<!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>
      <journal-title-group>
        <journal-title>ISEE</journal-title>
      </journal-title-group>
    </journal-meta>
    <article-meta>
      <title-group>
        <article-title>Combining Hardware And Software Development: A Case Study On Interdisciplinary Teaching Projects</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Ljube Boskovski</string-name>
          <email>ljube.boskovski@tum.de</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Mariana Avezum</string-name>
          <email>m.avezum@tum.de</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Technical University of Munich</institution>
          ,
          <addr-line>Munich</addr-line>
          ,
          <country country="DE">Germany</country>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2018</year>
      </pub-date>
      <volume>1</volume>
      <fpage>12</fpage>
      <lpage>15</lpage>
      <abstract>
        <p>-Studies have shown that students retain more information when learning by doing, with practical projects. After graduation, students go on to work with real world projects, which often involve more than one discipline, and the components not covered in their university courses must then be learned on the job. Interdisciplinary university projects are often hard to implement due to lack of collaboration between respective parties, as well as different work practices used in each organization involved. While many advances have been made in teaching agile development to software engineering students, members of other faculties often have different, incompatible, work practices that require structured design and testing processes. We will show how incentivizing students from such different backgrounds to work together can have a huge impact in their learning experience and analyze multiple case studies of such interdisciplinary projects. Decisions such as team structure, collaboration, interdisciplinary change management and task prioritization will get evaluated based on the necessary complexity, and good practices will be drawn based on the results of the presented case studies.</p>
      </abstract>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>I. INTRODUCTION</title>
      <p>
        Capstone courses have had several successes in teaching
agile project development to computer science students in
universities around the world [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ], as well as in introducing
students to the challenges and benefits of working with real
industry partners [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ]. These courses, however, mostly involve
one specific discipline, and few effort have been made to
motivate university students from different faculties or areas
to work together. This is especially difficult due to the
overspecialization that some students face in university.
      </p>
      <p>Different reasons can be found as to why such
interdisciplinary approaches are hard to implement. Beyond the fact
that the necessary project partners often have little experience
working with each other, they may not even know each other
and find it hard to communicate. This distance becomes even
bigger when applied to student teams. Moreover, when
designing interdisciplinary systems to be developed by students,
some design goals present themselves, which may be in direct
contradiction with each other. While it is desired for the
subteams in the different faculties to collaborate and work together
in order to enhance the working experience, it is essential to
limit the dependency (and thus effect) that they have on each
other, as it must always be assumed that part of the designed
system may fail.</p>
      <p>The following paper presents the challenges, and approaches
used in the WARR Hyperloop project, and deduce from that a
few lessons learned from the team at the Technical University
of Munich, where an attempt was made to allow 30 students
from different levels and faculties to work together, in a project
that heavily involves both hardware and software components.
The project involves different iterations with evermore refined
design requirements. After receiving instructor approval that
the proposed design is viable to build, the students ultimately
construct a physical hardware prototype, which participated in
an international competition. All the while the team members,
who were from seven different faculties, had to work
collaboratively with very tight deadlines, which made sure that
each individual had the highest interest in the rest of the team
completely understanding their work.</p>
      <p>Furthermore, we also give some project management
examples of when and where problems occurred during the
described course, as well as which techniques were used to
mitigate it. Section II describes further details on how problem
decomposition into smaller tasks can reduce the impact one
group of students has on another, while still ensuring that
they need to work together for the integration of the entire
system. We will then finish by presenting some lessons learned
from our experiences. As society changes, it is increasingly
important to know how to communicate the skillset learned
throughout life to people from different backgrounds, and to
be able to learn from the experiences from others, which means
that working with people from different backgrounds can be
both a teaching and learning experience to all parties involved.</p>
    </sec>
    <sec id="sec-2">
      <title>II. CHALLENGES</title>
    </sec>
    <sec id="sec-3">
      <title>One of the main challenges presented when organizing an</title>
      <p>
        interdisciplinary educational project, is how to organize the
project and team so that students from all different areas can
benefit from the interdisciplinary aspect and learn additional
material without being blocked by other team members. This
is only possible to attain with a clear team structure and
responsibility definition between sub-teams, as team structures
often define communication channels [
        <xref ref-type="bibr" rid="ref3">3</xref>
        ].
