liker and Choi, 2004). There are three
key dimensions of supply chain integration: information integration, co-ordination, and global linkage
(Lee, 2000). The best supply chains identify structural shifts, sometimes before they occur, by
capturing the latest data, filtering out noise, and tracking key patterns. They create profits, increase
market share, strengthen competitive positions and enhance the value of a company (Lee, 2004; Lee,
2000).

Successful large companies such a DELL, IKEA, Japan Seven-Eleven or Zara have the ability to
manage transitions – changing market conditions, evolving technology, different requirements as the
product moves through the life cycle. Companies like that have a triple-A supply chain that provides
companies with sustainable competitive advantage (Lee, 2004). They have agility, adaptability and
alignment. An example is the Spanish apparel company Zara. After September 11, the company was
able to co-ordinate its designers, supply chain partners, and manufacturers to launch a new line of
apparel featuring black within two weeks. The result was a tremendous sales shift in their favour.
8

(Supply chain challenges, 2003). Other major players such as Federal Express, Procter & Gamble or
Wal-Mart used the networked supply chain to dramatically transform the competitive landscape of
their markets (Lawton and Michaels, 2001).

However, seamlessly integrated supply chains are rather an exception than the rule. As communication
is taking place sequentially from one tier to another, OEMs lack information from the component level
and suppliers fail to respond quickly to changes propagated on a short term notice (Rabe and Mussini
in this book). To alleviate this situation, large companies such as Cisco install eHubs, which enable
networked rather than sequential communication. (Scherer, 2005, Stone et al in this book). In the
automobile chain however the introduction of collaborative networks and the modular approach
(Doran, 2004) is beginning to entail a shift of the added value from the vehicle manufacturers to
suppliers and to other business partners such as system integrators (Gerst and Jakobs in this book).
The now proliferating build-to-order strategies will have repercussions for all players in the
automobile supply chain. Considering a five-day car scenario with a manufacturing lead time of one
day, this would leave one day to receive orders and inform suppliers, and three days to deliver the
finished vehicle to the customer (Stone, Miemczyk and Esser in this book).

In recent years, companies have made significant investments in high level planning tools such as for
example advanced planning systems in order to optimise their supply chain management. However,
these tools have failed to deliver real-time visibility on process, product and resources: organisations
still cannot see what is going on in their supply chains (Gerst and Bunduchi in this book, Stone,
Miemczyk, Hellingrath and Witthaut in this book). A Forrester Research survey found that more than
half of the companies interviewed have SCM systems that failed to meet their expectations (Sherer,
2005).


Interoperability
Such integrated supply chains need collaboration, seamless information-flow and monitoring
including RFID and extended system processing capacities. A basic element of the system might be an
ASP (application service provider) approach. The presence is however characterised by a plethora of
isolated systems and applications with limited to no interoperability. From a standards point of view
the B2B space is rather characterised by a proliferation than a lack of standards (Ruggaber and Berre
in this book). In the automotive industry, suppliers are currently concerned about the significant costs
of IT system administration caused by the undisciplined approach by vehicle manufacturers (VMs)
and the implications of adopting new technology on an unregulated, individual supplier/VM basis
(Stone, Miemczyk, Hellingrath and Witthaut in this book). Ruggaber and Berre thus suggest an
interoperability framework. This baseline allows the viewpoint-based integration of enterprise specific
software.

As SMEs tend to do business with more than one company they have to adopt systems of several
larger customers or vertical networks. Accordingly a vast majority of SMEs see the lack of
standardisation as a main barrier to closer co-operation. Unfortunately, it appears that so far
development of IT standards has almost exclusively been technology driven; with standards created
solely reflecting providers’ and implementers’ priorities alike. Most other stakeholders and, most
notably, SME users hardly participate in the standardisation activity. The result might be that the large
companies agree on extremely complex specifications, leading to a failure of an initiative as their SME
suppliers are not able to implement and manage the complex specifications. Also, the non-use of many
standards-based services by SMEs is largely due to the fact that insufficient knowledge and resources
are available to employ these systems, which are perceived as being extremely complicated to deal
with. In fact, this perception may be considered as a major impediment to a more successful uptake of
standards based systems by SMEs. This exemplifies an urgent need for simpler standards (Gerst and
Jakobs in this book).

An interesting case is RFID as it promises to significantly improve the end-to-end product visibility of
supply chain operations. Despite the technology being 50 years old, the implementation is - from a
user point of view - in its infancy notwithstanding some more advanced industries (e.g. retail). The
9

discussion regarding RFID implementations often is driven by high promises in terms of cost
reduction and improved visibility of the supply chain. If these expectations are not fulfilled, potential
customers defect from the “RFID vision”. In this respect RFID standards are a major issue in securing
the high investments in RFID technology on different levels (e.g. interface protocol, data structure,
tags etc.). Not only different standards co-exist in parallel, but also different actors with sometimes
divergent interests influence the standardisation life cycle. However, if several competing RFID
standards exist at the same time, items would have to be re-labelled during their shipment which
would offset all the benefits that RFID are supposed to deliver (Gerst and Bunduchi in this book).

Apart from the technical issues surrounding the development of standardised business processes across
the entire industry, a range of organisational, social and economic factors have influenced the OEMs’
and the suppliers’ choices and actions. This has eventually led to the undesired outcome of failing to
accomplish the initial vision of industry wide collaboration supported by common industry wide
standards. (Gerst and Jakobs in this book). Thus, interoperability does not only include technical
issues of infrastructure and applications but “soft” issues such as co-operation cultures, trust, business
models and communities (Athena contribution to Interoperability, 2005).


More sophisticated buyers
Information transparency is an increasing challenge to producers. On-line auctions give customers
more information than ever about producers’ pricing and in particular reverse auctions tend to yield
lower prices. Although in some sectors the new sales channels have not caught on, for many industrial
products they have quietly become part of the normal way of doing business. In addition, downstream
consolidation of industrial distributors and large retailers makes information exchange easier because
it passes through fewer companies. This environment has an impact on prices that manufacturers can
charge (Abele, et. al., 2003).

An interesting case of a sectoral marketplace is Covisint, initiated by three large automobile producers.
Due to the distribution of power that historically characterised the relations between OEMs and
suppliers, the latter were apprehensive of Covisint. They saw it as just another exercise to intensify
OEMs’ power pressure as market places are a tool for reverse auctions and market testing against the
incumbent supplier (Sherer, 2005). Some suppliers also feared that Covisint would require significant
additional resources and investments from their side, whereas the benefits would mostly materialise at
the OEMs’ side. Suppliers already struggled with the administration of a number of such
‘standardised’ portals and their requirements were by and large neither part of the Covisint vision, nor
part of the development of the standardised technology. One result was the creation of another e-
marketplace, named ‘SupplyOn’, by a number of large tier-1 suppliers who defected from the Covisint
vision (Gerst and Bunduchi in this book).

Also the implementation of demand-side approaches (build-to-order or assemble-to-order) makes
OEMs and large retailers more powerful. A case in point is the Dell model. It relies on demand side
pull rather than supply side push - no computer is produced unless there is a corresponding demand in
the marketplace. Thus inventory of the OEM is virtually eliminated. Due to the direct made-to-order
approach, Dell can also see on a daily basis, if customer preferences are shifting, and which customer
segments are concerned. (Magretta, 1998). Other industries such as the clothing industry with vendor
managed inventory systems (Au and Ho, 2002) or efficient consumer response initiatives of the
grocery industry (Sherer, 2005) follow the example. A simple transfer of the Dell assemble-to-order
system to the automotive sector is however not possible. Dell’s supply chain keeps sufficient
component inventory on site in order to build computers to order once the order is received. The
inventory thus represents the de-coupling point in the supply chain that largely buffers component
supply from demand variability and lead time reduction requirements. Dell only handles 15-50 key
components per computer, whereas the average vehicle consists of 2.000-4,000 components, many of
which are currently customised by colour, engine size, etc. Consequently holding a component stock
in the same fashion is cost-prohibitive in the automotive industry. Rather BTO strategies depend on a
capable and responsive supply chain, and subsequently, on the logistics operations that connect
10

suppliers to the assembly plants, and assembly plants to dealers and customers (Stone, Miemczyk and
Esser in this book).

Cost reduction or optimal use of innovative capacity
Cost reduction is one highly desired result of supply chain integration, but not the only one and to
some not even the most important one. Do immediate benefits of low wage costs outweigh the long-
term benefits of investing in relationships? Certainly the spread of Internet technologies to get
suppliers to compete fiercely on cost efficiency, for instance introducing reverse auctions to get the
lowest price for components as Ford did, has enabled OEMs to push prices down (Liker and Choi,
2004).

However, examples that not all manufacturing is going to move to low-wage countries are the two
Japanese car manufacturers Toyota and Honda. They do not source much from low-wage countries.
Their suppliers’ innovation capabilities are more important than their labour costs. Accordingly the
two companies aim at a long-term relationship that involves trust and mutual well-being. At the same
time the relationship connotes discipline and the expectation of improvement and growth. When
creating the partnerships the two car producers try to learn as much as possible about the suppliers,
thus being able to co-operate effectively with the suppliers to make their processes leaner (Liker and
Choi, 2004, Jackson and Winkler, 2005).

Whatever approach is chosen, manufacturers require accurate and rapidly available cost estimates at
different points in the product life cycle for the decision process. An online learning and mentoring
service with a supporting cost engineering database developed by one of the projects promises to be a
valuable service in the future (Rios et al in this book).

SMEs in the more competitive environment
The Triple A supply chains of SMEs with high agility, adaptability and alignment exist, but remain
rare species in value chains of SMEs (these companies are also likely to outgrow their SME status
rapidly). While the majority of interactions between organisations are still limited to e-mail,
companies are, albeit selectively, upgrading technology. Throughout the supply chains, there tend to
be different levels of IT infrastructure and e-commerce solution complexity and upgrading does not
happen systematically and with all partners at the same time. For example, a retailer might use vendor
managed inventory and never out of stock (NOS) systems with a specific supplier, order occasionally
online from its buyer group and procure in a traditional way with a listed supplier of its buyer group.
However, others are moving quickly to a high level of sophistication, pulling their value chain partners
with them. More complex e-commerce solutions integrate automated value activity interactions. They
do not only automate value chain processes in a linear fashion, but they automate ‘bundles’ of value
activity interactions (Feindt et al. 2005).

In the IT market many solutions can be found which support the operation of supply chains. However,
they are mostly tailored to larger companies, requiring some substantial IT infrastructure, or they force
the participants into the use of specific ERP and/or APS systems (Rabe and Mussini in this book).
Typically, SMEs opt for readily available off-the shelf systems and services, which need to be cheap
and easy to install, maintain and use. Proprietary systems are also used frequently where SMEs are
compelled to do so by e.g. a major business partner -with all associated problems (Gerst and Jakobs in
this book).

Recent field research shows that each company’s IT requirements are related to the level of the
company in the supply chain. Of course, all detected requirements converge to a unique objective,
which can be synthesised as: maximum flexibility in manufacturing keeping stocks at the minimum
level. An ASP software solution based on these requirements of SMEs at different tiers in the supply
chain is structured into the four areas modelling, operation, integration and intelligence (Rabe and
Mussini in this book).



11



SME Networks
A trend in manufacturing is that original equipment manufacturers (OEMs) want to co-operate with
fewer suppliers but on a worldwide scale. As a result, small and medium sized suppliers will be
suppliers to integrators, that is, tier 1 or 2 suppliers. There are very few small suppliers that are
established in such specialised niche markets so that they can remain direct suppliers to, for instance,
the automobile or aerospace industries. A strategy that small companies pursue is to group themselves
in networks with the aim to be perceived as a larger organisation and to gain visibility. Companies join
forces in a co-operative SME network to bid for larger projects, subcontract to each other but also to
participate in a learning environment and benefit from increased purchasing power (Feindt et al,
2005).
Such networks are rather horizontal and are based on alternating customer-supplier relations, where
one time a company sub-contracts products or services from the network and another time supplies
products or services to another partner in the network. Accordingly, co-operative networks display a
peer-to-peer network topology. This type of network seems to correspond more than the vertical ones
to the actual business concept of a VO, where companies stay independent and act as equal partners
while working together (Nöster and Gruber in this book).

There is however no structural blueprint for successful co-operation in a network. The kind of virtual
structures in place eventually depends on the specific business, geographical and cultural factors. One can
distinguish externally initiated networks and strategically (internally) driven networks. The first cluster is
characterised by equal power distribution, regional dimension and clearly defined strategy. The second is
focal, international and aims at fulfilling a specific business case. There is no indication that networks that
were initiated with external support are less successful than those created on the companies’ own
initiative. The correlation of ICT with focal power shows that it is used to help the leading enterprises co-
ordinate activities and processes of the VE. Success factors for a network are business opportunities, trust,
commitment to co-operation, exchange of information and network management. ICT as such does,
however, not have a direct effect on success of co-operation (Nöster and Gruber in this book).

Similar horizontal SME networks have emerged from some of the traditional Italian districts in
response to increasing competitive pressures: the “Demand & Supply Network” (DESNET). In
practice, the new organisation emerged from an agreement among a set of SME and is structured as an
“extended virtual enterprise”. This structure makes the Demand and Supply Networks much more
robust against the increasing variability of the products and labour markets. Such a chain of SMEs is
structurally co-operative, and the naturally leading SME is the most efficient firm. The analysis of
existing Industrial Districts highlights how far most of the districts are away from the desirable target
of a collaborative network. A notable exception is the industrial district Torino Wireless. In this
context, the importance of non-competitive institutions, like Research Centres and Universities,
involved in the DESNET organisation cannot be underestimated. They have to play a core role in the
proliferation of the collaborative networks in the industrial districts in Italy (Villa, Antonelli and
Cassarino in this book).

Conclusions
Focusing on different types of networked organisations and their requirements, a number of
conclusions and research recommendations can be drawn from the articles:

- Following the shift of the value creation from the OEMs in some value chain, also research
needs to shift from OEM to their suppliers and beyond and their specific needs with respect to
IT systems. Sector oriented projects need to contain a critical mass of partners to make an
impact, not forgetting the smaller players in the chain.

- As in the areas of ICT for businesses, technical issues are closely linked to a range of
organisational, process, social, economic and cultural issues, projects need to be
12

interdisciplinary, bringing together researchers from various cultures around a core set of
ideas. Research into soft factors/issues should also be included in technology projects.

- With the creation of new businesses high on the agenda in Europe, SMEs which, after all,
form the employment and growth engine of the EU, need to take part in shaping the IT
infrastructure upon which they very much rely. Consequently they need to be an essential part
of research projects (e.g. as dedicated user groups to align context specific user requirements
and in validation through application trials).

- Similarly SMEs need to be involved in the standardisation process, if a wide implementation
of less complex standards is aspired. The latter might be achieved through the participation of
SME umbrella organisations in the standardisation process or the change of procedures to
facilitate SME participation, for instance the increased use of electronic communication means
in order to minimise travel and time budgets.

- Further research also needs to go into different types of SME networks. There is, for instance, a
need for developing more accurate means to identify and measure the value added for the
companies, as well as the costs, in order to ultimately assess the success of a virtual organisation.

- A number of projects will produce interesting results which the Commission might want to
support with technology diffusion or take-up initiatives, market validation, and initial
deployment.


References
Abele, J.M./Caesar, W.K., Roland, J.H. (2003): Rechanneling Sales, McKinsey Quarterly, Issue 3, pp.64-78
Athena contribution to Interoperability. Policy Action Plan Version 1, March 2005, www.athena-IP.org.
Au, K.F. and Ho, Danny C.K. (2002), Electronic Commerce and Supply Chain Management: Value-Adding
Service for Clothing Manufacturers, Integrated Manufacturing Systems 13/4, pp.247-254.
Doran, D. (2004): Rethinking the supply chain: an automotive perspective, in: Supply Chain Management,
Vol.9/1, pp.102-109.
Feindt, S., Jeffcoate, J., Chappell, C. (2005): E-commerce links for SMEs within the industry value chain,
Nabeel A. Y. Al-Qirim (ed.), Global Electronic Business Research in Small to Medium-Sized Enterprises:
Contemporary Issues, Implications and Future Trends, London.
Gruber, M., Koeszegi, S., Nöster, M. (2005): Initiated Networks – A strategic Alternative for SMEs, presented at
the 21
st
EGOS-colloquium, 30
th
of June – 2
nd
of July, Berlin.
Jackson, B., Winkler, C. (2005): Building the Advantaged Supply Network, in: Supply Chain Management
Review Jay/June, pp.42-49.
Lawton, T.C., Michaels, K.P. (2001): Advancing to the Virtual Values Chain: Learning from the Dell Model, in:
The Irish Journal of Management, Vol. 22/1, pp.91-112.
Lee, H.L. (2004): The Triple-A Supply Chain, in: Harvard Business Review, October, pp102-112.
Lee, H.L. (2000): Creating Value trough Supply Chain Integration, in: Supply Chain Management Review,
September/October, pp.30-36.
Liker, J.K., Choi, T.Y. (2004): Building Deep Supplier Relationships, n: Harvard Business Review December,
pp. 104-113.
Magretta, J. (1998): Fast, Global and Entrepreneurial: Supply Chain Management, Hong Kong Style, Harvard
Business Review, September-October.
Norman, R., Ramirez, P. (1993): From Value Chain to Value Constellation: Designing Interactive Strategy, in:
Harvard Business Review, July-August, pp 65-77.
Sherer, S.A. (2005): From Supply-Chain Management to Value Network Advocacy: Implications for e-Supply
Chains, in: Supply Chain Management, Vol 10/2, pp 77-83.
Supply chain challenges: Building Relationships (2003) Harvard Business Review, July, pp.65-73.
13

2 2. . O Op pe er ra at ti io on n o of f t th he e s su up pp pl ly y c ch ha ai in n

2.1 Analysis and comparison of supply chain business
processes in European SMEs

Markus Rabe, Fraunhofer IPK, Markus.Rabe@ipk.fraunhofer.de
Bruno Mussini,

Joinet S.P.A., bruno.mussini@joinetspa.com

In many European regions, SMEs are part of intense customer-supplier networks. These companies
survive, only, because of the excellent skills of their employees and the extreme flexibility of their
processes. Co-ordination mechanisms are poor, and the extremely lean organisations do not allow
for any substantial IT overhead. SCM tools are too complex, expensive and personnel-intensive for
these enterprises.
The SPIDER-WIN project (IST 507 601) aims to achieve efficient, simple and context-aware SME
co-operation with low-level local software requirements and adapted to the typical availability and
quality of resource data in very small enterprises, focused on the exchange of order status changes.
This will be achieved by an ASP platform with asynchronous data exchange between the platform
and the enterprises (but, never between the enterprises, directly).
In the first eight months of the project, in order to analyse the specific constraints and potentials
within the targeted enterprise types, a substantial analysis of the supply business processes was
performed at three supply networks in Europe. The study was based on a reference model, including
guidelines, reference classes, template models and other auxiliary documents. The results from the
single companies have then been merged, first into supply network models and finally into one
“general model”, which mirrors the specifics and potentials.
Analysing these potentials and prioritising them with respect to the economic benefit, the major
methods have been identified which will support the processes.


Introduction
On the IT market many solutions can be found which support the operation of supply chains.
However, they are mostly tailored to larger companies, requiring some substantial IT infrastructure, or
they force the participants into the use of specific ERP and/or APS systems. The SPIDER-WIN
project’s goal is to provide solutions for networks including SMEs with very low level IT
infrastructure. Furthermore, there should be no need to change from existing IT systems, nor should
there be the requirement for any external online connection, which might give the feeling of a “big
brother” to the companies. The means provided will clearly depend on the specific (mostly low)
quantity and quality of data available.

This paper is reporting major results of the first year of the SPIDER-WIN project. This first period
was mostly dedicated to analyse in great detail the specific requirements of the enterprises in focus, to
define typical workflows which can be supported, and to investigate the available data and interfaces.
For this purpose, several tasks had to be performed:

• Analysis of the workflow within the companies, with respect to the supply chain processes.
• Determination of the (economic) potentials, both qualitative and quantitative, which could be
provided by the new co-operation platform.
• Analysis of the data which could be provided (or maintained) by the companies in order to
support new mechanisms.
• Specification of a general workflow model, which serves as a base (a) for the definition of a new
co-operative supply chain model and (b) which supports the clear identification of work flow
peculiarities in the single companies or company networks.

14

For a substantial analysis, three supply networks from different European regions have been analysed.
As languages are a major issue in SMEs, where even the managers often are not very fluid with
English, the studies have been conducted by local companies (“coaches”). The companies involved in
the study have been:

• Emilia Romagna (Italy): Bentivogli, CIAP, Ducati, PM and Joinet (coach).
• Basque (Spain): Fibertecnic, GAMESA, Metaltrek and Sisteplant (coach).
• Mazovia (Poland): PZL Swidnik, PZL Warszawa and IMIK (coach).

However, it would not be very likely to achieve comparable data out of three different studies per-
formed by different coaches, nor would it be likely that any valuable general model could be derived
out of this procedure. Therefore, a process was started led by Fraunhofer IPK which integrated the
studies in three phases:

In the preparation phase, reference models and documents have been developed as the base for the
local studies, and the coaches trained in applying these means.
In the interview phase, preliminary results have been analysed and experiences exchanged in coach
workshops. As a result, the reference models have been adjusted where necessary.
In the analysis phase, similarities between the different workflows have been identified and (where
necessary) proved in discussions with the companies. A general model has been developed, which
includes relevant information about available documents and data as well as the potentials.


Modelling Techniques

Integrated Enterprise Modelling (IEM) Method
The subject of analysis requires a process oriented approach. Therefore, the Integrated Enterprise
Modelling (IEM) Method [Mertins and Jochem 1999] has been applied, with the tool MO²GO for its
efficient use. This method is very flexible, and the tool supports the application specific definition of
resource sets, evaluation schemes etc. Furthermore, through the object oriented approach of the IEM
the use of reference classes is very efficient, simplifying the task of defining common terms, structures
and attributes.

Reference Models for Distributed Business Processes
Reference models increase the efficiency of modelling, raise the quality level of the developed models,
and improve the reusability of the models. Such reference should include class structures, template
models and a manual which describes the correct and efficient use of the reference models as well as
the validity and constraints of the model [Rabe and Mertins 1998]. Especially for distributed systems,
where different persons perform the modelling task at different locations, reference models can
improve the work, significantly [Rabe and Jaekel 2003 and 2004].

The Supply Chain Operations Reference Model (SCOR) was designed for the effective communi-
cation among supply chain partners [Supply Chain Council 2002]. SCOR is a reference model, too,
and not just a BPM technique, as it incorporates additional elements like standard descriptions of
processes, standard metrics and best-in-class practices. Furthermore, SCOR has been turned out to be
a good base for enhanced models [e.g., cp. Stich and Weidemann 2002].

Modelling Languages for Distributed Modelling
When analysing the processes of interoperating enterprises, there are multiple challenges due to
different terms, structures and “philosophies”. Furthermore, sometimes there are already business
process models, but they have been set up using different modelling techniques and tools, which are
very often incompatible. There is recent research for the development of meta languages or “unified
languages”, which would enable the transfer of business models between different modelling
approaches [Berio, Anaya and Ortiz 2004].
15



Distributed Modelling Approach

Business Process Model Approach
As a base for comparable models, which could later be integrated into one general model, the IEM was
selected as a common modelling method. In order to ensure that the regional consultants Joinet,
SISTEPLANT and IMIK use the IEM in a proper and successful manner, the personnel of these
partners were trained in the due application of this method, as well as the use of the related MO²GO
software and the specific issues to be regarded when generating a model on a distributed basis.
Furthermore, techniques of process analysis, interview methodology and modelling were on the topics
list of the training sessions. This includes techniques of systematic and fundamental analysis of
process objects and their features like activities, products or services, orders, resources, users and
customers.

This task was directly followed by a “learning on the job”, where a short sample process study was
performed in each region. This ensured that (1) the reference model and the guidelines are operational
for this application case and (2) that the regional coaches have really internalised the IEM and the
required open interview technique. Based on these (distributed) experiences, the reference model was
adapted, improved and finally decided.

Guidelines
A Guideline Document Suite for open interviews was set up, adapted to the specific case, and related
to typical roles in the users’ enterprises. The suite includes a description of the processes, variables
and metrics to be considered as well as supporting documents and document templates. The guidelines
supported the structure and completeness of the interview as well as the comparison of the results from
the different supply networks.
The complete Interview Guideline Suite consists of the following elements:
1. “Open Interview Guideline”, including the interview structure and checklists of the major points to
address.
2. “General Procedure and Hints”, providing hints and tips for all steps of the interview, from the
preparation via the interviewing and modelling up to reporting. The interviewer is not expected to
use this document during the interview, but for the preparation phase of the interview and for the
rework of the interview results.
3. List of “Expected Documents and Information” developed as a checklist to compare the results of
the interview with the expected information. This checklist can be used by the interviewer himself
or herself to check after the interview whether all information has been collected during the
interview and whether all documents are collected.
4. “Interview Protocol Template”, created to ease the editing and formatting of all information which
is not included within the process model. One protocol should be gathered for each interview.


Reference Model
A common model sketch in form of different template models was developed by IPK, Joinet and
SISTEPLANT, in order to guarantee that all information and requirements detected can be syste-
matically documented within one single, consistent model.

Class Structure
The reference class structures define common terms for objects and enable overall evaluation
procedures of the model. Furthermore, these classes inherit specific attributes which enable to hold
two names for each object in parallel: An English name and the name in the native language. These
names can be switched, enabling the discussion based on the native terms within the companies, and
using the English terms when discussing and comparing the model in the coach team.

16

These classes should be used for modelling wherever possible. However, the objects based on these
classes may be extended with additional attributes. Additional sub-classes may be appended, if the
company specifics require the clear differentiation of such types of objects (e.g., because they include
different sets of attributes, or because they are handled by different processes). The high-level class
structures for resources and orders are illustrated in figure 1.


Single Enterprise Template Model
The Single Enterprise Template Model was developed with three objectives. First, the template can be
used during the interview (together with the guideline) to ask the interviewee each area of interest and
to structure the interview along the process under consideration.

The second objective of the template model is a common naming of the process elements. This is
important because there are four different regions involved in the SPIDER-WIN project (Germany,
Italy, Poland and Spain) but the project language is English and none of the project partners is a native
English speaker. The wording and structure of the template model is orientated at the Supply-Chain
Operations Reference-model (SCOR).



Figure 1: High levels of the reference class structures for resources and orders

Third, the template serves as the base for the individual process model of the interviewed company.
This individual process model is the core documentation of the interview results, as (different to the
interview protocol) this model is structured with respect to the processes as well as to the objects and
attributes.

Basing the single enterprise models on the template, each enterprise within the supply chain is
modelled in a similar structure. This will increase the comparability of the different models and also
enable a simpler merge of the models into a general overall model.

The main focus of the project is the communication between the companies and its support by IT
systems. Therefore, the order transfer between the enterprises is modelled in more detail than the
product transfer (fig. 2). The model structure is based on the highest level of SCOR (fig. 3).

Supply Chain Template Model
In opposite to the Single Enterprise Template, the “Template Model for the Supply-Chain” was
developed to support the modelling of the supply network as a whole, thereby connecting the single
models. This model shows the relation of different enterprises along the supply chain with respect to
their exchange of information and goods (fig. 4). However, this model is only used during the study, in
order to highlight the interfaces and to structure the discussions about interface details. At the end, the
full model will not just be a sequence, but a network of different customer supply relations.
17

INPUT INPUT OUTPUT OUTPUT
P
l
a
n
n
i
n
g

a
n
d

C
o
n
t
r
o
l
l
i
n
g
P
l
a
n
n
i
n
g

a
n
d

C
o
n
t
r
o
l
l
i
n
g
P
r
o
d
u
c
t
C
h
a
i
n
P
r
o
d
u
c
t
C
h
a
i
n

Figure 2: IEM template model for a single enterprise (level 0, main model areas marked)


Figure 3: IEM template model with SCOR related areas
Tier 3 Tier 2 Tier 1

Figure 4: Template model for the supply chain - overview


18

Metrics
The variables and metrics have to be defined, structured and mutually related in order to identify
solutions for cross-enterprise networking. As the result of a pre-analysis the following structure of
metrics was set up for the interviews:

• Category I: Figures or indicators which show the quantitative resources needed for the SCM
process itself (potentials of saving effort e.g. in terms of personnel effort, finance, or resources).
• Category II: Figures or indicators which can be used as input for the control or prediction
algorithms.
• Category III: Figures or indicators which are of interest to the end users, mainly in order to
identify processes in the supply chain which promise optimisation potentials.

Since the main influencing factor for most of the metrics is time, certain general points of the order life
cycle (e.g. order issue, order receive, order agreement, start of production, order change issue, etc.)
should be found at all companies within the supply chain. Those points should be marked on the
process models in order to achieve transparency about the calculation of the certain metrics and in
order to avoid different calculation and presentation of the metrics at various companies.


Conduction of the Study
The study was performed in each region, separately. The coaches exchanged the models, interview
protocols etc. by e-mail in this period, in order to get feedback, to synchronize the models and to
become aware of aspects detected in one region, which could be of importance to other regions. Due to
the careful preparation, very few physical meetings of the coaches were required in this phase.

First, the single company models were established. They were then merged to models of the three
supply networks, thereby identifying additional potentials and challenges at the company interfaces. In
total, 103 sub-models (“levels”) were established, with a total of 1852 process elements.

Then, a “general model” was derived from the as-is-models. This general model contains the “standard
processes” and identifies possible variants. Therefore, by comparison of a specific supply chain model
with the general model, the specifics of the supply chain can be identified. The result is the first step
for customising a supporting software system for the supply network, identifying the modules required
as well as the data and processes to be adapted.


Findings

Definition of Supply Roles
The driving companies of the enterprise networks (“prime contractors”) are in different supply chain
positions: they can be OEMs, or suppliers at first or second level (sometimes, the level depends on the
specific product, and not only on the company). In order to establish a common model for the supply
chain execution support, the prime contractors are defined as tier-1 in the following analysis: the tier-1
is the downstream end of the supply chain which is supported by the new methods and software.

Similar, the position of the suppliers depends strongly on the product. The same enterprise might
supply some parts in tier-2 to their customer, and (indirectly) other parts as tier-3 or tier-4. However,
one of the major distinctions in terms of the business processes is found to be the company’s role in
the product development process. Classically, enterprises producing on-design parts have been direct
suppliers (“tier-2”). Today, however, the supply chain execution process is often separated from the
design process. In such cases the supplier is selected by the tier-1, and the product design as well as
the production (especially in quality aspects) is conducted in direct interaction with the tier-1.
However, the order management is performed through an intermediate supplier, who will also manage
additional value generation (e.g. kitting, assembly).

19

Therefore, the SPIDER-WIN project has defined the following roles:
• tier-2a suppliers produce on-design parts (designed by or in close co-operation with the tier-1), on
orders directly communicated by the tier-1.
• tier-2b suppliers produce on-design parts on orders communicated by tier-2a, which do not
necessarily depend, directly, on orders from tier-1 to tier-2a (e.g., because of economic lot sizes
or longer lead times).
• tier-3 suppliers produce catalogue parts on orders which come from tier-1, tier-2a or tier-2b.

These definitions are applied (even if they need explanation), because from the process analysis they
turn out to classify the supply execution processes very precisely, and enable to describe the
requirements of the enterprises according to these roles. The level tier-3 is not subject of the further
study.

The results of the field study clearly show that each company’s requirements are related to the level of
the company in the supply chain. The tier-1s have some requirements close to the tier-2as’ ones, and
some requirements specific to their role. The same is for the tier-2as’ and tier-2bs’ requirements.

Of course, all detected requirements converge to a unique objective, which can be synthesized as:
maximum flexibility in manufacturing keeping stocks (wherever they are and whatever their status is,
raw material, subassembly parts, etc.) at the minimum level. In the following sections, the
requirements are presented classified by the supply level.

Requirements of Tier-1
The major challenge of the tier-1 enterprises is to follow the market demand in an agile way, which
implies to make significant short term changes to the production plan. The complexity of the supply
network makes it very hard to understand whether the supply network (tier-2as and tier-2bs) can
satisfy these requests or not. The tier-1 experience expressed in the interviews documents that quite
often short term changes, even if quickly propagated to tier-2as and tier-2bs, can not be satisfied by
the suppliers. Therefore, the main requirement is to create conditions which enable the whole supply
chain to satisfy tier-1’s requests even if they are frequently changed or updated, and in the case that the
supply chain can not satisfy these requests, tier-1s requires to be informed as long in advance as ever
possible. This requirement can be divided into a set of sub-requirements:

• Capability to know the exact quantity of a final product that tier-1 can manufacture when specific
parts expected from suppliers are missing.
• Capability to know supply problems as much in advance as possible, in order to be able to
elaborate alternative production plans.
• Capability to know whether an alternative production plan can be satisfied by parts (or sub parts
or raw material) existing along the supply chain.
• Push suppliers (2as and 2bs) in better managing stocks and the respective suppliers, in order to
have significant price reductions (that tier-2as and 2bs will be able to apply thanks to the
financial saving coming from their internal stock reduction and the better management of the
respective suppliers).
• Capability to know real stock values along the supply chain when they think to a part update or
substitution (phase out).
• Dispose of significant indicators which enable to monitor the suppliers’ behaviour in terms of
flexibility, in order to be able to push suppliers in improving flexibility and to measure real
improvements. These indicators are expected to push suppliers in improving their (generally
poor) organisation, based on the assumption that significant flexibility improvements are not
likely to be achieved without organisational improvements.
20

• Dispose of a unique communication tool with all suppliers, able to propagate significant
information directly to the tier-2bs. This tool will enable to save a lot of time which is currently
spent in very low added value activities.
• Dispose of a communication tool for all the partners of supply chain, including the transportation
providers, in order to reach information about status of transport, unexpected events, plan of milk
runs etc.

Requirements of Tier-2a
Tier-2as play the most critical role, as they interface the tier-1 with the tier-2b, and have to manage all
interfacing tasks like economic lot sizing, stocking with respect to lead times and possible product-
phase-out, handling different parts number and many other issues. Of course, their main requirements
would be to have more stable production plans, and production forecasts both on a longer term and
more reliable. However, tier-1s declare that they can not follow this request due to the customer-
oriented strategy.

Tier-2as put in place several organisational procedures for reacting to medium-short term tier-1s
changes, but there are no shortcuts for solving the problem of the tier-2bs delivery time, which very
often is significantly longer than the delivery time requested by the tier-1 (especially for raw material).
Tier-2as very often issue orders to tier-2bs before having the related tier-1’s orders, by making “own
risk” forecast based on their long-term experience.

Tier-2as have very often significantly reduced management capability with respect to the tier-1s, even
if they have a quite good organisation and use suitable IT tools. They do not have a possibility to
propagate to tier-2bs very quickly the changes generated by tier-1s. Often they do not want to do this,
anyway. The reduced tier-2b management capabilities make the direct propagation to tier-2bs not
convenient for the tier-2as. Tier-2as take care of “smoothing” market disturbs coming from tier-1s,
and so they send orders and plans to tier-2bs less frequently and less “hysterically” with respect to the
orders from tier-1 to tier-2a.
Therefore, the tier-2as’ requirements can be summarised as follows:

1. Dispose of longer term tier-1 material requests.
2. Dispose of tools able to support them in an intelligent propagation of information coming from
tier-1.
3. Dispose of a smart communication tool with all suppliers and with customers. This tool will
enable to save a lot of time currently spent in very low added value activities (the expected time
saving has been estimated by Italian tier-2as users, and considered significant by all the rest of
users).
4. Dispose of simple tier-2bs logistic/flexibility behaviour indicators, in order to push tier-2bs in
improving organisation, which is mandatory for achieving appreciable improvements in the
aforementioned simple indicators.
5. Dispose of simple tools which enable a deeper integration between planning (ERP) and
production management (scheduler).

Requirements of Tier-2b
Tier-2bs’ requirements mainly refer to the “visibility” on tier-1 requirements (and to the visibility of
tier-1 on real tier-2bs behaviour). They partially agree on the filter that tier-2as apply on information
coming from tier-1. However, they would anyway appreciate a better visibility and transparency with
tier-1. As described above, tier-2bs may for specific parts take also the role of tier-2as, and they are
aware of the delay introduced by the tier-2as in propagating the information.

For tier-2bs, also, this main requirement can be declined into sub requirements:
1. Direct visibility on tier-1 forecasts, possibly in longer terms than today.
2. Fast propagation of material request changes generated by tier-1s.
21

3. Dispose of a tool which enables the final customer (tier-1) to identify the real origin of the delays.
4. Centralise information related to orders, invoices, etc. so that it can be easily accessed.
5. Dispose of a tool which enables the management of “urgency based” planning.
6. Avoid the loss of information during the transmission of orders from tier-1 to tier-2b.
7. The contract terms tend to disappear due to the continuous changes made during the sourcing. In
this aspect, a tool could help to accelerate the transmission of these changes.
8. Improve the speed of new material requests from tier-2a and tier-2b to tier-1 when a sourced raw
material is damaged and has to be replaced.
9. Real time information on parts phase in and phase out.


Further Requirements Detected During the Study
Several significant additional potentials have been identified, which are not clearly related to the
supply level, but to the network as a whole:
1. Easier management of unexpected events. These events can be last minute demand changes,
unpredictable delays somewhere in the supply chain etc. Thanks to the availability of shared
multilevel communication tools (highly integrated with the companies’ IT legacy systems) the
management of these events related information will be more and more transparent and easier.
This information will be propagated faster up and down in the supply chain and the prediction of
the impact of these events will be more reliable. This is not only a matter of time saving but also a
matter of reliability of answers to the final customers.
2. Easier management of components phase-out. The visibility of lower tiers’ stock would enable a
careful and piloted components phase out.
3. Availability of independent logistic/flexibility performance indicators. These indicators will act as
strategic decision support tools for organisation changes all along the supply chain. Tier-1s could
more widely evaluate their suppliers (today mainly based on the evaluation of the quality of the
provided parts and with respect of the requested delivery time), and lower tiers will be enabled to
evaluate the stability of their customers’ delivery requests.
4. Managerial-cultural growth of smaller companies in the supply chain. This aspect mainly refers to
tier-2bs. The analysis carried out clearly shows that tier-1s prefer to have within the supply chain
very limited changes of suppliers. They prefer long term agreements, because very often lower
tiers have a deep knowledge of a certain manufacturing process which is critical for tier-1. When
these suppliers have limited managerial-organisational capabilities, it is worth to support them to
improve these aspects. The adoption of a multi-tier communication platform based on the ASP
concept will achieve two basic objectives in that direction:
• it will introduce to these companies an advanced but low impact IT tool.
• it will make logistic/flexibility independent performance indicators available, which can be
used by these companies as a starting point for internal projects targeted to improvements of
the logistic and organisational processes.
5. Availability of information related to transportation. A shared communication tool managing the
transportation documents enables all network members to see and monitor the transportation
status.


