Supporting Collaborative Product Development through Automated
Interpretation of Artifacts
Matheus Hauder
1
, Sascha Roth
1
, Florian Matthes
1
, Armin Lau
2
and Heiko Matheis
2
1
Lehrstuhl f
¨
ur Informatik 19 (sebis), Technische Universit
¨
at M
¨
unchen, Boltzmannstr. 3, 85748 Garching, Germany
2
DITF Denkendorf, Centre for Management Research, Koerschtalstr. 26, 73770 Denkendorf, Germany
Keywords:
Product Development, Collaboration Model, Business Modeling, Business Process Coordination.
Abstract:
Small and medium-sized enterprises collaborate with partners to develop highly innovative and knowledge-
intensive products. Collaboration models describe in detail how this goal can be achieved. These models
can be supported by an implemented and executable version thereof. This paper describes an implementation
of a collaboration model from the European research project SmartNets using an automated interpretation of
artifacts. We explain the collaboration model, describe the implementation and evaluate our proposed solution
by means of an example from one of the projects industrial networks.
1 INTRODUCTION
Increasing complexity of products and processes,
condensed product life-cycles, quickly changing mar-
ket requirements and globalized competition are ma-
jor challenges that companies, and in particular Eu-
ropean small and medium-sized enterprises (SMEs),
have to face. The key to mastering these challenges
is an effective and efficient development, production
and marketing of new products, processes, and ser-
vices. For many SMEs, due to their limited avail-
ability of resources and their concentration on core
competences and niche markets, it is very difficult to
elaborate such innovations from the idea to the final
product completely on their own. Therefore, more
and more, SMEs collaborate with trusted partners in
loosely coupled development networks to combine
their knowledge and resources and to share the as-
sociated risks (Mazzarol and Reboud, 2008). Even
though there are several examples of success, the lack
of appropriate methods and tools to support conjoint
innovation within networks is still hampering the ef-
fectivity and efficiency of such collaboration (Dooley
and O’Sullivan, 2007). The objective of the European
research project SmartNet is to develop and to evalu-
ate methods and tools supporting collaborative devel-
opment and production networks in the product de-
velopment, in network coordination and in the trans-
formation from development to production
1
. For that,
1
https://www.smart-nets.eu/, last accessed Dec. 6, 2012
SmartNet follows the concept of the smart organiza-
tion (Filos, 2006).
Social software solutions as information and com-
munication technologies could play a crucial role as
enabler for collaboration in future. They support the
conjoint documentation, exchange and elaboration of
information, and with the help of appropriate ICT sys-
tems, process models can be implemented and en-
acted, promoting and supporting the use of these pro-
cesses in the daily work. Although social software
provides manifold possibilities for a better support of
these challenges, current research is mainly focused
on the use of social software for advertising and mar-
keting purposes (van Osch and Coursaris, 2013). We
argue that social software can be used to endorse a
better support for collaborative product development.
In the following, the SmartNet Collaboration
Model, a model for collaboration in development
and production networks developed in the course of
SmartNet will be introduced. It will be demonstrated
how the model has been implemented on the projects
collaboration platform Tricia. Special attention will
be given to the SmartNet Navigator, a tool to assess
the process status of a development project. This pro-
cesses status is automatically computed with an own
substitution language that is utilized to present the
current progress of the involved artifacts in an inter-
active visualization. Its application in collaborative
product development will be evaluated with an exam-
ple from a project dealing with the conjoint develop-
ment of an innovative motorcycle helmet.
151
Hauder M., Roth S., Matthes F., Lau A. and Matheis H.
Supporting Collaborative Product Development through Automated Interpretation of Artifacts.
DOI: 10.5220/0004774901510156
In Proceedings of the Third International Symposium on Business Modeling and Software Design (BMSD 2013), pages 151-156
ISBN: 978-989-8565-56-3
Copyright
c
 2013 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

!
Figure 1: Reference process - key element of the SmartNet Collaboration Model (Lau et al., 2012).
2 SMARTNET COLLABORATION
MODEL
In the operative planning, execution and control of
collaborative innovation activities, there are several
questions, which will continuously come up through-
out the development. Six generalized key questions
have been identified which will emerge in one or an-
other form (Lau et al., 2009):
• What do I have to do, and when shall I do it?
• How can it be done?
• Who can do it?
• How can I do what, and when?
• What shall I do when and with whom?
• How can I collaborate with whom?
The SmartNet Collaboration Model provides answers
to these questions by connecting three key elements of
collaborative development: process model for devel-
opment and production, internal and external actors,
innovation management methods and techniques (Lau
et al., 2009). Each element is described in detail with
numerous instances, e.g. with a list of roughly eighty
actual methods and forty-eight process activities. Fig-
ure 1 shows the complete reference process, to which
we will refer later on when discussing the particular
part of the SmartNet Navigator.
