<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Managing Social Challenges in Cross-Organizational Event-Based Systems</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Laura S. Thiele</string-name>
          <email>laura.thiele@eah-jena.de</email>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Nico Brehm</string-name>
          <email>nico.brehm@eah-jena.de</email>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>German Aerospace Center (DLR), Institute of Data Science</institution>
          ,
          <addr-line>Ma ̈lzerstraße 3, 07745 Jena</addr-line>
          ,
          <country country="DE">Germany</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>University of Applied Science Jena, Department of Industrial Engineering</institution>
          ,
          <addr-line>Carl-Zeiß-Promenade 2, 07745 Jena</addr-line>
          ,
          <country country="DE">Germany</country>
        </aff>
      </contrib-group>
      <fpage>48</fpage>
      <lpage>55</lpage>
      <abstract>
        <p>During the last decade the manufacturing industry focused on the realization of industry 4.0 aspects. Besides the implementation of new technologies, existing software structures also need to be reviewed and adapted in this context. To stay competitive in the global market, especially small and medium-sized companies need to emphasize on better cooperation with other organizations. This leads to the implementation of cross-organizational distributed software system structures. The development of distributed systems faces different challenges - technical and code-centric as well as social challenges. This paper focuses on the social challenges that appear in distributed development processes. After defining the main challenges, the paper introduces a development approach that is based on the integration of a Federated Management System (FMS). FMS is a technical approach to minimize social challenges by the generation of system transparency and the provision of a platform for communication and interaction. It facilitates a distributed system development of cross-organizational event-based systems.</p>
      </abstract>
      <kwd-group>
        <kwd>Software Development</kwd>
        <kwd>Distributed Systems</kwd>
        <kwd>Event-Driven Systems</kwd>
        <kwd>Wiki-Based System Management</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>
        The manufacturing industry is subject to current trends in the market, such as
increasing product variety, custom and individual fabrication, as well as reducing
production and delivery times [
        <xref ref-type="bibr" rid="ref12 ref13 ref4">4,12,13</xref>
        ]. Due to these trends, the manufacturing
industry started to adapt traditional structures of their manufacturing in order
to implement the ideas of Industry 4.0.
      </p>
      <p>
        Most of the Industry 4.0 approaches are based on comprehensive data
acquisition and usage. By now, data is captured by several different devices and
systems. It needs to be collected and stored in order to process it for further
purposes. To support the manufacturing industry in its current challenges, several
research projects, e.g., [
        <xref ref-type="bibr" rid="ref10 ref15 ref5 ref6 ref7">5,6,7,10,15</xref>
        ], were started to develop system architectures
that enable comprehensive data acquisition and usage in the manufacturing
environment. Since research emphasizes that especially small and medium-sized
enterprises (SME) may increase their competitiveness through improvement of
cooperation with federated companies (suppliers, purchasers, and subcontractors)
[
        <xref ref-type="bibr" rid="ref11">11</xref>
        ], research projects started to consider an inter-organizational data exchange
in their developed systems [
        <xref ref-type="bibr" rid="ref10 ref5 ref7">5,7,10</xref>
        ].
      </p>
      <p>By analyzing the outcome of these researches, it is noticeable that all developed
architectures show distributed system structures. Which means that parts of the
systems are separated but work together like a single system.</p>
      <p>
        Mishra &amp; Tripathi [
        <xref ref-type="bibr" rid="ref8">8</xref>
        ] categorize distributed systems into systems where:
a) only software and hardware are distributed, b) users are distributed and c)
both – users, hardware and software are distributed. Since software
development strategies changed from a fixed group of developers towards an open and
community-based development, like in open source software projects, systems
can also have distributed developers. Hence, the distributed system schema of
Mishra &amp; Tripathi can be enhanced by the category of distributed developers, as
displayed in Figure 1.
      </p>
    </sec>
    <sec id="sec-2">
      <title>Distributed System</title>
    </sec>
    <sec id="sec-3">
      <title>Distributed Users Fig. 1. Distributed Systems Distributed Software &amp; Hardware</title>
      <p>
        Considering the schema that is displayed in Figure 1, a distributed system may
belong to one up to three of the displayed categories. Each category implicates
specific challenges that need to be considered [
        <xref ref-type="bibr" rid="ref2 ref3 ref8">2,3,8</xref>
        ].