Consequences for the SPIDER-WIN SCM Model Development
From the result of the field study, the SPIDER-WIN project was focused on providing methods and
new solutions in four main areas:

1. Modelling area for description, modelling and start-up of a SPIDER-WIN multi-tier network of
companies.
2. Operation area for running operations in a SPIDER-WIN multi-tier network.
22

3. Integration area for enabling members of a SPIDER-WIN multi-tier network to integrate the new
platform with existing ERP systems and/or shop floor management tools.
4. Intelligence area for providing significant performance indicators and consequently run
improvement projects.
In the following sections a preliminary description of the solutions is presented per area.

Modelling Area
This solution is expected to model all aspects of a SPIDER-WIN multi-tier network of companies. The
main aspects refer to:
• Roles of companies modelling (e.g. roles of companies, roles of users, expected communication
workflows).
• Document template description (e.g. orders, invoices, transportation documents).
• Interlink modelling, modelling of the information requested for running cross companies
solutions (namely propagation and prediction).
• Local models, defining relations between the interlink model and the companies’ operational
systems (e.g. Manufacturing Execution Systems).

Operation Area
Due to the specific requirements of IT systems for SMEs as described above, most of the functionality
will be provided in ASP mode. With respect to collaboration and negotiation, the related solution is
expected to address the following aspects:

1. New orders (purchasing/production).
2. Changes of orders (e.g. withdrawn, date/qty/price changes, delivery problems).
3. Other documents related to order, which are not related to production but affecting administrative
stuff (e.g. transport document or delivery note, invoices, goods entry sheet).
4. Forecast (publication of production plan). This information differs from order because (a) it is a
publication of a set of product-code to be delivered in a given time-range (time-bucket) for a
given amount. Each datum is a forecast and will become, or not, an order in the next future and
(b) due to the previous point, each datum making part of a forecast is not negotiable (it has not a
life-cycle ruled by a workflow as the lines of order have).
The solution is expected to propagate data about tier-1 delivery plans in the upstream direction
(“propagation”). The propagation process will be possible thanks to:
1. Incoming orders or forecast from downstream.
2. Incoming interlink data provided by tier-2a.

The way by which forecasts are propagated depends on the configuration (see modelling area). It will
enable tier-2bs to see real time information about tier-1 delivery plans without any tier-2a filtering and
in a format understandable by tier-2b (thanks to interlink information).

Furthermore, the related solution is expected to predict and signal possible problems/criticalities
affecting the tier-1 requests (order), not only due to changes made directly on that orders by tier-2a
supplier, but from a variety of further important data (“prediction”). The prediction process will be
possible thanks to:

1. Incoming order from tier-1 (and their current state on the platform).
2. Incoming order from tier-2a (and their current state on the platform).
3. Incoming intermediate production information data from tier-2a (and eventually from tier-2b).
The way in which all these data are managed and prediction alerts calculated depends on the
configuration (see modelling area).

23



Integration Area
A specific adapter will be required in order to enable a unified link between the SPIDER-WIN
platform (ASP mode) and the local control system(s). It specifies all the data and events managed in
this relation. Therefore, any software system which provides the required functionality can be linked
to the Platform through this adaptor. The adapter consists of two elements:

1. A guideline specifying which functionality is required, and how it should be accessed.
2. Local and web software modules for running information exchange.

The information to be collected from the ERP system will be constituted by manufacture orders and
phases. The manufacture orders describe the actions to be taken in the shop floor to manufacture the
products. These orders can be split into several steps (sections). Every section describes an action
performed at the shop floor which causes a change in the materials or semi-finished products
incorporated to the productive process.

In order to achieve a better stock control, material traceability and to know which material batches
have been incorporated to every manufacture order and phase, also information about the material
consumption can be transferred.

Intelligence Area
The solution is expected to provide relevant statistical information (Key Performance Indicators, KPI)
about the supply chain performance, in order to measure global supply chain performance and logistic
performances of the single companies. The major goal is to identify elements or sections of the supply
chain, which contain significant potential for improvement.


Conclusions
The study has demonstrated that a well-adapted reference model is an important base for the
conduction of cross-enterprise business process studies. There was significant additional effort
required for the preparation and adaptation of the reference model. However, this effort was by far
overcompensated, as the result was a smooth and efficient field study, which led to transparent and
comparable results. Achieving such a smooth process was not trivial, as the study was performed at
enterprises of minor size, and included three different languages (plus English as the lingua franca,
but not natively spoken by any of the persons involved in the study).

The Integrated Enterprise Modelling (IEM) Method turned out to be a very efficient means for this
purpose, allowing switching the terms between two languages (the native interview language and
English), providing glossaries of the company specific terms as a side effect. The reference class trees,
adapted for the study and then carefully maintained through the interview phase significantly
improved the development of models with comparable structures, without urging the interviewers into
pre-defined processes. SCOR was a good base to establish common understanding between the
coaches. However, it was not very helpful in the interviews themselves, as only rarely the interviewees
have been aware of SCOR in sufficient detail.

Based on the study results, a “general model” of the as-is-situation could be extracted, which describes
general and specific process elements, systematically documented within one single, consistent model.
It contains the SCOR compliant process names, specific “information categories”, relations between
processes and information categories and further application rules.
The analysis of the general model led to requirements, classified by the enterprises’ level in the supply
chain, which then have been merged to a description of the envisaged software solution, structured
into the four areas modelling, operation, integration and intelligence.

24

The modelling area comprises software to describe the company network with the roles of the
companies in the network (one model per network), the local models and interlink between the
different tiers (once per tier-1 product) as well as supportive information like auxiliary documents.

The operation area consists of three major modules. The collaboration and negotiation module
provides all features required for processing (legally binding) orders. Such orders are always bilateral;
therefore, this module involves exactly two partners from the network at each point of time. To the
contrary, the propagation module and the prediction module always include three or more partners.
Based on specific events in the supply chain, they filter relevant information for partners either
downstream (prediction) or upstream (propagation).

The integration area links local IT systems with the platform. As there might be different local IT
systems, the integration area includes a general guideline (independent from local software), plus
supportive software.

Finally, the intelligence area provides statistics about the performance of the supply chain. This
information is to be used to improve the supply chain, addressing sections which show insufficient key
performance indicators (KPI). In addition, the KPIs can be used to identify and improve potential
weak parts of the platform itself.

The methods and prototypes are now under development, and will undergo detailed tests in late 2005.
First prototypic installations are planned for early 2006, and the project partners expect to be able to
provide a commercial product series no later than 2007.


Acknowledgement
The SPIDER-WIN project is carried out with financial contribution of the European Commission under the 6
th

Framework Programme, IST Project 507 601. The partners are Bentivogli (I), CIAP (I), Ducati (I), Fibertecnic
(E), Fraunhofer IPK (D), Joinet (I), P.M. (I), PZL Swidnik (PL), Sisteplant (E) and IMIK/University of Warsaw
(PL).


References
Berio, Giuseppe; Anaya, Victor; Ortiz, Angel: Supporting enterprise integration through a unified enterprise
modelling language. Proc. 16
th
Conf. Advances Information Systems Engineering for Networked
Organisations, Riga (Latvia) 2004, Vol. 3, pp. 165-176.
McLoughlin, Michael; Heavey, Cathal; Rabe, Markus: Research into developing a training tool federate in the
manufacturing systems domain. In: Mertins, K.; Rabe, M. (Eds.): Experiences from the Future. Fraunhofer
IRB, Stuttgart, 2004, pp. 341-350.
Mertins, Kai; Jochem, Roland: Quality-oriented design of business processes. Kluwer Academic Publishers,
Boston, 1999.
Rabe, Markus; Mertins, Kai: Reference models of Fraunhofer DZ-SIMPROLOG. In: Bernus, P.; Mertins, K.;
Schmidt, G. (Eds.): Handbook on Architectures of Information Systems. Springer, Berlin, Heidelberg, New
York, 1998, pp. 639-649.
Rabe, Markus; Jaekel, Frank-Walter: The MISSION project – demonstration of distributed supply chain simulation.
In: Kosanke, K.; Jochem, R.; Nell, J.G.; Ortiz Bas, A. (Eds.): Enterprise inter- and intra-organisational inte-
gration. Kluwer, Boston, Dordrecht, London, 2003, pp. 235-242.
Rabe, Markus; Jaekel, Frank-Walter; Dassisti, Michele; Erriquez, Mauro: Generic adaptor for distributed
simulation. In: Proc. Simulation and Visualization, Magdeburg (Germany) 2003, pp. 205-216.
Rabe, Markus; Jaekel, Frank-Walter: Distributed simulation and supply chains. In: Proc. Int. Conf. Competitive
Manufacturing COMA’04, Stellenbosch (South Africa) 2004, pp. 437-442.
Stich, Volker; Weidemann, Martin: Decision support for improvement of logistics performance in production
networks. In: Stanford-Smith, B.; Chiozza, E.; Edin, M. (Hrsg.): Challenges and Achievements in E-business
and E-work. Amsterdam et al.: IOS Press: Amsterdam 2002, pp. 638-645.
Supply-Chain Council: Supply-Chain Operations Reference Model - Overview of SCOR 5.0. Supply Chain
Council, Pittsburgh, 2002.

25

2.2 Making Build to Order a reality: The 5-Day Car
Initiative

Gareth Stone, University of Bath, mnsgms@management.bath.ac.uk
Joe Miemczyk, University of Bath, mnsjrm@management.bath.ac.uk
Rene Esser, ThyssenKrupp Automotive, Rene.Esser@thyssenkrupp.com

The automotive industry in Europe faces fierce competition in all its major markets and is dealing
with a customer who is less patient and more demanding in terms of vehicle choice. The current
system of making vehicles represents a ‘vicious circle’ where forecast based production and push-
based selling using discounts and incentives is leading to lower profits, thus more volume is needed
to maintain the equilibrium. Build to Order (BTO) strategies may offer a new direction for
manufacturers who suffer in this climate of spiralling costs and punctured profits. The article
discusses current usage of BTO and requirements for its successful implementation in the
automotive industry.


Introduction
The automotive industry in Europe faces fierce competition in all its major markets and is dealing with
a customer who is less patient and more demanding in terms of vehicle choice. Whilst vehicle
manufacturers develop shorter product lifecycles and offer a greater variety of models, this provides
shorter ‘market windows’ in which to generate the sales volume necessary to support the massive
development costs of a new vehicle. The current system of making vehicles represents a ‘vicious
circle’ where forecast based production and push-based selling using discounts and incentives is
leading to lower profits, thus more volume is needed to maintain the equilibrium (Holweg & Jones,
2001). Build to Order (BTO) strategies may offer a new direction for manufacturers who suffer in this
climate of spiralling costs and punctured profits.

The global environment of the automotive market has forced many companies to return to their
operations strategies with new eyes. Companies are moving or have moved from centralised
operations to decentralised operations in order to take advantage of the available resources and in
many cases, simply to be closer to their markets. Automotive manufacturers have undergone
numerous changes in terms of strategies, tactics, and operations with the aim of meeting the rapidly
changing business environment. Currently, companies have to compete based on a range of
performance objectives such as quality, price, delivery, responsiveness and flexibility (Miemczyk &
Holweg, 2001).

Consequently, the automotive supply chain is undergoing a major transition. Current vehicle supply
systems are for the most part based on ‘stock push’, whereby the majority of vehicles are sourced from
existing finished goods inventory in the marketplace. In the UK for example, only 33% of all new
vehicles sold in 1999 were actually built to order (BTO), and the remainder was sold from the average
two months stock held at any one time (ICDP, 2000). This is not a bad strategy and has worked well in
the past. However, there is a major flaw in the thinking. Not all the cars built will have a ready and
willing customer. The industry can deal with this situation by discounting, effectively paying the
customer to take cars off their hands. Customers are far more sophisticated, better informed, and
demanding and competition is global and oversupplied. The industry now understands the benefits of
moving towards a build-to-order model where production exactly matches the market requirements. A
plethora of ‘Lean production’ programmes have been launched but many have failed to address a key
philosophy: let the customer pull and produce only when an item is needed. It is mass customisation,
rather than mass production has become a major objective of automotive companies. BTO is now
becoming a popular operations paradigm after the success of its notable implementation at BMW.

26

To tackle this issue in a comprehensive manner. the Intelligent Logistics for Innovative Product Technologies
(ILIPT)

is funded by the European Commission’s 6
th
Framework Programme, running from July
2004 for 48 months. With an overall budget of over €16 Million (€9 million from the EC), the
overall objective of the project is to develop new methods and technologies to facilitate the
implementation of a new manufacturing paradigm for the European automotive industry. This new
paradigm, ‘the 5-day car initiative’, will approach the building of ‘cars to order’ in a reduced time
scale. This includes customer order at the dealership (or even at home) through all manufacturing
processes through to delivery to the customer. With the automotive industry’s current 40-day
capability, achieving a 5-day order to delivery cycle in Europe is certainly a far reaching challenge.

Although the project has as a broad remit, it will focus on three specific themes to ensure a holistic
approach to developing technologies that will enable the ‘5-day car’. Modular Car - aims to develop a
digital prototype of a passenger vehicle using an innovative system and new approaches of flexible
space frame and skin panels and modules for rapid build to order based on a high level of modularity.
Flexible Supply Network - aims to make significant breakthroughs in information and material
logistics by radically reducing time lags, stock piles and unused capacity which are the trademark of
the dynamic automotive network environment. And finally ILIPT will look at the Integration of
Complex Product Processes which will develop a vision and roadmap of the future automotive build-
to-order (BTO) network structures and processes. It will also develop methods to validate innovative
concepts in product design, network design and BTO processes for achieving a 5-day car.

This research investigates the way in which product design, production and information systems must
develop in order for Original Equipment Manufacturers (OEM) to meet their intention of
implementing BTO strategies. This chapter reports on the results of a number of deliverables within
the ILIPT project that deal with the state of the art (and in some cases practices). The analysis is based
on extensive reviews of previous studies (including the 3DayCar Project), surveys and interviews with
practitioners.


Current usage of BTO
The idea of building cars to order is not new. More than two decades have passed since lean
production set out with the promise ‘to build cars at the rate the customers demand it' (Minden, 1983).
Minden stressed that the overall objective of the Toyota Production System, the ancestor of the lean
philosophy (Womack and Jones, 1996) was to build cars to order only and hence avoid the substantive
waste or ‘mud’ inherent in overproduction. While lean efforts have allowed undeniable improvements
in manufacturing efficiency, from the customers’ perspective they often failed due to a myopic focus
on the factories. Indeed, 15 years after the publication of the seminal work ‘The Machine that Changed
the World' (Womack et al., 1990), few volume vehicle manufacturers are able to build to customer
order, and only one builds solely to customer order. In Europe, the average customer has to wait 48
days for their European car to be built and delivered to order. For Japanese models built in Europe,
the time reaches 63 days, and for European built specialist vehicles the lead time is 43 days (ICDP,
2000).


3
The research consortium is comprised of both industry and research organisations from across Europe including the new
member states. The 29 partners, coordinated by ThyssenKrupp Automotive AG, represent all tiers of the automotive supply
chain from vehicle manufacturers, through first tier suppliers, logistics providers and specialists in manufacturing and supply
chain management including small to Medium sized Enterprises (SME).

27


Figure 1: Ideal Order to Delivery time for UK customers (Elias 2000)

Despite these long lead times, research in the UK has shown that 59% of customers want their vehicle
to be delivered within 14 days or less as shown in Figure 1, based on 1039 UK customers who had
recently bought a new car (Elias, 2000). Even in the US, where vehicles traditionally are sold from
dealer stock, 74% of consumers would rather wait and order the vehicle instead of buying one from
the dealer’s lot that is incorrectly equipped, but the majority of North American consumers would wait
no more than three weeks to receive their vehicle (Business Wire, 2001).

As a result of the vehicle manufacturers’ inability to supply vehicles within the two-week timeframe
that most customers find acceptable, very few vehicles in the world are actually built to customer
specifications, and large stocks of finished vehicles are held to enable a degree of choice for
customers, generally at dealerships, but also at national importers, despatching lots, and the like.

This becomes particularly apparent in the US, where the distribution system is geared at ‘instant
gratification’ or ‘one-stop shopping’. Manufacturers hold as many as 100 days of sales in the
marketplace, and 60 days is considered to be the optimal level. The US is also the only country where
stock levels are published by the manufacturers themselves (due to a mutual agreement), whereas in
Europe stock levels are carefully guarded. However, in the UK 370,000 new vehicles are stored in old
airfields and dealer forecourts at any one time, so there is visible proof that overproduction poses
serious inventory challenges in Europe as well.

Japanese producers in Japan are most likely to build-to-order for the domestic market. One reason for
this is the high percentage of cars destined for export (in some cases up to 60% of production). These
exports are based on sales forecasts and can be used to buffer customer-ordered production for the
domestic market. Dealerships are tightly integrated into the planning process and are aided by the
stringent Japanese vehicle inspection system, called ‘Shaken’, which makes driving vehicles beyond
three years of age less economical. This results in predictable replacement planning on the part of
consumers and their dealers.

There is a huge cost penalty in current vehicle distribution as a result of holding large volumes of
stock, at the highest value (finished goods) and discounting that stock in order to sell. It has been
estimated that across Europe £6 billion could be saved by eliminating the finished vehicle stocks, and
an additional £2.5 billion annual profit potential could be achieved by further operational savings and
increased product mix profits due to more profitable options such as alloy wheels, air conditioning or
higher specification trim levels (ICDP, 2000).

28

Recent industry declarations have illuminated these issues, and many European manufacturers have
proclaimed their intention to implement build-to-order strategies in order to reduce sourcing from
stock. Probably, the best known of these is the project ‘Nouvelle Distribution’ by Renault. The aim
here is to produce a built to order vehicle in 14 days from order to delivery by the end of 2001. VW
and Ford both have similar 14 or 15-day car programmes, with BMW even attempting a 10-day order-
to-delivery (OTD) lead time (Automotive-World, 2001). Volvo was also an early adopter of this
model with a plan to reduce OTD times from 6 weeks to 28 days in 1990, and even down to 14 days in
1995 – whilst having a 100% 'customer-based' production (Hertz et al., 2001). It has been recognised
that delivery speed for orders built at the factory are too slow. Although this is only partly due to
transport, in the US, Ford with average transport delivery of 15 days (and up to 30 days) aims to
reduce this to 6 days. Ford has enlisted the help of UPS to help co-ordinate this plan and achieve
inventory savings and use their tested tracking tools and software (Connelly, 2000). Table 1 shows the
targets vehicle manufacturers have set themselves in recent years. Although many vehicle
manufacturers (VMs) can achieve this timescale today, it is still for the minority of customers. A
significant delay is still likely if a customer orders an unusual combination of extras such as air
conditioning and sun roof on a small engined car.


Table 1. Order to delivery targets of Vehicle Manufactures (Source: Miemczyk and Holweg 2004)

Building vehicles to order as opposed to a forecast has a clear logic – both for the vehicle
manufacturers and the customers. Such a drastic change in strategy however would have ramifications
for all players in the supply chain – for suppliers as well as inbound and outbound logistics operators.
The implications for the vehicle manufacturers have been reported previously (Holweg and Jones,
2001, Holweg and Pil, 2001), showing that manufacturing could assemble a vehicle in less than one
day, yet the order takes over a month from the dealer to the factory. Considering a five-day car
scenario with a manufacturing lead time of one day, this would leave one day to receive orders and
inform suppliers, and three days to deliver the finished vehicle to the customer.

The lead-time reduction efforts will initially have to focus on the vehicle manufacturer, which
currently accounts for 88% of the OTD lead-time despite final assembly only taking 4.5-20 hours
(Holweg and PiI 2001), but subsequently affect all subsystems - component suppliers, vehicle
manufacturers, and logistics operators. Additionally, increased flexibility will be demanded from
suppliers and logistics operators to cater for the variability resulting from an increased percentage of
cars built-to-order in short lead-times. Thus, a shift toward BTO will require not only the
manufacturers to radically alter their processes, but has equally drastic ramifications for all players in
the supply chain, in particular for suppliers and logistics operators.
Vehicle
Manufacturer
Programme Name Order-to-Delivery Target
BMW COSP –Customer Oriented Sales
Processing
10 days
DaimlerChrysler FastCar / Global Ordering 15 days
Ford Order-to-Delivery 15 days
General Motors Order-to-Delivery 20 days
Renault Projet Nouvelle Distribution (PND) Initially 14 days, revised to 21
days in 2002
Nissan SCOPE (Europe), ANSWER (Japan),
ICON (USA).
14 days
Toyota N/a 14 days
Volkswagen Kunde-Kunde (‘Customer-to-Customer’) 14 days
Volvo Distribution 90 14 days
29

An interesting parallel here is the application of BTO in the consumer electronics sector, in particular
at Dell Computers, which has been very successfully applying the concept. The critical difference
between the automotive industry and Dell however is that Dell - strictly speaking - uses an assemble-
to-order (ATO) strategy (based on Mather 1988), whereby it keeps sufficient component inventory on
site in order to build computers to order once the order is received. This inventory therefore represents
a de-coupling point in the supply chain that largely buffers the component supply chain from demand
variability and lead-time reduction requirements. In comparison to the automotive industry however,
Dell only handles 15-50 key components per computer, whereas the average vehicle consists of 2,000-
4,000 components - many of which are customised by colour, engine size, trim level, and body-style
(Holweg and PiI 2001). Thus, holding component stock in the same fashion is cost-prohibitive in the
automotive industry, and BTO strategies are dependent on a capable and responsive supply chain, and
subsequently, on the logistics operations that connect suppliers to the assembly plants, and assembly
plants to dealers and customers. Renault's recent experience marks a case in point. In 2002, it had to
revise its order-to-delivery lead-time target from 14 days (set in 1999) to 21 days - not because of any
inflexibility in the manufacturing operations, but because their distribution logistics operations could
not cope with the lead-time reduction (de SaintSeine 2002). Initial research into the outbound logistics
operations (Holweg and Miemczyk 2002; Holweg, Miemczyk, and Williams 2001) and the order-to-
delivery process (Holweg and Jones 2001; Holweg and PiI 2001) has already been reported,
confirming the necessity of adopting a holistic approach to logistics as the connecting element in the
system.


Product architecture
ILIPT research has shown that by considering the vehicle as a ‘product architecture’, characteristics
such as body construction and vehicle complexity can affect total lead-time, not just in production.
This reinforces and broadens the significance of vehicle design, now included as part of the bigger
picture or seen as part of ‘the whole’ in terms of the order fulfilment process. Design engineers who
think their boundaries begin and end at production must reconsider, because vehicle construction and
the links with complexity and process reliability have a wide- reaching impact that stretches across the
delivery process. In the future, effective screening, management and integration of new technologies
by VMs and their partners is essential in order to maximize the potential for rapid BTO.

In the 3DayCar project Howard (2001) identified a number of requirements that would lead to the
capability to support build to order in short lead times shown in table 2.

Table 2. BTO requirement, technology and capability (source Howard 2001)
Requirement Technology/capability
Construction
• Develop body appropriate to
configuration requirements
• Steel or aluminium space frame
‘independent body and panels’

• Decouple paint • Colour thermoplastic panels
• Painted body storage tower
Complexity
• Minimise complexity in design and
production
• Modular body in white construction
• Multiplex / Controller area Network
• Roof, door front rear end modules.
• Late configuration
Capacity
• Increase volume flexibility • Modular body construction
• Platform / parts sharing
• Production Optimisation software




30

Construction requirements
Vehicle design must be scrutinised within the context of its impact on the bigger picture. It is not
limited to the design process alone but on the total supply chain. Minor improvements to the product
architecture may have far reaching and dramatic effects across the total supply chain (Clark &
Henderson 1990).

The welded steel monologue (the vehicle body frame with a structural skin) has served the industry
well and has provided the platform for the development of volume car production throughout the 20
th

century. High levels of economies of scale were achieved from the standardisation of parts and the
high volume production of common pressings. This all-steel body, abandoned the separate body &
chassis design of earlier models and developed the unibody or monocoque body structure, becoming
the single most important factor in the economics of vehicle construction and which is still applied to
this day (Wells & Nieuwenhuis 1997, 2000). However, developing a BTO process in 21
st
century
production using the conventional steel monocoque body presents a significant challenge. The
principle of the monocoque is based on a ‘structural skin’. Outer body panels are combined as part of
the overall load bearing entity of the vehicle, which then requires all operations to be performed on-
line and in a fixed operational sequence: press - weld - paint - final assembly. This method of body
construction, synonymous with large batch sizes and painted body buffers, limits responsiveness and
the ability to build-to-order (Howard & Miemczyk 2000). Despite the predominance of the
conventional steel monocoque body, there is growing interest by volume manufacturers in alternative
methods of construction.

Space frame construction is particularly relevant to successful BTO, because the structural frame and
non load-bearing panel design can ‘de-couple’ parts of the production process (Hoeskstra & Romme
1992). Customer orders can be attached later in production to a generic frame that is customised in
final assembly. Wasteful, environmentally unfriendly processes such as vehicle painting can be
removed from the assembly line almost entirely, where panels can be outsourced to suppliers or
moulded as coloured thermoplastic panels. Alternative vehicle construction can enable BTO by the
late configuration of finished exterior panels at final assembly or even at the dealer service centre.


The complexity issue
Many commentators have now converged and agree that automotive manufacturers can no longer
follow the traditional mass production model of capturing market share and high profits by producing
large volumes of standardised product (MacDuffie, 1996, Holweg & Pil, 2001). The automotive
industry is now required to offer ever increasing levels of product variety. This necessitates from
environmental factors such as, rises in energy prices, trade structures, internationalisation of the
market and importantly the increasing sophistication of customers. We are now witnessing the result
of these changes in the number of models and model variants that are available or being offered on the
global market. We find it important to distinguish between vehicle complexity and variety; both of
which represent a major inhibitor to rapid BTO. Complexity is defined as the level of internal
component variation handled by manufacturing and logistics operations sufficient to construct the
vehicle. Typical automobile complexity today comprises around 4,000 to 5,000 parts (Fine &
McDuffie 2000). Variety is defined as the level of product choice offered to customers (Batchelor
2000). High levels of variety are normally associated with an increase in parts complexity, leading to a
trade-off between the variety offered in the marketplace, the volume of production and the
effectiveness of manufacturing operations (Slack et al., 1998). Variety can range from 820
specifications: Nissan Primera, to 3 billion: Mercedes Benz S-class (Holweg & Pil 2001). A
dichotomy exists between product variety and inventory, which obstructs just-in-time production with
high variety environments (Bennet & Forrester 1993).

Two approaches can be used to minimize the impact of component complexity on production:

1.) Process based approaches, such as flexible manufacturing equipment.
31

2.) Product based approaches, which allow for high variety whilst reducing the level of
component variation, using product platforms, modular design concepts and component
standardisation.

On what basis in the future will VMs adopt a responsive manufacturing strategy: production / process-
based or innovative design and new technology? Whilst these questions are not mutually exclusive,
alternative body construction and innovative vehicle engineering in partnership with key suppliers is
emerging as an increasingly popular approach. Examples of flexible manufacturing equipment include
painted body storage towers. Painted body storage towers are considered a short-term enabler to BTO,
allowing unreliable upstream processes to be de-coupled and enabling the customer order to be
attached after vehicle painting.


Figure 2: A painted body storage tower (Source: Three Day Car Initiative)

The prospect of adopting BTO and coping with an increase in customised orders presents a serious
dilemma to VMs. Greater understanding is required of how future component assembly and supply
may operate and how complexity can be kept at a realistic level. Tools such as the ‘Production Variety
Funnel’ (Hines & Rich 1997) were originally used to compare the complexity and lead-time of
internal operations between industries. It is used to highlight the build-up of parts complexity in
vehicle production.

Modularity can be applied to the management of complex systems, by breaking them into parameters
and tasks that are interdependent within and independent across vehicle architecture (Baldwin & Clark
1997). Pine (1993) also argues that mass-customised products can be achieved through component
sharing and product modularity. Three types of modularity are of interest in rapid BTO: modularity in
design (MID), modularity in production (MIP) and modularity in use (MIU) (Sako & Murray, 1999).
Their benefits include, MID: reductions in complexity resulting from reduction in parts, MIP: leaner
production due to less operations performed on the line and MIU: higher product variety by offering
customers a choice of modules. Minimizing internal vehicle complexity enables BTO by reducing
inventories, part count and assembly lead time. The current emphasis on integration of vehicle
modules for specific models, conflicts with the popular notion that the automotive industry should
increase parts sharing and standardisation (Wallbank 2000). However, where styling is not an issue
this can often be overcome.

In addition to performance and packaging benefits, reduced complexity is achieved through the
integration of structural, exterior, cooling, electrical and lighting content.
32


Figure 3: Example of a front end module – Visteon

Nowadays almost every OEM faces the problem of increasing product complexity and reduced
lifecycle time, having to develop and manufacture a larger amount of variants within less time-to-
market.


The capacity requirements
In a true BTO environment customer demand may fluctuate on a daily, even hourly basis, requiring a
system that is more responsive but not significantly more costly in terms of capital investment or
inventory. Conventional capacity management involves long range forecast planning, management of
bottlenecks, smoothing of product schedules and balancing stocks of components to optimise the
peaks and troughs of demand. Whilst capacity has been traditionally governed more by production
planning and inventory management than by the configuration of product architecture, current
measures such as batch size, inventory and lead-time may be improved by applying principles such as
Design for Assembly (Boothroyd & Dewhurst 1987) and Design for Manufacture (Ettlie 1987).

Consensus over capacity remains divided, but remains a high priority on business and political
agendas. ‘Maximizing capacity’ in the automotive sector typically signals a productivity-driven
response. Traditionally this involves: either increasing plant labour levels, adjusting line speed and
adding extra shifts, or closing down older less efficient factories. The emergence of concepts such as
BTO, customer-pull and responsiveness matches the increase in specialist vehicles and niche markets,
represented by the sports utility sector. Is it possible that the start of the new century coincides with a
genuine shift in thinking, away from production-push and economy of scale, towards customer-pull
and economies of scope? In Europe, the emergence of innovative metal-forming techniques and
materials is a key factor in platform sharing, spreading the cost of investment across many models.
Also, the delivery of modules by 1
st
tier suppliers located in adjacent parks can reduce complexity and
lead-time in production. The scope of vehicle design appears to be increasing, therefore, including not
only product characteristics but also the means for its delivery.

Another related and important issue is that of reliability in the production process. Unreliable
production and delivery process encourages dealers to sell from stock, rather than place orders on the
factory and perpetuates the stock push system. Furthermore, customers do not know whether the car
they receive is factory fresh or has been in stock for months. Each customer order must become a
batch-size-of-one, meeting exact customer requirements in terms of specification and delivery date.
Therefore, a sea change in mindset is required to shift towards BTO, suggesting dramatic rises in
flexibility and responsiveness across supply chain partners. In production this is traditionally achieved
in two ways: either through increasing labour levels and investing in equipment, or by developing new
design solutions, such as adopting modular assembly.



33

Platform and parts sharing parts
Currently the platform strategy is probably the most widely
accepted answer to this challenge of maintaining
capacity levels despite growth in niche vehicles.
The platform strategy enables car
manufacturers to offer a complete range of
product families of vehicles with different
appearance, style and image, based on common
and standardised technology. In this way development costs
can be allocated to a larger amount of vehicles and
development time can be reduced by
simultaneous engineering. Production cost can
be reduced through a higher level of
standardisation. Moreover, vehicles that
belong to the same product family can easily be
manufactured on a shared production line, adding more flexibility to react to
fluctuations in demand, lowering the investment costs especially for lower volume cars.

At first, sharing was viewed as something to be kept as quiet as possible — Jaguar was initially keen
to play down connections between its recent cars and Fords—, but now car companies are throwing
caution to the wind and setting about sharing as never before. According to Richard Parry-Jones, in
charge of Ford engineering worldwide, it’s like Lego. Basically, each of the brands owned by Ford —
including Mazda, Volvo, Jaguar and Land Rover — can pick what they want from a shared box of
parts.

Platform strategy is perfectly in line with the trend in the automotive industry to merge into huge
conglomerates, continuously reducing the number of independent car manufacturers. Within a car
manufacturer group, each brand represents a different image, lifestyle or price segment.

Already introduced in the early nineties, Volkswagen officially embarks on the platform strategy since
1998. When it comes to sheer numbers of variations, Volkswagen AG is implementing the world’s
most ambitious common platform scheme. Each future model from Volkswagen, Audi, Skoda and
Seat will be based on one of four main platforms. The goal is to cut development time, reduce build
time, improve quality and slash prices.

The centrepiece of VWs strategy is the A platform that debuted on the Audi A3 and the Skoda
Octavia. Supporting both front- and four-wheel drive, the A (PQ34) platform is featured in the Golf
IV and several derivatives, including the Bora, Golf wagon, as well as the Volkswagen Beetle, the
Seat Leon and the Seat Toledo. Currently the A (PQ35) platform supports VW Golf V, VW Touran,
Audi A3, Skoda Oktavia II, VW Passat.

Some manufacturers described their development of new body structures as experimental. This is
surprising, particularly when the very high level of investment seems to be at odds with the relatively
low annual production volumes in most cases. Other factors may be at work here in addition to the
drive to streamline production and minimize delivery lead-time, such as the need to develop
lightweight, fuel-efficient vehicles for the future. A significant hurdle facing all manufacturers is
generating a suitable return in the notoriously competitive small car market. The principle of BTO
may seem sound in theory, but in reality the cost of converting existing production sites or borrowing
sufficient capital for new projects may stretch the rules of amortization too far for some
manufacturers. It is ironic that perhaps only the biggest will be able to fund the smaller, more flexible
plants of the future. Confronting the challenge of BTO requires more than product changes; support
ICT will be inevitable and is discussed in the next section.




34

Software solutions to increase capacity flexibility
Production Optimisation software provides real-time capacity planning, based on plant-specific
constraints. But neither Body storage towers, nor Optimisation software can ultimately be considered a
substitute for implementing best practice across the total production process. The emphasis of all
future developments within the ICT systems of the automotive industry will lie on collaboration,
seamless information-flow, and monitoring. New technologies such as Radio Frequency Identification
(RFID), and extended system processing capacities will allow the processes we have just described to
become a reality.

One of the biggest challenges to overcome, in order to make this vision reality, will be the
development of standards for planning and execution processes. Demands and capacity levels of all
relevant tier suppliers, the Vehicle manufacturers and Logistics Service Suppliers must be able to be
simulated throughout the whole supply chain. During the production, all execution processes must be
able to be monitored, allowing the tracking and tracing of the physical material flow. In order to do so,
standards will be needed concerning the handling of all these processes. Due to the lack of standards
within today’s automotive industry, isolated applications have developed. Future standards must be,
vendor independent taking the requisitions of all stakeholders within the industry into account,
worldwide applicable and generally accessible.

In the near future individual stakeholders within the supply chain will be using different ERP tools as
in today’s heterogenic IT-landscape. A system architecture is needed that is able to transform all
diverse formats into one system. Using this approach, it will be possible to handle the supply network
from every level of the supply chain. The basic element of the system might be an ASP approach as
tested during the LiNet Project in the Years 2001 – 2004. This approach allows a centrally hosted,
decentralised architecture with individualised levels for each participant.

Since all relevant components are equipped with Communication technology (CT: e.g. RFID), it is
possible for suppliers and customers to monitor and view the movement of parts within the supply
chain. In today’s world, IT systems are mainly used to document past activities, or simulate future
planning & execution processes. CT on the other hand will make it possible to see the process as
reality, and allow tracking the variance between planned and real-time execution processes. In the near
future each part will have its own IP-address, stating its module/part number, planned distribution
route, as well as its current physical location and status of production. Furthermore, the parts will be
able to communicate directly into the supply chain if interruptions appear. Overall, new
communication technologies (CT) will allow the connection of reality with IT will be made reality for
the first time.


Conclusions
In conclusion, new vehicle supply systems, still largely based on a mass production paradigm or
"push" logic, are proving increasingly less competitive in today's markets. Current "stock push"
vehicle supply, whereby the majority of vehicles are sourced from existing finished goods inventory in
the marketplace, requires holding stock at the most expensive point in the supply chain (Fisher 1997).
Furthermore, the sales incentives needed to shift this inventory incur a huge cost penalty in current
vehicle distribution.

The key obstacle for not building an increasing percentage of vehicles to customer order is the vehicle
manufacturer's inability to supply customised vehicles within timeframes customers find acceptable.
Average order lead-time for customised vehicles ranges from six weeks in Europe to ten weeks or
more in the U.S. - yet consumer research has shown that the majority of new vehicle buyers expect
their vehicle within two to three weeks, a figure which is fairly consistent across Europe, Japan, and
the U.S.

While production flexibility can benefit more responsive build to order production, designing vehicles
for BTO has perhaps the greatest potential to maximize on customer preferences. In identifying the
key technological opportunities for BTO in vehicle design we have shown that by considering the
35

vehicle as a ‘product architecture’, characteristics such as body construction and vehicle complexity
can affect total delivery lead-time, not just in production. This reinforces and broadens the significance
of vehicle design, now included as part of the bigger picture or seen as part of ‘the whole’ in terms of
the order fulfilment process. In the future, effective screening, management and integration of new
technologies by VMs and their partners is essential in order to maximize the potential for achieving
the 5-day car.