The relations between these three elements will be
attributed with values indicating support potential, i.e.
for which step in the process, which kind of actor is
best suited to be involved (Lau et al., 2012). A ded-
icated processing model is provided which explains
how these structures can be applied in the practice of
collaborative networks (Lau, 2012).
2.1 Implementation of the Model on a
Collaboration Platform
We implemented the above-described collaborative
process utilizing the Hybrid Wiki extension (cf.
e.g. (Matthes et al., 2011)) of the Enterprise 2.0 plat-
form Tricia. Hybrid Wikis are based on a non-rigid
typed system. They allow users to alter respective
models at runtime by modifying the contained data
on which it is based. Attributes and entities can be
added or removed by end-users in such an intuitive
manner that the actual process of modeling, steps into
the background and users are first and foremost con-
cerned about their data (data first, schema second).
2.1.1 Tricia - Hybrid Wiki for Enterprise
Collaboration
With Hybrid Wikis, collaborative work that is com-
monly based on a model in particular can profit from
the lightweight modeling approach. Hybrid Wikis do
not require a special syntax for modeling or contribut-
ing data, i.e. everyone may add and manage infor-
mation/knowledge, which is commonly referred to as
crowd sourcing. Besides the typical unstructured con-
tent, a Hybrid Wiki page also includes structured con-
tent consisting of key value pairs (e.g. management
Third International Symposium on Business Modeling and Software Design
152

activity type and process phase in Figure 2). By de-
fault, an attribute added by an end-user is of the type
string.
The type can also be assigned to a value explic-
itly, which can be regarded as the schema definition.
A small icon appears to indicate that this is a type de-
fined by a so-called type definition. The actual type
assignment of a wiki page is as simple as adding ordi-
nary tags to a web 2.0 page. By assigning such a Type
Tag, the type definition refers to this concept (here:
activity type). In such a definition, each attribute may
be assigned a type, i.e. string, number, date, enumer-
ation, or a link to another page. In the latter case, also
the type of the target page can be specified. With these
type definitions, Hybrid Wikis cover the range from
not at all typed to strictly typed (mandatory attributes
with cardinalities) meta-models as in common rela-
tional database systems. Once defined, the type defi-
nition is applied to all instances of this type of pages.
!
!
Figure 2: Attributes of the activity type.
2.1.2 Elements of the Collaboration Platform on
Tricia
For the implementation of the SmartNet Collabora-
tion Model the extended functionalities of Hybrid
Wikis are essential. These above described func-
tionalities enable a contextualization of simple pages
to combine textual descriptions of process activities,
methods and actors and relations between these enti-
ties. The implementation of the collaboration model
was started with the reference process in Figure 1 be-
cause of its guideline characteristic. Every single ac-
tivity that is mentioned in this process is implemented
as a page of a special type, which is called activity
type. All activity types consist at least of an explana-
tion on what this activity is about and its position in
the reference process. The position in the reference
process is described by the process phase to which an
activity belongs and by its management activity type
from other activity types to identify possible previous
steps in the process. Accordingly, reverse relations
that can be found under Incoming Links in Figure 2
point out potential following steps. The storage of
that information enables the implementation of a kind
of workflow to suggest possible next steps on the ba-
sis which activity types were already finished. This
kind of structuring allows adapting the reference pro-
cess very flexibly to specific needs of different fields
of application. Due to the fact that these activity types
are one of the key elements of the collaboration model
these adaptations affect all other derived tools like the
SmartNet Navigator, which is deeply related to the
reference process. The second element of the col-
laboration model, which is already implemented on
Tricia, is a methodology to support the development
and the collaboration in the networks. In a first step,
around eighty methods - suitable for the application
in enterprise networks and across companys borders
- were identified and described on the project plat-
form. These descriptions consist of basic information
about the methods with strengths and weaknesses and
application guidelines to support the project partner
in implementing the methods into their development
projects. Also actors are implemented in a similar ap-
proach as type tag with various attributes, identifying
mostly their positioning in the supply chain. By this,
it will be possible to extract information about actors
from models of the network topology of the project
networks. Finally, the relations between the key el-
ements of the collaboration model were built. For
that, more information is required. For example, to
describe methods sufficiently, it is necessary to iden-
tify:
• Which results can be reached by applying the
methods,
• The activity types for which the methods are suit-
able, and
• Their contribution to knowledge management, in-
cluding required and created knowledge.