      </p>
      <p>
        Effective collaborations within distributed systems require adequate technical
solutions. But especially for systems that have distributed users and/or developers,
an adoption of good organizational practices and development processes is also
necessary [
        <xref ref-type="bibr" rid="ref3">3</xref>
        ]. In this paper we do not focus on the technical and code-centric
perspective of distributed system development, instead we are focusing on the
social part of the development process and how to address it with technical
solutions.
2
      </p>
      <sec id="sec-3-1">
        <title>State of the Art</title>
        <p>
          The integration of system users and developers with different knowledge in
the development phase of a software is not a new aspect. All agile methods
already implemented this aspect to enhance the field of knowledge and to get
feedback early in the development phase [
          <xref ref-type="bibr" rid="ref3">3</xref>
          ]. In most agile software projects
the participating developers are known. Therefore, it is easier to manage the
cooperation between them. Software projects that are developed by distributed
developers, e.g., via an open platform like GitHub, show the problem that
system users and developers are unknown. This complicates the coordination
and cooperation.
        </p>
        <p>
          Allaho &amp; Lee [
          <xref ref-type="bibr" rid="ref1">1</xref>
          ] analyzed software projects on GitHub in their research. They
recognized that project documentation is very important to help developers to
understand the project. It helps to create transparency of the project structure
and behavior.
        </p>
        <p>
          Transparency is a very important aspect, also for Carbot et al. [
          <xref ref-type="bibr" rid="ref3">3</xref>
          ]. They
analyzed open source projects and occurring problems in their work and
determined that most of the projects and its leadership miss transparency for the
contributors. Furthermore, they recognized that most projects hardly ever follow
any kind of democratic practices, which makes it difficult for contributors to
influence the development of the project. During their research, Carbot et al.
found out that many projects are only developed by a few members. It seems to
be difficult to motivate others to contribute.
        </p>
        <p>
          Promoting contribution not only refers to system developers, it also includes
system users. Wikipedia is probably the most known example. There, a high
amount of system users are motivated to contribute to the system [
          <xref ref-type="bibr" rid="ref14">14</xref>
          ]. Wikipedia
provides a framework that makes it easy to contribute to the system even if
the user is not a software developer. The principle of a wiki-based platform for
communication and coordination of software projects was already implemented in
different tools that support agile software development projects. It is a convenient
method to exchange information between participants.
        </p>
        <p>
          The importance of a medium for communication between system developers
was also addressed in the work of Mu¨ller [
          <xref ref-type="bibr" rid="ref9">9</xref>
          ]. He pointed out that
communication supports collaboration within a system’s environment and motivates new
contributors to get into the community.
3
        </p>
        <p>
          Social Challenges in Distributed System Development
Based on the observations that were made, the social challenges in distributed
software project can be summarized as:
C1 how to provide transparency (motivated by [
          <xref ref-type="bibr" rid="ref1 ref14 ref3">1,3,14</xref>
          ]),
C2 how to attract and support new contributors (mot. by [
          <xref ref-type="bibr" rid="ref14 ref3 ref9">3,9,14</xref>
          ]) and
C3 how to optimize collaborations (mot. by [
          <xref ref-type="bibr" rid="ref3 ref9">3,9</xref>
          ]).
        </p>
        <p>Transparency (C1) needs to be established in the entire distributed system.
That means, architectural aspects (relating the hardware and software) need
to be transparent as well as system contributions from users (e.g., feedback) or
developers (e.g., system extensions). System transparency helps to understand
the systems structure and behavior. Hence, it can be seen as the main social
challenge in distributed systems.</p>
        <p>System contributions (C2) are very important as they expand and improve the
system. System transparency helps to minimize entry barriers for new contributors.
Furthermore, the knowledge about a system’s functional and operational range
may inspire people to contribute new extensions and improvements.</p>
        <p>The optimization of collaborations (C3) is important to maintain efficiency
in system development and system operation. The system, or its environment,
should provide a possibility (e.g., a tool) for users and developers to manage and
optimize their collaborations.
4</p>
      </sec>
      <sec id="sec-3-2">
        <title>Context and Assumptions</title>
        <p>This section provides an overview of the context and the assumptions under
which the development strategy, presented in the following section, was developed.