References
Abrahamsson, M., (1999): Time-based distribution", in: Hadjiconstantinou, E., Quick Response in the Supply
Chain, Springer-Verlag, Heidelberg.
Automotive-World, Build-to-Order.
Batchelor, J. (2000): Human Performance in Planning & Scheduling: Fieldwork Study, Methodology and
Research Lines, Taylor Francis.
Bennett, D., & Forester, P., (1993): Product variety and just in time: conflict and challenge, in: Proceedings of
the 1st International Symposium on Logistics, University of Nottingham.
Boothroyd, G., & Dewhurst, P. (1987): Product Design for Assembly, Wakefield.
Bowersox, D., Closs, D. (1996): Logistical Management: The Integrated Supply Chain Process, McGraw-Hill,
New York, NY.
Business Wire, "Gartner survey shows US consumers prefer concept of build-to-order when buying an
automobile", Business Wire.
Christopher, M. (1992): Logistics and Supply Chain Management. Strategies for Reducing Costs and Improving
Services, Prentice-Hall, London.
Clark, K. & Henderson, R. (1990): Architectural Innovation: The Reconfiguration of Existing Product
Technologies and the Failure of Established Firms, Administrative Science Quarterly, March
Connelly, M. (2000): UPS helps Ford speed deliveries, Automotive News, 74, 5860, p.3.
DTI (2000): Report of the Alternative Fuel Group of the Cleaner Vehicles Tack Force. An Assessment of the
Emissions Performance of Alternative and Conventional Fuels.
ECG (2001): Survey on Vehicle Logistics, European Car-Transportation Group of Interest, Brussels.
Elias, S. (2000): New vehicle buyer behaviour - quantifying key stages in the consumer buying process.
Ettlie, J., & Reifeis, S. (1987): Integrating Design and Manufacturing to Deploy Advanced Manufacturing
Technology, Interfaces 17/6 November / December.
Fine, C., & MacDuffie, P (2000): E-volving the Auto Industry: E-Business Effects in the Automotive Industry,
Research proposal to the Sloan Foundation on behalf of the International Motor Vehicle Programme
Fisher, M. L. (1997): What is the Right Supply Chain for your Product? Harvard Business Review 75, March-
April (2), pp.105-116.
Gregory, J., 1999, "Vehicle logistics and the three-day car programme".
Hall, R. (1993): The three-day car challenge, Target, 9, 2, pp.21-30.
Hertz, S., Johansson, J.K., Jager, F. (2001): Customer-oriented cost cutting: process management at Volvo,
Supply Chain Management: An International Journal, 6, 3, pp.128-41.
Hines, P. & Rich, N. (1997): The Seven Stream Mapping Tools, International Journal of Operations &
Production Management. Vol 17, No 1.
Hoekstra, S. & Romme, J. (1992): Integral Logistics Structures: Developing Customer Orientated Goods.
McGraw Hill, London.
Holweg, M., J. Miemczyk, and G, Williams. (2001). The 3DayCar Logistics Study. How to organise automotive
logistics in a build to order environment. Solihull, 3DayCar Programme.
Holweg, M., Jones, D.T. (2001): The challenge of building cars to order: can current vehicle supply systems
cope?, Taylor, D.T., Brunt, D.C., Manufacturing Operations and Supply Chain Management, Thomson
International, London.
Holweg, M., Judge, B., Williams, G. (2001): The three-day car challenge: cars to customer order, Logistics
Focus, pp.36-44.
Holweg, M., Pil, F. (2001): Successful build-to-order strategies start with the customer, Sloan Management
Review, 74-83.
Howard, M. & Miemczyk, J. (2000): Painting the 3DayCar, World Automotive Manufacturing No. 28 August.
Howard, M., Young, K. and Graves, A. (2001): Towards the 3DayCar: Vehicle Design and its impact on Rapid
Build-to-Order, International Journal of Vehicle Design. Vol.26, No 5. pp. 445-468.
ICDP (2000): Fulfilling the Promise: Is there a Future for Franchised Car Distribution?, International Car
Distribution Programme, Solihull.
MacDuffie, J-P., Sethurman, K., & Fisher, M (1996): Product Variety and Manufacturing Performance:
Evidence from the International Automotive Assembly Plant Study, Management Science. Vol. 42, No 3.
36

Miemczyk, J. and Holweg, M. (2004): Building cars to customer order - what does it mean for inbound logistics
operations? Journal of Business Logistics, Volume 25, Number 2 pp.171-197
Miemczyk, J., Holweg, M. (2001): Building the `three-day car': implications for inbound logistics operations.
Monden, Y. (1983): Toyota Production System, Industrial Engineering and Management Press, Norcross, GO.
Pine, P. (1993), Mass Customization, Harvard Business School Press, Boston.
Sako, M & Murray, F. (1999): Modules in Design, Production and Use: Implications for the Global Automotive
Industry, A paper prepared for the IMVP Annual Sponsors Meeting, October.
Wallbank, E. (2000): Commonisation and the Effects on ‘World’ Cars: IT Integration Between Suppliers and
OEMs, Proceedings from International Automotive Conference, Sunderland, April.
Wells, P. & Nieuwenhuis, P. (1997): The Death of Motoring.’ John Wiley.
Wells, P. & Nieuwenhuis, P. (2000): Why big businesses should think small, Automotive World. July/August.
Womack, J., Jones, D.T. (1996): Lean Thinking: Banish Waste and Create Wealth for Your Corporation, Simon
& Schuster, New York, NY.
Womack, J., Jones, D.T. (1996): Lean Thinking: Banish Waste and Create Wealth for Your Corporation, Simon
& Schuster, New York, NY.


37

3 3. . E Es st ta ab bl li is sh hi in ng g a an nd d m ma ai in nt ta ai in ni in ng g S SM ME E n ne et tw wo or rk ks s
3.1 Co-operative SME networks in manufacturing
Matthias Nöster, SFC, noester@sfconsulting.net
Marita Gruber, University of Vienna, School of Business, Economics, and Statistics,
marita.gruber@univie.ac.at

The theory about Virtual Organisations (VOs) is well advanced and provides numerous concepts
and classifications, but the practical understanding of structure and success factors of VOs is still
limited. This research aims to identify common and diverging structural aspects of VOs and
discusses a number of success factors for a particular type of a VO – the co-operative business
network. Using mainly qualitative research methods, we used a four step approach for our research.
We developed a Virtual Profile model and classified 30 VOs. Then we carried out in-depth
interviews with selected co-operative networks. Finally, both data sets were analysed using
qualitative methods. The results of the analysis reveal the following factors as the most crucial in co-
operative networks: business opportunities, trust, commitment to co-operation, exchange of
information and network management.


Introduction
In theory, “co-operation and concentration on core competencies are two complementary strategies
regarded as a possible loophole from inflexibility and inefficiency” (Schräder in Wüthrich et al., 1997,
p.43). Virtual Organisations (VOs) have been proposed as an organisational form suitable to combine
the two aspects while at the same time guarding a maximum of flexibility for the individual
participating companies.

While Byrne et al. (1993, p.36) determine a VO as “(…) a temporary network of independent
companies – suppliers, customers, and even rivals – linked by information technology to share skills,
costs, and access to one another’s markets.”, Sieber/Griese, (1998, p.213) define it as “(…) an
appropriated kind of co-operative organisation to explore business opportunities that one enterprise
itself would not be able to work out.”

In an extensive analysis, Gruber/Koeszegi/Nöster (2005) identified two clusters of VOs: the more
focal, initiated networks, comparable to vertical networks and the more equal, initiated networks,
comparable to co-operative business networks.

Dell as one of the most cited examples of a Virtual Organisation (see for instance Saabeel et al., 2002,
Davidow / Malone, 1992, Wüthrich et al., 1997) is an example of a vertical network with the original
equipment manufacturer (OEM) as the general contractor. This kind of networking structure is
characterized by unilateral business relations, i.e. a fixed buyer and supplier position (Albers et al.,
2002), and a supply chain network topology (see Figure 1). The buyer can often exert significant
market power on the suppliers given the fact that the switching costs within the network are kept low
(minimal legal and organisational ties). While this structure may be very beneficial for the prime
contractor, it can leave the suppliers in a critical state of dependence (Davidow / Malone, 1992, p. 164)
or even exclude SMEs that cannot implement the dominant company’s processes and meet its tight
standards.

The VERITAS project, however, aims to strengthen the competitiveness of SMEs and therefore the
focus was placed on a different model of a VE, the co-operative business network. Such networks are
rather horizontal and are based on alternating customer-supplier relations, where one time a company
sub-contracts products or services from the network and another time supplies products or services to
another partner in the network. Accordingly, co-operative networks display a peer-to-peer network
topology. This type of network seems to correspond more than the vertical ones to the actual business
38

concept of a VO, where companies stay “… economically self-determinant and autonomous and act as
equal partners while working together.” (Picot, A./Reichwald, R./Wigand, R.T., 2003).

In the framework of the groundwork research performed for the VERITAS project, we identified a
large number of VO cases and analysed whether structural aspects have an impact on the success of
the co-operation. A second objective was to understand the processes in the networks in more detail in
order to derive some critical success factors of co-operation.

The aim of this specific paper is to provide practitioners with some indication of critical success
factors in the practical implementation of co-operative business networks. After a discussion of the
structural settings of VO, based on the analysis of 30 VO cases, we therefore turn to the interpretation
of critical success factors, identified in in-depth interviews with selected VOs.


Research approach

Sample
A sample of 30 Virtual Organisations was drawn from a list of more than 65 identified virtual
networks, discussed in recent academic or industrial literature (e.g. Wüthrich et al., 1997; Mertens/
Faisst, 1997 or Sieber/Griese, 1998). According to the focus of VERITAS project and the purpose of
this analysis, the sample selection was based on the following criteria:

(1) Concentration on network organisations in the manufacturing sector,
(2) Integration of examples of different countries, and
(3) Focus on SME
4
-networks.
Industry Sector
number of
examples percentage
Manufacturing (in a wider sense) 5
Manufacturing of machinery 10
Manufacturing of motor vehicles 3
Manufacturing of Aircraft 2
Manufacturing of Furniture 1
Manufacturing of special machines 2
Manufacturing total 23 76,67%
Construction 1 3,33%
Wholesale Trade 1 3,33%
Energy sector 1 3,33%
Environmental services 1 3,33%
Software publishing 1 3,33%
Entertainment Industry 2 6,67%
30 100,00%
Table 1: Industry sample

The resulting sample (Table 1) highlights a strong emphasis on the manufacturing sector (77%). 57%
of the examples contain mainly SMEs and the networks are located in 10 different countries. Two
thirds of the VOs consist of a maximum of 50 companies. From this list, we selected seven VOs for
additional in-depth interviews according to their affiliation to the manufacturing sector and a
dominance of SME members.




4
SME=Small and Medium Enterprises
39

Method
We applied qualitative research methods based on the approach of Grounded Theory (Glaser / Strauss,
1967 or Douglas, 2003) and followed a four-step-approach:
1. Firstly, based on literature review, we developed a “Virtual Profile” including 10 items to characterize
a typical VO. This analytical framework was established in order to determine the structural setting of
VOs.
2. In a second step, two independent coders profiled each of the cases in our sample according to the
established scheme. To determine the values expert rating based on information available from
desktop research and from personal communication with members of the VOs was used.
Furthermore, coding results were compared and discussed in order to establish consistency between
coders.
3. Third, we carried out in-depth interviews with knowingly successful network organisations out of our
sample, following a semi-structured interview guideline. Through the interviews, we gained
additional qualitative data about the structure, co-ordination and management of these networks.
4. Following the Grounded Theory Approach (Glaser / Strauss, 1967 or Douglas, 2003) we clustered
and re-clustered the established profiles according to different external (e.g. country of origin,
industry) and internal (e.g. coordination processes, distribution of power) dimensions for comparison.
Additionally, the data from the interviews was analysed to identify factors most common and most
diverging across the examples.
For the in-depth interviews, seven networks were selected according to their affiliation to the
manufacturing sector and a dominance of SME members. We interviewed the network managers/coaches
and in some cases also the participant firms, of four German, two Swiss and one Mexican networks.
Except for one Swiss network, the networks were created after 2000. In total we carried out nine
interviews, recorded them on tape and transcribed the tapes afterwards. The interviews lasted between
one and three hours and covered the aspects listed in Table 2.

Main category Sub-category
Network Description
Vision/strategy of network and members
Number and types of projects
STRUCTURE
Role of broker/coach
Factual co-ordination/management aspects COORDINATION/
MANAGEMENT
Personal co-ordination/management aspects
Network manager BUSINESS CASE/ADDED
VALUE
Network member
Table 2: Methodological framework for interview analysis

The interview data was then analysed in a content analysis based on the principles of (Mayring, 1983)
and cross comparison between the networks was used in order to identify similarities and differences
in the various categories.

While the whole analysis of the 30 VO cases was based on quantitative and qualitative methods,
according to the Generalisation Model (Srnka/Koeszegi, 2004), we focus on the results of the
qualitative analysis here. The latter turned out to provide more relevant results for the practical
implementation of co-operative business networks, which are presented in the following chapters.


Findings

Virtual profiles of VOs
In parallel with identifying VO examples, we developed a profiling model in order to determine the
structural settings of VEs and to relate these to success or failure of the VOs where possible. The
scheme was based on the assumptions of a Virtual Degree (see for example Bauer/Koeszegi, 2003 or
Scholz, 1996), where it is assumed that in reality no organisation can be a perfect virtual organisation,
but displays a specific virtual degree on a continuum with two extremes. Drawing on earlier research
40

such as Bultje/van Wijk, 1998, Bauer/Koeszegi, 2003 and Gulati et al., 2000, we developed our own
framework consisting of 10 criteria
5
, shown in Table 3. Using an additive model, the assumption was
that the more towards the left hand side the value was, the more the example would correspond to the
virtual concept. Table 3 shows a typical profile for each network type – the vertical network and the
co-operative business network. While some differences in values are less significant, a rather clear
distinction is possible in a number of criteria, such as evolution, aim/vision, scope, dispersion of
power and ICT. The criteria are discussed in detail below.

DIMENSION ITEM
Vertical
Network
Cooperative
business
network
5 4 3 2 1 5 4 3 2 1
Evolution (emerged - initiated) X X
General Industry Ties (strong - weak) X X
1. Set-up
environment
Ties between Partners (strong - weak) X X
2. Purpose Aim / Vision (well defined - non defined) X X
Focus on Core Competencies (high - low) X X
Scope (international - local) X X
Standardization of GS (low - high) X X
Dispersion of Power (equal - focal) X X
3. Organizational
framework
Use of ICT (extensive - limited) X X
4. Time horizon Duration (temporary - long-term) X X

Table 3: Comparison of two typical Virtual Profiles

According to our data set, two possible VOs development paths exist in practice (evolution): (1),
through the active support of a third party, (2) when companies identify a direct business opportunity
and initiate a network with partners based on previous business relations (compare also to Hannus,
2004). In our data set, half of the networks emerged, while the others were supported and also funded
by an external institution (industry association, university, consultant, regional authority). We could
not find any patterns that initiated networks have fewer chances to survive than emerged ones.

Two thirds of the investigated VOs had been established in an industry where companies tend to form
clusters regularly (industry ties), indicating a positive influence on readiness to co-operate from the
outside. Of our sample of 30 VOs 11 had even participated in a cluster or network before. The in-
depth interviews (see below) confirmed that often the general managers of the network members had
known each other before (existing ties between partners), which helped them to establish trusted
relationships more easily.

More than two thirds of the VOs had formulated at least a vague vision/aim, a factor that seemed to
gain importance in recent years. The latter might be due to the fact that funded VO projects (these
initiatives have increased since 2000) require the companies to have developed a clear strategy in
order to succeed with their proposal. Another reason might be that the newer networks have learnt
their lesson from some of the earlier failures.
The following five of the criteria from our model belong to the group of organisational factors and
describe the core structure of a Virtual Organisation.

Focus on core competencies (high vs. low) describes the degree of functional specialisation in the
network considered essential to create a best-of-breed VO. The possibilities for companies in networks
to focus on their core competencies were also emphasized in the theory of the resource based view of
strategy (Prahalad, C.K./Hamel, Gary, 1997). A strong focus on the core competences could be
identified in 24 out of 30 cases.

5
For detailed explanations, please refer to the VERITAS deliverable D1.1, available for download at
www.veritas-eu.org.
41


Scope (international vs. local) aims at classifying the geographical extension of a network, as it is
often argued that VOs are borderless and span across countries or even continents. We investigated in
12 local, 4 regional and 14 international networks and from the interviews concluded that most of the
VOs aim at international markets and customers, but are networks of locally or regionally linked
partners.

Standardisation of governance structure (low vs. high) refers to the extent of formalisation of roles
and rules in VOs. The latter are supposed to act with a minimum of rules or contracts and a great
amount of mutual trust, which is seen as an instrument to reduce opportunistic behaviour (Granovetter,
1985 or Koeszegi, 2001). The standardisation of the governance structure was found to be one of the
most diverse factors in our sample. While half of the VOs had established rules and defined roles, one
third disposed of very few standardised mechanisms for governance.

We also included the criterion Dispersion of Power (equal vs. focal), describing the influence and the
dependency between network partners. As discussed in the introduction, our assumption was that
equal power dispersion would be more feasible for SMEs than a focally dominated structure. The first
also corresponds more to the concept of a VO, as being based on value-orientated relationships where
all partners have equal rights and no partner decides about another (Wüthrich et al., 1997). 60% of our
sample shows an equal power distribution among the partners, 30% of the networks were driven by
one dominating company.

Finally we had a look at the Use of ICT (extensive vs. limited) in VOs, because information and
communication technology is usually seen as a driving force for Virtual Organisations facilitating
processes such as communication and resource sharing (e.g. Davidow/Malone, 1992 or Mews, 1997).
Of our sample of 30, 17 VOs generally used ICT for communication and collaboration, but only seven
cases use a common Intranet or more sophisticated ICT tools. 20% of the cases co-ordinated their
work solely using traditional communication media like telephone, fax, and sometimes e-mail. The
lack of interest among the network members in ICT tools was confirmed by several network managers
during the interviews.


Success factors for co-operative networks
The in-depth interviews carried out with 7 co-operative business networks provided us with rich data
covering the areas shown in Table 2. The cross-analysis and interpretation of this data revealed a
number of factors or issues that, for the interviewees, seemed to be decisive for either success of
failure of network co-operation. We therefore assume that critical success factors for co-operative
networks are to be found in the following areas, which are interrelated:
1. Business opportunities
2. Trust
3. Commitment to co-operation
4. Exchange of information
5. Network-Management

Business opportunities
Any network can be an environment for social networking, which may be perceived as pleasant or
even useful by some of the members, but no co-operation will endure in the medium-term if no
business opportunities arise. Particularly in the set-up phase, all the network members should therefore
actively seek for such opportunities and not just rely on the manager to acquire business. The manager
can facilitate the tendering process and the splitting of orders between the network members, but his
time resources are limited. Only in the later stages of the network – growth and maturity- the manager
can free enough time to focus more on acquisition and new business for the network.

42

Gruber/Koeszegi/Nöster (2005) show that externally initiated clusters are not driven by one company
and are not necessarily pursuing a specific business opportunity. Therefore, the careful search for and
development of a strategy is essential in order to identify and realise business opportunities in the
future. This was confirmed in the in-depth interviews where the network managers agreed that the
definition of a sound network strategy was essential for the success of a VO.

Trust
In line with much of the academic literature (Granovetter, 1985, Handy, 1995, or Koeszegi, 2001) all
interviewees agreed that trust was a very important precondition for co-operation in the network. Trust
replaces to some extent the more rigid factual management tools in VEs and determines the speed of
setting up a project (“trust is speed”). The less negotiation and contractual agreements necessary, the
quicker the project will be set up and the better its chances in the market. There is a challenge,
however, as co-operation can induce opportunistic behaviour. According to our interview data, the
development of trust depends on a number of factors, including time, soft factors (sympathy) and the
means to institutionalise trust (co-operation agreement).

Commitment to co-operation
A network only succeeds if all network members are actively involved in its development, because all
companies need to identify new business opportunities and take them to the network. In addition, they
all have to agree on the strategic direction of the network and decide on its activities and management.
As soon as one company loses touch with the network for a while, it might find it hard to catch up
with the latest developments of such a flexible and agile organisation. If these developments do not
correspond to the company’s strategic interests then it will definitely lose interest in the network. The
requirement for all members is therefore to be constantly involved from the beginning in order to
shape the network’s strategic development in unison. This in turn requires that companies are
committed (compare also Sherer, 2004, p. 329) at senior level and ensure that the necessary resources
– time and personnel - are available.

Commitment, however, is not just a cause for success, it is also an effect of a number of other factors.
Not only are business opportunities and trust essential for commitment over the long term, it is also
crucial that each co-operation partner has specific capabilities which increase the chances for the co-
operation to function and partners to be committed.

First there are the functional capabilities - the expertise of the company in a number of specific
business functions. Only if all companies know and bring in what has been termed core competence
(Prahalad/Hamel, 1990, Barney 1991) into the network, they can:

1. Offer these competences without risk of losing their competitive advantage.
2. Focus on the development of their core competencies and rely on sourcing the best
complementary competences from the other members.
Second, potential members need to be in a stable financial situation and have a fixed customer stock.
Companies that appear to be financially unstable will create suspicion among the other companies in
the network. Co-competition is certainly vital for a VO, ensuring some market dynamics within the
network and a certain redundancy of competencies for periods of high demand. However, companies
that compete for the same customer will not be able to co-operate.

Exchange of information
Open and transparent communication is another critical success factor the co-operation. Secrecy about
certain projects and hidden agendas create suspicion and negative feelings. The interviewed cases
agreed that the network manager plays a crucial role in the open exchange of information and that,
contrary to some of the theory (Davidow/Malone, 1992 or Mews, 1997) exchange of information is
often informal and face-to-face. The use of ICT is limited and cannot replace conversations on the
phone and physical meetings. Diverging is the extent to which these meetings are obligatory. One of
the interviewed network managers stated that their meetings are always attended by all members,
because they are voluntary and combined with a social event. Other networks aim to oblige their
43

members to attend by including a respective clause in the consortium agreement. To penalise
companies that do not attend, however, seems to be contra-productive and not feasible as it does in
other respects of co-operation “misbehaviour”.

A specifically sensitive issue is to share information about prices for products and services sold to
other network members. In order to avoid surprises when a project needs to be set up quickly,
networks aim to clarify these price issues in the consortium agreement. The prices have to correspond
to real costs without a margin in order for the network to provide full solutions at the most competitive
prices. If each partner adds a certain margin to his contribution than the whole product or service will
certainly be more expensive than that of an integrated company as a competitor.

Network management
A common understanding of how to handle business processes as well as conflict situations is crucial
for success. It is therefore necessary to define basic roles and rules for the co-operation.
A set of management tools is available for network management to facilitate or control processes and
the selection of these tools can have a strong influence on the success of the co-operation. The
challenge thereby is to remain as flexible as possible while at the same time minimising the risks
involved in co-operation with actual or potential competitors.

Regarding the management processes within the network, a clear separation between the level of the
stable network and the level of the individual project is useful. Our data indicates that more emphasis
has to be placed on the network level. While a VO obtains a stable organisational form through the
network, the actual VEs are the individual projects of setting-up, managing and completing temporary
business ventures. There are of course some critical aspects in this process such as the selection of the
partners and coping with questions of liability and price agreements, but all interviewees agreed that
they could improve the whole VE life-cycle process over time. The more effective the management of
the stable network, the easier to implement are the individual projects.

In order to classify management processes on each level, a further distinction was made between
factual management and personal management issues and tools in reference to Albers et al., 2002.
Factual management comprises the rather abstract and rigid management tools of contracts, rules and
common ICT infrastructure, while personal management issues relate to more flexible, situation-based
processes. The latter category includes issues of trust and personal interaction. Table 5 provides an
overview of the management tools available for networks.

Compared to other organisational forms, the VO’s specific advantage lies in the quick and
straightforward set-up of individual projects the moment a window of opportunity opens. It therefore
needs to rely heavily on personnel management, though some factual management tools are
indispensable. Success of a VO will partly be determined by the right balance between the two types
of tools used on the two levels.

Factual management Personnel management
Network level
• Selection criteria for members
• Portfolio/ competence analysis
• Co-operation agreement
• Price definitions
• Competence data base
• Internet and other ICT platforms
• Common meetings to select partners
• Network manager/ coach as
moderator and promoter
• Trust building activities
• Open and transparent
communication

Project level
• Sub-contracting
• Detailed co-operation contracts
• Brokering tool
• Coach/ network manager supports
the partner selection
• Information sharing
• Conflict management
Table 4: Factual and personnel management tools


44

Conclusions
The research presented here allows some conclusions that provide a number of high-level
recommendations for practitioners interested in implementing the VO concept in the form of a co-
operative SME network.

The Virtual Profiles of 30 VOs showed is that there is no blueprint for the structure of a VO as they
adopt different approaches of organisation and co-ordination, depending on their specific environment.
An additional analysis of the same sample discussed in Gruber/Koeszegi/Nöster (2005) however
identifies two clusters of VOs with different network structures (see Table 3): externally initiated
networks and strategically (internally) driven networks. The first cluster is characterized by equal
power distribution, regional dimension and clearly defined strategy. The second is focal, international
and aims at fulfilling a specific business case. The analysis also indicates that both types of networks
can be successful.

Assuming that the co-operative network is considered the desired organisational form for SMEs, this
research and in particular the in-depth interviews provide indication of a number of common aspects
leading to a more successful co-operation:

1) Members of a co-operative network must share a common business need or case and the network
must provide a way (strategy) to address this need. If this is not the case, some companies will be
reluctant to co-operate and business opportunities will not be identified.
2) Trust is a very crucial factor for a co-operative network. Though trust develops largely on the
personnel level, it can be institutionalised through, for instance, the careful selection of partners
and a co-operation agreement.
3) Active commitment at general manager level is essential for the co-operation and depends on
business opportunities, the match of the companies’ capabilities, as well as available time and
financial resources.
4) Exchange of information must be open and transparent and even sensitive issues such as prices of
goods and services must be discussed with all partners. ICT tools can support but not replace
direct personal exchange of information.
5) All networks have defined rules and roles, particularly on the network level, but one needs to
keep in mind that too many formalised rules tend to curtail the VO’s agility.
6) ICT is merely an enabler in the implementation of a co-operative network, the business and
organisational factors have far more weight.

Though not discussed here, our analysis also revealed that it was important to consider the value added for
the network member in a wider sense. It includes indirect, qualitative and longer-term effects that cannot
be measured with traditional budget decision or accounting tools. Further research needs to identify more
accurate means to identify and measure this value added for the companies, as well as the costs, in order
to ultimately assess the success of a VO.

Acknowledgement
We would like to thank the members of the EU-project VERITAS (IST 2004-511013) for their
support, feedback and critical reflection of the VO-concept. The project served as the basis for our
investigation and analysis and therefore provided a significant impetus for our work.


References
Albers, S., Bisping D., Teichmann K., Wolf, J. (2002): Management Virtueller Unternehmen, in: Albers, S.,
Wolf, J. (Hrsg.): Management Virtueller Unternehmen, Wiesbaden, S. 3-60.
Barney, J.B. (1991): Firm resources and sustained competitive advantage, Journal of Management, Vol. 17,
pp.99–120.
Bauer, R., Koeszegi S. T. (2003): Measuring the Degree of Virtualization, eJOV Electronic Journal of
Organisational Virtualness, 5, 2: http://www.virtual-organisation.net, 2003, accessed 27.8.2004.
45

Byrne J., Brandt R., Port O. (1993): The Virtual Corporation, International Business Week, February 8th, p. 36-
41.
Bultje, R., Van Wijk, J. (1998): Taxonomy of Virtual Organisations, based on definitions, characteristics and
typology in: VoNet: The Newsletter: http://www.virtual-organisation.net, 1998, accessed 28.8.2004.
Davidow, W.H., Malone, M. (1993): Das virtuelle Untenehmen: Der Kunde als Co-Produzent, Campus,
Frankfurt.
Douglas, D. (2003): Grounded theories of management: A methodological review, Management Research News,
26, 5, p. 44-52.
Glaser, B. G., Strauss, A. L. (1967): The Discovery of Grounded Theory: Strategies for qualitative research, De
Gruyter, New York.
Granovetter, M. (1985): Economic Action and Social Structure: The Problem of Embededness, American
Journal of Sociology 91, 481-510, found in: Achrol, R. S.: Changes in the theory of inter-organisational
relations in marketing: Towards a Network Paradigm, (1997) Academy of Marketing Science, Vol.25, no.1,
p.65-71.
Gruber, M., Koeszegi, S., Nöster, M. (2005): Initiated Networks – A strategic Alternative for SMEs, to be
presented at the 21
st
EGOS-colloquium, 30
th
of June – 2
nd
of July, Berlin.
Gulati, R., Nohria, N., Zaheer, A. (2000): Strategic Networks in: Strategic Management Journal, Vol. 21, p. 203-
215.
Handy, C. (1995): Trust and the Virtual Organisation, Harvard Business Review, May/June 95, Vol. 73 Issue 3,
p. 40.
Hannus, M. (2004): VO Guidelines, Deliverable D54.1 of VOSTER, http://voster.vtt.fi, accessed on 31 August.
Koeszegi, S. (2001): Vertrauen in Virtuellen Unternehmen. Wiesbaden: Dt. Universitätsverlag, Gabler, Edition
Wissenschaft.
Mayring, P. (1983): Qualitative Inhaltsanalyse: Grundlagen und Techniken, Weinheim, Deutscher Studien
Verlag.
Mertens P. , Faisst W. (1997): Virtuelle Unternehmen: Idee, Informationsverarbeitung, Illusion in: Scheer, A.-
W.: 18. Saarbrücker Arbeitstagung für Industrie, Dienstleistung und Verwaltung 1997, Physica-Verlag,
Heidelberg.
Mews, M. (1997): Virtuelle Unternehmen zwischen Anspruch und Wirklichkeit, IT Management, 3, pp 12-17.
Picot, A. , Reichwald, R. , Wigand, R.T. (2003): Die grenzenlose Unternehmung: Information, Organisation und
Management, 5. Aufl., Gabler, Wiesbaden.
Porter, M. (1985): Competitive Advantage – Creating a Sustaining Superior Performance, New York, The Free
Press.
Prahalad, C. K., Hamel, G. (1990): The Core Competence of the Corporation, Harvard Business Review, vol. 66,
pp. 79-91, 1990.
Saabeel, W., Verduijn, T.M., Hagdorn L., Kumar K. (2002): A Model of Virtual Organisation: A Structure and
Process Perspective, in Electronic Journal of Organisational Virtualness, eJOV 4 (2002) 1, pp. 1-16
Scholz, C. (1996): Virtuelle Organisation: Konzeption und Realisation. Zeitschrift für Führung und Organisation
(zfo), 65(4), 204-210, 1996
Sherer, S. A. (2003): Critical success factors for manufacturing networks as perceived by network coordinators,
Journal of Small Business Management, 41, 4, p.325-345. .
Sieber P., Griese J. (Eds.) (1998): Organisational Virtualness, Proceedings of the VoNet-Workshop, April 27-28,
Simowa, Bern.
Srnka, K.., Koeszegi, S. (2004): Integrated Qualitative and Quantitative Research: A Structured Framework and
Exemplary Study on Electronic Negotiations, Working Paper OP 2004-05, University of Vienna.
VDI Verein Deutscher Ingenieure (ed.) (2004): Innovationsnetzwerke, VDI Düsseldorf.
Wüthrich H. A., Philipp A. F., Frentz M. H. (1997): Vorsprung durch Virtualisierung: Lernen von virtuellen
Pionierunternehmen, Gabler, Wiesbaden.

46

3.2 Issues in the Management of COllaborative DEmand
and Supply NETworks

Agostino Villa, Politecnico di Torino, agostino.villa@polito.it
Dario Antonelli, Politecnico di Torino, dario.antonelli@polito.it
Irene Cassarino, Politecnico di Torino, irene.cassarino@polito.it

Supply Chain Management is a strategic issue for the extended logistic organisation of modern
companies. It evolved from the traditional administration of the firm’s suppliers to a necessary tool
for the company survival. A further evolution has been the introduction of a collaborative approach.
Presently, a DEmand & Supply NETwork (DESNET) is an organised group of enterprises, each one
driven by a proper management but all aiming to collaborate together in order to increase their
respective impact on the final and intermediate markets.
Therefore, the manufacturing process has become more and more dependent on a strict co-
ordination and co-operation among a large number of trade partners. Presently there is a lack of
knowledge on the subject, because it deals with cross-sectional issues, requiring a strict correlation
and integration among researchers belonging to different traditional engineering fields and non
traditional disciplines like organisation management, knowledge management, negotiation and
information technologies. It is necessary to found a common research framework specific to the
supply network management.
This issue boosted a European VI Framework project named CO-DESNET: an international project
co-ordinating research centres and companies worldwide, aiming to link together the provider of
management solutions with the users, namely the companies applying outsourcing strategies.


Introduction
A major issue for many Small-Medium Enterprises (SMEs) in Europe is currently the necessity of
being competitive in a global market, that means: on one hand, to be flexible such to be able to
produce a large variety of product volumes and to offer a wide mix of final products; on the other
hand, to be recognisable and able to make their trade mark recognised in every market place. As an
individual unit, an SME is prevented from operating in such a way, owing to its very small dimension
and its reduced possibility of developing significant industrial and marketing efforts.

This situation results to a push towards aggregation. A first type of aggregation of SMEs occurred
about twenty years ago, and “agglomerations” of SMEs arose, mainly in Italy and to a lesser extent in
some other European countries. Some SMEs operating in the same territory, either a region or an
administrative district, and on similar product types, were forced to define together common
marketing strategies. Examples are the Italian “industrial districts”. Several industrial districts became
very successful as soon as they were able to promote their own trade marks, often related to consumer
products. Several references can be found in a catalogue of Italian industrial districts reported in the
web.

As a matter of fact, the initial success of the Italian industrial districts contains the reason of their
current crisis: the lack of investments in innovation, mainly due to the lack of a real management
centre. Except from some cases, industrial districts have not been able to transform themselves into
“networks”. This great difficulty might originate in the problem that SMEs do not know how to
operate and effectively manage a “network of SMEs”.

This contribution will discuss this issue, aiming to present a model describing how several SMEs can
co-operate in a collaborative network, thus preventing them from being strongly competitive (and
consequently, destroy the network itself). The “road map” of the contribution can be summarised as
follows.

Section 2 will introduce some preliminary concepts about “clusters”, “networks” and “supply chains”,
usually adopted when dealing with management issues related to groups of SMEs. Section 3 will
47

present a model of an SME network describing how SMEs can interact through negotiation of prices
and volumes to be produced. Section 4 will discuss some conditions for the initiation of collaboration
between SMEs participating in the same network: the goal is to clarify if and when a co-ordination
centre is necessary and fruitful. Section 5 will finally analyse a few examples of Italian industrial
districts, in order to discuss their actual robustness or weakness.

Clusters, Supply Chains and Networks: some preliminary concepts
As mentioned above, the first attempt of some SMEs to collaborate to increase their potential in a
larger product market was to search for a common interest in either some production activities or in
marketing efforts. This initial type of co-operation gave rise to clusters, i.e. groups of small firms
operating in a common territorial environment. The first idea was that a number of small firms should
be able to grow together, just as a grape, as they could share common abilities and opportunities
offered by the same town or territory, as well as common administrative rules and reference persons.
The second surrounding idea was that a common place could also be used for creating a trade mark,
related to the territory, facilitating a large circulation at least all over the country, possibly
internationally. The sole condition which has been recognised as necessary for the cluster assembling
was to have a common “political committee” with the aim of promoting contacts of the SMEs with
local authorities, such as dedicated legislative measures and funds for development.

A typical example of a “cluster” is the Manzano District (Friuli Venezia Giulia, Italy), producing 80%
of the Italian and 50% of the European chair industry. The Manzano District is mainly centred on three
towns of Udine: Manzano, San Giovanni al Natisone and Corno di Rosazzo, forming the so called
“Chair Triangle”. There are 1,200 companies concentrated in only 100 square kilometres with 15,000
employees, of which 37% produce finished products, 27% semi finished or components, and 26% are
sub-contractors. This sector started in the 1950s: in 1951 it had already reached 137 plants with 1,237
employees, and ten years later the number of enterprises had almost doubled.

During this period the Friuli district started to see the importance of strategic factors as image and
design and an association with the Milan Italian Style representatives including Gio Ponti, Carlo De
Carli and Vico Magistretti began. An interesting synergy between manufacturing companies and
designers was thus created. In 1977 the first Salone della Sedia di Udine was inaugurated, and today it
still represents the only fair in the world specialising in seating. As a matter of fact the competitiveness
of the district organisation is boosted by the cost saving deriving from networking a number of
functions among many specialised companies. The network provides costs reduction, flexibility, and
knowledge diffusion in the district.

Agglomerations of firms focused on a specific manufacturing sector evolved in regions where the
gradual but relevant growth of a single enterprise was occurring. In these situations, the creation of
some SMEs was supported by the opportunity of offering side or ancillary production to the larger
enterprise, which played the role of “leader” of the group. Usually, this leading condition consisted in
the particular connections between the main enterprise and the others, being the former the client of
either raw parts or of services supplied by the SMEs. This gave rise to the large number of “supply
chains” operating in the most developed regions. The same motivation is also originating the
allocations of new industry networks in the developing countries. Now, the main idea, which
motivates the generation of a new supply chain, is to assure that a “production line” could have its
necessary supplies “just in time” and with the required precision and quality. Small ancillary firms are
forced to assure that condition: to this aim, they are required to allocate either their plant or, at least, a
sufficient store, close to the leading enterprise. As it can be easily understood, in this case the
condition which has been recognized as necessary for the supply chain organisation was to have a
strong management centre, exercised by the leading enterprise, and a hierarchical organisation able to
assure that all SMEs will operate in support of the leading one.
Recently, a more structured organisation of SMEs has emerged: the “Demand & Supply Network”
(here mentioned by the acronym DESNET). In practice, the new organisation emerging from the
agreement among a set of SMEs is a DESNET, which is structured as an “extended virtual
enterprise”, that means a temporary network of several small firms which decide to co-operate in a
48

common given value chain for a limited time horizon. The usual example of a DESNET is a set of
different enterprises able to produce different parts, which are used in a common family of final
products, and to apply co-ordinated complementary production programmes for a common
industrial goal. A special interest of the industrial bodies in DESNETs is geared to the “co-operation
agreements” signed for a finite time horizon, and stating that each firm can partially interact with
the other network partners (it means that each firm can also be involved in the network by a portion
of its own market share). The resulting DESNET can have a finite life and it does not completely
reduce the autonomy of any component firm, as each can still produce items for its own clients, and
operate in a proper market segment. More precisely, all small firms, which agree to be active in a
DESNET must sign an agreement to co-operate in defining common production plans for specific
products.