      </p>
      <p>An example of how important such communication is, can
be found in the battery management system of the second
WARR Hyperloop pod, developed at the Technical University
of Munich. The interior of the battery compartment had
to be pressurized to about atmospheric pressure, since the
cuboid shaped battery shells would not withstand the force to
inflating in vacuum and thus possibly incinerate themselves.</p>
      <p>Furthermore, the batteries power the sensors and actuators of
the pneumatic system, which controls the pressure levels of the
battery compartments, among other subsystems. Notice how Fig. 1. The WARR Hyperloop II Team at their team building event at
there is a cyclic dependency chain between the pneumatics Tegernsee, Bavaria. The team divided into sub-teams which competed in
system and the battery management system, which automati- tbougiledthinegr. the most precise and powerful catapult out of limited resources
cally leads to a high information flow between the according
engineers. All in all, an alternated iterative work flow can be
observed, which consists of requirements being passed along e 40,000 and an additional delay of two weeks. While our
from one part of the team to the next, while always ensuring experience shows that students can become very motivated to
that the latter has enough action items to work on indepen- solve problems when others depend on the results of their
dently. With this information cycle, not only the development work and vice versa, this may not always be possible. If a
efficiency is kept high, but also reciprocal functionality of the team later in the dependency chain is prevented from working
software and hardware is assured constantly. because of the results of their predecessors, this may have</p>
      <p>While this collaborative work is great for the educational big consequences not only for motivation, but possibly also
process, much attention is necessary so that it doesn’t extrap- on several organizational aspects. Therefore it is important to
olate time and budget limitations. Although time constraints strengthen the feeling of belonging to the team and build up
are usually given by the semester dates at any given uni- trust and respect to the others very early, for example with a
versity, budget may be harder to manage. Any project that team building event as seen in Figure 1.
involves (buying) hardware components inevitably requires All of these challenges go a long way to show how
more money than software-only projects. How much money important communication is, but this can become quite hard
actually is necessary obviously depends on the complexity when you have team members with different backgrounds, that
of the project and the number of students involved, but if may understand completely different concepts under similar
this complexity starts to get too big, it may make sense to nomenclature. In the history of the WARR Hyperloop team,
introduce some finance and/or business students to the project, there were in total 18 different nationalities among all
particto help manage and acquire any financial resources necessary. ipants.</p>
      <p>This, in turn, adds one more interdisciplinary aspect, by having There are not only social diversities, but also differences
students manage their own budget, although it could add more in experience. The study advancement of the students differ
external dependencies. from the first semester of the bachelor’s degree to almost the</p>
      <p>
        Our experience shows that one of the biggest challenges end of the PhD study. The high gaps in experience between
in setting up an interdisciplinary teaching project is how to collaboratively working students might seem antipathetic, but
manage the risk that inevitably remains due to dependencies it actually can be utilized to benefit the learning process of
inside and outside of the project. As any given risk can be the students, which will be discussed in the Section III.
accessed by considering its probability and severity [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ], it Furthermore, it is often the case that each study
discigenerally is the potential of gaining or loosing some kind of pline has its own work culture or style, which may be very
value. Even if the probability is very low, bad fortune can lead incompatible with each other. While the iterative software
to a massive drawback. Software engineering courses often engineering approach does handle changes well with repetitive
solve testing by integrating automatic testing frameworks into prototyping, that is not the case with the incremental hardware
the development process, which ensure that the code developed design approach. Hardware cannot be physically adjusted as
by students adheres to the necessary outputs and standards [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]. easily as software code can be changed fundamentally. This
Such continuous testing, however, is not completely possible is proven by the fact that projects outside of the computer
for hardware components, which can become a big depen- science department very hardly adhere to agile development
dency source. techniques.
      </p>
      <p>
        For example, an external hardware supplier canceled the An example of projects where hardware design and software
delivery of multiple essential components for the prototype development are deeply combined can be found in real time
of the WARR Hyperloop II team at a late point of time. systems where dozens of sensors are needed to autonomously
As a direct consequence, the team had to quickly contact a monitor, control, and react to the environment (e.g. the
protonew supplier, which lead to a price advancement of about type of the WARR Hyperloop II team. When time is limited by
deadlines, the system has to be presented, verified and assessed
[
        <xref ref-type="bibr" rid="ref6">6</xref>
        ] extremely early, which is especially difficult when software
technologies have to be compatible with the components on
the printed circuit boards and sensors and actuators have to be
integrated into the structure.
      </p>
    </sec>
    <sec id="sec-4">
      <title>III. REALIZATION</title>
      <p>We mitigated the lack of contact between various
faculties by forming sub-teams, which are composed of software
engineers as well as hardware designers. The flawless
integration of both software and hardware components in the end
product is only possible when direct communication between
software developers and hardware designers is established,
easiest archived by putting them in a small group of people
who are in constant interaction, analogously to communication
flows found in Scrum projects. A good example would be the
electronics sub-team of the WARR Hyperloop III Team, which
not only included electrical engineers who worked on the
printed circuit board designs, and mechanical engineers, who
worked in physical structure integration, but also of computer
scientists, who developed the system’s architecture. This was
only possible due to the big variety of studies in the team.