Since processes in the manufacturing industry are mostly triggered by occurring
events (e.g., start and finish of tasks, error announcements, attainment of states,
such as temperatures) it is assumed to have a system that has an event-driven
architecture: information is published to a broker that routes it to subscribing
services.</p>
        <p>To support collaborations among small and medium-sized companies, an
inter-organizational system usage should be supported by the system. That may
enable companies to integrate machines and devices to the system and share
information with a federated group of organizations.</p>
        <p>Furthermore, the system should enable decentralized system development,
where system components are developed by autonomous groups of software
developers. They may develop new system components (services) that can be added
to the system.</p>
        <p>In order to develop a strategy that supports the development of such systems, in
summary, we assume that:
– the system shows a distributed system structure in all categories (software &amp;
hardware, users, and developers)
– the system is based on an event-based approach (publish-subscribe pattern)
– the system is situated and developed in a trusted domain (no security
considerations)
– the system will be developed in an agile manner (continuous system
development and enhancement)
5</p>
      </sec>
      <sec id="sec-3-3">
        <title>The Federated Development Strategy</title>
        <p>To support the development of cross-organizational event-based systems, we
developed a federated system development strategy that helps to handle the
social challenges as they were stated in Section 3. To manage the challenges we
message
exchange</p>
      </sec>
    </sec>
    <sec id="sec-4">
      <title>Federated</title>
    </sec>
    <sec id="sec-5">
      <title>Management System (FMS)</title>
    </sec>
    <sec id="sec-6">
      <title>Broker GUI</title>
      <p>developed an architecture of a Federated Management System (FMS) that needs
to be integrated to the systems architecture.</p>
      <p>Figure 2 displays an architectural sketch of our federated development strategy
for cross-organizational event-based systems. In the upper part, the figure displays
a system architecture that exchanges event messages between two companies
via a broker (the assumed system situation). In this situation, social challenges
appear because of the distributed system usage.</p>
    </sec>
    <sec id="sec-7">
      <title>Company A</title>
      <p>E challenges E
social</p>
    </sec>
    <sec id="sec-8">
      <title>Company B</title>
    </sec>
    <sec id="sec-9">
      <title>Developer</title>
    </sec>
    <sec id="sec-10">
      <title>Service</title>
    </sec>
    <sec id="sec-11">
      <title>Service</title>
    </sec>
    <sec id="sec-12">
      <title>Developer Fig. 2. Federated Management System</title>
      <p>The lower part of the figure shows the FMS that needs to be integrated. It
consists of a Federated Message Archive (FMA) component that includes an
event-database and a graphical user interface (GUI). The FMA component is
connected to the systems broker. Whereas other services only listen to specific
events, the FMA subscribes to all messages that might be posted in the system.
Consequently, the broker forwards all messages to the FMA which stores them
in the event-database. In doing so, the FMA collects several information about
posted events, such as message type, message type’s publishing frequency, message
content and more.</p>
      <p>The GUI of the FMS, see Figure 3, provides access to the information in the
FMA. It is the operative component that uncovers system information to users
and enables transparency in the system environment. As displayed in Figure 3,</p>
    </sec>
    <sec id="sec-13">
      <title>Federated Message Archive (FMA)</title>
    </sec>
    <sec id="sec-14">
      <title>Event</title>
      <p>DB
access</p>
      <p>to
system
information
a user may search for specific information that is shared via the system, e.g.,
information related to orders. Hence, the system displays a list of messages that
are related to the search term. In this example the user can see that there exist
more than one message type that is related to orders: topic ’order/new’ and topic
’order/accepted’. Furthermore, the user can explore the messages content, e.g.,
the single attributes, in order to get the information he or she is locking for.</p>
      <p>By searching for a specific message topic, the user can obtain further
information about a specific message type. After forwarding a search query of a
message topic to the FMA, all collected messages of the specific message type
are compared and analyzed by the FMA component. Findings are formatted and
sent to the GUI were they are displayed to the user. Hence, the user can obtain
specific information about a message type, such as the frequency in which it is
used or specific information about attributes that are held in the message content
(e.g., range of attributes values or most common value).</p>
      <p>The GUI of the FMA can be accessed by every developer that belongs to the
trusted group of organizations in which the system is implemented (see Figure 1).