This new character of the emerging DESNETs makes them much more robust against the increasing
variability of the products’ and labour markets. Some existing industrial districts are going to
gradually change their initial very lean organisation into a DESNET one. An example of this
evolution can be the new district Torino Wireless. An increasing interest in this new organisation is
also emerging from the political framework: the potentially higher robustness of a DESNET and its
potential to better cope with a changing environment is the reason for the search of new funding
criteria and programmes from national and international institutions.


A Comprehensive Model of a DEmand & Supply NETwork
In principle, the design and management of a “temporary DESNET” requires the application of new
models of network operations, based on the concepts of multi-agent organisations and defined in formal
terms according to the theory of optimisation of large-scale dynamic systems. However, the approach
here proposed will be more “practical”: a model of a simple DESNET will be defined, which will
support the analysis of the DESNET main characters and the comparison of DESNET features with the
ones of an “industrial district” and of a “supply chain”.

Accounting for the presence of several individual SMEs, a DESNET can be formally modelled as a
virtually connected chain of service stages, each one containing either a small firm or a set of parallel
firms, each firm with its proper autonomous decision-maker, denoted “agent”. As in usual practice, it
will be assumed that any agent aims to co-operate with the others, but also wants to obtain the best
profits for his own enterprise.

To deal with a simplified organisation of SMEs, it will be considered that each stage is connected to the
up- and downstream stages through a virtual market place. This means that each small firm will
negotiate contracts for producing goods with downstream (buyer) firms as well as contracts for
acquiring materials with upstream (supplier) firms. This negotiation opportunity is a qualifying
character of a DESNET. Since each firm aims to maximise its income, it utilises its DESNET as a frame
to conduct a “good negotiation”. Here “good negotiation” means that an agreement between each pair
of “consecutive agents” (belonging to two consecutive stages of the chain) is concluded, assuring a
sufficient income to both of them.

The formal model of a DESNET proposed is based on the idea that such a networked industrial system
consists of an “open commercial system” within which any two member SMEs interact by exchanging
material and financial resources. For each material resource (i.e., parts of a common final family of
products), a proper market place exists on which this resource is negotiated through a proper monetary
value, between the supplier (i.e., the resource producer) and a customer (i.e. the purchaser who will
utilize the resource in its own production sequence).

In a DESNET, considering the complete production cycle of a final product, each buyer will purchase at
least a resource to apply his own manufacturing operations in order to transform it into a new item with
more added value, to be sold in a downstream market place. Thus, the principal functions of any
component firm are: purchasing, transforming (through either manufacturing or servicing), and selling
of items.
49


In a Collaborative DESNET6, this set of dynamic commercial interactions occurs within a co-
ordinated protected industrial network. Partners indeed are connected together through collaboration
agreements, which reflect the effectiveness of the commercial negotiations. These agreements improve
the possibilities of survival for each member, because the network becomes the interface with the
large-scale enterprises and it has a stronger contractual power with respect to the single SME.

The model presented in fig. 1 aims at allowing an easy but correct evaluation of costs and advantages
of the above sketched regulated industrial system. To this aim, a DESNET composed by 4 enterprises,
belonging to 3 different stages, is considered, including a firm playing the role of the final client. Two
market places define the interactions among the agents: one is concerned with the negotiation between
a supplier firm and the internal stage, the other is related to the negotiation between the internal stage
and the final client.

Figure 1: Schematic representation of DESNET dynamics

The operations and decisions at each stage are described according to the following steps:

1st step: the client firm C asks for a volume y of final products to the internal working stage, and it
offers a price p;
2
nd
step: the two SMEs S
K
. i=1.2, belonging to the internal work stage ask for a volume m of raw
materials to the supplier firm S. and they offer a price q;
3
rd
step: within the internal work stage, the SME S
1
searches for an agreement with SME S
2
for
defining the two volumes y

and y

to be delivered to the client.

Each step consists of a negotiation, with a reciprocal exchange of proposals. Within such an exchange
of proposals, the decisions of all players can be modelled as follows.

a) The two SMEs included into the internal work stage operate in order to maximize their
respective income, according to the following conditions:

(a-1) the expected profit G
i
, i=1, 2, which they want to maximize, results from the balance
between the return from sale and the costs for product manufacturing and the material
purchase:

; 0
1 1 1 1 1 1
≥ × − × − × = m q y r y p G with ;
1
p p ≥ ;
1 1 1
y m × ≥ α (1)


6
Denoted in the following CO-DESNET, as the European project to which the paper is referred to the
Coordination Action (CA) project n° IST-2002-506673 / Joint Call IST-NMP-1, Collaborative Demand and
Supply NETwork, supported by the European Commission, Information Society Directorate-General,
Communication Networks, Security and software, Applications
Supplier
Firm 2
Firm 1
Client
DESNET
Price Price
Quantity Quantity
50

; 0
2 2 2 2 2 2
≥ × − × − × = m q y r y p G with ;
2
p p ≥ ;
2 2 2
y m × ≥ α (2)

under the constraint:
_
q q ≤ (3)

(a-2) the capacity constraints to which both SMEs are obliged to satisfy:

) , min(
1 1
C y y ≤ (4)

) , min(
2 2
C y y ≤ (5)

y y y = +
2 1
(6)

Note that the following notations have been adopted:

p
i
, i=1,2, price which the SME i wants to obtain when selling its products to the client;
y
i
, i=1,2, volume of products the SME i wishes to manufacture;
r
i
, i=1,2, unitary production cost of SME i;
m
i
, i=1,2, volumes of raw materials which SME i wants to purchase from the supplier S;
q, cost of the raw material which the two SMEs wish to pay to the supplier S;

_
q , price of the raw material which the supplier S wants to obtain from its sale;
α
i
, i=1,2, rate of material required to produce a final product at each SME:
C
i
, i=1,2, production capacity of SME i.

b) The supplier S presents a proper objective of income, to be maximized, and it can assure a
bounded production capacity to its own clients:

;
_
m q G
S
× =
; 2 1
m m m + = (7)

.
S
C m ≤
(8)

where:
G
S
is the expected income of the supplier S;
C
S
is its available production capacity.

c) Finally, the client firm C has the objective to minimize its costs and an a-priori known demand
of final products (note that the demand for products must be either known or forecasted, as in
any practical situation):

; y p J
C
× = . _ demand known Y y = ≥
(9)

The way of operating and of taking decisions in the modelled DESNET can be described by analysing
the individual strategies that each agent wants to apply. Each strategy can be derived by analysing
51

each decision-making problem associated to one of the three types of agents (and discussing the
admissible solution of the related Linear Programming optimisation).

I. Each SME S
i
, i=1,2, inside the internal work stage aims to maximize its own profit Gi.
This goal can be reached in different ways, depending on the price offered by the client C
for purchasing final products:
- in case the price offered by the client C is greater than the unitary costs for
manufacturing and material purchasing, then a SME aims to maximize its production
volume and accept the offered price, in order to maximize (1) and (2), i.e.

if ; q r p
i i
× + ≥ α then ) ( max
i
y imize and accept p; (10)

- otherwise, if the offered price does not balance the unit costs, a SME is interested in
reducing its production to the minimum level possible, and to increase the price of
products, in order to reach a balance between production costs and income:

if ; q r p
i i
× + < α then ) ( min
i
y imize and ask for . p p
i
> (11)

II. The supplier S will show a complementary behaviour, depending on the request for raw
materials from the SMEs:
- in case the request for material, m, which the supplier receives, is larger than its own
production capacity, it increases the material price in order to maximize (7):
if ;
S
C m ≥ then ) maximize(
_
q ; (12)

- otherwise, it can accept the offered price and supply the maximum possible material:
if ;
S
C m < then .
_
q q = (13)

III. At the end of the DESNET, the client C plays a complementary role with respect to the
supplier, by acting on the price of the final products:
- in case the offer of products from the internal SMEs is at least equal to the client
demand, then the client itself will reduce the offered price in order to minimize (9):

if ;
2 1
Y y y = + then ) ( min p imize ; (14)

- otherwise, in case the offered volumes are low, the client is compelled to increase the
offered price:

if ;
2 1
Y y y < + then ) ( max p imize . (15)

Depending on the individual management strategies which each SME adopts, the group of SMEs
could operate either in co-operation or in competition, depending on the offered price and required
costs, as well as on the offered or requested volumes. The issue here is to recognise situations in which
the group of SMEs can be “autonomously collaborative”, thus giving rise to a CO-DESNET without
necessity of any co-ordination. “Potentially competitive” situations require a co-ordination centre.
Depending on the situation, complementary types of CO-DESNET could be organised.



52

Conditions to Promote Collaborative Networks
The investigation of conditions assuring a collaborative interaction among SMEs in a DESNET is a
challenging task for the designers and organisers of industrial districts. Obviously, the existence of
collaborative co-operation is a predisposition that facilitates the creation of a network. If spontaneous
co-operation is not possible, the necessity for some kind of co-ordination arises.

In order to identify conditions assuring co-operation of SMEs, the individual management strategies
derived in the previous section will be analysed under the condition of perturbations with respect to
price or volume. The reasoning behind this “perturbational approach” is the following: if a
perturbation forces the agents to modify their strategies in such a way that the original perturbation is
emphasized, then the strategies themselves cannot be considered stable, and the set of SMEs will
evolve towards competition.

a) First, the effect of a variation of the price p offered for final products is considered, in order to
analyse the interaction between the client C and the internal SMEs.

Assume that, for a given requested volume y, the client C will offer a price p lower than the ones
expected by the two internal SMEs, i.e.
2 1
p p p ≈ < .
In this case, the two SMEs S

and S

will co-operate together to reduce volumes to be delivered to
the client C (see condition (11)), thus assuring a collaborative behaviour.

In the opposite case, the client C offers an augmented price p for a given volume y, i.e. ,
1
p p < and
2
p p >
In this second situation, SME S

will try to reduce its production y

and to ask a greater price p

.
whilst the other SME S

will maximize its produced volume y

in front of the acceptable price p (as
prescribed by conditions (11) and (10) respectively).
In practice, this second situation shows a competition between the two SMEs, since a disequilibrium
occurs in front of the price perturbation.
One can counteract this undesired situation by two different (and complementary) co-ordination
actions:
by a different specialisation of the two SMEs depending on the manufactured products, so that
different production costs could be obtained in the two SMEs for different final products;
by an oriented innovation of the two SMEs, such as to obtain in the near future production costs with
the above mentioned balancing effect;
in both cases, the “balancing (either specialisation or innovation) strategy” should act as follows:

if ,
2 2 1 1 2 1
q r q r p p × + > × + ⇔ > α α one must find two values
2 1
, α ∆ ∆r such that

q r q r r × + = × ∆ − + ∆ −
2 2 1 1 1 1
) ( α α α ;

First Remark. Both above mentioned situations can be boosted by public initiatives: examples are the
“calls for proposals” for research and technological innovation programmes which can supply funds to
DESNETs in terms of low-rate financing.

I. by “absorption” of the weaker SME S
1
by the stronger one, and consequent
reorganisation; this last action obviously is going towards the transformation of the
DESNET first into a supply chain, then into a holding centred on the stronger
enterprise.

A second type of considerations can be drawn when the interactions between the internal SMEs and
the supplier S are considered. In this case, the effect of a variation of the price
_
q requestedfor the
raw materials to the internal SMEs is accounted for.
53

Assume that the two SMEs ask for a quantity of materials m, and the supplier reply is the request of
a price
_
q greater than the one allowing to SMEs a positive income. In this situation, SMEs realise
that

2
2
1
1
_
α α
r p r p
q



> (16)
Then, both SMEs are forced to reduce their production. But, owing to a reduction of required
materials, also the supplier must lower its own supplied volumes m, so that it can realise that a
lower price should be offered in order to increase the delivery of materials to SMEs (as one
can see from conditions (12) and (13)).

Second Remark. The above consideration shows that the connection between the supplier and the two
internal SMEs is “stabilised” because, in the considered simplified model (1) – (9), the two SMEs
adopt a common decision for what concerns the requests of materials to the supplier: their actions are
co-operative “against the supplier avidity”.

Third Remark. If, for any need, the two SMEs should reduce their production volumes, then the
client C will raise the offered price p. If so, a further “stabilising” action will result, since the increase
of price p induces a corresponding increase of the right-hand-side member of condition (16), thus
driving a global increase of all prices in the chain of stages.

The two remarks suggest an interesting and fruitful preliminary conclusion: a chain of SMEs is
structurally co-operative, and the naturally leading SME is the most efficient firm. This consideration
justifies the development of “chains” within a network of SMEs: its practical interest has been recently
recognised by several economic dossiers of newspapers, where the necessity for a stronger
organisation of Italian industrial districts in terms of “networks of chains” is recommended as effective
criteria for obtaining future funding.


Analysing an Example of Italian Industrial District
The considerations developed to present conditions, which support the organisation and support of
innovative networks of small firms can now be used for analysing the actual state of some existing
industrial districts in North-Italy. The section starts with a presentation of the main characters of
Italian districts as stated both in national and regional laws, and gradually specified in the history of
the districts themselves.

An industrial district is a production system concentrated in a geographic area where a large number of
firms operate, all related to the various stages of a common production chain. Then, specialisation of
manufacturing phases is a main feature of a district, as well as the small dimension of firms involved:
both features are strictly related, and depending on the subdivision of the production cycle into its
phases, and the allocation of different phases to different firms. An industrial district is also a socio-
economic frame, and its success depends on the social structure of the area. The robustness of a district
is thus a consequence of a common system of values and rules of the social community. Besides these
elements, the aim of the district is to reach a dimension sufficient to reach economies-of-scale, and to
have a wide recognition on international markets.

A typical example is the “district of gold” in the area of Valenza Po, a small-sized town in the
Piemonte Region in North-West Italy. The skills of this particular cluster are presented below using
the analysis model of the CODESNET project.






54

Table 1: The Valenza Po case study representation.

DESNET:
 Valenza Oro
Address:
c/o CNA di Valenza, Palazzo Servizi
15048 Valenza (AL) Italy
Web-site / Mail:
www.valenza.cna.it

1
st
Step: Aggregated analysis of the DESNET current status (a)
Analysis Drivers:
Type of DESNET: ID or SC ID: Industrial Districts with many micro enterprises
Spatial allocation & distribution of firms
in a geographical area:
Province of Alessandria, over a space of about 50 Kmq
Personnel employed and main
professional skills
About 1300 persons, with specialised skills for gold and jewel
manufacturing (based on on-the-job and special school formation)
Coverage of the product market The amount of gold processed is about 30 tons/year, corresponding to
80% of precious stones worked up in Italy
Performance indicators:
Estimated annual sales 1.550 million Euro/year
Area description 1300 firms (only 5-6 of mid scale); about 34.000 resident people in
this area
Average firm dimension 5-6 persons per firm, over a total of 7000 employed
Export volumes 760 million Euro/year, with a percentage of about 50%
Phase 2.1. Analysing the DESNET Operational Structure
Analysis Drivers:
Division of labour among firms Four types of main working activities: pre-working gold raw
material; jewel production; precious stones cutting; packing and
marketing
Organisation of logistics and distribution External logistic services are usually requested
Information pattern in the network Very “naïve” within the network
Performance Indicators:
Production stages;
N. firms per stage & N. personnel per
stage;
Degree of concentration of production
Variable number of stages, since different types of final products are
sold; e.g. jewel = 4 stages; gold ingots = 1;
Not defined;
The five most important firms control 70% of jewel production
Logistic bodies; carriers; personnel Not applicable, owing to external services
Applied ICT technologies Usual PC-based (main administrative functions)
Phase 2.2. Analysing the DESNET Organisational Arrangement
Analysis Drivers:
Agreements and control mechanisms
among partners
Agreements concerning promotion and marketing;
Periodic meetings to plan exhibitions
Responsibilities in the network Reference is made to the SMEs Firms’ Association, which acts as
coordinator of any event
Existence of collective agreement with
external bodies
No special agreement
Existence of larger leading firms in the
DESNET
A few leading firms operating in jewel production and sale
Performance Indicators:
Type of co-ordination body & function Committee (including representatives of local municipalities and
employers’ association); mainly a political function
Type of organisation structure “Lean” organisation, since all partners refer to the common ID
Committee
55

N. leading firms, average size of leading
and ancillary firms
5-6 leading firms; their average size is about 30-60 personnel;
average size of ancillary firms is less than 5 persons.
Phase 2.3. Analysing the DESNET Interactions
with its Socio-Economic Environment
Analysis Drivers:
Marketing strategies Mainly individual strategies are applied, but with common promotion
events (competition still remains high)
Types of investments from external
enterprises
Not available today
Degree of integration of the DESNET
with its local labour market
Very high, mainly due to the family-based structure of the small
firms
Performance Indicators:
Percentage of employees in DESNET
firms over the population (ready to work
people)
About 70%
N. DESNET firms over N. firms in the
area
About 80%, when accounting for production and service firms (no
shops)

The district is now suffering a crisis because of the current difficult circumstances for the economic
systems worldwide. The crisis is pushing some firms towards a re-organisation of the district, by
searching for an effective and efficient co-ordination centre without greatly affecting the individual
firms’ autonomy. A first attempted solution was building a “co-ordination committee”, composed of
representatives of local authorities (the local municipalities, the regional government), and local
associations of enterprises. The committee scope was to be a space for discussing proposals and
agreeing actions on industrial policy. The main goal of the co-ordination committee in the Valenza
district has been the promotion of proposals for funding new services for the small firms including
support from the local authorities for improving the e-connections between firms, and to force co-
operation even in front of a real competition (which involves the entire gold sector).

The necessity of a better co-ordination has recently been recognised by the most important firms (5-6
firms appear to be the leading ones in the district). Following considerations similar to those above
presented, the Valenza committee is going to enforce the co-ordination structure through two
complementary initiatives:

a) by promoting the creation of consortia of SMEs inside the district, in order to generate larger
industrial bodies, among which co-operation events and proposals should be easier;

b) by organising a new structure to co-ordinate the leading firms in order to obtain ISO certifications.
This last initiative has been launched by the Association of Valenza Operators (AOV) and by
ASPERIA (Azienda Speciale della Camera di Commercio di Alessandria per la Promozione
Economica), and is going to generate a new hierarchical structure which includes technical and
administrative support (named “technical coordinator” and “technical secretary”). This initiative is
scientifically and technically supported by the Politecnico di Torino, as shown in the following Fig. 2.
The common certification appears to be a significant thrust towards a larger co-operation of the firms,
even if individuality still remains really strong.

56

A S P E R I A A . O . V .
T e c h n i c a l c o o r d in a t o r
T e c h n i c a l s e c r e t a r y
W o r k T e a m
C . N . R . - T E M P E
P o l y t e c h n i c o f T u r i n
P i l o t F i r m P i l o t F i r m P i l o t F i r m … …
A S P E R I A A . O . V .
T e c h n i c a l c o o r d in a t o r
T e c h n i c a l s e c r e t a r y
W o r k T e a m
C . N . R . - T E M P E
P o l y t e c h n i c o f T u r i n
P i l o t F i r m P i l o t F i r m P i l o t F i r m … …


Figure 2: The coordination structure of the Valenza district.

Conclusions
The aim of this study is to propose mixed approach to the description of the complex way in which
several enterprises interact inside a cluster, borrowing and merging contributions from different study
fields: Linear Programming optimisation, empirical observations and benchmarking, evolutionary
concepts. The analysis of existing Industrial Districts highlights the distance that split up present
Districts from the target of a collaborative network, a CO-DESNET. To this aim, it must be stressed
the important role played by non-competitive institutions, like Research Centres and Universities,
involved in the DESNET organisation. This role will become more and more significant for the
successfully introduction of a diffuse co-operative behaviour inside the DESNETs.

References
Villa, A., Cassarino, I (2004): Management of a Multi-Agent Demand and Supply Network, Proceedings of
IFAC Multitrack Conference on Advanced Strategies for Social and Economic Systems.
Samarra A. (2003): Lo sviluppo nei distretti industriali, Percorsi evolutivi fra globalizzazione e localizzione,
Carocci, Roma.
Paniccia, I. (2002): Industrial districts: evolution and competitiveness in Italian firms / Ivana Paniccia -
Cheltenham : Elgar, cop.
Zheng, L., and Possel-Dolken, F. (2002): Strategic Production Networks, Springer, Berlin.Villa, A. (2001): “
Introducing some supply chain management problems, Int. J. Production Economics, 73, pp.1-4.
Huang, C. Y. and Nof, S. Y. (2000): Formation of Autonomous Agent Networks for Manufacturing Systems,
International Journal of Production Research, Vol. 38, No. 3, pp. 607-624.
Brandimarte, P., Villa, A. (1995): Advanced Models for Manufacturing Systems Management, CRC Press, Boca
Raton.
Sethi, S.P., and Zhang, Q. (1994): Hierarchical Decision Making in Stochastic Manufacturing Systems,
Birkhauser, Boston.
Regional law n. 24 12 May 1997

Codesnet project web portal: www.codesnet.polito.it
Industrial District: www.dipp.it
CNA: www.cna.to.it
Services to enterprises: www.comune.pinerolo.to.it
Report about main Italian industrial districts: www.aster.it
Italian industrial districts: www.distretti.org
Italian industrial districts: www.clubdistretti.it
District of Valenza: www.valenza.cna.it
Province of Torino: www.provincia.torino.it
Piemonte industrial districts: www.ice.gov.it
Torino Wireless ICT district: www.torinowireless.it
www.repubblica.it
www.sole24ore.it
57

4 4. . I In nt te er ro op pe er ra ab bi il li it ty y a an nd d s st ta an nd da ar rd di is sa at ti io on n
4.1. Interoperability of Enterprise Systems and
Applications

Rainer Ruggaber, SAP Research, SAP AG, rainer.ruggaber@sap.com
Arne-Jørgen Berre, SINTEF ICT, arne.j.berre@sintef.no

Organisations are engaging in more and more sophisticated business networks to improve
collaboration. These business networks can range from more static relationships like Supply Chains
to very dynamic networks like virtual organisations. A prerequisite to enable business networks is
the interoperability of the participants systems and applications. ATHENA is an Integrated Project
funded by the European Commission under Framework Programme 6 that addresses
Interoperability of Enterprise Systems and Applications proposing a holistic approach. ATHENA
will provide technical results like reference architectures, methodologies and infrastructures
consolidated into the ATHENA Interoperability Framework (AIF) complemented by business results
that provide ROI calculations and impact predictions for new technologies. This contribution
provides insights into ATHENA, the industry scenarios considered, the proposed approach to
address the identified interoperability issues, and expected as well as achieved results from the
ATHENA programme of work.


Introduction
One of the trends in the global market is the increasing collaboration among enterprises during the
entire product life cycle. This trend requires that enterprise systems and applications need to be
interoperable in order to achieve seamless business interaction across organisational boundaries, and
realise networked organisations. IEEE (IEEE, 1990) defines interoperability as “the ability of two or
more systems or components to exchange information and to use the information that has been
exchanged”.

The European Commission considers the development of interoperability of enterprise applications
and software as a strategic issue for European companies to strengthen their co-operation and gain
competitiveness in the global market. In the context of the EU Framework Programme 6 (FP 6) the
integrated project ATHENA (Advanced Technologies for Interoperability of Heterogeneous Enterprise
Networks and their Applications) has been funded (ATHENA, 2004). It consists of a set of projects
and is to lead to prototypes, technical specifications, guidelines and best practices.

ATHENA takes a holistic approach to solving the Interoperability problem taking a technical as well
as a business viewpoint into account. Previous activities in that space led to fragmented solutions
addressing only part of the problem. From a standards viewpoint in the B2B space there is rather a
proliferation than a lack of standards.

ATHENA itself is driven by industry scenarios and requirements. Industrial users in the consortium
provide scenarios from aeronautics, automotive, telecommunications and the furniture industry. The
approach of ATHENA is to go from the specific scenarios provided industrial users to generic
requirements applicable to a whole industry. ATHENA looks at the scenarios provided by industrial
users in detail and then tries to abstract to industry scenarios common for an industry by identifying
commonalities and differences. Section 2 is providing a description of the scenarios considered in
ATHENA. Based on these scenarios, ATHENA defines pilot cases which are used for evaluation and
testing of the solutions provided by research. The pilot cases are also described in Section 2.

ATHENA proposes an interoperability framework for enterprise applications and software systems
called ATHENA Interoperability Framework (AIF). The AIF including its three main integration
areas, namely conceptual, technical and applicative integration, is introduced in Section 3.
58

Section 4 provides a brief description of the ATHENA programme of work. It introduces the
ATHENA Action Lines focusing on research and technology development, community building and
management as well as their interrelation.

ATHENA defines thirteen key results which are introduced in Section 5. These key results are
achieved incrementally over time. An overview of the results that have been achieved by month 12 of
ATHENA is given as well.

Scenarios
This section presents the current results of ATHENA related to industry short-term needs and
requirements regarding interoperability of enterprise systems and applications (Anastasiou et al,
2004). The scenarios and the specific requirements form the basis for developing test cases and
validation scenarios through which the evaluation and testing of the solutions provided by the R&D
technology providers will be performed.

Industrial Scenarios in ATHENA
In the ATHENA approach to requirements management specific interoperability requirements are
extracted through the analysis of real life scenarios. The initial set of scenarios in ATHENA has been
formed based on the needs and interests of the industrial users in the programme. The four industrial
sectors and their representatives, currently participating in the programme, are:

- EADS CCR - Aeronautics and Aerospace
- Fiat CR - Automotive
- AIDIMA - Furniture
- INTRACOM S.A. - Telecommunications

The industrial scenarios initially defined cover the four areas that are presented in the following
subsections. Although these scenarios are identified and developed by the industrial users, their
analysis, extraction of requirements and identification of interoperability solutions is performed
through the collaborative work of multi skilled teams including industrial users, methodologists and
solution providers. This collaborative work aims to close the gap between research and industry by
facilitating users’ deeper understanding of research trends, as well as researchers’ deeper
understanding of industrial needs. It will lead to the development of “realistic” ATHENA global to-be
scenarios from which interoperability requirements will be derived and that will form the basis for the
evaluation and piloting activities.

Scenario 1: Product Data Management in a Virtual Networked Enterprise
The concrete case concerns sharing and exchange of technical information in the aeronautics sector for
the change management process and for different types of supply chain relationships. Important
factors specific to the Aeronautics and Defence sector are:

• The long lifecycle of an aircraft (around 50 years long). This means that information about the
product should be useable during the whole lifecycle, which is longer than the lifecycle of an
application, software product or even an organisation.
• The complexity of the product, due to the high level of reliability that is required, and the
number of disciplines involved.
• Change management is very important; a product is defined as a set of ordered modifications.
• This scenario highlights a family of cases mainly to deal with the first factor.

Scenario 2: Collaborative Product Development
This scenario focuses on the Automotive sector and on the Product Development Process portion that
prescribes the suppliers involvement in the objectives definition and in the product planning:
Collaborative Product Development (CPD). CPD can be depicted as a process with three phases that
starts with the Target Setting and the nearly contemporary supplier selection process or Sourcing. And
a third phase, once the suppliers’ panel is defined, called Product Design. Along the whole CPD
59

process, the interaction between OEM and suppliers consists in a heavy exchange of information,
sometimes conveyed through the Net, and sometimes directly transferred face to face during meetings.
Also although in most cases data are managed by suitable Enterprise Information System, human
action is needed in order to transfer data from one system to another. Some document management
applications that guide the business processes exists (e.g. Open Plan application that support the
sourcing execution), but most of the documents are sent by e-mail. Business logic is embedded within
the applications, so every time the business model changes it is not easy to update the software
accordingly. A large number of stakeholders participate in this scenario and until now the definition of
a common format for data exchange has not been achieved.

Scenario 3: e-Procurement
The objective of the e-Procurement scenario is to facilitate e-Business services interoperability and
implementation of integration mechanisms by analysing the current e-Business implementation level
in furniture sector and promoting multi-sector international agreements.

Product suppliers sell goods over the Internet through sell-side e-Commerce applications. Buying
companies purchase goods over the Internet through buy-side e-Procurement applications. The
scenario focuses on an e-Procurement application, through which members of a buying company
purchase goods from multiple suppliers.

Scenario 4: Product Portfolio Management
The focus of this scenario is on the management of the portfolio of new product development projects
(NPD Projects), including also projects related to the development of new versions of already released
products. The Product Portfolio Management is of significant importance especially to large enterprise
with many business units and complex products.

The efficient performance of the product portfolio process requires federated information coming from
marketing, project execution, as well as from the product life cycle management. It is also a
knowledge intensive process, as it presupposes a very good and holistic view of the enterprise:
strategy and objectives, skills and competences, as well as experience coming from previous projects.
It requires therefore, many different aspects of interoperability to be covered: Business aspects,
Knowledge Aspect and ICT aspects. The Product Portfolio Management scenario studied in ATHENA
focuses on intra-enterprise level and on the telecommunication sector.


ATHENA Pilot Cases
The pilots will cover the Aerospace, Automotive, Telecom and Furniture Industries. The piloting
activities include the identification and implementation of test cases, test scenarios and test procedures
in a real context of industrial users. The test cases and test scenarios are defined taking into
consideration the to-be scenarios developed in the ATHENA Dynamic Requirements process.
Currently, the following four test cases have been defined:

Test Case 1: Product Data Management in a Virtual Networked Enterprise
The case is based on establishment of collaboration between two organisations, that use different
specific change and configuration management processes, and nevertheless need to interconnect their
process, in a way that allow interoperability between the heterogeneous Product Data Management
Applications and software products. The federation of the processes, applications and software is
based on the usage of consensual business (CMII, Manufacturing STEP application protocols) and
technical standards at the level of the considered network. Integration of Product models is based on
the usage of neutral information model related to several disciplines: AP233 for System Engineering,
AP239-PLCS for customer support, AP214 for geometry, AP209 for calculation etc.

Test Case 2: Collaborative Product Development Case
The essential test case is centred around the lifecycle of a Request for Quotation (RfQ) document and
focuses on the management of process-critical events during the processing of the RfQ and allocated
discussion and – if necessary – modification of the technical specification attached to the RfQ.
60


The test case is an integral part of collaborative product development and it is located in the early
phase of the CPD, during which an OEM and its 1st and 2nd tier suppliers collaborate in order to
verify (and where necessary to amend) an RfQ.

In the automotive sector, major portions of RfQ related input specifications are defined by the OEM.
A major problem and obstacle to interoperability between an OEM and its 1st and 2nd tier suppliers
are modifications made to the specifications after publishing them together with the RfQ.
There are different reasons for these modifications:

- Availability of new business-level information that change business parameters and may affect
the technical specifications.
- Inconsistencies and technical problems observed as the RfQ is discussed between the OEM
and its suppliers may force changes in the specification.
- In some cases, the availability of new technology may lead to changes being made.
- Sometimes, business relationships change during the process, and the specification needs to be
adopted to the capabilities of new suppliers.

Test Case 3: e-Procurement Case
The Test Case for the e-Procurement Scenario is based on a major Spanish office furniture
manufacturer. Currently they are looking at implementing new technologies to assist its interactions
with both customers and suppliers. The scenario is divided in two parts. In each case, apart from the
manufacturer there is another company involved: either a supplier or a retailer.

The supplier side of the e-Procurement scenario deals with the raw material procurement and the client
side deals with Quotations, Orders and Products delivery.

The issues identified in this case are the following:
- Repetitive manual process for regular bulk orders.
- Confusion resulting from poor product descriptions.
- Missing information.
- Lag time from product order to delivery could be shorter.
- Time spent rating supplier.

Test Case 4: Product Portfolio Management Case
The PPM case investigates into the use of Model Generated Workplaces to support simultaneous
project, resource and results management, performance measurement through work management
views, and the provision of shared project and work monitoring views. The overall objective is to
support collaborative work by providing the actors in the enterprise with the tools, information and
communication support they need to efficiently perform their work. This could be split up into the
following expectations:

- Ability of integrated execution of multiple applications using custom-made, adaptive and
model-generated environments.
- Provision of (near) real-time aggregated views of key business information. These aggregated
views could be provided as services to the roles and actors required, accessing and integrating
data in existing legacy systems. Such aggregated views will enable actors to take more accurate
and timely decisions, exploiting to the full extent the capabilities of existing ICT systems.
- Product related knowledge sharing within and between product life cycle phases.

Approach
This section presents the ATHENA Interoperability Framework (AIF) for enterprise applications and
software systems. This is achieved through a set of reference models that addresses interoperability
issues for conceptual integration, technical integration and applicative integration of software systems.
61

The goal of the AIF is to provide a framework that provides guidance for addressing the
interoperability issues identified in the scenarios presented in the previous section.


Background
Model-driven development (MDD), and in particular OMG’s Model-Driven Architecture (MDA)
(Object Management Group 2003), is emerging as the state of practice for developing modern
enterprise applications and software systems. We believe that there is a need for an interoperability
framework that provides guidance on how MDD should be applied to address interoperability.

MDD as a new architectural approach for developing software systems based on requirements derived
from enterprise and business models. Interoperability solutions should be driven by business needs
first and software solutions second.

Model-driven development represents a business-driven approach to software systems development
that starts with a computation independent model (CIM) describing the business context and business
requirements. The CIM is refined to a platform independent model (PIM) which specifies services and
interfaces that the software systems must provide independent of software technology platforms. The
PIM is further refined to a platform specific model (PSM) which describes the realisation of the
software systems with respect to the chosen software technology platforms. In addition to the
business-driven approach, a model-driven framework should also address how to integrate and
modernise existing legacy systems according to new business needs (Architecture-Driven
Modernisation, ADM).


ATHENA Interoperability Framework (AIF)
The ATHENA Interoperability Framework presented in this paper is designed to fulfil these design
rationales:
- Identification of interoperability issues by interrelating software architectures and enterprise
architectures.
- Identification of the relevant software architecture components.
- Integration of model-driven software development processes.
- Structuring of software technologies, frameworks and methodologies.
The interoperability framework itself is structured according to three main integration areas defined in
ATHENA (ATHENA, 2004):
- Conceptual integration which focuses on concepts, meta-models, languages and model
relationships to systemise software model interoperability.
- Technical integration which focuses on the software development and execution
environments.
- Applicative integration which focuses on methodologies, standards and domain models. It
provides us with guidelines, principles and patterns that can be used to solve software
interoperability issues.
For each of these three areas we developed a reference model to describe and support the application
of model-driven development of software systems.


Reference Model for Conceptual Integration
The reference model for conceptual integration has been developed from a MDD point of view
focusing on the enterprise applications and software system. A computation independent model (CIM)
corresponds to a view defined by a computation independent viewpoint. It describes the business
context and business requirements for the software system(s). A platform independent model (PIM)
corresponds to a view defined by a platform independent viewpoint. It describes software
specifications independent of execution platforms. A platform specific model (PSM) corresponds to a
view defined by a platform specific viewpoint. It describes the realisation of software systems. Figure
62

1 shows this relationship with respect to an enterprise system. It shows how MDA and ADM could be
perceived as a “top-down” and a “bottom-up” approach to software development and integration.
The models at the various levels may be semantically annotated using ontologies which help to
achieve mutual understanding on all levels. We also see the usage of interoperability patterns for
horizontal and vertical integration.

We have identified four categories of system aspects where specific software interoperability issues
can be addressed by conceptual integration. These four aspects can be addressed at all three CIM, PIM
and PSM levels.
- Service aspects: Services are an abstraction and an encapsulation of the functionality
provided by an autonomous entity.
- Information aspects: Information aspects are related to the messages or structures
exchanged, processed and stored by software systems or software components.
- Process aspects: Processes describe sequencing of work in terms of actions, control
flows, information flows, interactions, protocols, etc.
- Non-functional aspects: Extra-functional qualities that can be applied to services,
information and processes.
All of the elements discussed above are integrated into figure 1 where we look at horizontal
and vertical integration between two enterprise systems A and B or between multiple enterprise
systems.
Computational System A
Enterprise System A
(MDD Abstraction)
Execution Platform A
Ontologies
Model-Driven
Architecture (MDA)
&
Architecture-Driven
Modernisation (ADM)
Platform Independent
Model (PIM)
Computational Independent
Model (CIM)
Architecture-Driven
Modernisation (ADM)
Semantic
Annotation
Semantic
Annotation
Model-Driven
Architecture (MDA)
Platform Specific
Model (PSM)
Semantic
Annotation
Computational System B
Enterprise System B
(MDD Abstraction)
Execution Platform B
Ontologies
Model-Driven
Architecture (MDA)
&
Architecture-Driven
Modernisation (ADM)
Platform Independent
Model (PIM)
Computational Independent
Model (CIM)
Architecture-Driven
Modernisation (ADM)
Semantic
Annotation
Semantic
Annotation
Model-Driven
Architecture (MDA)
Platform Specific
Model (PSM)
Semantic
Annotation
S
e
r
v
ic
e
A
s
p
e
c
t
s
In
fo
r
m
a
tio
n
A
s
p
e
c
ts
P
r
o
c
e
s
s
A
s
p
e
c
ts
N
o
n
-
F
u
n
c
t
io
n
a
l
A
s
p
e
c
t
s
Horizontal Integration
V
e
r
t
i
c
a
l











I
n
t
e
g
r
a
t
i
o
n
Reference
Ontology
MT
MT
MT
MT Model Transformation
MT
MT Model Transformation
MI
MI
Interoperability
Patterns
Model Interoperability

Figure 1: Reference model for conceptual integration
We will use this reference model to address model interoperability, where meta-models and ontologies
will be used to define model transformations and model mappings between the different views of an
enterprise system.


Reference Model for Technical Integration
The reference model for technical integration has been developed from a service-oriented point of
view where a software system provides a set of services required by the businesses and users of the
enterprise.