All this information was documented using the hy-
brid attributes of the Tricia system either by building
relations to other information (e.g. for the different
degrees of suitability), using prepared enumerations
(e.g. to categorize the method type and the build-
Supporting Collaborative Product Development through Automated Interpretation of Artifacts
153

Figure 3: Meta-model of the described SmartNet collaboration model.
ing blocks of knowledge management (Probst et al.,
1998)) or normal text (e.g. to specify input and output
knowledge). By using all this data it will be possible
to answer the six key questions presented in Section 2
in future.
2.1.3 SmartNet Navigator - Tool for Automated
Status Analysis
The automated status analysis is performed using a
substitution language developed within the Hybrid
Wiki project that can be used to query the data (at-
tributes) of the collaboration model. Their compu-
tation is based upon a meta-model that is shown in
Figure 3. The collaboration model consists of sev-
eral meetings that are part of a development project.
Meetings can contain activity types that might require
preceding activities as input. They can also be related
with a management activity type, e.g. project man-
agement, network management, and a development
phase, e.g., production and marketing.
An example instance for an activity type is shown
in Figure 2. Result pages are used to document the
findings from different activity types in a develop-
ment project. Activity types also contain tasks that are
assigned to respectively by users. Every task contains
a starting date and a deadline until the task has to be
finished. The progress of the status is tracked with
an attribute task status. In addition, a development
project can have several users who are responsible for
the project.
Figure 4 shows an example for the computation of
states for the process phase. Variable allStati is com-
puted containing all states from a particular process
phase by selecting all activitiy types from this phase.
The state of an activity is retrieved with a function
called statusOfActivityType. In the second variable
valid combinations for the visualizations of the result-
ing state are stored. Only in case both values are fi-
nalized (or open) the aggregated value is finalized (or
in-progress). In case the variable allStati is empty, the
process phase is regarded as finalized. Alternatively,
the variable is not empty and a higher-order function
is used to compute the aggregated value of all con-
taining activities.
3 PRACTICAL APPLICATION
The SmartNet Navigator is applied in all three indus-
trial networks of the SmartNet Project. From the be-
ginning of the project all partners started document-
ing their development activities such as meetings and
tasks by using the Tricia platform. Using the possibil-
ities of the hybrid attributes, provided by the platform,
enables the industrial partners to extend the basic in-
formation with additional data as for instance start-
ing date and deadline for tasks and participants and
location for meetings. In addition to this very con-
tent specific information they related all their devel-
opment activities on the one hand to certain develop-
ment projects and on the other hand to activity types.
The following chapters give an exemplary view of
the situation in one of the industrial networks with a
short description of the network and the targeted prod-
uct and the realization of the SmartNet Navigator.
Third International Symposium on Business Modeling and Software Design
154

!
statusOfProcessPhase
1: let allStati = find("activity type", "Process phase", ph)
2: .select(?(at) (this.statusOfActivityType(at))) in
3:
4: let combine = ?(s1,s2) (
5: s1.equals("finalized").and(s2.equals("finalized")) ? "finalized"
6: : s1.equals("open").and(s2.equals("open")) ? "open" : "in-progress") in
7:
8: allStati.isEmpty() ? "finalized" :
9: allStati.aggregate(combine, allStati.first())
statusOfActivityType
10: let meetingStatus = this.statusOfMeetings(at) in
11: let taskStatus = this.statusOfTasks(at) in
12:
13: this.noResultsForActivityType(at) ? (
14: meetingStatus.equals("open").and(taskStatus.equals("open")) ? "open"
15: : ( meetingStatus.equals("finalized")
16: .and(taskStatus.equals("finalized")) ? "finalized" : "in-progress")
17: ) : "finalized"
Figure 4: Code listing computing states of the process phase executed when loading the SmartNavigator.
!
Figure 5: The SmartNet Navigator of the Innovative Helmet Network with an automated interpretation of artifacts.
3.1 Introduction to the Innovative
Helmet Network
The Innovative Helmet Network (IHN) mainly con-
sists of three SMEs that are supported by a research
organization and a consultancy. Together they are
developing a new motorcycle helmet, which will be
made out of completely new materials with special
properties. This new helmet will reach a new level in
protection and will give more freedom to the designer
especially in shape and size. Besides the development
of this new helmet, the network also develops new
production technologies and processes to increase the
effect of the new material.
To handle the arising challenges, the core of the
network consists of a chemical company which is re-
sponsible for the development of the material and its
processing, a product engineering and manufacturing
company which has a lot of experience in the develop-
ment and production of protection gears for different
fields of application and which will be the manufac-
turer of the final helmet.
The third partner in the network is an engineer-
ing consultancy with main focus on technical de-
sign, which supports both other partners and co-
ordinates the collaboration during the development
phases. Both associated partners support the network
in general issues, e.g., management and protection of
intellectual property rights (IPRs), the application of
innovation management methods and the SmartNet
Collaboration Model.