This ensures a high degree of transparency within the whole system community
(C1).</p>
      <p>New developers that are interested in system contribution and join the
crossorganizational system environment get access to the FMS. Due to transparency
that is provided by the FMS, new system developers are able to get to know
the systems structure and behavior. By analyzing events that were stored in the
FMS, new developers can learn how messages are structured and how they can
be subscribed in order to use their information for further purposes.</p>
      <p>Besides the provision of system information, the FMS furthermore provides
a wiki-based area where system developers are able exchange information. By
this, system users get the opportunity to ask for special services and for system
extensions within the system community. Concrete demands on system extensions
may animate developers to contribute to the system (C2). Especially for new
contributors it might be easier and more motivating to start contribution on a
concrete task.</p>
      <p>The wiki-based area of the FMS supports communication among the system
participants. This is very important to optimize collaborations (C3). The
wikibased area may not only be used for placing demands, but also for coordinating
contributions. Developers can quote on which functionality they are currently
working and several developers are able to coordinate their collective work
via, e.g., Kanban boards. Furthermore, new developers can ask for help within
the community and senior system developers may share their knowledge and
experiences.
6</p>
      <sec id="sec-14-1">
        <title>Conclusion and Future Work</title>
        <p>In this paper we presented an approach to handle social challenges in
crossorganizational event-based systems. First, we presented a classification model
of distributed systems which shows that distributed systems can be categorized
into three categories: hardware &amp; software distributed systems, as well as users
distributed and developers distributed. Depending on their category of
distribution, distributed systems face different challenges. Systems that have distributed
users and/or developers have to handle not only technical and code-centric issues,
but also social challenges. In this paper, we focused on social challenges and
summarized them as follows: 1) provision of transparency, 2) attraction and
support of new developers and 3) optimization of system collaborations. Finally,
we presented FMS and how this system approach that can be used to mitigate
social challenges in distributed systems.</p>
        <p>In the future work it is planned to evaluate FMS in real system environments.
To enable such evaluation, cross-organizational event-based systems will be
implemented in the manufacturing environment and system developers will be
asked to enhance the system by components that supports the manufacturing
process. The evaluation of FMS’s usability will be done in user studies that will
be undertaken with the developers of the involved companies. There will be a
mixture of developers with more and less experiences to establish the different
needs and requirements to the FMS. Based on the evaluations outcome, FMS
will be further adapted and improved.</p>
      </sec>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          1.
          <string-name>
            <surname>Allaho</surname>
            ,
            <given-names>M.Y.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Lee</surname>
            ,
            <given-names>W.C.</given-names>
          </string-name>
          :
          <article-title>Trends and behavior of developers in open collaborative software projects</article-title>
          . In: 2014 International Conference on Behavioral, Economic, and
          <string-name>
            <surname>Socio-Cultural Computing</surname>
          </string-name>
          (
          <year>BESC2014</year>
          ).
          <source>IEEE (Oct</source>
          <year>2014</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          2.
          <string-name>
            <surname>Braubach</surname>
            ,
            <given-names>L.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Pokahr</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          :
          <article-title>Addressing Challenges of Distributed Systems Using Active Components</article-title>
          . In: Intelligent Distributed Computing V, pp.
          <fpage>141</fpage>
          -
          <lpage>151</lpage>
          . Springer Berlin Heidelberg (
          <year>2011</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          3.
          <string-name>
            <surname>Cabot</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Izquierdo</surname>
            ,
            <given-names>J.L.C.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Cosentino</surname>
          </string-name>
          , V.:
          <article-title>Community-based software development for MDE tools</article-title>
          .
          <source>In: Joint Proceedings of EduSymp 2016 and OSS4MDE</source>
          <year>2016</year>
          (
          <year>2016</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          4.
          <string-name>
            <surname>Denkena</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Dittrich</surname>
            ,
            <given-names>M.A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Uhlich</surname>
            ,
            <given-names>F.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Maibaum</surname>
            ,
            <given-names>L.</given-names>
          </string-name>
          , Mo¨rke, T.:
          <article-title>Das gentelligente Werkstu¨ck</article-title>
          .
          <source>In: Handbuch Industrie 4.0</source>
          , pp.
          <fpage>295</fpage>
          -
          <lpage>321</lpage>
          . Hanser Fachbuchverlag,
          <string-name>
            <surname>Munich</surname>
          </string-name>
          (
          <year>2017</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          5. DIN:
          <article-title>Erweiterung des EPCIS-Ereignismodells um aggregierte Produktionsereignisse zur Verwendung in betrieblichen Informationssystemen (Jan</article-title>
          <year>2016</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          6.
          <source>DIN: Reference Architecture Model Industrie 4.0 (Apr</source>
          <year>2016</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          7.
          <string-name>
            <surname>Fuchs</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Oks</surname>
            ,
            <given-names>S.J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Franke</surname>
          </string-name>
          , J.:
          <article-title>Platform-based service composition for manufacturing: A conceptualization</article-title>
          .