The architecture of the enterprise applications and software systems can be described according to a 4-
tier reference architecture where each tier provides different software services required by the
enterprise. The software system itself is coupled to a service bus that provides the necessary
communication infrastructure. Infrastructure services such as composition, mediation, matchmaking
63

and transformation that enables interoperability between software systems should be provided. We
recognise the need for a model repository for managing models of various kinds, a service registry for
managing naming, directory and location of services, an execution repository for managing
information and state needed in the execution of software services and processes, and a data repository
for managing results and traces of the executions.

Figure 2 shows how a service bus comes into play when integrating two (or more) enterprises systems.
The service bus will make use of infrastructure services, and registry and repository.
Figure 2: Reference model for technical integration
We have defined a reference architecture that separates the architecture of a software system into four
logical tiers (see Figure 3). The reference architecture consists of a local user-space called the user
service domain, and a shared transactional business-space called the business service domain. The four
tiers are as follows:
- User interface tier provides presentation and user dialog logic.
- User service tier provides the user’s model, which may include user session logic and
user-side representations of processes and information. It is an abstraction for a set of
business services.
- Business service tier provides components that represent business functionality and
pervasive functionality (vertical vs. horizontal services). This tier provides enterprise-level
services, and is responsible for protecting the integrity of enterprise resources.
Components in this tier can be process-oriented, entity-oriented or workflow-oriented.
- Resource services tier provides global persistence services, typically in the form of
databases. Resource adapters (e.g. JDBC or ODBC drivers) provide access, search and
update services to databases and its data stored in a database management system (DBMS)
like Oracle or Sybase.
In addition to these four tiers we need a service communication bus so that services deployed at the
various tiers can interoperate both within a tier and across tiers.
64

Legend
Service
Inter-service
communication
U
s
e
r
S
e
r
v
i
c
e
D
o
m
a
i
n
B
u
s
i
n
e
s
s

S
e
r
v
i
c
e
D
o
m
a
i
n
LS
Resource Adapter RA
LA Local Adapter
Local Storage Database
Resource
Service
Tier
Business
Service
Tier
User
Service
Tier
User
Interface
Tier
LS
RA RA
LA
Service Bus
(Middleware
Services)

Figure 3: 4-tier reference architecture for software system architectures


Reference Model for Applicative Integration
The reference model for applicative integration has been developed based on work related to
enterprise architecture frameworks and software architecture frameworks [ENARS; 2004]. Enterprise
and software models can be related in a holistic view, regardless of modelling language formalisms,
by the use of meta-models. This is important in order to understand the dependencies between the
different models and views to achieve interoperability.

The MDD methodology needs to follow a structured approach where interoperability requirements
from business operations in a networked enterprise drive the development of software solutions. This
means that MDD methodology needs to be related to enterprise architectures. A specific part needs to
address how the MDD concepts and the technical software components are reflected in a model world
of the enterprise. Figure 4 shows how the model world, reflecting the applicative integration, is related
to the reference models for conceptual and technical integration. Enterprise and software models can
be built to understand and analyse the physical world of an enterprise.

An enterprise model describes a set of enterprise aspects, which includes descriptions of the business
operations of the business models. These business models provide a context for the software solutions
that needs to be developed and integrated.

Software models describe how software systems are used to support the businesses of an enterprise.
The software models further refine the business models in terms of software realisation models. All
these models should include descriptions of the four system aspects identified in the reference model
for conceptual integration. The software models can be classified as CIM, PIM or PSM models
according to a MDD abstraction.

65

Software System
Enterprise Architecture A
(Model World)
Software
Model
Specification
Models
Business
Context
Business
Enterprise A
(Physical World)
Business
Transactions
Business
Processes
Business
Collaborations
Models of Service, Information
Process and Non-Functional Aspects
Enterprise
Model
Business
Models
Realisation
Models
Users
Business
Tasks
Vertical
Integration
Computational System A
Enterprise System A
(MDD Abstraction)
Execution Platform A
Ontologies
Model-Driven
Architecture (MDA)
&
Architecture-Driven
Modernisation (ADM)
Platform Independent
Model (PIM)
Computational Independent
Model (CIM)
Architecture-Driven
Modernisation (ADM)
Semantic
Annotation
Semantic
Annotation
Model-Driven
Architecture (MDA)
Platform Specific
Model (PSM)
Semantic
Annotation
MT
MT
MT Model Transformation
U
s
e
r

I
n
t
e
r
f
a
c
e

S
e
r
v
i
c
e
s
U
s
e
r

S
e
r
v
i
c
e
s
B
u
s
i
n
e
s
s

S
e
r
v
i
c
e
s
R
e
s
o
u
r
c
e

S
e
r
v
i
c
e
s
S
e
r
v
i
c
e

B
u
s
MT
MT
MT MT
MT
Models of other Enterprise Aspects

Figure 4: Reference Model for Applicative Integration
In our interoperability framework we have identified a set of models that we see useful in achieving
interoperability. However, we also acknowledge the fact that this is just a baseline. Different
enterprises must be able to develop their own software views that they see purposeful. It is important
that the applicative integration supports the development of a set of shared views amongst different
stakeholders and provides means for managing the dependencies between these views. We believe a
viewpoint-based integration approach must be chosen. This allows incorporating viewpoints, which
are implicitly or explicitly defined by other enterprise or software modelling approaches into an
applicative framework.


The ATHENA project
The ATHENA IP aims to enable interoperability by providing reference architectures, methods and
infrastructure components. In ATHENA Research & Development will be executed in synergy and
collaboration with Community Building: research will be guided by business requirements defined by
a broad range of industrial sectors and integrated into Piloting and Technology Testing as well as
Training. ATHENA consists of three action lines in which the activities will take place (Chen et al,
2004). In Action Line A, the research and development activities will be carried out. Action Line B
will take care of the community building whereas Action Line C will host all management activities.
In this Section we focus on the research activities of ATHENA.

R&D Projects
In Action Line A six research topics/projects were defined for the first stage of the IP. The definition
of these R&D projects was based on the needs to support and realize the ATHENA Interoperability
Framework:

- Enterprise Modelling in the Context of Collaborative Enterprises (A1) aims at developing
methodologies for management and modelling of situated processes, flexible resource
allocation and assignment. Furthermore, it investigates methodologies for work
management and execution monitoring.
- Cross-Organisational Business Processes (A2) deals with modelling techniques to
represent business processes from different organisations on a level that considers the
privacy requirements of the involved partners. Such models need to be executed through
IT systems and need to operate efficiently in an architectural environment that adapts to
particular business scenarios.
- Knowledge Support and Semantic Mediation Solutions (A3) aim at the development of
methods and tools for the semantic enabled enterprise. A key objective is to build an
66

integrated software environment that is able to manage the semantics of different
abstraction levels that can be found in an enterprise.
- Interoperability Framework and Services for Networked Enterprises (A4) is concerned
with the definition of reference architectures and infrastructures supporting
interoperability of enterprise systems and applications.
- Planned and Customisable Service-Oriented Architectures (A5) is to develop the
understanding, tools and infrastructures required for service-oriented architectures which
can be achieved more easily through the planning and later customisation of solution.
- Model-driven and Adaptive Interoperability Architectures (A6) is to provide new and
innovative solutions for the problem of sustaining interoperability through change and
evolution, by providing dynamic and adaptive interoperability architecture approaches.

Based on the experience with the abovementioned R&D Projects of Action Line and the update of the
Interoperability Roadmap two new projects are defined. These new projects will also run in the course
of the ATHENA programme of work and are addressing additional topics that go beyond the initially
planned activities. These research projects are currently under finalisation:

- Business Documents and Protocols (A7) is to analyse current industry practice for
business documents and protocols handling, to develop an ATHENA approach for
creation and usage of business documents and protocols, and to create business content
using the ATHENA approach to business document and protocol handling.
- SME Interoperability in Practice (A8) is to analyse of interoperability situation and needs
of SMEs and current approaches, to evaluate differences in interoperability requirements
of SMEs and large enterprises, and to define a characteristic (SME) interoperability
scenario as a guideline and common problem definition for development of an integrated
and targeted solution based on ATHENA results.

Research Environment
ATHENA builds on results and experience of the thematic network IDEAS - Interoperability
Development of Enterprise Applications and Software (IDEAS, 2003). Goal of IDEAS was the
definition of a roadmap for Interoperability of Enterprise Systems and Applications. This roadmap was
used in ATHENA to define the R&D Projects described in the previous section.

Furthermore, besides ATHENA, the European Commission approved a Network of Excellence
addressing a similar problem space: INTEROP (Interoperability Research for Networked Enterprises
Applications and Software) (INTEROP, 2004). It aims at integrating expertise in relevant domains for
sustainable structuration of European Research on Interoperability of Enterprise applications.
ATHENA is closely related to INTEROP.

Results
This section gives an overview of the results that are foreseen by the ATHENA project. Right from the
outset of ATHENA, all research activities defined the results that they plan to achieve. To make this
list manageable, a set of thirteen key results was identified.


ATHENA Key Results
The thirteen key results will be achieved in various areas covered by the holistic approach of
ATHENA. Each of the key results is incrementally achieved over time during the project lifetime and
further in the 5 years intended horizon. For more details regarding the ATHENA key results please see
the ATHENA deliverable D.C1.1 & D.C2.1, which will be available from the ATHENA website
(ATHENA, 2004). The key results are structured into one of the three result types: ‘technical results’,
‘strategic results’ and ‘business results’. In the following the key results are summarised:


67

Business Results:
- Key result 1 – “Business Interoperability framework”, is expected to address
interoperability systematically in a business – as opposed to technical – context and by
identifying the key business drivers for interoperability.
- Key result 2 – “Interoperability requirements for applications” will provide an
understanding for the individual interoperability requirements of enterprises derived from
industry scenarios.
Strategic Results:
- Key result 3 – “Interoperability Impact Analysis Model” is expected to provide a general
model for determining the impact of interoperability on businesses and to apply the model
to assess ATHENA’s impact.
- Key result 4 – “Policy Action Recommendations” will be based on the ATHENA
experiences/results and intended to provide recommendations to the European
Commission in this policy area.
- Key result 5 – “EIC Services and Infrastructure” will represent the identification, design
and implementation of a portfolio of services supporting the interoperability community
through (and in) the Enterprise Interoperability Centre (EIC).
Technical Results:
- Key result 6 – “Reference Architecture” provides the conceptual integration of research
results achieved in ATHENA and is embedded into the overall ATHENA Interoperability
Framework.
- Key result 7 – “Guidelines and Best Practices” represents the knowledge and experience
gathered during validation of research results in industry/application scenarios.
- Key result 8 – “Interoperability infrastructure” and associated tools represent the
technical integration of research results.
- Key result 9 – “Collaborative Enterprise Modelling Platform” will contain at least three
key elements: Exchange, federation and integration of enterprise models; extended
modelling language and execution environment; and methodology to establish EM in
companies in a benefit-driven way.
- Key result 10 – “Cross-Organisational Business Process Modelling and Enactment” will
provide a suitable modelling methodology and execution capability for cross-
organisational business processes.
- Key result 11 – “Ontology-based Semantic Annotation and Reconciliation
method/language/tool” provides a framework including languages and tools for ontology-
based semantic annotation and reconciliation.
- Key result 12 – “Service Composition Framework” supports a service-oriented ICT
environment, focusing at the situation where compositions are planned a priori and later
customised (or refined).
- Key result 13 – “Model-driven and Adaptable Interoperability Framework and
Infrastructure” provides a model-based and adaptable interoperability framework and
infrastructure based on a model-driven architecture approach combined with dynamic
service-oriented architecture, agents and peer-to-peer (P2P) technologies.





68

Results achieved so far
This section provides descriptions and evaluations of the results that ATHENA has achieved in its first
12 months of existence.

In the first 12 months of ATHENA 18 individual results have been achieved. These results contribute
to the 13 key results of ATHENA described in the previous paragraph. Achieving the key results is an
incremental process, which builds on the achievement of results at each milestone. Insofar, achieving
the results at month 12 is important to be able to achieve the overall results. Table 5 provides the list
of achieved results in month 12.


ATHENA results achieved up to month 12
Business
Results
• (B4) Method of defining the dynamic requirements of industrial end users and a system for
implementing the method
• (B4) Initial set of to-be scenarios and common requirements based on involved end users
• (B5) International Technology Testing co-operation plan
• (B5) Technology Testing planning, and validation framework
• (B5) Defined test plans for involved industries
• (B6) Training services plan and Curriculum defined
Strategic
Results
• (B1) The EIC established
• (B1) EIC services and infrastructure defined and started
• (B2) Knowledge sharing strategy and website available
• (B3) ATHENA Contribution to Policy Action Plan (V1)
Technical
Results on
Integration
• (A4) Definition of an initial reference architecture and integration requirements
• (A4) Specification of interoperability infrastructure and tool interfaces
Technical
Results on
Business
• (A1) Views and constructs to represent collaborative enterprises
• (A1) Common collaborative enterprise repository definition
• (A2) Modelling method for cross-organisational business processes (CBPs)
Technical
Results on
Knowledge /
Semantics
• (A3) Ontology authoring and management system with Ontology modelling language for
informational knowledge
Technical
Results on ICT /
Systems
• (A5) Models of service description and usage
• (A6) Specification of basic architecture reference model for a model-driven and adaptive
interoperability architecture and enhanced repository infrastructure


More details on the ATHENA results together with an evaluation of the achieved results can be
found in the deliverable D.C2.2, which can be found on the ATHENA public web site [ATHENA,
2004].

In a first phase the research activities in Action Line A addressed the Collaborative Product
Development scenario. The results listed in Table 1 were applied to address the following aspects of
the scenario:

- The first aspect addressed the exchange of test plans between design partners. In the given
scenario the OEM is providing test plans to the design partner that have to be used to test the
developed components. The basic assumption is that the collaborating partners do not share the
same modelling environment for test plans. The results of A1 were used to show the exchange of
such test plans, between modelling environments of different providers that do not share a
common data model.
69

- The aspect of cross-organisational business processes was demonstrated using a business process
modelling tool developed in ATHENA that is able to directly support the concepts of private
processes, public views and collaborative processes. The modelling method developed in A2
explicitly addresses the information hiding details of the private process in the public view.
Furthermore, the developed business process model was simulated to support the process modeller
in refining the business process.
- The semantics dimension of the scenario was covered by showing the definition of a draft
Ontology for Request for Quotation (RFQ) messages. The features of ontology management
system ATHOS developed in ATHENA include the multi-user support together with its unique
approach to achieve a higher level of domain adequacy.
- The results of A5 were shown by a joint definition of service interfaces and messages exchanged.
Two aspects were of interest: Firstly, the annotation of exchanged SOAP messages to evaluate
their conformance with WS-standards by raising the abstraction to allow non-technical experts to
identify and deal with incompatibilities. Secondly, the dynamic supplier choice was shown using
an agent framework demonstrating the necessity of being able to handle a dynamic, not a-priori
known set of suppliers.
- The aspect of adaptive interoperability architectures was covered using the Business Resource
Management Framework (BRMF) that used a P2P approach to distribute business documents both
internally and between enterprises. The BRMF makes distributed access to managed documents
available and furthermore provides an automatic notification service for updates to documents
managed by the BRMF – one significant problem in Collaborative Product Design, when business
partners work with outdated information.

It can be clearly seen that the Collaborative Product Design scenario has multiple interoperability
aspects to it that must be addressed by dedicated, specific concepts, methods and tools. It has to be
kept in mind that the solutions presented in this contribution reflect the status of ATHENA after
twelve months. Until the end of ATHENA it is expected to complete the solution offering and to
provide even more powerful interoperability solutions.


Acknowledgement
The work published in this paper is (partly) funded by the E.C. through the ATHENA IP. It does not
represent the view of E.C. or the ATHENA consortium, and authors are solely responsible for the
paper's content.

The authors thank and acknowledge the members of the ATHENA consortium:
SAP (D), AIDIMA (E), COMPUTAS (NO), CR FIAT (I), DFKI (D), EADS (F), ESI (E), FORMULA
(I), FHG IPK (D), GRAISOFT (F), IC FOCUS (UK), INTRACOM (EL), LEKS (I), SINTEF (N),
TXT (I), UNIV. Bordeaux I (F), UNINOVA (POR), IBM (UK), SIEMENS AG (D).

References
IEEE (1990): IEEE Standard Computer Dictionary: A Compilation of IEEE Standard Computer Glossaries,
Institute of Electrical and Electronics Engineers, New York, NY.
ATHENA (2004): ATHENA Public Web Site, http://www.athena-ip.org/, (accessed: 2004).
Object Management Group (2003): MDA Guide Version 1.0.1, Object Management Group omg/2003-06-01,
12th June.
Elvesæter, B., Neple, T., Aagedal, J. Ø., Rolfsen, R. K. and Stensli, O. Ø. (2004): MACCIS 2.0 - An
Architecture Description Framework for Technical Infostructures and their Enterprise Environment,
presented at Command and Control Research and Technology Symposium, San Diego, USA.
Chen, D., Stevens, R., Schulz, K., Doumeingts, G. (2004): European Approach on Interoperability of Enterprise
Applications – Thematic Network, Integrated Project and Network of Excellence, IFAC04, Brazil.
IDEAS Project Deliverables (WP1-WP7), Public Reports, www.ideas-roadmap.net, 2003.
INTEROP, "INTEROP Portal", http://www.interop-noe.org/, (accessed: 2004).
Anastasiou, M., Núñez, M.J., Garcia, O. (2004): Towards Interoperability of Heterogeneous Enterprise Networks
and their Applications – Requirements Handling and Validation activities, presented at I-ESA Conference,
Geneva.
70

4.2. E-Business Standardisation in the Automotive Sector
– The situation of SMEs

Martina Gerst, University of Edinburgh, martina.gerst@ed.ac.uk
Kai Jakobs, Aachen University, kai.jakobs@cs.rwth-aachen.de


The use of ICT related technologies, particularly e-business systems, facilitates the creation of a
network of relationships within a supply chain. Yet, such inter-organisational integration requires
interoperability that cannot be achieved without agreed standards. But who has a say in the
standardisation process?
Using the automotive industry as an example, this paper looks at two approaches towards
standardisation, both of which involve large companies and SMEs. One approach is based on a
standardised, albeit sector-specific portal, the design and implementation of which was pushed by a
group of large car manufacturers. The alternative could be the use of international standards, and
pro-active participation in the open standards-setting process by all relevant stakeholders. It turns
out that the situation of SMEs is not very favourable in either case – both processes are largely
dominated by the big guys. Nonetheless, the paper makes some recommendations how this situation
may be changed for open standards setting.


Introduction
The automotive industry is facing a number of challenges to the established relations between its
players. Issues to be addressed include, for instance, shorter product life cycles, increasing cost
pressure in stagnant markets, and higher complexity of the embedded electronic systems. To meet the
associated production requirements, standardisation of processes, systems, and data is inevitable.

This has already led to a range of transformations in the structure of the automotive supply chain.
Large OEMs
7
have been forced to create networks to replace the existing one-to-one relations with
their suppliers (which are typically SMEs
8
).

Also, in many cases an SME supplier does business with more than one OEM. In this situation, bi-
lateral standardisation to improve the co-operation between OEMs and suppliers, and between
different suppliers, respectively, is inefficient. Still, this has been the approach of choice in many
cases. However, possible alternatives are available, including sector-specific harmonisation (e.g., in
the form of an electronic market place) and, particularly, international, committee-based
standardisation.

According to a study of Nexolab in 2001, standards were a major headache for SMEs. 75% of the
suppliers saw the lack of standardisation as a major obstacle for closer collaboration. Therefore, it
might be useful for companies to rethink their standardisation strategies.

The automotive industry is characterised by vertical integration in terms of the business relationship
structures between OEMs and suppliers (Lamming, 1993; Adolphs, 1996). A current trend in
manufacturing is that OEMs attempt to co-operate with fewer suppliers, but on a world-wide scale. As
a result, small and medium sized suppliers become suppliers to tier 1 or 2 suppliers, rather than
directly to the OEMs.

The pressure for collaboration has led organisations in the automotive sector to become involved in a
range of projects by means of inter-organisational systems (IOS). Examples include electronic
collaboration projects, the integration of engineering processes, and electronic catalogue projects to
present product and service data. Such IOSs are adopted not only to achieve operational effectiveness

7
Original Equipment Manufacturers.
8
Small and Medium-sized Enterprises.
71

by reducing co-ordination costs and transaction risks (Kumar & van Dissel, 1996) but also to improve
communication and information presentation. Collaboration and integration shift the emphasis from
“stand-alone” initiatives to the development of standardised and integrated solutions (Koch & Gerst,
2003). In this context, one form of IOS that fulfils the criteria of collaboration and integration are
business-to-business/supplier portals incorporating standardised business processes. Covisint, an e-
marketplace, founded in 2000, by large OEMs, is a very good example to analyse the standardisation
process in an industry which is characterised by a large number of SMEs. The theory of social shaping
of technology (SST) of standardisation studies understand standard setting as locally constructed and
negotiable between the actors involved in the process (Graham et al, 1995; Williams et al, 1993). SST
studies have focused on identifying the socio-economic factors that shape standard development
arguing that these factors and the relevant actors cannot be reduced to a narrow economic or technical
explanation.

In this paper, the SST perspective is applied to two case studies to analyse the use of IT to standardise
business processes in the automotive industry. The cases explore standardisation processes in its social
context and identify and discuss the factors that shape the development and implementation of the
standards. The remainder of the paper is structured as follows: following a very brief outline of the
research methodology, the development and implementation of the Covisint processes is described in
section 3. Section 4 looks at the role of SMEs in today’s formal standardisation processes, which
might be an alternative to the proprietary, sector-specific approach standardisation previously
discussed. A discussion of the results concludes this paper.


Research approach
This paper discusses two different approaches to standardisation primarily based on empirical
research. A case study will analyse the adoption of Covisint by a Large Automotive Company (LAC),
and will discuss the factors that shaped the development process and the various issues related to the
standard’s implementation. This analysis adopts a single case-study design based on qualitative
research. Data have been collected through semi-structured interviews, an online user survey, direct
observation, and extensive secondary data research. A mixture of quantitative and qualitative methods
have been deployed to analyse the data.

The study on the role of SMEs in committee-based standardisation is based on desk research and
several (smallish) studies. Here, data were collected through different questionnaires, each comprising
a number of open-ended questions. Qualitative methods have been deployed to analyse the data.


The Covisint case study

Context
According to a study by McKinsey (2003), the automotive industry will in the next ten years be
shattered by a third ‘revolution’, following the invention of the assembly-line production by Henry
Ford and the lean production of Toyota. Customers are expecting better value for the same money,
resulting in continuous cost pressure and innovation marathons for OEMs.

This has led to a range of transformations in the automotive supply chain. For example, in order to
improve customer satisfaction and increase revenue growth and shareholder value, large OEMs and
their suppliers started establishing large automotive networks. Yet, the added value of these
collaborative networks is beginning to shift from the OEMs to suppliers and to other business partners
such as system integrators (see figure 1).

72

•In den 90er Jahren
Tier 1 Tier 3 Tier 2 OEM Tier 1 Tier 3 Tier 2 OEM
•Heute
Tier 1 Tier 3 Tier 2 OEM


•M3
Tier 1 Tier 3 Tier 2 OEM


•M3
•In den 80er Jahren
Lieferant 1
Lieferant 2
Lieferant n
Lieferant 3
Lieferant 4
OEM
Supplier 1
Supplier 2
Supplier n
Supplier 3
Supplier 4
OEM
The future
The future
OEM 1
OEM 1
Strategic
specialists
Strategic
specialists
Development
services
Development
services
System
integrators
System
integrators
OEM ...n
OEM ...n
OEM 2
OEM 2
•In den 90er Jahren
Tier 1 Tier 3 Tier 2 OEM Tier 1 Tier 3 Tier 2 OEM Tier 1 Tier 3 Tier 2 OEM Tier 1 Tier 3 Tier 2 OEM
•Heute
Tier 1 Tier 3 Tier 2 OEM


•M3
Tier 1 Tier 3 Tier 2 OEM


•M3
Tier 1 Tier 3 Tier 2 OEM


•M3
Tier 1 Tier 3 Tier 2 OEM


•M3
•In den 80er Jahren
Lieferant 1
Lieferant 2
Lieferant n
Lieferant 3
Lieferant 4
OEM
Supplier 1
Supplier 2
Supplier n
Supplier 3
Supplier 4
OEM
The future
The future
OEM 1
OEM 1
Strategic
specialists
Strategic
specialists
Development
services
Development
services
System
integrators
System
integrators
OEM ...n
OEM ...n
OEM 2
OEM 2
OEM 1
OEM 1
Strategic
specialists
Strategic
specialists
Development
services
Development
services
System
integrators
System
integrators
OEM ...n
OEM ...n
OEM 2
OEM 2

Figure 1: Automotive networks determine future collaboration. Source: BMW

In the 1980s, the relations between one OEM and all suppliers were similar. In the 1990s this changed
to a ‘tier-x’ structure, where the main collaboration partners of an OEM were the tier-1 suppliers
which, in turn, collaborated through tier-2 suppliers, and so on. Today, due to the higher market
demands, OEMs are not only collaborating with their supply base but also with other business
partners, for example system integrators. In the future, the relations between OEMs and their suppliers
are expected to change dramatically (Gerst & Bunduchi, 2004).

Apart from shifts in the value chain, the industry is confronted with a number of transformations that
challenge the established relations between industry players. The automotive industry is characterised
by extremely complex processes, and the standardisation of processes and data is inevitable in order to
meet production requirements. Driven by challenges such as shorter product life cycles, increasing
cost pressure in stagnant markets, and higher complexity of the electronics embedded in modules and
systems, OEMs gradually increase the outsourcing of manufacturing, which is expected to rise from
25% to 35% within the next 10 years (McKinsey 2003).

The supplier community is also undergoing major changes as the result of this pressure. Increasingly,
platforms and model varieties require advanced deals and project management capabilities. This
means that in terms of innovation management suppliers have to be able to provide leading-edge
technology and efficient simultaneous engineering processes. This change affects primarily the tier-1
suppliers which are taking over systems integration responsibility and management of the supply chain
from the OEMs. At the same time, they also take an increasing share of risk, which used to be incurred
by the OEMs. As a result, the industry is forced to collaborate more closely, e.g. by adopting portal
technology.

Development
The initial phase of the portal project was characterised by the Internet hype, and by internal and
external cost pressure. The automotive industry saw the use of standardised technology as an
opportunity to streamline and extend their business processes to reduce costs and risks, and to enable
them to work seamlessly across organisational borders. For OEMs, the vision behind Covisint was to
connect all players in the automotive industry into one global market space, based on standardised
business processes. The approach of the initial Covisint stakeholders, Ford, GM and Daimler Chrysler,
was to consolidate their market power and set a de-facto industry standard. The founders expected that
the other OEMs, and the tier-1 suppliers, would buy into the Covisint vision. However, only two other
OEMs joined Covisint, Renault/Nissan in 2000 and PSA in 2001.

This original vision of the Covisint founders was not necessarily shared by the suppliers. There were
two reasons for this. First, suppliers already struggled with the administration of a number of such
‘standardised’ portals, and the suppliers that had been approached at an early stage had mixed feelings
regarding the OEMs’ approach to volume bundling and pricing. One result of this controversy was the
creation of another e-marketplace, named ‘SupplyOn’, by a number of large tier-1 suppliers who
73

defected from the Covisint vision. Second, most suppliers were excluded from the development
process, with only a few of the largest and most powerful tier-1 suppliers being invited at a later stage
of the development phase. As a result, suppliers’ requirements were by and large neither part of the
Covisint vision, nor part of the development of the standardised technology. Therefore, despite the
claimed goal to address the costs and risks reduction pressure across the entire industry, the
development stage only included the requirements and visions of a limited number of OEMs.

In the first instance, following the ‘tradition’ of all IT projects in the industry, standards development
was related to best practices in the industry and was worked out by a limited number of specialists
from those OEMs that were involved in Covisint. At a later stage, this informal small group approach
to standard development was replaced by a formal consortium of the Covisint stakeholders and the
software companies which delivered pieces of software to complete the necessary Internet hub. The
consortium approach resembled the typical approach to IT standards development, following specific
procedures and employing different working groups. Additionally, industry experts were invited to
presentations and workshops to contribute to the standard development. This change in the approach
to standards development was supposed to assure suppliers of the legitimate character of the standards,
which in turn would benefit the adoption rate. At the same time, OEMs were confident that they could
influence suppliers during negotiations to adhere to their interpretation of the Covisint standards.
However, the development of the standards was tightly controlled, as participation in the consortium
was severely restricted with only the most powerful suppliers being invited, and with opaque working
procedures.


Implementation
The implementation was characterised by severe cost pressure, as the process was driven by the need
to integrate into the solution each new user requirement (within budgetary limits). For example,
tensions appeared within the implementation project team between IT departments and the business
units (BUs) regarding the inclusion of organisational specific functionalities into the standard. The
negotiations, as well as the solutions, were heavily influenced by budget restrictions. Therefore, the
implementation of the standardised portal involved ‘work around solutions’ to address both users’
needs and the limitations in the budget. Such a pragmatic approach called into question the vision of
the best practice solution’, as ‘best practice’ was sacrificed to comply with financial limitations.

Apart from cost pressure significant technical difficulties challenged the implementation of the
standardised technologies. For example, the design of the user interface had to support personalised
and integrated access for suppliers. Moreover, the planned approach was to re-use the content of
already existing OEMs’ web sites in order to reduce content management cost. However, two
problems were encountered at this stage. First, it was technically impossible to adopt the layout
approach of the OEMs portals’ web sites using the portal technology that Covisint provided. Second,
due to security concerns, it was not possible to integrate much of the already existing portlets
9
. The
complexity of the technology and the need to integrate legacy systems meant that security concerns
made it highly difficult to implement the ‘single sign-on’ functionality for the overall IT infrastructure.
The longer into the project the more difficult the co-ordination of OEMs and Covisint became.
Covisint’s prime objective was to offer standardised processes. However, during implementation it
became clear that even the participating OEMs, let alone the suppliers, had difficulties to adopt this
standard which they had already agreed upon during the development phase.

Organisational and social challenges added to the technical and economic pressure to shape the
implementation process. For example, the integration of applications was a major implementation
issue. These applications were either standard software such as SAP, with separate login processes, or
legacy systems specifically programmed to meet the needs of individual BUs, and not able to deal
with web-based frames. As a result, the cost for additional programming to ‘web-enable’ these

9
A portlet is a Web component, usually managed by a container that processes requests and generates dynamic
content. Portals use portlets as pluggable user interface components to provide a presentation layer to
information systems (Sadtler et al., 2003).
74

applications had to be discussed with the different application owners. As IT budgets were limited, the
negotiations focused around the question which department had to cover the extra costs.

Covisint was bought by Compuware in 2004. The former version of the LAC supplier portal has been
completely integrated into the Covisint supplier portal. Today, and despite the massive initial
problems, the Covisint LAC supplier portal is enjoying something like a second lease of life.


SMEs and Standardisation
So far, we have looked at the role SMEs may play in the context of largely proprietary, sector-specific
standardisation processes, driven and dominated by large companies. There is, however, an alternative
route towards standards – through standards setting bodies (SSBs) producing open specifications. In
the following, we will have a closer look at the prospects of SMEs in this environment. This chapter
will, therefore, analyse what would have to be done to make standards setting in the ICT domain more
accessible, and useful, for small and medium enterprises.

Motivation
Today, the standards setting processes in the Information and Communication Technologies (ICT) and
e-business sectors are very much dominated by the large companies and other financially potent
stakeholders. As a consequence, there is a real danger that standards – and thus, ultimately, policies –
are based on the needs and requirements of a comparably small – albeit powerful – group of
stakeholders. The action plan for innovation ‘Innovate for a Competitive Europe’ – rightly – says
“Voluntary standards, properly used, can help establish the compatibility of innovative concepts and
products with related products and so can be a key enabler for innovation. … SMEs should be more
involved in standardisation, to exploit their potential for innovation and to enhance the accountability,
openness and consensus-based character of the European standardisation system.” (European
Commission, 2004).

Yet, the Working Groups (WGs) of almost all standards setting bodies are populated by
representatives of large, multi-national companies. The comparably few representatives of Small and
Medium Enterprises (SMEs) typically come from highly specialised vendors or manufacturers. (SME)
users – i.e. those who ‘merely’ deploy ICT systems – are hardly represented at all, and neither are their
umbrella organisations.

Today, SMEs are under enormous pressure from their – frequently large – customers to deploy e-
commerce systems (including the necessary underlying ICT infrastructure) which are compatible with
the customer’s respective systems. Yet, as these systems typically differ, SMEs accordingly have to
set up and maintain a number of different such systems. This is hardly a sustainable option, and the
use of standards-based systems is an SME’s only chance to keep both its ICT environment
manageable and all customers happy.

Some Background
There seems to be general agreement that participation of all stakeholders – particularly including
users – is a sine qua non for an ICT standardisation activity to be successful. In fact, increased user
participation is often considered as the panacea for all problems.

In fact, the importance of the involvement of all stakeholders in the standardisation process is nicely
illustrated by the case of MAP
10
. The failure of this potentially very useful multi-million dollar
initiative may not least be attributed to the complete absence of SMEs in the standardisation activity.
The extremely complex specifications were done solely by very large companies, for which this
complexity apparently was not such a big problem. Unfortunately, they failed to realise that the

10
Manufacturing Automation Protocol’, an initiative started by General Motors and other large companies in the
mid-eighties to provide for inter-operability of IT systems in production environments (see e.g. (Dankbaar,
1992)).
75

situation was very different for their SME suppliers. For them, the specifications were way too
complex to be implemented and managed. Eventually, the whole initiative collapsed.

Typically, SMEs opt for readily available off-the shelf systems and services, which need to be cheap
and easy to install, maintain and use. Proprietary systems are also used frequently where SMEs are
compelled to do so by e.g. a major business partner (with all associated problems). The non-use of
many standards-based services by SMEs is largely due to the fact that insufficient knowledge and
resources are available to employ these systems, which are perceived as being extremely complicated
to deal with. In fact, this perception may be considered as a major impediment to a more successful
uptake of standards based systems by SMEs. This exemplifies an urgent need for simpler standards.

The procedures adopted by the individual standards setting bodies suggest that the degree of control
over, and influence on, the standards setting process is about equally distributed between the different
stakeholders (see Figure 2).

Standard
Committee
End
users
produces
Others
Government
representatives
Implementors
Service
providers
Corporate
users

Figure 2: The Naive View of a Standards Setting Process (Jakobs, 2004)

Unfortunately, this does not quite capture reality. Especially the assumption of an equal influence of
all stakeholders appears to be flawed (Swann, 2000). In fact, it appears that so far development of IT
standards has almost exclusively been technology driven. This can largely be attributed to the fact that
relevant standardisation committees have typically been dominated by vendors and service providers.
Accordingly, a more realistic model is called for, which will be presented in section 5.

SMEs in Standards Setting – A Small Study
As part of a project co-funded by the European Commission one of the authors did a small study of
selected ITU and ISO working groups, to learn about some issues relating to SME users in standards
setting
11
. In summary, it became clear that both ITU and ISO are indeed dominated by large
companies. SME representation (if any, that is) occurs primarily through small consultancy firms, as
opposed to actual users. Also, the influence ‘real’ SMEs (i.e., excluding consultants) have on the
process is said to be very limited in most WGs. At the – higher – decision level it is virtually nil.

Respondents’ opinions were split about SMEs’ influence at the technical level. A sizeable minority
basically stated that in many cases influence is related to market power. The majority, however, noted
that the individual capabilities of the representatives (i.e., e.g., technical skills, language proficiency,
enthusiasm) are the deciding factor.

SME participation would broaden technical expertise of a WG, as they are frequently closer to state-
of-the-art technical development than big companies, and less bound by internal rules and
administrative procedures. Also, they would be welcome as a counter-weight to the interests of the big
companies. This holds particularly if they represented fora or some other form of umbrella
organisations. However, it was also noted that the typical sporadic/infrequent participation of SME
representatives may lead to inadequate familiarity with both technical aspects discussed as well as
procedures, thus causing unnecessary delays to the process.


11
The full Report may be found at http://www-i4.informatik.rwth-aachen.de/~jakobs/grant/Final_Report.pdf.
76

Cost of participation is considered the major obstacle SMEs will face if they want to become active in
standards setting. Suggestions how this could be overcome include increased deployment of electronic
media to replace meetings, lower or waived fees for SMEs, provision of dedicated travel money. In
addition, it was suggested that SMEs join forces and co-sponsor representatives.


Discussion
The Covisint study shows that standardisation efforts are triggered by a complex array of non-
technical and technical considerations. The case illustrates that ICT standardisation is not only about
bridging the gap between the technologies and business processes of different companies, but also
about bridging across complex social processes.

As suggested by the SST perspective, this vision of industry wide collaboration has been actively used
by OEMs to mobilise resources internally and to attract suppliers into buying into Covisint. However,
a number of factors have shaped the OEMs - and suppliers’ - choices during the development and
implementation of the standardised technology, which have eventually led to a very different outcome
than that envisaged initially by the founding OEMs.

Each of the founding OEMs has an extensive network of suppliers. They, in turn, frequently supply
more than one OEM. In this situation, bi-lateral standardisation of the complex processes and
technology that enable collaboration both between OEMs and their suppliers, and between the
different suppliers is less than effective, as it would leave suppliers with the need to maintain one
system per each OEM. Moreover, market pressures were forcing OEMs to reduce costs, increase the
efficiencies in the industry and enhance collaboration with their suppliers. Therefore, the idea to join
forces in order to provide a ‘single point of entry’ and set an industry standard seemed advantageous
for both groups. Furthermore, when the Covisint idea emerged in late 1999, the use of leading-edge
Internet technology to reorganise internal and external business processes to support collaboration
across the entire supply chain was on every company’s agenda. Consequently, the foundation of
Covisint was a natural step to increase the effectiveness of the industry through a collaborative effort
of the largest industry players. Indeed, such collaboration was required to share the risks and costs
between a number of players.

The three founders showed their commitment to the Covisint vision through an initial investment of
about $500 million. However, due to the distribution of power that historically characterised the
relations between OEMs and suppliers, the latter were apprehensive of Covisint. They saw it as just
another exercise to intensify OEMs’ power pressure. Some suppliers also feared that Covisint would
require significant additional resources and investments from their side, whereas the benefits would
mostly materialise at the OEMs’ side.