3.2 SmartNet Navigator Supporting the
Development Network
Product development is a complex and challenging
topic in particular in networks of several enterprises.
In this network context the enterprises need some
Supporting Collaborative Product Development through Automated Interpretation of Artifacts
155

space where they are able to collaborate, share knowl-
edge and information, document and coordinate their
work with each other. For this specific network that
participates in the SmartNet Project the partners use
the project collaboration platform ’Tricia’ to face all
these issues.
Typically for the industry sectors, documentation
will only be done if it is either recommended or re-
quired due to quality or risk assurance or the enter-
prises really can benefit from it. Thus, the SmartNet
Navigator has to consolidate the documentation of all
development activities to show the current status of
the collaborative development project. This automat-
ically interpreted status supports the coordination of
the collaborative product development project.
For the IHN this means that all meetings, tasks and
results that were documented on the project platform
are related to activity types and to the development
project they contribute to. Based on the rules exem-
plarily described in Section 2.1.3 the SmartNet Nav-
igator application calculates the current status of the
development project and visualizes (cf. e.g. (Hauder
et al., 2012)) the result as shown in Figure 5.
This example shows that the IHN has collected
and formulated their ideas (M2 is finalized; colored
green) on how this new motorcycling helmet should
protect its bearer. They are currently working on the
processing of the new material to set up several proto-
types of the complete helmet. Parts of it like the inner
shell have already been tested against the homologa-
tion criteria (see Prototype test (α-test) in Figure 5).
Obviously, the SmartNet Navigator in Figure 5
reveals several open (grey) or in-progress (orange)
modules and activity types owing to insufficient docu-
mentation or switching between several activities and
forgetting to finalize or start them. In both cases the
SmartNet Navigator reminds the team on the one hand
to finalize the activities and on the other hand to col-
lect all information they may need in later activities
when it is easy to get them. A second big advantage
for the network or single decision makers is to get a
quick overview on the current status of the develop-
ment.
4 CONCLUSIONS
The SmartNet Navigator helps enterprises acting in
collaborative networks to identify states of develop-
ment projects. Network partners highly esteemed this
aggregated view, as it gives a good overview at a sin-
gle glance and serves as a nice instrument to evaluate
the information basis. The information about the cur-
rent state will be used for further implementation of
tools supporting the application of the collaboration
model itself, e.g. by using the information to propose
methods and techniques of innovation management to
the users or to recommend network partners.
REFERENCES
Dooley, L. and O’Sullivan, D. (2007). Managing within dis-
tributed innovation networks. International Journal of
Innovation Management (ijim), 11(03):397–416.
Filos, E. (2006). Smart organizations in the digital age. Inte-
gration of Information and Communication Technolo-
gies in Smart Organizations. Idea Group Publishing,
Hershey, pages 1–38.
Hauder, M., Matthes, F., Roth, S., and Schulz, C. (2012).
Generating dynamic cross-organizational process vi-
sualizations through abstract view model pattern
matching. Architecture Modeling for Future Internet
enabled Enterprise.
Lau, A. (2012). Methodisch-technologische Unterst
¨
utzung
kollaborativer Innovationsprozesse in Smart Net-
works. Verlag Dr. Hut.
Lau, A., Fischer, T., Hirsch, M., and Matheis, H. (2012).
Smartnet collaboration model-a framework for collab-
orative development and production. In Engineering,
Technology and Innovation (ICE), 2012 18th Interna-
tional ICE Conference on, pages 1–10. IEEE.
Lau, A., Fischer, T., Rehm, S.-V., and Hirsch, M. (2009).
Collaborative innovation in smart networks–a meth-
ods perspective. In 2009) Proceedings of the 15th In-
ternational Conference on Concurrent Enterprising:
ICE.
Matthes, F., Neubert, C., and Steinhoff, A. (2011). Hybrid
wikis: Empowering users to collaboratively structure
information. In 6th International Conference on Soft-
ware and Data Technologies (ICSOFT).
Mazzarol, T. and Reboud, S. (2008). The role of comple-
mentary actors in the development of innovation in
small firms. International Journal of Innovation Man-
agement (ijim), 12(02):223–253.
Probst, G., Raub, S., and Romhardt, K. (1998). Wissen man-
agen: wie Unternehmen ihre wertvollste Ressource
optimal nutzen., volume 2. Aufl. Gabler, Wiesbaden.
van Osch, W. and Coursaris, C. K. (2013). Organizational
social media: A comprehensive framework and re-
search agenda. Hawaii International Conference on
System Sciences HICSS 46.
Third International Symposium on Business Modeling and Software Design
156