          <source>Procedia CIRP 81</source>
          ,
          <fpage>541</fpage>
          -
          <lpage>546</lpage>
          (
          <year>2019</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          8.
          <string-name>
            <surname>Mishra</surname>
            ,
            <given-names>K.S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Tripathi</surname>
            ,
            <given-names>A.K.</given-names>
          </string-name>
          :
          <article-title>Some Issues, Challenges and Problems of Distributed Software System</article-title>
          .
          <source>In: (IJCSIT) International Journal of Computer Science and Information Technologies</source>
          , Vol.
          <volume>5</volume>
          (
          <issue>4</issue>
          ) (
          <year>2014</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          9. Mu¨ller, M.:
          <article-title>Agile Challenges and Chances for Open Source: Lessons Learned from Managing a FLOSS Project (Nov</article-title>
          <year>2018</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          10.
          <string-name>
            <surname>Otto</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Lohmann</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Steinbuß</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Teuscher</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          :
          <article-title>IDS Reference Architecture Model Industrial Data Space</article-title>
          .
          <source>Tech. rep</source>
          .,
          <source>International Data Space Association</source>
          and
          <article-title>Fraunhofer-Gesellschaft zur Fo¨rderung der angewandten Forschung e</article-title>
          .V.,
          <string-name>
            <surname>Dortmund</surname>
          </string-name>
          (
          <year>2018</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          11.
          <string-name>
            <surname>Peris-Ortiz</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Ferreira</surname>
            ,
            <given-names>J.J.</given-names>
          </string-name>
          :
          <source>Cooperative and Networking Strategies in Small Business</source>
          . Springer-Verlag GmbH (
          <year>2016</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          12.
          <string-name>
            <surname>Reinhart</surname>
            ,
            <given-names>G.</given-names>
          </string-name>
          , Zu¨hlke, D.:
          <article-title>Handbuch Industrie 4.0, chap</article-title>
          .
          <source>Von CIM zu Industrie 4.0</source>
          , pp.
          <string-name>
            <surname>XXXI-XL. Hanser</surname>
            <given-names>Fachbuchverlag</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Munich</surname>
          </string-name>
          (
          <year>2017</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          13.
          <string-name>
            <surname>Schuh</surname>
            ,
            <given-names>G.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Reuter</surname>
            ,
            <given-names>C.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Hauptvogel</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Brambring</surname>
            ,
            <given-names>F.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Hempel</surname>
            ,
            <given-names>T.</given-names>
          </string-name>
          :
          <article-title>Ergebnisbericht des BMBF-Verbundprojektes PROSENSE : hochauflo¨sende Produktionssteuerung auf Basis kybernetischer Unterstu¨tzungssysteme und intelligenter Sensorik</article-title>
          ,
          <source>chap. 1 Einleitung</source>
          , pp.
          <fpage>1</fpage>
          -
          <lpage>13</lpage>
          . Apprimus Verlag Aachen,
          <string-name>
            <surname>Aachen</surname>
          </string-name>
          (
          <year>2015</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          14. van Steen,
          <string-name>
            <given-names>M.</given-names>
            ,
            <surname>Pierre</surname>
          </string-name>
          ,
          <string-name>
            <given-names>G.</given-names>
            ,
            <surname>Voulgaris</surname>
          </string-name>
          ,
          <string-name>
            <surname>S.:</surname>
          </string-name>
          <article-title>Challenges in very large distributed systems</article-title>
          .
          <source>Journal of Internet Services and Applications</source>
          <volume>3</volume>
          (
          <issue>1</issue>
          ),
          <fpage>59</fpage>
          -
          <lpage>66</lpage>
          (
          <year>Nov 2011</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref15">
        <mixed-citation>
          15.
          <string-name>
            <surname>Theorin</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Bengtsson</surname>
            ,
            <given-names>K.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Provost</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Lieder</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Johnsson</surname>
            ,
            <given-names>C.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Lundholm</surname>
            ,
            <given-names>T.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Lennartson</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          :
          <article-title>An Event-Driven Manufacturing Information System Architecture for Industry 4.0</article-title>
          .
          <source>International Journal of Production Research</source>
          <volume>55</volume>
          (
          <issue>5</issue>
          ),
          <fpage>1297</fpage>
          -
          <lpage>1311</lpage>
          (
          <year>Jul 2016</year>
          )
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>