However, on the OEM side significant resources involving not only additional budget but also extra
human resources were required to address the pending integration issues. The need for these additional
resources led to negotiations concerning their allocation across different BUs within the OEMs. As a
result of these negotiations, some application owners (the BUs within the participating OEMs)
abandoned the idea of adopting standardised business processes, and started blaming Covisint for not
providing mature, workable solutions. It was even claimed that suppliers already working with the
applications did not see any of the benefits. Consequently, far from reaching stabilisation and closure,
the choices made by the OEMs further deepened the disagreement regarding the approach to an
industry wide standardised portal, which was deserted not only by suppliers but even by some of the
BUs within the founding OEMs.

The above seems to indicate that SME suppliers were not particularly satisfied with the standardised
solution developed by their large customers. Yet, it would appear that SMEs do not necessarily fare
any better in today’s open standards setting processes.

Today, not unlike the Covisint case, active participation in ICT and e-business standards-setting is
largely limited to large, multi-national companies. In particular, SMEs hardly stand a chance to make
77

their voice adequately heard. As standardisation and policy-making are mutually dependent, this is an
extremely unsatisfactory situation. Ultimately, it means that the influence of globally acting multi-
nationals on European policy is out of proportion with e.g. the number of jobs they provide in Europe.
In a way, SMEs are part of a modern-day ‘Third Estate’ with respect to their capability to influence
standardisation and thus, ultimately, policy making. This holds despite the fact that there are over 20
million SMEs in the EU.

Standardisation processes should provide a platform where opportunities of technologies,
requirements of various types of companies from all sectors, consumer preferences, and other societal
needs, e.g. protection of the environment, are efficiently mediated. Standards that are useful for all
relevant stakeholders should be the outcome of these processes.

Unfortunately, it appears that so far development of IT standards has almost exclusively been
technology driven; with standards produced solely reflecting providers’ and implementers’ priorities
like, for example, manageability rather than usability. Most other stakeholders, including the general
public, consumer organisations and, most notably here, SME users, constitute what one might call the
‘Third Estate’ of IT standards setting (see Figure 3).

.
................
SME
users
General
public
Consumer
organisations
Product
user groups
Professional
umbrella
organisations
Manufacturers
Service
providers
Standards
Committee
End
users
Large
corporate users
Government
Business
partners
very little influence
strong influence
dialogue, impact questionable
limited influence
The 'third estate'
SME
umbrella
org.
...............
hardly any influence

Figure 3: Relations between stakeholders in standardisation

The figure shows that the members of the ‘Third Estate’ – and specifically SMEs – are largely
separated from the key players, with SME umbrella organisations perhaps located somewhere in
between. Although they represent the vast majority of standard users these groups have an extremely
little say in the standards setting process. This holds despite the fact that organisations such as ANEC,
the ‘European Association for the Co-ordination of Consumer Representation in Standardisation’, and
NORMAPME, the ‘European Office of Crafts, Trades and SMEs for Standardisation’, are actively
participating in selected standard working groups on behalf of their constituency.


Conclusions
The task of developing and implementing standardised business processes to collaborate more
effectively across the full supply chain is more challenging than ever. Supplier portals are one of the
options to collaborate more closely and harmonise cross-company business processes. Apart from the
technical issues surrounding the development of standardised business processes across the entire
industry (i.e. the complexity of technology, integration issues and security concerns), a range of
organisational, social and economic factors have influenced the OEMs’ and the suppliers’ choices and
actions, which have eventually led to the undesired outcome of failing to accomplish the initial vision
of industry wide collaboration supported by common industry wide standards.

78

However, given the failure of the large portals, the industry should at least consider to turn to
committee-based standards in the future instead. Such standards could be developed under the
responsibility of a Standards Setting Body, based on consensus and due process, and with all
stakeholders having the chance to participate and to contribute their ideas and needs.

Regarding the role of SMEs in open standards setting – “Standardisation is a prerequisite for a broad
deployment and use of ICT, and will trigger and enable new business.” ((PWC, 2004); see also (Blind,
1999) and (Swann, 2000) for similar accounts). With the creation of new businesses high on the
agenda in Europe, it would be extremely unhelpful if SMEs which, after all, form the employment and
growth engine of the EU, were excluded from shaping this infrastructure upon which they very much
rely.

One possible approach here would be to provide funding for suitable SME umbrella organisations (we
are not even starting to think about the potentially resulting, or at least claimed, distortion of
competition). It would then be their task to identify those standards committees whose work is of
particular relevance to SMEs, and to represent their constituency’s interests there. Yet, in this case two
problem areas need to be addressed.

For one, SME users are not a homogeneous group. Accordingly, something needs to be done about the
problem of diverse and context-specific user requirements (see e.g. (Jakobs et al., 1998). In particular,
there is a need for a mechanism to align these requirements. This should ideally happen prior to the
actual standardisation process. Dedicated ‘SME user groups’ might be an option worth considering,
despite the problems that have to be associated with this approach.

Along similar lines, sector-specific standards may be a way to raise the interest of SMEs to actively
participate in standards setting, as such standards might be closer to their specific business interests.
This would, of course, imply the need for a mechanism to guarantee inter-sector interoperability.
Another, related option would be to deploy the national standards bodies to a greater extent as SME
representatives in the – far more important – international arena. Lower travel budgets and the
prospect of communicating in their native language might be an incentive for more SMEs to
participate in standards setting, and let the national bodies represent them in the international/global
arena. This might also, at least partly, resolve the problem of requirements alignment.


References
Adolphs, B. (1996): Stabile und effiziente Geschäftsbeziehungen – Eine Betrachtung von vertikalen
Koordinationsstrukturen in der deutschen Automobilindustrie, Dissertation, 1997, Köln.
Blind, K. et al. (1999): Economic benefits of standardization (in German). Beuth Publishers, ISBN: 3-410-
15066-, 1999.
Dankbaar, B., v. Tulder, R. (1992): The Influence of Users in Standardization: The Case of MAP. In: Dierkes,
M.; Hoffmann, U. (eds): New Technologies at the Outset – Social Forces in the Shaping of Technological
Innovations. Campus/Westview.
European Commission (eds) (2004): Innovate for a competitive Europe: A new action plan for innovation. http://
europa.eu.int/comm/enterprise/ innovation/consultation/docs/innovate.pdf (accessed April 13, 2005)
Gerst, M., Bunduchi, R. (2004): The Adoption of Standardised Technology in the Automotive Industry, in: e-
Adoption and the Knowledge Economy: Issues, Applications, Case studies, edited by Paul and Miriam
Cunningham, IOS Press, Amsterdam, pp. 287-294.
Gerst, M., Schulze, H. (2003): e-Business in der Umsetzung- das globale DaimlerChrysler Lieferantenportal. In:
Beschaffung Aktuell, Heft 8/2003, pp. 50-51.
Graham, I., Spinardi, G., Williams, R., Webster, J. (1995): “The Dynamics of EDI Standard Development”. In:
Technology Analysis & Strategic Management, 7(1), pp. 3-20.
Jakobs, K.; Procter, R.; Williams, R. (1998): Infrastructural Technologies to Enable Electronic Commerce. Proc.
3rd International Conference on the Management of Networked Enterprises, ICMNE ‘98, Montreal 1998.
Jakobs, K. (2004): (E-Business & ICT) Standardisation and SME Users – Mutually Exclusive? Proc. Multi-
Conference on Business Information Systems, Track 'E-Business - Standardisierung und Integration,
Cuviller Publishers, Göttingen, 2004.
Kumar, K., van Dissel, H.G., (1996): Sustainable collaboration: Managing conflict and co-operation in
interorganisational systems. MIS Quarterly 20 (3), pp. 279–300.
79

Koch, O., Gerst, M. (2003): E-Collaboration-Initiative bei DaimlerChrysler. In: Integrated Supply Management
– Einkauf und Beschaffung: Effizienz steigern, Kosten senken, Hrsg. Prof. Dr. Bogaschewsky, R., pp. 207-
234.
Lamming, R. (1993): Beyond Partnership, Strategies for Innovation and Lean Supply, Prentice Hall,
International (UK).
McKinsey (2003): Study HAWK 2015 – Wissensbasierte Veränderung der automobilen Wertschöpfungskette,
VDA 30 Materialien zur Automobilindustrie.
PricewaterhouseCoopers (2004): Rethinking the European ICT agenda: Ten ICT-breakthroughs for reaching
Lisbon goals. https://www.ictstrategy-eu2004.nl/ pdf/Rethinking_the_European_ICT_agenda_def.pdf
Sadtler, C., Ganci, J., Griffith, K., Hu, D., Marhas, D. (2003): IBM Webshpere Product Overview, Redbook
paper, IBM.
Swann, P.G.M. (2000): The Economics of Standardization. Final Report for DTI.
http://www.dti.gov.uk/strd/economic%20benefits%20of%20standardisation%20-%20EN.pdf. Accessed
June 21, 2004.
Williams, R., Graham, I. and Spinardi, G. (1993): “The Social Shaping of EDI”, Proceedings of the PICT/COST
A4 International Research Workshop, Robin Williams (eds.), European Commission, Edinburgh, pp.1-16.
80

4.3 Challenges in the adoption of RFID standards

Martina Gerst, University of Edinburgh, martina.gerst@ed.ac.uk
Raluca Bunduchi, University of Edinburgh, r.bunduchi@ed.ac.uk

The present needs for cost reductions and higher efficiency throughout the supply chains have led to
the re-emergence of a 50 years old tracking technology – RFID - as one of the most popular “must
have” technologies today. Despite the age of the RFID technology, the current market is immature,
with IT vendors struggling to keep the pace with the spectrum of functionalities demanded by the
customers. One of the major stumbling blocks for the wide adoption of RFID is the lack of global
standards. This paper discusses the challenges surrounding the development and adoption of RFID
standards. The paper argues that the competition between the two global RFID standardisation
bodies reflect the dual nature of the RFID standards themselves with the component manufacturers
driven ISO focusing on generic technology standards, and the end-user driven EPC focusing on data
specific standards. The study finds that the two RFID standardisation efforts are marked by a
notable absence of cross industry end-user involvement. Although end-users recognise the need for
common RFID standards, at this early stage of RFID development they are not prepared to invest in
global standardisation efforts when the current technology does not even address their specific
business requirements. At the same time, they feel that industry specific RFID standard consortia
should play a more significant role, or else their needs will not be addressed. The danger is that
future RFID technologies will be shaped only by the interests of component manufacturers, and
large retailers’ standards through their participation in ISO and respectively the EPC process.


Introduction
Although Radio Frequency Identification - RFID - technology is more than 50 years old, it became
very popular for automatic identification and supply chain applications only during the last couple of
years. This development is in a way reminiscent to the Internet hype and is triggered by several
factors: first, the RFID mandates from large organisations like Wal-Mart and the U.S. Department of
Defence have forced suppliers to adopt RFID technologies and tag their products. Second, consulting
companies and market research firms such as Gartner or Forrester have joined the RFID frenzy and
have predicted huge benefits to adopters (McGinity, 2004).

In recent years, companies have made significant investments in high level planning tools such as for
example advanced planning systems (e.g. SAP APO) in order to optimise their supply chain
management. However, these tools have failed to deliver real-time visibility on process, product and
resources: organisations still cannot see what is going on in their supply chains. The lack of end-to-end
product visibility is crippling the effectiveness and efficiency of supply chain operations. For retailers
and other vertical industries, RFID promises to significantly improve the visibility of their supply
chain. RFID lets companies know exactly which item is in which truck, and when it arrives, in which
warehouse. It also solves problems that bar codes could never do because it eliminates the “line of
sight” requirements and it stores much more information than a conventional bar code (Schindler,
2003).

A 2004 Forrester study of the RFID middleware vendors (Leaver, 2004) unveiled a very immature
RFID market moving at a very fast pace, shadowed by two significant challenges:

(1) RFID mandates have ignited demand for a wide spectrum of functionalities, but the RFID
standardisation process struggles to keep up the pace with the users’ demands.
(2) The vendor market is highly fragmented, dealing with a technology in a very early stage of
development. Despite the vendors PR efforts that reflect the RFID frenzy and their attempts to
jump into the RFID bandwagon, most of current RFID developments are merely trials, pilot-
stage projects, initial participation in standard making councils i.e. EPC or industry specific
consortia such as ATA, and laboratory-based simulations of real life RFID based scenarios
(Jakovljevic, 2004).
81


This paper focuses on the challenges surrounding the development of RFID technologies, in particular
RFID standardisation. The study begins with a very brief overview of the RFID technology, followed
by an overview of the major issues that characterise the RFID market. The rest of the paper is
structured in two parts, one exploring the standard supply side, the standardisation bodies, and the
following describing the standard demand side, the standard adopters. The paper will be concluded by
a discussion of the RFID standardisation process and of the challenges faced by potential RFID
adopters.

The study is part of NO-REST - an EU study to analyse the dynamics of e-business standards and their
impact for the network organisations (http://www.no-rest.org). The empirical research is based on a
qualitative case study methodology. The primary source of data was expert interviewing,
complemented with secondary source data such as documentation, newspaper articles and information
gather from the website of a range of organisational actors operating in the RFID market (vendors,
early RFID users, standard setting bodies, and consultancies). In total, nine in depth, semi-structured
interviews have been done with RFID experts including one representative of the ISO committee on
RFID technology, two representatives of the EPC process representing US and UK, one representative
of the Auto-ID centre which preceded the EPC initiative, two RFID users: a car maker and a
component supplier, an RFID system vendor, and two consultant companies which provide services in
the RFID area.


RFID Technology
RFID is a method of identifying unique items using radio waves (http://www.rfidjournal.com). RFID
technology allows the automatic collection of product, place, time and transaction data quickly and
easily without human intervention. An RFID system includes a reader (or interrogator), a transponder
(or a tag), and their associated antennas. The reader transmits the radio signal, through its antenna, that
the transponder receives via its own antenna. The transponder converse with the reader to verify and
exchange the data. Once the reader receives and eventually verifies the data, it sends it to a computer
for processing and management (http://www.hightechaid.com/stdsupdate/stds_update1.htm). This
process is illustrated in the figure below:


Figure 1: RFID system

RFID readers are automatic locks, fixed or mobile hand held scanners. They are usually connected to a
computer and serve the same purpose as a barcode scanner. The transponder, also called tag, smart
card or smart label, consists of a chip containing a processor and a receiver, and an antenna to
broadcast and receive data via radio frequency. Dependent on the installation size of the antenna and
the air interface protocol, the coverage reaches up to several meters. In contrast with the barcode, the
transponder can be read without direct visibility and is contact-free thus eliminating the “line of sight”
requirement. Additionally, transponders can store more information as the code used to identify
products with RFID (for example the EPC code) is much larger than the conventional bar code
(McKinney and Barraclough, 2004). RFID tags are also safer in terms of staining or abrasion. A huge
advantage of RFID is the parallel data collection: a RFID reader can read up to 200 tags.


82

There are two types of RFID tags available:
- Active tags are powered by an internal battery and are typically read/write, i.e., tag data can be rewritten
and/or modified.
- Passive tags operate without a separate external power source and obtain operating power generated from
the reader. Passive tags are consequently much lighter than active tags, less expensive, and offer a virtually
unlimited operational lifetime.

The antenna connected to the RFID reader activates the RFID tag and transfers data by emitting
wireless pulses (Knospe & Pohl, 2003; http://www.rfidjournal.com;
http://www.hightechaid.com/stdsupdate/stds_update1.htm).

Common applications of RFID technologies include transportation and distribution through tagging
of commercial trucks to monitor access and egress from terminal facilities and rail applications to
improve fleet utilisation, industrial applications such as identification of high-unit-value products
moving through an assembly process, animal identification for example tracking valuable breeding
stock and security and access control to monitor the movement and use of valuable equipment and
personnel (http://www.aimglobal.org/technologies/rfid/common_applications_rfid.asp).

However, despite its obvious advantages over bar code, RFID implementation has not reached the
same momentum such as the Internet 6 years ago. The reasons that explain for such a delay in the wide
spread deployment of RFID technologies are listed below:

1. The RFID market is congested, with a massive amount of diverse players such as chip makers,
transponder manufacturers, system integrators or consultancies, all of whom offer different, and
generally proprietary, products and services. Available systems consist of different frequency
ranges, transfer modes, etc. For a potential customer, it is difficult to acknowledge the distinct
benefits and disadvantages of these different RFID solutions (Jakovljevic, 2004; Leaver, 2004).
2. Currently, RFID technologies cannot offer a so-called “killer application” which is an off the shelf
standard solution. A selection of different RFID systems has to be done by the users depending on
the organisational specific process and technological requirements (Leaver, 2004).
3. Due to the fragmented market (a variety of individual RFID products and services), the total costs
of RFID implementation are not transparent. Apart from the fact that transponder prices range
from 50 Eurocent to 80 Eurocent, the exact price calculation as part of a cost-benefit analysis is
difficult because of the number of unknown variables. A RFID implementation cost analysis has
to take into consideration not only the investment in the transponders and readers, but also the
other cost drivers, such as peripheral systems, software and integration effort. Beyond that fact, the
barcode still is a much cheaper alternative option (Jakovljevic, 2004; McGinity 2004;
www.rfidjournal.com;).
4. The discussion in the media regarding RFID implementations often is driven by high promises in
terms of what can be expected as benefits, that is cost reduction and improved visibility of the
supply chain. If these expectations are not fulfilled, potential customers defect from the “RFID
vision”. There are a number of technological challenges which are still not adequately addressed
by existing RFID products, for example the misread rate or “collision” of tags is a significant
problem in real-world situations, and the diversity of possible interfaces (metal, liquid etc)
requires careful positioning and failsafe testing of tags and readers alike (Jakovljevic, 2004).
Additionally, there are a number of privacy and health issues concerning the wide spread adoption
of RFID (Knospe, & Pohl, 2003).
5. RFID technologies require a huge effort in terms of standardisation. RFID standards are a major
issue in securing the high investments in RFID technology on different levels (e.g. interface
protocol, data structure, etc.). Not only different standards co-exist in parallel, but also different
actors with sometimes divergent interests influence the standardisation life cycle (Jakovljevic,
2004; Leaver, 2004).




83


RFID standardisation bodies
RFID standardisation landscape is characterised by two competing global standardisation initiatives –
ISO and EPC Global – complemented by a plethora of industry specific RFID standardisation efforts
such as the American Trucking Association in the transport industry, the NFC forum in the in
consumer electronics, mobile devices and computer industry or the Automotive Industry Action Group
in the automotive industry. This section will focus on the two global approaches to RFID
standardisation, which shape the future direction of RFID standards and consequently the evolution of
RFID technologies.

In 1995, a joint ISO IEC JTC1 committee –SC31 – was set up to develop standards for automatic
identification techniques. The initial work was based on the earlier work on RFID standards within
CEN and the UCC and EAN 2000 GTag initiative. The members of the SC31 committee are the
representatives of the national standard bodies such as the BSI IST34 committee on bar coding in UK.
They represent either internal consultants within large corporations in particular RFID manufacturers
such as Internet Corporation and Paxar which are part of the US delegation, or Texas instruments from
France (http://usnet03.uc-council.org/SC31/member_directory). Apart from large corporations, SC31
members also include external consultants that represent different smaller companies which cannot
afford to send their own representative to the meetings. Their work on the committees is primarily
voluntary.

RFID ISO standards cover 4 different areas: technology (e.g. ISO 18000 series including the air
interface standards, which are developed within the SC31 committee), data content (e.g. ISO 15418),
conformance and performance (e.g. ISO 18046), and application standards (e.g. ISO 10374). The ISO
standards are defined at a very high level, focusing on the interface rather than on the data which is
transported. As a result, ISO standards are generic, being able to be supported by any system and in
any context, irrespective of the data that is being carried.

In parallel with the ISO standardisation efforts, MIT and UCC together with a number of industrial
partners including Procter & Gamble, Gillette and Wal-Mart set up the Auto-ID consortium in 1999 to
research RFID technologies and standards. The members included end users, primarily from consumer
packaged goods, large retailers and solution providers, including hardware and software providers and
consultants. The Auto-ID members included large retailers such as Wal-Mart or suppliers such as
Gillette (http://archive.epcglobalinc.org/ aboutthecenter_oursponsors.asp). In October 2003, a new
entity was created - EPC Global - a joint venture between UCC and EAN, which took over the
standardisation and commercialisation responsibilities. The AUTO-ID centre continued to focus on
R&D in the area of RFID technologies.

EPC standards describe the tag and the air interface depending on the data being carried. EPC
standards prescribe the physical implementation of the tags and readers, rather then specifying their
generic characteristics. EPC standards are more limited in their scope than ISO standards, for example
where the ISO standards for air interface cover all the frequency range, EPC operates only within the
UHF between 860-930MHz with one standard for 13.56MHz (http://www.infomax-usa.com/rfid.htm).
The figure below describes the main RFID ISO and EPC standards depending on the different
frequency bands.
84

ISO 18000-2 ISO 18000-3 ISO 18000-7 ISO 18000-6 ISO 18000-4
ISO 11784
ISO 11785
ISO 14223
ISO 14443
ISO 15693
EPC HF 1
ISO 18092
EPC UHF 0
EPC UHF 1
135 kHz 13,56 MHz 433 MHz 2,4 GHz
Item
management
Animal
identification
Contactless
chipcards
Near field
communication
EPC
Active
transponder
900 MHz

Figure 2: RFID technology standards and frequency bands
Source: Knospe & Pohl, 2003, pg. 43

In conclusion, whereas EPC Global is a commercially driven initiative dominated by the large end
users, in particular retailers, ISO is driven by RFID manufacturers and follows a generic approach to
standards. In contrast with EPC standards which are heavily retail focused, ISO standards address
generic rather than industry specific requirements. The table below compares the two standard settings
in terms of their organisational characteristics (membership, procedures), and their approaches to
RFID standardisation.

Table 1: ISO vs. EPC
Characteristics ISO EPC
Membership Driven by RFID manufacturers Driven by large users (retailers and their
large suppliers)
Resources Volunteers
internal consultants from large companies
external consultants that represent different
smaller companies + national standard bodies
Full time employees
Academics (MIT)
UCC – is a trade association funded by
industry members (worked on bar codes)
Process “Formal” standard body, characterised by
openness, transparency, due process => slow,
bureaucratic process
“Standard consortia”, driven by the interests
of its members => fast process
Approach High level, generic approach, focusing not on
the data itself, but on how to access it:
the technical building blocks, not the
applications
air interface
high level data access techniques, data
object definitions
Case level identification
Specific, focuses on the data itself:
data carrier, data access and product
mark-up language
similar with the bar code system
(central to EPC is the GTIN)
=> It can identify a specific item
Air interface Cover the entire range of frequencies Only UHV
Chips Bigger, smarter, active chips => more
expensive
Smaller chips => cheap enough to make
economic sense for the package good
industry
85

Whereas ISO can claim that it reflects the global requirements into a legitimate process (equal footing
and consensus based), EPC focuses on speed and emphasises the broad support it receives from the
industry community. The ISO and EPC processes can be seen as complementary. However, for both
EPC supporters and for ISO the need for a single, global standard is impetuous. The benefits of
standardisation would be lost if in different parts of the globe, multinationals would have to invest in
different technologies for RFID.

The adoption of RFID standards
There are four categories of actors involved in the deployment of RFID standards: component
manufacturers, technology vendors, consultants, and end user companies ranging from retailers and
car makers to pharmaceutical companies and governments. The table below summarises the
major concerns of users regarding RFID standardisation.

Table 2: RFID standardisation and RFID users
Standardisation
issues
RFID
component
supplier
RFID system vendor RFID consultants
RFID end user
(auto industry)
Importance Very important Indirectly important,
because of its
relevance to customers
Important Important
Areas of
standardisation
considered relevant
Air frequencies
ICs
Reader
Air frequencies
(indirect relevance, as
vendors focus on
applications, not
hardware)
Tag
Reader
Air frequencies
Applications
Reader
Data
Perception (reasons
why standards are
considered
important)
Significant in
areas where
company is
market leader
Dual approach:
standards and
participation in
standard setting is
important to
collaborate towards a
common accepted
standard to address
customers needs,
- but otherwise work
to impose their
product as a de facto
standard
- as their focus is on
the application side,
the hardware concerns
their partners
There is a lack of
commonly agreed
standards which hampers
RFID market diffusion,
especially for external use
(as currently most RFID
project concern internal
deployment)
Standardisation is crucial
for the deployment of
RFID in inter-
organisational
relationships, but not so
much for the internal use
of RFID
The only
concerns is that
the technology
(embedding the
standards) has to
work (anywhere)
Participation in
standard settings
- Extent
- Reasons
ISO
- depending on
the importance
of application
- exercise
influence in
market areas
where double
development is
expensive
ISO and EPC
- limited involvement,
but visible member
- gain awareness,
study the (potential)
market, influence the
standards
EPC
- active involvement
- awareness
None
86

Importance of RFID standardisation
Standards are considered important by all four categories of actors for the reasons discussed at the
beginning of this paper: the global diffusion of RFID technologies at low costs is not possible in the
absence of common, global standards that prescribe the major components of the RFID system: the
reader, the tag and the air interface. In addition to these generic components, the different actors are
interested in the specific areas of standardisation that directly affect their businesses.

RFID component suppliers are interested in RFID hardware standardisation such as integrated circuits
(ICs) for example used in smart cards because the type of IC defines the purpose of RFID use which
directly affects their products. The area were the RFID chip is implemented defines what information
has to be incorporated onto the chip. Each application translates in different IC characteristics such as
in a different air interface protocol or a different performance (read/write range and rate).

RFID is also widely embraced by software vendors who incorporate RFID into their software to
support a range of supply chain operations for their customers. System vendors expect that RFID will
change the process of inter-organisational collaboration. It is seen as a completely different way to
input data into a business system. To enable this change in collaboration, standards are seen as highly
important. Without standards, the benefits of deploying RFID externally cannot be realised. However,
the interest of the vendors is limited as their businesses consist in providing the RFID applications to
customers, whereas the hardware – which represents the major part of RFID technologies - is provided
by their partners, i.e. component manufacturers. Consequently, their interest in RFID hardware
standardisation (tags, readers and interfaces) is only indirect, to the extent that it affects their
customers’ willingness to adopt their products embedding RFID applications.

Most of the end users implement RFID in pilot projects, focused solely on closed-loop scenarios
12
.
The end user understands standardisation as important to avoid lock in, but then again RFID
applications development by a vendor is expected to be a solution to a specific business problem,
rather than providing a global, standardised solution. For users, what matters is the specificity of RFID
solutions, rather than its global (standardised) character: the RFID solutions provided by the IT
vendors have to fit the users’ specific business needs and has to work as expected. Therefore, currently
end users tend to avoid direct participation in standardisation activity, adopting more a “wait and see”
attitude of keeping track of what is going on generally in the RFID market.

Relevant areas of standardisation
Standardisation interests reflect the area of operations that the actors are involved in, and their
strategic interests. Standardisation of air frequencies and readers is a common concern across most
actors, as it impedes their ability to deploy RFID on a global scale. The reason is that RFID frequency
ranges are generally governed by government legislation and as such may differ in some countries.
Consequently, reaching a consensus is essential to ensure that RFID technologies work on a global
scale. Component suppliers focus on hardware standardisation such as IC as it affects the product it
produces, whereas end users would be interested in standardised tags and OTS13 applications which
do not exist so far.

Actors’ perceptions of RFID standards and their strategies concerning RFID standardisation
The perceptions of RFID standardisation differ across the RFID community. The RFID component
manufacturers address standards only in areas where they position themselves as market leaders,
justifying the importance of standards by customer needs together with economic considerations such
as cost savings and process efficiency. In such areas, they pursue a strategy of active involvement in
the ISO process in order to maximise their input on the future form of technology which they view as
shaped by the ISO standards.


12
Closed-loop scenarios refer to in-house supply chain implementations and use of RFID; for example between
two different business units. Currently, RFID is not used in the context of inter-organisational collaboration, for
example with suppliers who would force end users to concern themselves with the subject of standardisation.
13
Off-the-shelf
87

RFID system vendors pursue a two-directional standardisation strategy. In order to maximise sales,
collaboration in standard setting is seen as important to align interests and eventually agree on a
common standard which has the potential to attract a wider acceptance within the potential RFID user
community. System vendors are well aware of the global nature of supply chains which represent one
of the major applications of the RFID technology. The emergence of global markets and global
sourcing during the last decades has stretched these supply chains over intercontinental distances. The
accumulated demand variety, uncertainty, costs, distances, and time lags on a global scale make it
even more imperative that these long chains be managed efficiently and effectively. The use of RFID
technologies promises to increase the visibility of the supply chain, hence reducing costs and
increasing efficiency in the global supply chain management. However, if several competing RFID
standards exist at the same time, items would have to be re-labelled during their shipment which
would offset all the benefits that RFID are supposed to deliver.

Consequently, collaboration for the development of a commonly agreed upon RFID standard is seen
as the best strategy. However, in the event that no compromise is reached between the vendors and
users involved in committees such as ISO or EPC global, system vendors have the alternative to set a
de facto standard, either by itself (if they have significant market power), or together with other
software vendors. From a software vendors’ point of view, the standardisation focus is rather on the
application side whereas standardisation of hardware components such as readers is left to the
respective manufacturers of these devices.

In what concerns the end user, as already mentioned before, the reasons why standards are considered
to be important are related to RFID technology implemented which embeds standards. For them, the
concern is to avoid lock in on a particular hardware and/or software vendor.

Participation in standard settings
Participation in standard setting bodies reflects the type of standard that the actors are interested in
(see point 2). As component manufacturers focus on generic technology standards, they invest their
resources into ISO rather than EPC. The major reasons for participation are development of generic
standards and cost reductions as lack of involvement would lead to duplication of efforts which is not
only costly but it also increases the danger of higher competition.

System vendors were involved in EPC since the very early stages. Some of them were initially part of
the ISO process as well, however the very slow uptake of RFID standards in the market and the pursue
of proprietary protocols and IP rights led to their defection from the ISO process and their exclusive
focus on EPC global. EPC was seen as counterbalance to the slow and component manufacturers
driven ISO process, which would enable a faster development of RFID standards that would address
their specific interests. Despite their focus on EPC, the system vendors were aware that there is a
strong need for the EPC driven RFID standards to be accepted within a wider, global RFID
community. Such a global recognition can be only obtained if the standards are ratified by ISO, which
is seen as the legitimate producer of global standards.

RFID end users are more in a “wait and see position” in what concerns standardisation and the
standardisation bodies engaged in RFID standardisation. First reason, as already mentioned in the
introduction, RFID still is in its infancy. RFID technologies are not used as a common place
technology to support every day business. Most of current RFID projects are implemented in-house
and as there is no need to integrate external parties, end users simply do not feel the urgency in getting
involved in whatever standardisation activity. As RFID is implemented internally, the standardisation
that enables cross company RFID implementations is not high on the agenda. The major current
concern for end users is that the technology őjust has to workŒ. Consequently, a second reason that
explains the lack of involvement of end users into RFID standardisation initiatives is that such
standardisation is not on end user companies’ agendas. Most of the prospect RFID customers either
are too small to consider themselves as able to influence standardisation and consequently do not
include standardisation in their business strategy. Therefore, no budget is available to participate in
whatsoever standardisation body activity regularly. Until the basic technological issues surrounding
RFID development are not solved, standardisation will be only a remote concern for end users.
88


The extent and the reasons to participate in standardisation bodies depend on the business strategy of
the actors, for example in terms of market positioning. The more a standard is seen as important
related to the market position achieved or to be on target, the stronger is the will to exercise influence
in certain areas in order to stabilise market share and avoid double development and extra cost.

In terms of standardisation engagement and reasons, for example market leading component suppliers
played a strong role in the development of the ISO 14443, the standard for smart cards. This can be
explained based on the interests that component suppliers have in a special market such as the smart
card market segment, because there is a potentially commercial need in this area which a company
wants to exploit. Therefore, the significant involvement of component suppliers in the standardisation
effort to ensure that their vision will shape the future of the standard, and thus of the technology that
embeds the standard.

System vendors in contrast, see their involvement limited to an observer position, although being
regarded as visible in the RFID community. The system vendors focus is more on gaining awareness
and on studying potential markets as well as on influencing standards to a certain extent where
necessary. But system vendors are not forcing their engagement in terms of pushing standardisation on
a higher level of awareness.

Currently, end users do not participate in RFID standardisation. Apart from informing themselves
from the Internet or asking business partners about their experiences, their engagement in
standardisation is fairly visible. This might change when RFID experiences an uptake as recognised
technology and companies are forced to give RFID standardisation a serious thought.

To sum up, one can say that standardisation in RFID is seen important by the actors but there is a
certain lack of urgency for it due to the minimal experience with RFID technology in inter-
organisational situations. However, there is a serious interest of the actors for alignment of interests in
order to get the market acceptance of RFID standards, intended to give RFID as technology of the next
generation an uptake in the real practice world.


Discussion and conclusions
The RFID standards realm includes two distinct and yet overlapping standardisation efforts – the ISO
RFID standardisation and the EPC Global. However, for both EPC supporters and for ISO, there is the
need for a single, global and stringent standard. The benefits coming from standardisation would be
lost if in different parts of the globe, multinationals would have to invest in different technologies for
RFID. RFID standards range from technology standards such as IC and air frequencies to data
standards (EPC) which directly affect organisation business processes. Whereas technology standards
development tend to be driven by IT vendors (Jakobs, 2000), data and business process standards tend
to emerge as a result of direct end user involvement (Markus, 2003). In a similar way, whereas the
former tend to be generic, the later are industry specific (Markus, 2003). The standardisation landscape
of RFID thus reflect this dual nature of the standards themselves with the component manufacturers
driven ISO focusing on high level, generic and technology standards and the end user driven EPC
focusing on data specific standards. ISO standards cut across industries, whereas EPC has a strong
retailing industry focus. The interaction between these two standardisation processes which are both
marked by a notable absence of heavy, cross industries end-user involvement (apart from the large
retailing companies) shape the evolution of the RFID technology.

Today, from a user point of view, the implementation of RFID is in its infancy despite all the
promising announcements of RFID technology vendors and consultancies predicting a boost in sales
figures for RFID technology and related services. Due to market and cost pressures or regulatory
interferences on the part of the public sector, some industries are more advanced than others, e.g. retail
and the automotive sector.

89

Although organisations are well aware of the benefits that RFID provides, there are some questions
that remain still unanswered. As the barcode has gained a key position in the market and it is very well
accepted, potential RFID users are on one hand aware of the benefits RFID can provide, but on the
other hand complain about the lack of a concrete business case. Apart from the question of cost and
the respective investment, the technology still shows weaknesses, e.g. antenna breakdown, small
coverage. Additionally, RFID tags are far more expensive than barcodes. As most of the applications
are only in use to support internal processes, the ROI of RFID in an internal so-called “closed-loop
scenario” is mostly negative.

Moreover, the RFID market/arena is populated by different actors with different interests in and
approaches to the technology. Standardisation and/or the involvement in standard setting bodies
depend on the role an actor takes up in the RFID game. There is definitely a gap between the
development of RFID standards and their implementation. Whereas RFID component suppliers
participate in EPC as well as in ISO, users are only interested whether a RFID solution fits their
business requirement. Additionally, it appears that from the users perspective industry specific
standard consortia should play a more significant role in RFID standardisation, otherwise the industry
faces the danger that their voices will not be heard in the IT vendor dominated RFID standard bodies
such as EPC.


References
Jakovljevic, P. J. (2004): RFID – A new technology set to explode?, TechnologyEvaluation.Com, April 2004,
pp. 12, http://www.technology-
evaluation,com/Research/ResearchHighlights/Scm/2004/04/research_notes/TU_SC_PJ_04_22_04_1.asp,
Access Date: February, 2005;
Knospe, H. & Pohl, H. (2004): RFID Security, Information Security Technical Report, 9(4), pp.39-50;
Leaver, S. (2004): Evaluating RFID Middleware, Forrester, August 2004, p. 21;
Jakobs, K. (2000): User participation in standardization processes - impact, problems and benefits, Vieweg
Publishers.
Markus, M. L., Steinfield, C. W., Wigand, R. T. (2003): The Evolution of Vertical IS Standards: Electronic
Interchange Standards in the US Home Mortgage Industry, Paper presented at the Proceeding of the
Workshop on Standard Making: A Critical Research Frontier for Information Systems, Seattle.
McGinity, M. (2004): Staying Connected, Communications of the ACM, 47(1), pp.15-18;
McKinney, J., Barraclough, G. (2004): Radio Frequency Identification Technology Overview, Pharmatech,
pp.110-112;
Schindler, E. (2003): Location, Location, Location, netWorker, 7(2), pp.11-14;

Web sites:
http://www.aimglobal.org/technologies/rfid/common_applications_rfid.asp
http://www.epcglobalinc.org
http://www.hightechaid.com/stdsupdate/stds_update1.htm
http://www.no-rest.org
www.rfidjournal.com
http://usnet03.uc-council.org/sc31/


90

5 5. . T To oo ol ls s a an nd d t te ec ch hn no ol lo og gi ie es s s st tr re en ng gt th he en ni in ng g
c co om mp pe et ti it ti iv ve en ne es ss s
5.1 Virtual Cost Engineering Studio (V-CES): a new service
for cost engineering professionals

*1
Rajkumar Roy,
*1
Jose Rios,
1
Jeevan Sagoo,
2
Ingo Hussla,
2
Lars Hansen
*
Corresponding authors,
1
Cranfield University, Enterprise Integration (Bldg 53), MK43 0AL, UK,
E-mail: r.roy@cranfield.ac.uk & j.rios@cranfield.ac.uk
2
IZET Innovationszentum Itzehoe, Fraunhofer Str. 3, 25524 Itzehoe, Germany, E-mail: hansen@izet.de


The application of engineering practices and scientific principles to problems of cost estimating,
analysis/assessment, control, planning, scheduling and management, throughout a project life-cycle
is considered to be the basis of Cost Engineering. Cost engineers and cost estimators can be
characterized as professionals with an educational background ranging from medium level
vocational training to university degrees, with different professional experiences prior to their
involvement in cost practices, mainly in manufacturing, design, management or accounting, and
with involvement in other activities apart from cost estimating itself, but also part of the cost
engineering discipline, for instance: risk analysis, value analysis, cost modelling, design to cost, etc.
These professionals are responsible for creating ‘cost estimates’ that in general are used as
fundamental criteria to make business decisions. In fact, cost engineering has become essential for
survival of many companies, and predicting the cost of a product is a challenging task. The
achievement of an estimate requires a lot of experience, data, and knowledge of different techniques
and methodologies. And, as it has been already recognised by companies, professional bodies and
associations, there is a need to train Cost Professionals so they can develop the necessary
competence to succeed in their tasks.
In this context, V-CES aims to develop and deploy a set of virtual services to the Cost Engineering
Community, basically in the form of e-Training. The access to such services will be done via Internet
and a Web browser. This implies the possibility for the cost professionals of exercising a just-in time
training, estimating and consulting directly related to their work. The main purpose of this paper is
to present the identified findings and solutions already adopted in the project.


Introduction
Assuming that both quality and functionality are satisfied, cost is probably the most influential factor
in the outcome of a product or service within many of today’s industries. The determination of a
product/service cost evolves along its life cycle and it demands the creation of different estimates.
Cost engineers and cost estimators are responsible for creating such ‘cost estimates’ that in general are
used as fundamental criteria to make business decisions. As it is widely recognised, predicting cost of
a product is a challenging task. The achievement of an estimate requires a lot of experience, data, and
knowledge of different techniques and methodologies, being ‘expert judgment’ one particular element
to be considered as part of the required experience, and in general it demands the involvement of
several professionals [Rush & Roy, 2001]. Traditionally, designers did not consider cost as a
‘requirement’ with the same importance as functional and quality requirements, quoting from [BSI,
1981]: “A design that fails to meet its cost specification is no better than one that fails to satisfy its performance
requirements. Furthermore. when all other factors are equal. the decision by the customer whether or not to buy
a product is largely determined by its cost’. In fact, designers were not really involved in any cost
estimating activity, their awareness of the cost implications of their design decisions was not explicit,
when requested the cost estimates were performed by other personnel and by the time a cost estimate
was available the design solution had likely changed or had been finally adopted. These issues have
been widely recognised by industry and they are currently being addressed through ‘Cost
Engineering’. Quoting [Greeves & Joumier, 2003]: “The alternative offered by cost engineering is to have
cost information available when design choices are being made. so that they will be made in the knowledge of
91

approximately what the different potential solutions are likely to cost. This awareness of the likely cost is
essential to be able to make effective cost/benefit trade-offs.”

In consequence, cost estimates are important as a decision making criteria, and in that sense they have
to be available ‘Just-in Time’ when the design solutions are evaluated, and they have to provide a level
of accuracy suitable for the stage in which the decisions have to be made. Accuracy and speed in
estimate development are directly linked. In a particular product design stage a higher accuracy may
demand a deeper search and collection of data, and the interaction with a larger number of technology
experts, which in fact may demand more time for the estimate completion. The quantification of the
needed accuracy is quite difficult and it depends heavily on the kind of product [Layer, et al., 2002].
But there are also some other factors like: who was involved in the creation of the estimate, how the
estimate was done, and the data available at the time of creation that basically depends on the product
design stage where the estimate is developed [Roy, 2003] [Roy et al., 2003]. Typical average values
for accuracy range from ±50% in the feasibility study to ±15% in the detailed design stage. In general,
the level of uncertainty in the estimate is compensated with an increase in the contingency part.

In brief, to compete and qualify, companies are increasingly required to improve their quality,
flexibility, product variety, and novelty while consistently reducing their costs. In a collaborative
engineering environment, ‘cost estimates’ are needed at different points in the product life cycle, and
the faster they are available and the more accurate they are the better for the decision process in which
they are used. In this process, human resources are a key element as it is recognised by companies, and
in this direction, in addition to the efforts made by OEMs, professional bodies and associations in the
USA have developed some certification programmes to provide indicators of cost engineering
expertise in different industrial sectors, mainly in construction and process industry. At the European
level, the development of cost engineering competence is being addressed mainly by OEMs (e.g.:
ESA, Airbus, EADS, Rolls-Royce, Ford, Daimler-Chrysler, Fiat, just to name a few), which promote
cost engineering practices among their suppliers to achieve cost-effective performance.

In this context, the V-CES project is aiming to develop and provide a set of services to the Cost
Engineering Community: Cost Engineering Mentor, Training in Cost Engineering, a Cost Engineering
virtual community and Cost Engineering help line, and a Cost Engineering Database (see figure 1).
The access to such services will be done via Internet and a Web browser.

V-CES Team research has identified the requirements, that cost professionals and companies, mainly
OEMs, suppliers and SMEs, have. The objective was to define the framework where all the services
should be deployed. In this research process, several factors are being investigated and evaluated:
o Industrial sectors, company and professionals profiles, paying special attention to the OEM-
supplier collaboration, in the form of practices sharing, and to the SMEs as a distinctive kind
of company. Sectors: aerospace, automotive and electronic equipment manufacturing.
o E-training, which demands defining the process of gaining the knowledge and the skills for
developing cost estimates, supported by the use of ICT. Mentoring and Community
development are relevant elements of the web-based training, which also demands the
consideration of the pedagogical approach to adopt in the process.
o Cost Estimating Mentor, which basically demands the definition of the cost estimating process,
the integration of ‘mentoring’ and its link with the training services.
o Cost database, which implies four different aspects, the identification and collection of data,
the creation and implementation of a particular Database, security issues, and its integration
with the other services.
92

WEB Browser
USER’S MACHINE
Back end servers:
Database (MySQL)
Training &
Community (Moodle)
V-CES Databases
to support e-Training
V-CES
TRAINING
V-CES
COMMUNITY
V-CES COMMUNITY
e-Training by discussion
WEB server
(Apache)
PHP
Scripting
engine
V-CES SERVER
request
request
V-CES Training
e-Training by lecturing
V-CES Mentor
e-Training by doing
V-CES
MENTOR
Front-end:
e-Mentor GUI
Moodle
XML document with cost data
HTML/XML training material
Ms Producer presentation
Virtual community response
Interaction with e-Mentor GUI
INTERNET
Knowledge & skills
E
x
p
e
r
i
e
n
c
e
DATA
COST
ESTIMATE
t
a
r
g
e
t
s
targets
targets
t
a
r
g
e
t
s
t
a
r
g
e
t
s
targets
demands
demands
demands

Figure 1: V-CES services

The network framework in which such services will be deployed demands a hybrid client-server and
peer-to-peer interactions. In this sense, there is a group of network technologies, Web protocols and
development frameworks that have to be considered. In brief, the following sections present the
research already conducted, in particular, the current findings, the problems and challenges noticed in
the project, and the solutions adopted for the deployment of the V-CES services.

Scope of Cost Engineering in V-CES
Probably the first aspect to consider is: What is Cost Engineering? In essence, cost engineering is a
discipline with a wide scope of concern, according to Roy [Roy, 2003]: “cost engineering is
concerned with cost estimation, cost control, business planning and management science, including
problems of project management, planning, scheduling, profitability analysis of engineering projects
and processes”. In particular, V-CES is focused on ‘cost estimating’, which can be considered the first
stage and function of cost engineering, performed at different product life cycle stages, and with the
objective of approximating, in an independent, objective, accurate and reliable way, the true cost of
producing a product.

When considering the elaboration of a cost estimate, there are different techniques or methodologies
that can be used in its development. As it is presented by Roy [Roy, 2003] and in [Layer, et al., 2002],
“where and when” each of the possible techniques can be used, depending on which stage of the
product life cycle the estimate is carried out. For the purpose of V-CES, three main cost estimating
methodologies are considered: bottom-up, analogy based, and parametric. Among the three
methodologies, parametric is clearly distinctive. This methodology is based on the identification and
mathematical formulation of relationships between cost and product parameters, known as cost
estimating relationships. This approach demands the access to historical data and assumes that the
relationship will continue in a similar way in the future. The bottom-up methodology, also known as
detailed and grass roots, demands access to detailed cost data, and the allocation of the cost elements,
mainly: labour, material, and equipment. The analogy-based or case-based requires the identification
of the similarities and differences of items, the identification of a reusable case, and its adaptation if
necessary. The availability and kind of data needed in each case made the parametric approach more
suitable in the early design stages, and the bottom-up and analogy-based more suitable in later design
stages.
93

The creation of a cost estimate for a particular product demands the use or the development of cost
functions, which constitute a cost model. The model can be created using any of the cost estimating
methodologies. Commercial software is available to help cost engineers and estimators to create cost
estimates. However, some of these systems are a sort of black box; the user is trained in the use of the
model, which demands the understanding of the input variables and the interpretation of the final
result. Layer in [Layer, et al., 2002] provides an overview of some commercially available cost
estimating applications. One particular aspect that is evident to the Cost Professional Community is
the need to have a transparent and clear description of the cost model used to create an estimate. This
issue has been explicitly expressed by several professionals and is also made explicit in [Layer, et al.,
2002]. In this sense, the approach adopted in V-CES is process driven by the three methodologies
previously mentioned: detailed, analogy and parametric. The objective is to provide a service (Cost
Estimating Mentor) that guides the user to develop a cost model for any product making use of any of
such methodologies.


Market relevance and findings
The exposure of V-CES in different forums and to companies external to the project has given to the
V-CES consortium a clear confirmation that the services proposed: Mentor, Training, Community and
Database are relevant and interesting for the cost engineering professionals. However, when
considering the commercialisation of a service, a market study is a clear need, and with this idea a first
UK Market Research has been conducted [V-CES UK-MR, 2005]. Based on the feedback received
and on the referred market research, the following main findings can be pointed out:

o Only the potential UK market including SMEs (with more than 50 employees and Internet
access) and OEMs accounts for approximately 3,000 firms.
o Cost engineering methods are clearly used more and more; the ‘design-to-cost’ approach
demands a better understanding of the cost drivers in the design stage as well as the
familiarisation of engineers with cost concepts.
o There is a need to have a transparent description of the cost structure, showing the origin of the
cost drivers, to have a better visibility of the basis upon which a decision is based.
o Cost engineering and management has been identified as one of the most important topics by
companies, and the need for education/training in this area in order to achieve business
competence has been mentioned.
o The e-Learning market is growing globally, in particular the European business skills market
has an estimated value of 15 billion Euros by 2006, and it is estimated that around 30% of this
market will be delivered via e-Learning.
o In terms of e-Learning, the e-Learning market is shifting to be content-led, within a
collaborative environment, with a higher preference for asynchronous communication and with
the use of ‘learn by doing techniques’.
o Cost estimating software is mainly developed to calculate cost estimates for particular kinds of
products, but without providing a ‘step by step’ guidance in how to develop the cost functions.
o Only one cost estimation tool is accessible via web, ‘Cost Estimator’ from HomeTech
Information Systems, Inc., which is suitable to the construction sector.
o In general the license fee per seat ranges from £1,600 to £31,800 per year. One particular
company (Tecolote Research Inc.) allows installing and accessing the software in their own
server, reducing the cost for the client to maintain the application by approximately 15%.
o When looking at the cost engineering training courses content, most of them are related to
general concepts on cost engineering and cost management, in fact there is no company
providing any service online based on any kind of cost estimating methodology. In terms of
price, it ranges from £195 for a 1-day course to £1,750 for a 5-day course. When looking at
online training the prices range from £135 for (2 to 4 hours) to £740 (2 days). One particular
company MindLeaders provides an online training course on ‘Project Management: estimating
and scheduling resources’ 2 hours long, including asynchronous mentoring for £35.
o SMEs demand inexpensive ways of e-training, but they are willing to make use of it, quoting
from [Bridge, 2004]: “More than a million employees from 100.000 small and medium-sized
companies have taken Learn Direct courses since the service was set up in 2000.”
94

o At the European level the training should be provided making use of the language specific to
each particular country or region, this aspect is particularly relevant for SMEs.

Industrial sectors, company and professionals profiles
The industrial sectors addressed in the project are: automotive, aerospace and semi-conductor
equipment manufacturers. From the perspective of Cost Engineering, these three sectors are quite
distinctive, mainly because of the kind of products they produce or manufacture. The project for a new
airplane, car, or equipment, is quite different in terms of complexity, duration, risk and definitively
cost. The particular characteristics of these products, and also the profile and practices of the
companies involved in the project completion play an important role when considering the cost
engineering practices used. In this sense, a few conclusions can be derived: each company has its own
way of implementing cost engineering techniques and processes; cost engineering depends heavily on
the availability of reliable cost data; cost engineering as any other engineering activity demands a
strong commitment of man-effort and skilled cost estimators; and more collaboration between the
Original Equipment Manufacturers (OEM) and the suppliers is needed, but when considered within
the ‘cost engineering process’ it becomes more critical and more complex to achieve. These issues are
the raison d’être of V-CES: there is a need to train employees in cost engineering practices, there is a
need to provide companies with cost or at least ‘price’ data, and there is a need to provide a space for
collaboration in cost engineering practices both in a forum environment and in a peer-to-peer basis.

As it is clearly stated in the V-CES project description, one particular profile of company to be
specially addressed is the ‘SME’. The fact that SMEs need more attention to increase and improve
their technological level is widely recognised by the European Commission, which promotes specific
actions for this kind of companies. In particular, when referring to E-training or E-learning, the picture
is quite similar. As it is presented by [Attwell, 2003] from a research project jointly conducted by
Cedefop, the European Centre for the Development of Vocational Training, and the European
Commission DG Education and Culture, ‘what perhaps emerges most clearly is a somewhat chaotic
picture’. Quoting from his report, the explanation for this ‘chaotic picture’ comes from:
o Many SMEs lack a basic learning culture or organised training plans and infrastructure.
o There is little support from managers for introducing e-learning/training.
o There is no organised and planned action to take up e-learning/training.
o Broadband access seems to be a problem.
The report is extremely relevant, and it provides an insight into the problematic of e-learning/training
in the SMEs context.
In addition to the literature review, the V-CES project has conducted a survey to identify the potential
users’ technical capabilities [V-CES D1.3, 2005]. The survey is based on a questionnaire and in
interviews with experts. A questionnaire was developed to assess the background, cost engineering
activities and computing environment of the Cost Estimating professionals, and it was distributed by
the project partners among different organisations and at various events, e.g.: Manufacturing Day and
EACE Workshop at Cranfield, and Association of Cost Engineering Seminar at Coventry (UK). The
main aspects of the professionals’ profile can be summarized in the following points:
o They range from medium level of vocational training to professionals with a university degree.
o Most of the participants came from a manufacturing background, design and management.
o There is a limited awareness of different cost estimating techniques apart from the ones used in
their daily work.
o Most of the professionals are users of cost models, but they are not involved in cost functions
definition and cost models development.
o In addition to pure cost estimates creation, other activities in the cost engineering discipline are
performed by the participants such as: Risk Analysis, Value Analysis, Cost Modelling, and
Design to Cost.
o Many of the professionals do not have a clear picture of the benefits of sharing practices within
and outside the company. However, professionals from OEMs involved in the evaluation of
bids are the ones with a clearer picture of the benefits.
o Most of them have a User level in IT.
95

E-training and e-Mentoring in V-CES
The reason for using the term ‘training’ in V-CES derives from the definition of the term itself: “the
process of learning the skills needed for a particular job or activity”, being ‘learning: the process of
gaining knowledge through studying’, and being ‘skill: the knowledge and ability that enables a
person to carry out an activity properly’, combining all these definitions and the project context, we
could state that training in V-CES is “the process of gaining the knowledge and the ability needed for
developing cost estimates”. Adapting the definition of e-learning provided by JISC
(http://www.jisc.ac.uk/), e-training can be defined as “training facilitated and supported through the
use of information and communication technology (ICT)”. The e-training term is not widely used,
instead the term Computer-based training (CBT) seems to be the one more accepted, which is together
with: On-the-job training (OJT), games and simulations, and lectures and discussions, one of the four
basic classes of training [Blanchard & Thacker, 2004].

Based on the idea of: ‘available at any time and anywhere an Internet connection exits’; V-CES will
allow implementing a kind of Just-in Time Training (JITT) in Cost Engineering. The idea of JITT
relays on research that shows that a relevant part of the information learnt on a training course is
forgotten by the time the trainee needs it, a solution to this would be to take the training course as
close as possible to the time when the skills are needed, Just-In Time. Since, many of the training
needs could not be predicted with a relatively long time period, the availability of the training courses
should be permanent. This concept was already defined in [Lewis & Romiszowski, 1996].

To implement JITT in Cost Engineering, V-CES considers that the trainees will use computers and a
Web browser to access the service, where courses and training material will be available on-line. To
develop and deploy this service a Virtual Learning Environment (VLE) will be used, after the
evaluation of the applications available, and considering that the cost of a VLE system will have an
important impact on the final cost of the service, an Open Source solution has been selected: Moodle
[V-CES D1.4, 2005]. Moodle will allow managing the access of the trainees to the courses, activities,
and material, assisting in the creation of the content for the courses, facilitating the development of
material to achieve effective training, and assist in the management of the trainee records. It is relevant
to mention that one of the current findings of the project is the lack of availability of on-line material
in Cost Engineering, in particular very few sources have been identified: AACE International
(www.aacei.org), Al Dell’Isola (www.vecourse.com), and West Virginia University (elearn.wvu.edu).
Additionally, SCEA has similar training on a CD-ROM, called COSTProf.

In the case of V-CES the cost estimating process (bottom-up, detailed or parametric) is the kernel of
the training process itself. In particular, considering the basic types of training named by [Blanchard &
Thacker 2004], three kinds of e-Training actions are provided by V-CES (see figure 1):
o E-Training by doing. It will be available through the V-CES Mentor, where the cost estimating
tasks drives the training process.
o E-Training by lectures. It will be available through the Cost Engineering Courses, where
training material relevant to the three main cost estimating techniques will be available.
o E-Training by discussions. It will be available through the V-CES community, where
client-server and peer-to-peer interactions, in the form of synchronous and
asynchronous communications can take place.
Special attention has to be given to the ‘GVTCKPKPI D[ FQKPIŏ, implemented in the form of the Cost
Estimating Mentor, it demands three main actions:
o Definition of a general cost estimating process.
o Definition of a cost estimating processes for each of the methodologies to be used: bottom-up,
parametric and analogy (figure 2), to identify the different tasks to be conducted in each case.
o Identification of practices for such tasks, and of any help needed for the user to carry them out.
In principle, it seems that the definition of cost estimating processes and practices could depend on the
kind of product, on the positioning of the estimate within the product life cycle, on the accounting
96

practices of the company, etc. However, we should wonder ourselves, is there a way of defining a cost
estimating process independently of all these factors? Apparently, it depends on which level of detail
we may want to define the process. Following an activity decomposition practice we could start from a
general level of activities, with the possibility of decomposing each activity into lower detail level
ones. Based on this idea, the cost estimating process steps proposed by the Society of Cost Estimating
and Analysis (SCEA), and the Model proposed by NASA were selected, analysed and used as a
starting point to gather information from companies through an ad hoc questionnaire [V-CES D1.2,
2005]. When looking at processes in companies, it is also important to assess the capabilities of each
particular organisation to capture a picture of the current situation. Based on this principle, the US
Dept. of Defence developed the idea of ‘process maturity’, which has been applied to different
disciplines in the form of ‘Capability Maturity Model’, and in particular to cost engineering recently
by the European Aerospace working group on Cost Engineering (EACE) [Lewis & Pickerin, 2001].
Simplifying such Cost Engineering Capability Improvement Model (CECIM), in V-CES three main
levels have been considered:
[Level 1] Initial Level. In this case the organisation does not have a defined and documented cost
engineering process. The organisation depends on the skills and experience of individuals
for preparing, analysing and validating the cost estimates. The process is not really
repeatable. The software applications used in the process are individually built and used.
Different people may carry out the process in different ways.
[Level 2] Repeatable Level. The organisation has a defined and documented cost engineering
process. It makes use of tailored base practices, policies and standard processes for cost
estimating activities. The company may have implemented software tools to assist them
in the process. The process is repeatable, it can be tracked, and it can be qualitatively
verified.
[Level 3] Defined Level. The organisation has a cost engineering process model, with clear
measurable objectives defined, and information gathering to quantitatively assess the
fulfilment of such objectives. An improvement process is in place to make objective
assessments of new methods and techniques.
The Maturity Levels and a tentative Cost Estimating Process where created and included in a
questionnaire that was responded by the V-CES industrial partners and external companies. As a
result, a general cost process and the three cost estimating methods used in V-CES have been defined,
making the kernel of the Cost Estimating Mentor. Along the process the user will be requested to
make use of the cost database or to input his particular data, a ‘help’ option is provided as well, so the
user can familiarise himself with the concepts to be used.
A1
Create System
Breakdown
Structure
A2
Identify New
Components
A3
Identify &
Collect Prior
System Data
A4
Develop
Cost Factors
A5
Develop NSC
cost improvement
slopes
A6
Develop
Productivity
improvement
factors
A7
Collect all factors
and calculate new
system cost
A41
Develop
complexity
cost factor
A411
Assess
requirement
differences
A412
Assess design
differences
A413
Assess
design
maturity
A414
Assess
Technology
Readiness
A31
Identify data required,
data sources, assess
data availability
A32
Collect prior system
components design data
A33
Collect prior system
components cost data
A34
Normalize prior
system cost data
Optional
Optional

Figure 2: Analogy based cost estimating process
97

The relevance of being ‘process driven’, and the combination of three different kinds of e-training
approaches lead to the pedagogical approach adopted in V-CES, which is an integrated approach based
on the ‘Social constructionist’ (http://moodle.org). Social constructionist is based on four main
concepts:
o Constructivism: people actively construct new knowledge as they interact with their
environment. This idea is embedded in the Cost Estimating Mentor, which demands from the
user to interact with other professionals to accomplish his tasks.
o Constructionism: learning is particularly effective when constructing something for others to
experience, teaching peers. This idea is embedded in the Mentor ‘process driven’ approach.
o Social constructivism: extends the constructivism idea into a social group, where people
collaboratively create shared artefacts with shared meanings. This is implemented in the
Community and in the possibility to share practices.
o Connected and separate: deals with how people behave within a discussion. Connected
behaviour is when an individual tries to listen and ask questions in an effort to understand the
others’ point of view. Separate behaviour is when an individual tries to defend his/her own
ideas using logic to find holes in his/her opponent’s ones. A combination of both behaviours is
called ‘constructed’. This is partially addressed in the Community through the interaction in
forums of discussion.


V-CES cost database
To support the Cost Estimating Mentor, V-CES will provide a cost database. The user of the service
can request data from the V-CES database in different tasks of the cost estimating process, or he/she
may decide to provide his/her own data. The cost database development has been divided in four main
subtasks: data sources identification, data collection, database design and database populating. The
cost data to be provided is related to the following areas: labour, material, equipment, inflation rates
and currency exchange rates, which support the calculations of labour cost, material cost and
equipment cost to be developed when conducting a detailed or bottom-up cost estimate. Following are
the most important findings, challenges and actions in relation with the cost database development:

Labour:
[I] Different issues have arisen in relation with the labour database:
The existence of different names for the same position, which depends on sector,
country, and even companies.
In general, there is no information about the skill level required for the position.
Salaries are negotiated in an individual basis.
Fringe is different depending on the country and company.
Web based salary calculators rely on the number of inputs received by users, and the
number of data inputs used is unknown.
In some countries minimum salaries are agreed at national level for different industrial
sectors, but this practice is not worldwide.
[II] Considering the previous issues, the labour data collection is being conducted based on the
following actions:
Structuring of job positions by Departments.
Contacting European and National Statistic Offices and Inland Revenue Offices.
Use of appointment announcements in Internet, newspapers, magazines, and
employment agencies.
Contacting Human Resources Dept. from companies.
Gathering information from Trade Unions regarding minimum wages.
Material:
[III] Different issues have arisen in relation with the material database:
Different designations of the same material exits, basically depending on the country
and/or standard used.
98

Apparently, only one material company: CORUS provides public cost data related to
steel products. Additional data can be obtained from material resellers.
The relationship between raw material price and pre-elaborated product cost is unknown.
The final price of the material is negotiated between the supplier and the buyer.
New grades of material are permanently in development, over all for plastics.
[IV] Considering the previous issues, the material data collection is being conducted based on the
following actions:
Use of International Standards for material and product designation, which are used in
the definition of the material database structure, e.g.:
BS EN 10020:2000 Definition and classification of grades of steel
BS EN 10079: 1993 Definition of steel products
Direct contact with companies and attendance to industrial trade shows are being used to
collect data. The success of this action depends heavily on the interviewed people.
CORUS and material reseller prices are used as basic for metal products.
The prices provided by the material resellers depend on the size of the material product,
in order to reduce the amount of data to be included in the database, for each particular
kind of material product a single average value is provided (mean) plus a range of
variation (standard deviation).
Price list values are used to populate the cost database, but in order to consider the fact of
a possible negotiation of the final price, during the material cost estimating process a
discount value is requested to the user.
Equipment:
[V] Different issues have arisen in relation with the equipment database:
The final price depends on the machine configuration, and in general for each particular
machine there are a very high number of variants and options, which cannot be
generalised to all the same kind of machines.
There is not an internationally accepted taxonomy/ontology of equipment, and this is a
key aspect when thinking in creating a database.
The machines’ configurations evolve, after a particular time old machines are not
available and new ones are put into the market.
Companies are afraid of possible comparison with competitors, only when a quotation is
requested they are willing to provide precise data.
The operating cost per hour of a piece of equipment depends on where it is used.
[VI] Considering the previous issues, the equipment data collection is being conducted based on
the following actions:
Direct contact with companies and attendance to industrial trade shows.
Classification of machines based on Professional Associations and Best in the World
trade shows catalogues, e.g.:
VDW – German Assoc. of Machine Tools builders, and EMO Hanover.
Creation of ‘typical configurations’ to get a ‘basis price’, these values will be used to
calculate an average price.
In addition to these three basic groups of data, inflation rates are also provided. Mention that the
availability of ‘inflation rate data’ is not similar to all the countries, whereas in some countries like the
UK data area available for different categories of products, in other ones only the ‘consumer inflation
rate’ is publicly available.

Apart from the data collection, the update of the database is the second big challenge of this part of the
project. Due to the different nature of the data included in the database, its update frequency varies; the
policies to consider for this particular task are being defined in a data management report. However,
the approach to be considered when possible is to collect data from publicly available web sites and to
automate the data collection and updating by using specific software for web integration.


V-CES network framework and enabling technologies
One of the key aspects in the V-CES project is the network framework in which the outcomes will be
deployed, which basically demands a hybrid client-server and peer-to-peer interactions. Because of the
99

kind of services V-CES aims to provide, and the kind of interactions expected to happen, there is a
group of network technologies that need to be considered: SOAP, XML Technologies, HTML, Web
protocols, Virtual Learning Environments, and development frameworks like the Java based J2EE and
multi-agent systems middleware like JADE. Figure 3 depicts the technical solution adopted for the
project services. The first prototype is based on: Moodle, MySQL and PHP, and the second one will
incorporate a more advanced solution in a Java environment.

Back end
servers:
Database
Training
Commnunity
V-CES COST DATABASES
Front End: PHPMyAdmin, PHP, SQL
Conceptual Schema: IDEF1X
Physical level: MySQL / KAPOW
V-CES
TRAINING
requests
V-CES
COMMUNITY
V-CES COMMUNITY
Front End: PHP, HTML, Java, XML
Services: e-mail, chat-rooms, forums, wikis,
notice boards
Back-end: Moodle, MySQL / J2EE, JADE
WEB
server
PHP
Scripting
engine
V-CES SERVER MACHINE
requests
HTML
HTML
XML
XML
requests
V-CES TRAINING
Front End: PHP, HTML
Services: site management, user management, course
management, assigment module, chat module, choice
module, forum module, journal module, quiz module,
resource module, survey module, workshop module
Back-end: digital learning objects, Moodle, MySQL /
Digital learning objects
V-CES Mentor
Front End: PHP, HTML
Services: Bottom-up, Analogy based, &
Parametric cost estimation processes.
Based on activities/processes/practices
definition.
Back-end: Moodle, MySQL / J2EE,
JADE, JESS
V-CES
MENTOR
WEB Browser
USER’S MACHINE

Figure 3: V-CES services technical solution


Conclusions
This communication has presented the technical framework for V-CES project, the identified business
opportunity, the intended audience and a description of the main components of the web-based
services: e-Training, e-Mentor, and its associated Cost Database. The need for such services has been
justified, but there are several challenges in the way, some of them being addressed already. Briefly, it
can be concluded that:
o Cost estimating processes are the kernel to the e-Training and e-Mentor services, the current
level of definition achieved in the project of such processes is sufficient in terms of the
requested service from the professionals. However, further analysis is still being conducted in
order to define practices in all the different tasks. This approach is not completely innovative
in terms of the cost estimating solutions commercially available, but it is innovative in terms of
the e-training content currently available. However, what is more important, it complies with
the customer requirements of better understanding and more visibility and transparency in the
creation of the estimates.
o The creation of a cost database is a big challenge, over all when thinking about creating ‘real
cost estimates’. In this sense, the stage of the product life cycle for which the service is
intended is ‘early design stages’, only when the production facility has been selected is
possible to create a kind of ‘real cost estimate’. Moreover, the update of the data structure: new
job positions, materials and equipment must also be considered in addition to the update of the
cost data itself. This leads to the fact that cost databases for ‘real cost estimates’ can only be
developed and maintained by clusters of OEMs and their suppliers, however, for many
organisations it is still interesting to have access to estimated values when estimating in the
early design stages. V-CES aims to increase the awareness of the benefits of sharing cost
databases among the SMEs community.
100

o The security and confidentiality in the creation and storage of the cost estimates and the use of
the V-CES services is an important aspect that still has to be addressed in the project. Even
though currently, each user has his own database, the administrator of the system still has
access to all the information.
o The multilingual aspect is another challenge that has to be addressed, since the success of the
training part, over all for the SMEs in Europe, is very much dependent on the language in
which is deployed.


Acknowledgement
This work is funded by the European Commission through IST Project V-CES: Virtual Cost Engineering Studio
(No.: IST-2003-507041). The authors wish to acknowledge the Commission for their support. We also wish to
acknowledge our gratitude for their contribution to other members of our organisations: IZET, Decision Engineering
Research Team at Cranfield, Centro Ricerche Fiat, Price Systems and DAS.

References
Attwell, G. (2003): The challenge of e-learning in small enterprises: issues for policy and practice in Europe, Report
from Cedefop Panorama Series, Luxemburg: Office for Official Publication of the European Communities.
Blanchard, P. N., Thacker, J. W. (2004): Effective training: Systems, strategies, and practices, 2
nd
Ed, Prentice Hall.
Bridge, R. (2004): Online training, anytime, anywhere, The Sunday Times, October 31.
BSI PD 6470:1981: The management of design for economic production.
Greves, D., Joumier, H. (2003): Cost Engineering for Cost-effective space programmes, ESA bulleting 115,
August.
Layer, A., Ten Brinke, E., Van Houten, F., Kals, H., Haasis, S. (2002): Recent and future trends in cost estimation,
Int. J. Computer Integrated Manufacturing, Vol. 15, No. 6, pp. 499-510.
Lewis, J. H., Romiszowski, A (1996): Networking and the Learning Organisation: Networking issues and scenarios
for the 21st century, Journal of Instructional Science and Technology, Vol. 1, No. 4, November. e-JIST.
Lewis, D., Pickerin, H. (2001): A capability improvement model for cost engineering. A white paper on the EACE
Cost Engineering Capability Improvement Model (CECIM), EACE.
Roy, R. (2003), Cost Engineering: Why, what and how?, Decision Engineering Report Series, Cranfield University,
ISBN 1-861940-96-3.
Roy, R., Souchoroukov, P., Kopponen, H. (2003): Information requirements for cost estimating and the automotive
industry, The 8th Annual ASME Design for Manufacturing Conference, Chicago (USA), September 2-6,
Paper Reference: DETC2003/DFM-48153.
Rush, C. and Roy, R. (2001): Expert judgement in cost estimating: Modelling the reasoning process. Concurrent
Engineering: Research and Applications (CERA) Journal, 9 (4), pp. 271-284.
V-CES Consortium D1.2: Data collection templates and user requirements, January 2005.
V-CES Consortium D1.3: Findings on technological and human resource capabilities of users, January 2005.
V-CES Consortium D1.4: Report on comparative study on the capabilities of commercial software, January 2005.
V-CES Consortium D1.5: Report on functional and technological specification, January 2005.
V-CES Consortium: UK Market Report, created by Devendra Babooa, March 2005.




101

5.2 ICT as an Enabler to the 5-Day Car: A Central
Challenge to the ILIPT Project

Gareth Stone, University of Bath, mnsgms@management.bath.ac.uk
Joe Miemczyk, University of Bath, mnsjrm@management.bath.ac.uk
Bernd Hellingrath, Fraunhofer IML, bernd.hellingrath@iml.fhg.de
Markus Witthaut, Fraunhofer IML, markus.witthaut@iml.fhg.de

Within the ILIPT project ICT is seen as an enabler to the radically new logistic processes necessary to
realise a 5-day-car. The current situation in the ICT systems throughout the automotive supply and
distribution network is characterised by a lack of EDI standards necessary to support the complex
processes. If there are standards a lot of the ICT systems don’t adhere to theses standards making a flexible
and fast integration of the different systems impossible. As a result current ICT solutions show a lack of
integration. Besides the technologies of Enterprise Application Integration work underway under the
heading of interoperability seems to be the most promising approach to come to flexible integrated ICT
systems. Here the ILIPT project has undertaken an assessment of the currently active different
standardisation activities in order to generate an interoperability solution capable of matching the demands
stemming from a 5-day-car concept and on the same time adhere to standards as much as possible. But the
ICT integration won’t be enough to realise a flexible collaboration amongst the companies in an automotive
supply and distribution network. To achieve this, the simultaneous integration on the organisation the
process and finally the system level have to be continued in order to come to a 5-day-car.

Introduction
Today, customers often wait two months or more for delivery of a new car. As we suggest in chapter 2
of this book, this has now been recognised as a major driver for change in the automotive industry.
Here, we will discuss how the ILIPT project focuses on a breakthrough in information and material
logistics by radically reducing time lags, stock piles and unused capacity and the role ILIPT will play
in this dramatic transformation.

ILIPT considers ICT issues as a major enabler of new logistics concepts within a flexible production
network. This is allied with new forms of functional co-operation between the companies involved
along the value chain. Unfortunately, the development of ICT infrastructure and systems applied
within the automotive industry do not fulfil the new requirements.

Let’s take a closer look at the barriers facing the 5 day car system. Figure 1, outlines the main ICT
systems involved in the automotive supply chain from dealer systems through to supplier systems. The
arrows indicate where communication links are required. It is the proliferation of these communication
links (using different protocols and exchange formats) that often cause delay in information
processing.
102


Dealer
Mgt
System
Dealer
Comm’s
System
Dealer
Order
Administrat ion
-Planning
-Parts req’s
Order Bank
Order Extraction
Scheduled
Orders
Order
Sequencing
Build
Sequences
MANUFACTURING
Long term
Forecasting
1
st
Tier
Supplier
2
nd
Tier
Supplier
Supplier
Scheduling
MRP
Operational
scheduling
Electronic call-
off to
Warehouses &
Suppliers
Inventory
Cont rol System
Inventory
Files
Sequenced
supply message
forwarder
EDI
MRP
Parts
req’s
Despatch
Note/
Barcode
Monthly
Weekly
Daily
Sub
daily
MACHINING
EDI / FAX
MRP
Monthly
Weekly
Daily
Sub daily
VM Traffic
Control
Logistics
Resource
Planning
Logistics
O/n
O/n
O/n
O/n
O/n
EDI
Road
Transport
Sea
Transport
Hand held
computer
Mail
Customer Interface,
Website…
Vehicle Manufacturer
Generic IT Systems map of the
automotive supply chain


Figure 1: Generic IT systems map of the automotive supply chain

Legacy systems were originally built for a ‘different world’ of IT development, specific tasks (not
integrated) and where technology was associated with ‘control’. The generic IT Systems map (figure 1
illustrates the ‘stovepipe/chimney mentality’ of Vehicle Manufacturers). IT Systems follow internal
individual functional requirements and are therefore not driven by true customer order fulfilment,
inhibiting smooth information flow. The extent of IT legacy means that the ability of Vehicle
Manufacturers (VMs) to move to a build to order environment is severely limited. VMs are largely
governed by a centralised ‘package’ mentality, built around an inbound logistics optimisation view
rather than outbound, customer delivery. There is some evidence of a growing emphasis being placed
on removing internal stovepipes and increasing system visibility. Where this is the case, however,
change will be slow, we suggested on a timescale of perhaps 5 years.

Batch Processing also represents the major IT Systems barrier to a 5-Day Car. The current
configuration of VM’s systems results in individual mainframe systems updating once a day,
processing batches or ‘buckets’ of orders in a time intensive cycle that adds 4 to 5 days to complete.
This means that there cannot be a real time order interface with the customer. Legacy systems were
originally built for a ‘different world’ of IT development, specific tasks (not integrated) and where
technology was associated with ‘control’. IT Systems follow internal individual functional
requirements and are therefore not driven by true customer order fulfilment, inhibiting smooth
information flow.

Central management systems are popular amongst VMs, because of the ease of maintenance and the
purchasing advantage gained through economies of scale. However the time lags introduced at
regional plant level, where central batch processing cannot allow for local time differences, can result
in higher levels of inventory. Material and plant optimisation drive the VMs’ business process. IT
systems are currently geared towards the purchasing and production aspects of material supply and in-
bound logistics (pull to production) rather than towards flexibility to respond to individual markets
(pull to customer demand).


ICT Systems Down-Stream from the OEM
Throughout the development of this project, we have asked ourselves ‘where are ILIPT's new
processes required, in order to support the existing ICT infrastructure? Previous research shows the
following limitations.
103

• There is a lack of integration between Dealer Management Systems (DMS) and Dealer
Communication Systems (DCS) is causing high levels of hand keying and information duplication.
Dealers operate two distinctly separate systems.
• Although there is good visibility of Distribution Centre stocks, there is a lack of guarantees or
information on delivery times. Poor visibility of orders.
• Order visibility beyond stocks held in VM compounds and Distribution is highly variable. If the
vehicle required by the customer is visible in the stock locator, then response is either instantaneous
or reasonably quick. Once a factory order has been placed, some versions of DCS can provide data
further upstream in production, but feedback can be slow and can take 24 to 48hrs for the
information to be passed to the dealer.
• In the future, more dealers will sell cars of different brands. Currently, for each brand/OEM an own
DMS is needed.
• DMS do not consider the availability of critical parts from 1st and 2nd tiers when confirming the
due date for an ordered car configuration to a customer.


E-procurement as an Up-stream Enabler
VM’s have recognised the power of e-commerce in removing administrative costs from the supply
chain. It is expected that ‘Business to Business’ (B2B) will continue to grow across all markets world-
wide. Whilst the concept of industry exchanges is still in its infancy, there is evidence to suggest that
VMs recognise the need for integrating operating procedures and standards throughout the industry,
unlike their approach to EDI.

In November 1999, Ford and GM both launched their own version of Internet industry exchanges
called AutoXchange and TradeXchange. In February 2000 due to pressure from Visteon and Delphi
over concerns of spiralling IT expenditure for Suppliers, both sites were merged and renamed as
‘Covisint’ to be shared by Ford, GM and Daimler Chrysler. Toyota, Renault and Nissan are also
considering joining.


eCAP as an Example for Capacity Collaboration
The capacity management application “eCap” is part of the B2B supplier platform
“VWGroupSupply.com”, which was developed by the Volkswagen Group and set up with IBM,
ARIBA, i2-Technologies, ebreviate, Hewlett Packard, the Volkswagen subsidiary gedas and the
consulting company AT Kearney.

The eBusiness activities of the Volkswagen Group started in 2000 and soon expanded from a B2B
marketplace to a B2B supplier platform. Now, under the domain "VWGroupSupply.com", the most
important components of collaborative planning and execution have been introduced to all brands and
regions of the Volkswagen Group.

104


Figure 2: Applications of eCAP (http://www.vwgroupsupply.com/)
eCAP supports the flow of demand information from the OEM to its suppliers as well as the capacity
and forecast collaboration between these echelons of the automotive supply chain. In respect to
planning and execution processes, eCap covers the supply planning and the call-offs from the OEM
and the production planning of the first tiers.


Lack of EDI Standards
Electronic Data Interchange (EDI) emerged in the early eighties as a bespoke, dedicated
communication link between two organisations for logistical and technical messaging. Suppliers
perceive the major IT system barrier as a lack of adherence to EDI standards by VMs in terms of
protocol and data format (label layout). Protocol means the language used during transmission: i.e.
OFTP (Odette file transport protocol) and TCPIP (Internet protocol). Data format refers to the visual
interface containing predetermined fields or subsets into which information is entered: i.e., Supplier
name, address, date, material, & quantity.

Currently, EDI format changes are made by VMs up to 3 times a year. Suppliers are already receiving
messages in about a dozen different formats, all of which must be converted to a common standard
before they can be processed internally. This all causes delay and disruption to the system, particularly
in the event of a system malfunction. It has been calculated that each format change costs a supplier
around two ‘IT manager weeks’ worth of labour.

Considerable time is spent by suppliers with IT software consultants who are understandably reluctant
to build and maintain a customised system with diverse inputs from around a dozen demanding
customers. Specialist companies such as ‘Trusted Link Enterprise’ and ‘We Supply’ are popular with
smaller suppliers lacking IT infrastructure, who re-configure the EDI dialect and convert data to a
standard format that can be read directly by the supplier’s systems. Suppliers are currently concerned
about the significant costs of IT system administration caused by the undisciplined approach by VMs
and the implications of adopting new technology on an unregulated, individual supplier/VM basis.


Lack of Adherence to existing Standards
Suppliers perceive the lack of adherence to EDI standards by VMs, in terms of data format and
protocol, to be the major barrier to the current order to delivery process. Industry opinion is divided as
to whether ‘EDI is dead’ or merely whether the arrival of the Internet has rejuvenated existing EDI
formats by allowing them to be posted on the web. Uncertainty also exists over what new technologies
should be adopted, resulting in a rather haphazard uptake of technologies such as Odette, Edifact, Web
enabled PC’s and membership of Internet trade exchanges.
105

There is some concern by suppliers who have adopted web-enabled systems that they do not offer
sufficient reliability or security to conduct transactions between businesses. In the event of a system
failure this cannot be buffered by the low stock levels typically held at most plants today. Furthermore,
there is little understanding of vehicle sales and customer ordering by suppliers. Their end customer is
seen as the VM, not the new car buyer. IT systems reflect this: there is no customer delivery date
attached to any parts ordering.


Lack of System Integration
The lack of system integration means that different Tiers have to connect to the separate systems of
each of their customers. It is unlikely that there will ever be a single system for the whole automotive
industry. Therefore, tiers have to maintain connections to several systems. Today, standards are
lacking in order to support the necessary information exchange for the newest approaches (e.g.
demand capacity management). ILIPT has identified a number of processes which are necessary for
the requirements of the project such as a the need for a Virtual order bank, and capacity utilisation
handling and planning that require the exchange of data (such as capacity offers) for which no
standards currently exist.

Company X
Business
Partner A
Business
Partner B
Business
Partner C
Business
Partner D

Business
Partner A
Business
Partner B
Business
Partner C
Business
Partner D

business partners
that offer a solution
will have one
standard backend
integration
business partners
that offer a solution
will have one
standard backend
integration
Company X will inte-
grate own back end
system with B2B-SCM
solution of Business
Partner A
poor cost/benefit ratio
due to limited number of
transactions
Company X will inte-
grate own back end
system with B2B-SCM
solution of Business
Partner A
poor cost/benefit ratio
due to limited number of
transactions
user needs to log-in, check
situation and effect trans-actions
in the solutions of Business
Partners B, C and D (browser
interface)
inefficient due to high manual
effort in day to day business
user needs to log-in, check
situation and effect trans-actions
in the solutions of Business
Partners B, C and D (browser
interface)
inefficient due to high manual
effort in day to day business
m
a
n
u
a
l

p
r
o
c
e
s
s
m
a
n
u
a
l

p
r
o
c
e
s
s
m
a
n
u
a
l

p
r
o
c
e
s
s
m
a
n
u
a
l

p
r
o
c
e
s
s
m
a
n
u
a
l

p
r
o
c
e
s
s
m
a
n
u
a
l

p
r
o
c
e
s
s

Figure 3: Motivation for a better interoperability (Hofbauer, Alicke, Hellingrath 2004)

EDI or the Web
There is some misunderstanding over what impact the introduction of Internet based communication
will have for traditional, dedicated Intranet links such as EDI. Whilst bespoke EDI and supporting
protocol will become out-dated, the customer interfaces, data formats and information fields can
remain unchanged. This represents a major opportunity for 2nd and 3rd tier suppliers, a high
proportion of whom are not connected to 1st tier suppliers by EDI and rely on the fax, phone or post.
The cost of installing EDI value-added-networks such as the AT&T mailbox system is expensive.
Monthly rental in the 1980s was around £2000 - £3000 and was a significant factor in determining
whether some suppliers could tender for business with new customers who demanded an EDI link.
Suppliers such as Valeo are now beginning to perceive traditional EDI as complicated and costly,
experimenting with others such as Dana, TRW, Delphi and Motorola on a joint study of business-to-
business Internet-enabled technology.

Despite the current Internet experiments, uncertainty exists amongst many European suppliers over
what technology they should be investing in for the future. To date, suppliers have taken a rather
106

haphazard approach to IT system strategy where, according to the manufacturer they are supplying,
they have adopted a mixture of various EDI systems, fax or post. In the future, Web enabled PC’s and
membership of the relevant industry Internet exchange sites will be all that is required. However, this
too will require significantly more inter-VM co-operation than exists at present, in order to achieve
and maintain a universal operating standard. It is estimated that it will take Europe’s motor industry up
to 5 years to share the same level of e-commerce benefits currently enjoyed by American car
manufacturers.

Enterprise Application Integration
Most companies have various heterogeneous enterprise applications running at one or multiple sites.
Furthermore, especially the medium- and large-scale enterprises have experienced deficiencies in the
support of their business processes by their ICT. This disparity has created boundaries for
communicating and sharing information within the whole enterprise. Additionally, many companies
search to extend their value-chains on the Internet. Based on this, Enterprise Application Integration
(EAI) has evolved as an approach to share and/or exchange the information among various enterprise
systems in order to improve the support of the business processes by the ICT. Furthermore, EAI is
necessary in a business-to-business (B2B) model for sharing and integration the respective information
among the various application entities of different departments and/or different enterprises.
The respective EAI solutions have been developed in several stages:

Data-Sharing Approaches
The use of a shared data source (data-level EAI). This approach either
accommodates the data of the different systems in a database or builds
an application for file transfer in order to aggregate the data at a single
location. These centralised data sources can then be accessed by all the
integrated systems. It is suitable as long as the involved applications do
not have complicated business processes.
Application-to-
Application-based EAI
This approach enables interaction among the different applications and
concentrates on interfacing among application logic components that
implement business processes within or across enterprises. This
approach allows the business processes of one application to invoke
the respective functions of other applications.
Broker-Based EAI
The pure A2A integration approach has the disadvantage that the
number of interfaces grows with the number of application to be
integrated. Thus, the broker approach has been created, where a broker
holds the software components for integrating the various enterprise
applications.
Web-Service-Based EAI The web service-based EAI approach has been created to realise
application integration by loosely coupling software components,
which can be dynamically located and integrated on the Web. This
web-service-based approach is of major interest for the interoperability
tasks within ILIPT.


EDI and Web-based EDI
EDI is a collective term for different technologies and standards. For the automotive sector, several
competing standards are of relevance (e.g., Odette, VDA, UN/EDIFACT, ANSI X.12). However, all
these standards have in common that they define electronic data interchange (EDI) as a form of
electronic commerce through the exchange of structured data from one company's computer system to
another company's computer system. Hence these standards define first the type of messages
exchanged between the involved systems (e.g., an order, a delivery schedule, or an invoice) as well as
the format of the message. These format specifications define the fields of the respective message as
well as for each field its size and data type.

The established EDI standards have been designed at a time when communication bandwidth was still
a scarce resource. Therefore, EDI messages have neither been designed to be understood nor to be
generated by users. The exchange of EDI messages between applications of different companies
requires that sender and receiver agree on the standard to be used as well as the interpretation of
certain message fields. The latter is necessary since all EDI standards are not specific enough on the
107

semantics of certain fields. Furthermore, due to the fixed message format it is often not possible to add
the information that is necessary to perform the business processes in a sufficient manner.

Enabling Technology: Tracking and Inventory Management
Inventory management, product identification and vehicle tracking is currently possible through the
use of technology such as bar coding and RFID. In the future, as well as all parts within the product
having an identifier, the finished car could also be given an embedded chip for use during the life of
the vehicle as a replacement for a chassis number.


A single, on-line Logistics Trade Exchange
It is suggested that there is a need for information sharing ‘pool’ between VMs, 1st tiers and Logistics
companies for genuine multi-franchise delivery. All players’ IT Systems would allow limited access
(by contract) to each other, enabling everyone to maximise their transport through combining overall
volumes and utilising cross-docking. This could create a single logistics trading exchange, similar to
Covisint.


Interoperability
Interoperability is “the capability of two or more devices, systems or networks to exchange data
seamlessly and interpret this information within and between firms”. This definition is based on those
provided by the US DoD and EU.

Figure 4: The viewpoints of information systems interoperability (Fileto & Medeiros 2003)

The key findings of the ILIPT state of art analysis on interoperability are:
• From a technological point of view, the W3C standards and especially the existing semantic web
technologies offer the highest potentials to contribute to the interoperability objectives of ILIPT.
• Business processes relevant for interoperability are best captured within RosettaNet, Odette, and
AIAG.
• Nearly all of the investigated commercial products and commercial platforms have the drawback
that they are proprietary solutions.
• Agent-based approaches, e.g., MASCOT, (cf. Sadeh 1996) can provide valuable input for advanced
interoperability features within the ILIPT project.
• Among the on-going EU funded research projects, only Athena appears to cover interoperability
aspects according the objectives of the ILIPT project.
108

Emerging Interoperability Standards: W3C (XML, SOAP, WSDL, UDDI)
The World Wide Web Consortium (W3C) develops interoperable technologies like specifications,
guidelines, software and tools. W3C promotes interoperability by designing and promoting open (non-
proprietary) computer languages and protocols that avoid the market fragmentation of the past. This is
achieved through industry consensus, encouraging an open forum for discussion and the definition of
the Web as the universe of network-accessible information. The following table summarises the key
standards.

XML (eXtensible Markup
Language)
(EXtensible Markup Language) was intentionally created for using
richly structured documents over the web. XML allows anyone to
design their own document format and then create a new document in
that format.
Semantic Web
The Semantic Web is an initiative of the World Wide Web
Consortium (W3C) with the goal of extending the current Web with
semantics to facilitate Web automation and to provide universally
accessible content. The main idea is to enable computer programmes to
manipulate web content meaningfully by encoding semantics
information into web pages. This would allow data to be shared and
reused across application, enterprise, and community boundaries.
Resource Description
Framework
The Resource Description Framework (RDF) can be seen as the
basic idea of the Semantic Web as it is well suited to describe all kinds
of resources in a general way. Another very important idea of the
Semantic Web is the usage of ontologies, which typically consist of a
hierarchical description of important concepts in a domain, along with
descriptions of the properties of each concept. Ontologies will play an
important role in the Semantic Web because they are means to improve
interoperability by providing a source of shared and precisely defined
terms that can be used in meta-data.
Simple Object Access Protocol The Simple Object Access Protocol (SOAP) is a lightweight and
simple XML-based protocol, designed to exchange structured and
typed information on the Web by invoking methods on servers,
services, components and objects. The latest version, SOAP 1.2, has
the status of a W3C recommendation.
Web Service Description
Language (WSDL)
The Web Service Description Language (WSDL) has been published
as a working draft by the W3C and it is expected to have an official
recommendation before the end of year 2003. A WSDL document is
written in XML. The WSDL document describes a Web Service,
specifies the location of the service and the operations (or methods) the
service exposes. The programme sends the results of its process back
to the request originator in another SOAP message. Because HTTP is
one of the transport mechanisms used by SOAP, a Web Service
method call can be made to and from any Web enabled computer
anywhere in the world.
Universal Description,
Discovery and Integration
(UDDI)
The Universal Description, Discovery and Integration (UDDI)
protocol is one of the major building blocks required for successful
Web Services. UDDI creates a standard interoperable platform that
enables companies and applications to quickly, easily, and dynamically
find and use Web Services over the Internet.
Pragmatics of the W3C
standards
XML Technology is already an integral part of today’s applications.
Web Services provide seamless connection of internal and external
services based on universally accepted standards, like SOAP, WSDL,
and UDDI. The interoperability between heterogeneous platforms can
be significantly improved, especially by including Semantic Web
concepts such as standardised Ontologies. The combination of all these
technologies enables uniform interfaces for administration and
maintenance including remote administration and controlling, as well
as dynamic software updates.



109

Assessment of the Emerging Interoperability Standards
XML based applications are nowadays widespread and are often the basis for industry standard
interoperability solutions. Furthermore development and application of web services, also based on
XML technology, are driven by industry. The ideas and technologies related to the Semantic Web
prepare the ground for applications serving the need for semantic inter-enterprise co-operation.

Supply Chain Management, for example, requires a tight integration of many different parties across
the supply chain and might benefit from the Semantic Web integration capabilities. While many
discussions of the Semantic Web focus on emerging standards and what else needs to adhere before
the Semantic Web becomes useful, the fact is that the key standards are already in place. The existing
Semantic Web technologies can be used to build a Semantic Web infrastructure, whereas services,
applications and agents that are able to use this infrastructure are rather rare.

Interoperability at business level should be seen as the organisational and operational ability of an
enterprise to factually co-operate with other, external organisations, whether these organisations are
enterprises or public institutions. Co-operation requires a certain degree of compatibility of business
procedures; in order to understand the business compatibility of two organisations, it is necessary first
to understand each other. To this end, each organisation must achieve a deep and wide understanding
of itself.

Interoperability at knowledge level should be seen as the compatibility of the skills, competencies, and
knowledge assets of an enterprise with those of other, external organisations. This layer addresses the
methods and tools that support the elicitation, gathering, organisation and diffusion of business
knowledge within an enterprise. In business collaborations, two organisations need to verify their
inherent compatibility at the knowledge level, while at the same time preserving their competitive
advantage and confidentiality issues. If their knowledge repositories faithfully mirror their
organisations, it is possible to verify in a preliminary stage their mutual compatibility and, if
necessary, improve it.

Interoperability at ICT Systems level should be seen as the ability of an enterprise’s ICT system to co-
operate with those of other, external organisations. It is concerned with the usage of ICT to provide
interoperation between enterprise resources. Co-operation between humans, machines and software
programmes has to be established by supplying information through inter- and intra-system
communication.

The semantic dimension cuts across the business, knowledge and ICT layers. It is concerned with
capturing and representing the actual meaning of concepts and thus promoting understanding. To
overcome the semantic barrier, which emerges from different interpretations of syntactic descriptions,
precise, computer-process able meaning must be associated with each concept. It has to be ensured
that semantics are exchangeable and based on a common understanding to be indeed a means to
enhance interoperability. This can be achieved using an ontology- and an annotation- formalism for
meaning.


Why ICT is not enough
The flow of material and information, planning and control processes as well as supporting IT
systems, play a central role, on the road to a 5-Day Car, within the ILIPT project. Today, as described
before, the flow of information between partners in a supply chain is frequently limited. Many
disturbances are currently found in planning and execution processes. The flexibility needed in supply
networks to fulfil the demand of a 5-day-car can only be realised by coming to a new level of
collaboration among all partners of the supply network. Therefore three levels have to be taken into
account:
• Organisational Level
On this level the organisational structures of the supplier network have to be designed. This
includes the definition of roles and responsibilities for the partners in the network, giving specific
110

attention to new tasks like collaboration management or planning and control duties for a network
segment.
• Process Level
On this level the basic inter-company collaboration processes for planning the supply network,
executing individual customer orders and enable a sophisticated monitoring and control have to be
designed taking into account the defined organisation. These collaborative processes have to be
built up from standardised building blocks, enabling the rapid and cost-effective built-up and
change of operative networks. Integrated here are defined workflows for the solution development
and disturbance handling between the collaboration partners.
• System Level
The work on this level focuses first onto the design and development of ICT-systems enabling the
collaborative processes across the network. Furthermore the achieved solutions have to be viable
especially for SMEs. In order to achieve this interoperable connection and integration of company
specific planning and control systems throughout the network with these systems has to become
reality. Thus the solutions developed will be structured decentralised, keeping the autonomy of all
companies interacting in the network.


Concluding Remarks on the Approach taken within ILIPT
Taking the example of a virtual order bank (see Figure 5), an integral part of the 5-day car build to
order system, there are a number of implications for new ICT systems. It is currently feasible for ICT
systems to make appropriate checks in the order process, but only in the way cars are ordered today.
Today these checks are possible through a complex system of constraint algorithms that allow an
approximation of whether a given product (in this case a customised vehicle) could be assembled
(from 2000 - 3000 parts) and delivered exactly when required. Today this system does not respond
well to unexpected changes and provides poor delivery reliability.

Figure 5: Virtual order bank

The leap required in ICT capability is the possibility to check actual capacities for critical parts of the
supply chain to allow maximum flexibility in the supplying systems. The current approaches to
capacity management are simply not responsive enough to allow this type of functionality, in
timeframes that customers are willing to accept. One approach is to check the hundreds of suppliers
for each individual order, but this requires new organisational structures (managing capacities), new
processes (to exchange the right information at the right time) and new system solutions that can
include all critical suppliers (from major module suppliers to small SMEs delivering low volume, but
critical items). All these requirements have to build on the current ICT infrastructures that have
developed over the past decades and be able to cope with the variability in the systems that have been
111

implemented. This is the major challenge for the ILIPT project, to make existing ICT infrastructure
enable the 5-day car ideal.


References and Bibliography
Burmeister et al. (2005): State of the Art of Interoperability Standards and Technology, public deliverable D6.1
of the ILIPT project, available through www.ilipt-ip.org
Hofbauer, P. (Ed.), Alicke, K. Hellingrath, B. (2004): Supply Chain Interoperability Recursive Model, ITA
Presentation, July, http://sci.ita-int.org
Holweg, M. (2002): 'The 3DayCar Component Supplier Study' 3DayCar Sponsor Report S4 - 2/02
Holweg, M. (2000): The Order Fulfilment Process in the Automotive Industry – Conclusion of the Current State
Analysis, Sponsor report, S1 – 7/00
Holweg, M, Miemczyk, J., and Williams, G. (2001): How to organise automotive logistics in a build to order
environment, 3DC Sponsor report.SE4 - 5/01 3/02
Howard, M. (2000): IT Barriers report, Sponsor report MT4
EICTA (2004): Interoperability White Paper, European Industry Association (Information Systems,
communication technologies and consumer electronics), Brussels.
Fileto, R., Medeiros, C.B. (2003): A Survey on Information Systems Interoperability, Technical Report IC-03-
030, Instituto de Computação, Universidade Estadual de Campinas, December.
Keller et al. (2005): State of the Art of supporting ICT systems for collaborative Planning and Execution within
Supply Networks, public deliverable D4.2/5.2 of the ILIPT project, 2005. Available through www.ilipt.org
Mercer Managemement Consulting and Fraunhofer Gesellschaft (2004): Future Automotive Industry Structure
(FAST) 2015 – die neue Arbeitsteilung in der Automobilindustrie, Verband der Automobilindustrie e.V.
(eds.), Frankfurt/Main.
Sadeh, N. (1996): MASCOT An Agent Architecture for multi-level mixed initiative supply chain coordination,
Intelligent Coordination and Logistics Laboratory, Carnegie Mellon University, Internal Report, 1996.
Waller, B., Howard, M (2001): IT Systems Requirements for 3DayCar: Organising Responsive Information
Technology Systems for a Build-to-Order Environment, Sponsor Report, MT5
Waller, E., Williams, G. (2000): Downstream Mapping, 3DayCar Sponsor Report. M5
Williams, G. (2002): Production Planning, 3DayCar Sponsor Report. S5
112

About the Authors

Born in 1966, Dario Antonelli is Associate Professor in the research field - Metalworking Systems
and Technologies. He works for the Politecnico di Torino at the Department of Production Systems
and Economics since 1992 after a time period as a researcher for Fiat Research Centre.
His scientific activity is mainly related to the numeric modification of metalworking productive
processes and to the experimental laboratory identification of process parameters. His research is
documented by more than 40 papers on national and international journals and conferences. He also
wrote a book on "Production Systems". He has been visiting professor in the University of Eindhoven
and in the University of Darmstadt. Presently he works in the field of microforming of thin metal foils.
He co-ordinated a research project on this field for FATA Hunter.
Member of AITEM, he is in the co-ordination board of the EU-funded project CO-DESNET.

Arne-Jørgen Berre is chief research scientist at the Interoperable Enterprise Systems Technology
group. He received his Ph.D. from the Norwegian University of Science and Technology in 1993, with
the thesis "An object-oriented framework for systems integration and interoperability". He is Associate
Professor II at the University of Oslo Institute for Informatics since 1996, and has had scientific stays
at Xerox PARC, Tektronix Smalltalk Lab and Oregon Graduate Institute. He has served as technical
manager and scientific co-ordinator for a number of IST projects, including the Eureka Software
Factory Software Bus as early as 1988-1991, COMPASS, OBOE, COMBINE in the EU 5th
framework and ATHENA IP, INTEROP NoE, and the scientific board of the MODELWARE IP in the
EU 6th framework programme. He is involved in Interoperability and Model Driven standardisation in
OMG (UML2.0/EDOC and MDA, Financial domain task force) and OASIS, and has written a number
of scientific publications related to model driven development and systems integration and
interoperability.

Raluca Bunduchi is a Research Fellow in the Research Centre for Social Science at the University of
Edinburgh, funded under the United Kingdom Economic and Social Research Council’s e-Society
Programme. Her current research focuses on the development and adoption of e-business standards
and technologies. She studied economics and computer science at the Babes Bolyai University in Cluj-
Napoca, and obtain her PhD degree in Management at Strathclyde University in Glasgow in 2004. Her
PhD research addressed the strategic use of Internet technologies and their implications for the nature
of business relationships within and across organisational boundaries.

Born in 1978, Irene Cassarino got her university degree in Industrial and Management Engineering
in 2003 at the Polytechnic University of Turin, where she is presently following the doctorate in
Production Systems and Business Economics. Her research interests concern the dynamic and
collaborative forms of inter-organisation among enterprises, such the supply networks, covering
economical, organisational but also epistemological aspects. On these issues she wrote four
contributions at international conferences and two national publications. From 2004 she works in the
co-ordination board of the CO-DESNET (Collaborative DEmand and Supply NETwork) European
Coordination Action.

René Esser is manager for new products at ThyssenKrupp Automotive AG, where he is responsible
for the development of new products across the ThyssenKrupp group. He has over 15 years of
experience in the automotive industry, and has held senior management positions as a design engineer
at INA Waelzlager Schaeffler KG, and as a group leader to support key customers in the area of CAE
analysis for MacNeal Schwendler GmbH, before taking over the management of SDRC´s Ford
programme office with responsibilities for the whole CAD/CAM CAE and PDM implementation at
Ford Motor Company in Central Europe. He then joined ThyssenKrupp Drauz GmbH as director of
the automotive engineering department in Cologne, before taking over his current responsibilities in
2005. René Esser graduated as mechanical engineer from the University of Cologne in 1990.

113

Sylvie Feindt is currently managing director of SFC which she founded in 1997. Her experience
focuses on Ecommerce, virtual organisations, value chain integration and the socio-economic impact
of the Information Society.
After completion of her master’s degree in social science and history at the University in Bonn, she
obtained a degree from the College d’Europe in Bruges. She worked for Deutsche Telekom before
joining the European Commission as an expert in 1993. She worked for the horizontal Unit "General
affairs and co-ordination" of DG Internal Market. After three years in the Commission she changed to
a small German telecommunications company to work on and develop European research projects on
Ecommerce, information and communication systems for European parliaments.

Martina Gerst holds a degree as Diplom-Kauffrau from the University of Saarbruecken, Germany
and the Grande Ecole Institut Commercial de Nancy, France. After starting her career at the central
purchasing department of Siemens, Germany, she joined the KPMG procurement practice in Paris,
France, where she worked in the area of procurement and supply, followed by an engagement such as
Content Director for Commerce One and consultant for the Global Procurement & Supply department
of Daimler Chrysler.
Currently, Martina is a Research Fellow in the Research Centre for Social Science at the University of
Edinburgh. In parallel, she is writing up her PhD thesis studying the development of supplier portals in
the automotive industry. Current research interests include the development and adoption of RFID
technologies and more generally Internet-based technologies, and their outcome on inter-
organisational relationships and business processes.

Marita Gruber works as a Research Assistant at the School of Business, Economics, and Statistics at
the University of Vienna, where she is co-ordinating the project VERITAS (Virtual Enterprises for
Integrated Industrial Solutions IST 2004-511013). Her main tasks in this project are to support the
creation of an Austrian SME-network and to evaluate the efficiency of trust building activities.
Besides her research activities, Marita also gives lectures in Human Resources Management.
She completed her studies of International Business in February 2003, mainly focusing on the topics
of Organisation Theory and Human Resource Management. Her strong interest in the social
interaction of individuals and groups led her to the decision to work in the field of Human Resources,
starting in the year 2000. She worked for Competence Call Centre AG in the area of recruiting and
development and later for T-Systems Austria GesmbH where she was responsible for personnel
development of 800 people, including trainings and workshops.
Marita Gruber lived and worked temporarily in Switzerland, France and Chile and is fluent in German,
English and Spanish. She is currently participating in a Ph.D. programme to be completed in the year
2006.

Bernd Hellingrath studied computer science and mathematics at the University of Dortmund. Since
1988 he is employed as a research assistant at the Fraunhofer Institut für Materialfluss und Logistik
IML (Institute for Material Flow and Logistics) at Dortmund, Germany. Since 1995 Bernd is head of
the Department Enterprise Modelling. Besides the direction and execution of numerous research and
industry projects in the area of Supply Chain Management within several business sectors he is a
lecturer at the University of Dortmund. His PhD study concerns the decision support for the order
sequence planning.

Kai Jakobs has been with Aachen University's of (RWTH) Computer Science Department since 1985.
His current research interests include various aspects of IT standards and standardisation processes. In
addition, he has been working on a number of projects on various aspects of information networks.
Kai is (co)-author/editor of a text book on data communication and, more recently, three books on
standardisation processes in IT. He has been on the programme committees of numerous conferences,
and has also served as an external expert on evaluation panels of various European R&D programmes,
on both technical and socio- economic issues. He holds a PhD in Computer Science from the
University of Edinburgh.


114

Joe Miemczyk is a research fellow in the Innovative Manufacturing Research Centre based at the
University of Bath, School of Management. His main research activities involve exploring the
logistical and environmental implications of automotive build-to-order in the 3DayCar Programme.
Joe has wide academic experience first studying environmental science, setting up environmental
programmes in Poland and then assessing business environmental impacts at Masters level. The
Unipart Group of Companies was Joe's first chance to apply these skills in industry, where he designed
and implemented environmental management systems for automotive component manufacturing. This
led to facilitating policy deployment strategies to integrate planning across business functions.
Within the 3DayCar Programme, which started in 1999, Joe has undertaken much of the logistics and
environment research. As a result he has published widely in the area of automotive logistics and
presented at a number of conferences in operations management, logistics and environmental
management. His PhD study examines collaborative capabilities and the impacts on reverse logistics
and environmental.

Bruno Mussini graduated from the University of Bologna with a degree in Electronic Engineering in
1984. He is director and co-founder of Joinet S.p.A., where he has performed the role of Manager of
Service Development since the company's founding. At the beginning of his career he filled technical
positions in manufacturing companies in the mechanical and automation sectors. Subsequently, he
dedicated his professional services to data processing engineering companies, dealing with the design
of information systems for logistics, production management and quality control. For three years,
beginning in 1996 he managed international research projects destined to the development of e-
business applications designed for the creation and management of enterprise networks at the Modena
DemoCenter. Before participating in the foundation of Joinet SpA, he worked for about a year for the
KPMG Consulting Group filling the position of practice manager in the group dedicated to the
development of Information Technology projects in the context of supply chain management.

Matthias Nöster is a project manager for the German management consulting company SFC, a
business intelligence provider in the areas of eBusiness, networked organisations, entrepreneurship
and SMEs. For several years SFC has been strongly involved in socio-economic research and Matthias
has contributed to the development and implementation of European projects such as ASP-NET
(Growth), JANUS (IST) and VERITAS (IST-NMP) on behalf of SFC. As an economist with a
particular interest in technology and technology companies Matthias carries out surveys and
qualitative analyses within projects and derives strategic recommendations for industry and policy. He
holds an Austrian degree in pure economics and an MBA certificate in IT and Telecom Management.
He is currently doing a Master of Science in Manufacturing: Technology and Management at Open
University, UK.

Markus Rabe, born in 1961, studied physics at the University of Konstanz and made his PhD in
engineering (Dr.-Ing.) at the Technical University of Berlin. Since 1986 he is with Fraunhofer IPK in
Berlin. His main interests are factory planning, business processes, logistics engineering and
simulation. He is head of the department ‘enterprise processes and logistics’. Markus Rabe is active in
the German Simulation Society ASIM, has been member in several conference programme
committees, chair of ASIM work groups, chair of the conference „Simulation in Production and
Logistics“ in 1998, 2000 and 2004 and member of the “model building process” guideline group of the
German engineers’ association VDI. He published several books and more than 100 journal and
conference papers.

Jose Rios, Research Fellow in Cranfield University since September 2003, with a background in
manufacturing and information modelling. He obtained his PhD degree from the Polytechnic
University of Madrid in Spain, where he gained the University PhD Award in 1997. He obtained a
lecturer position in the Mechanical Engineering and Manufacturing Dept. in that University in 1998.
He has been involved in different projects related to manufacturing, information modelling, KBE, cost
engineering and product requirements, mainly with companies from the aeronautical sector, and from
1997 to 1999 he was the Spanish representative in the ISO TC 184/SC4.

115

Rajkumar Roy is Senior Lecturer in Cranfield University, with a background in manufacturing
engineering and artificial intelligence. He started his professional career in manufacturing industry
back in 1987, and worked in the area of knowledge engineering, decision support and shop floor
implementation of expert systems. His research projects, in UK and European level, have a strong
focus on industrial applications. He is currently leading the research in Decision Engineering area at
Cranfield. The research theme includes Engineering Cost Estimating, Design Optimisation and Micro-
Knowledge Management. He is a Chartered Engineer, and member of international organisations such
as IEEE, IED, and ACostE and has co-authored four books in his area of research.

Rainer Ruggaber holds a PhD and a diploma in computer science from the University of Karlsruhe
(Germany). He works for SAP Research in the Research Programme Enterprise Services and
Semantics. Before that he was a research assistant at the Networking Research Group at University of
Karlsruhe (TH) and involved in national research projects. Currently he is involved in the technical
coordination of the FP6 IST Integrated Project ATHENA. His experiences are in middleware
technologies, Web Services and mobile infrastructures. Rainer has been involved in several R&D
projects on Middleware and Interoperability.

Gareth Stone is a research Officer at in the Innovative Manufacturing Research Centre based at the
University of Bath, School of Management. His main research areas are associated with the
exploration of logistical and policy implications of automotive build-to-order concepts in Europe.
Gareth has a broad ranging industrial background gained in the Aerospace Industry over a period of
fifteen years. He has lectured in Operations Management at Southampton Business School and has
most recently undertaken a 3 year research programme on the UK Lean Aerospace Initiative (UKLAI)
which is partnered with MIT and focused on the implementation of lean processes within the UK
Aerospace Industry.

Born in 1949, Agostino Villa has been nominated Full Professor in 1990 and, since 1991 he teaches
Production Planning and Control at Polytechnic University of Torino.
Research lines developed by A. Villa mainly refer to modelling of large-scale distributed
manufacturing systems and material handling networks and supply chains, to developing methods and
tools for organising and managing industrial innovation processes, and to developing procedures for
organising human resources in industrial systems and public services.
In these areas, A. Villa published over 170 papers in international journals and conference proceedings
and 7 books with international and national publishing companies.
He is currently President of IFPR, Vice-Chairman of the Technical Committee MIM of IFAC, member
of the Working Group 5.7 (Integrated Production Management) of IFIP.

Markus Witthaut studied computer science at the University of Dortmund. Since 1992 he is
employed as a research assistant at the Fraunhofer Institut für Materialfluss und Logistik IML and is
working in the Enterprise Modelling Department. His main research areas are information and
communication technology related aspects of Supply Chain Management. Besides the direction and
execution of research and industry projects within several business sectors he is lecturing on Software
Engineering at the Applied University of Dortmund.

116
Index of Keywords (pages)

Aerospace, 59, 60, 97, 116
APS systems, 11, 14
ASP, 9, 11, 14, 22, 23, 24, 35, 115
ATHENA, 7, 58, 59, 60, 61, 62, 66, 67, 68, 69, 70,
113, 116
Automotive, 3, 7, 8, 9, 10, 12, 13, 26, 27, 29, 30,
31, 32, 33, 34, 35, 36, 37, 58, 59, 60, 61, 71, 72,
73, 79, 84, 89, 92, 95, 101, 102, 103, 105, 106,
107, 112, 113, 114, 115, 116

Batch Processing, 103
Build-to-order, 8, 27, 30, 35, 36, 103, 111, 112

Cluster, 3, 4, 7, 8, 38, 41, 43, 45, 47, 48, 100
Co-DESNet, 7
Collaborative
product design, 70
Complexity, 30, 31
Components, 64
Construction requirements, 31
Cost
database, 92
engineering, 94
Covisint, 10, 72, 73, 74, 75, 77, 108

Distributed modelling, 15, 16

E-Business Standardisation, 71
eCAP, 104, 105
EDI, 79, 80, 102, 104, 105, 106, 107
Enterprise
application integration, 102, 107
resource planning, 11, 14, 21, 23, 24, 35
EPC, 81, 82, 84, 85, 86, 88, 89, 90
E-Training, 96
European Research Area, 3

Factual management, 44

ILIPT, 7, 27, 30, 102, 106, 107, 108, 110, 111, 112
Industrial
district, 57
scenario, 59
Interoperability, 9, 10, 13, 58, 61, 62, 66, 67, 68,
70, 108, 109, 110, 112, 113, 116
framework, 68
ISO, 56, 76, 81, 82, 84, 85, 86, 87, 88, 89, 90, 115

Manufacturing
automation protocol, 75
Manzano District, 48
Model-driven development, 62

Network-Management, 42
NO-REST, 7, 82

OEM, 10, 12, 27, 33, 38, 60, 61, 69, 71, 73, 77, 92,
95, 103, 104, 105
Ontology, 68, 69, 70

Performance Indicators, 24, 55, 56
Personnel management, 44
Product portfolio management, 60, 61

Radio Frequency Identification (RFID), 3, 8, 9, 35,
81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 108, 114
standards, 3, 81, 82, 84, 85, 86, 87, 88, 89, 90
technology, 10, 81, 82, 83, 85, 88, 89
Reference model, 15, 16, 25, 62, 63, 65, 66

SME network, 3, 7, 8, 12, 13, 38
SPIDER-WIN, 7, 14, 17, 20, 22, 23, 24, 25
Suppliers
tier-1, 20, 22
tier-2, 21
Supply chain, 8, 9, 13
integration, 8
management, 14

Trust, 42, 43, 44, 45, 46

V-CES, 7, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100,
101
Vehicle manufacturers, 35
VERITAS, 4, 7, 38, 39, 41, 45, 114, 115
Virtual organisation, 7, 13, 38, 39, 40, 41, 42, 46,
58, 114

Web-based services, 3
117



European Commission
Strengthening competitiveness through production networks — A perspective from European ICT
Luxembourg: Office for Official Publications of the European Communities
2005 — 117 pp. — 21 x 29.7 cm
ISBN 92-894-9924-9
research projects in the field of ‘Enterprise Networking’