Figure 2 reflects this.</p>
    </sec>
    <sec id="sec-5">
      <title>This partitioning of small sub-teams including only 3-4</title>
      <p>
        people is also used in agile projects [
        <xref ref-type="bibr" rid="ref7">7</xref>
        ], promising an overview
on the project progression despite heavy and frequent changes
in requirements, as well as changes in the team composition.
In order to sensibly subdivide the whole team into
subteams tackling only one or two specific problems, the system
has to be analyzed and decomposed into smaller subsystems
[
        <xref ref-type="bibr" rid="ref8">8</xref>
        ]. With this technique an at first seemingly impossible
problem gets decomposed into a collection of far more easily
manageable sub-problems.
      </p>
      <p>
        Problem decomposition and team subdivision reduces the
negative reaction of requirements adjustments by resulting in
a minor interaction in terms of communication and
coordination and supporting adaptability to external changes [
        <xref ref-type="bibr" rid="ref7">7</xref>
        ].
Nonetheless, fundamental system modularity [
        <xref ref-type="bibr" rid="ref9">9</xref>
        ] is essential in
dynamic projects. In the requirement elicitation phase of the
agile projects life cycle the grade of cohesion and coupling
of the subsystems [
        <xref ref-type="bibr" rid="ref10">10</xref>
        ] should be defined to find a golden
mean between cost, quality of production and amount of time
needed. In theory, quality of a module increases with
increasing cohesion and coupling decrease, but more requirements
regarding modularity also mean that more time is needed for
developing and testing.
      </p>
      <p>
        A good example for the importance of interchangeable
and/or excludable subsystems is the first iteration of the
WARR Hyperloop pod prototype. The so called Pusher Vehicle
built by SpaceX promised to accelerate the student’s prototype
to speeds up to 100m/s, which would allow for magnetic
levitation using electromagnetic suspension [
        <xref ref-type="bibr" rid="ref11">11</xref>
        ] over the
subtrack consisting of aluminum plates mounted onto a concrete
fill bed. The problem is that when the relative movement
between magnets and conductor is too low, a drag force against
driving direction occurs by the shift of the magnetic field.
Hours before the final competition ride, SpaceX informed the
students that their Pusher Vehicle would not be able to
accelerate the prototype to the promised speed, and the students
then calculated that at the lower speed, the drag force created
by the electrical currents in the conductor would be too big. In
foresight, the students team modified their modular levitation
system, even at a very late point of time, hence preventing
repulsive forces against the direction of movement and, as a
consequence of making the subsystems interchangeable, won
the first SpaceX Hyperloop Pod Competition in January 2017.
      </p>
      <p>
        Nevertheless agile methodologies used in software
engineering are not always applicable in hardware design and
manufacturing due to limitations in hardware adjustments.
Therefore a forethought technical architecture and real time
management of risks are of crucial importance. As stated in
[
        <xref ref-type="bibr" rid="ref12">12</xref>
        ], risk management should be integrated into the project’s
development life cycle. This includes starting off each sprint
or iteration with a critical review of each previous decision’s
consequences, as well as what are the next possibilities, and
how these can affect the system as a whole. Additionally, it is
necessary to communicate well with all involved stakeholders
to minimize likelihood of risks as soon as possible.
      </p>
      <p>
        That said, the terms cohesion and coupling in [
        <xref ref-type="bibr" rid="ref10">10</xref>
        ] are used
to only depict software modules. In our case, we consider
both hardware and software components of the vehicle to be
independent modules, that should affect each other as little as
possible. It is essential that when one student’s components
gets changed, it doesn’t completely destroy or eliminate the
work done by another team, as this would deeply affect other
participant’s motivation. Creating modular systems goes a long
way into solving this problem, but can only do so to a certain
extent, as hardware and software components always need
some type of integration.
      </p>
    </sec>
    <sec id="sec-6">
      <title>IV. LESSONS LEARNED</title>
      <p>
        Combining the incremental hardware design approach and
the iterative agile software development life cycle adds to
the student’s understanding of project management. These
principles are often taught to the students only based on
theory, and when a student then receives a good grade in
the final of the lecture, it is said that he/she has understood
the subject material and learned something useful. Although
this approach certainly has its advantages when referred to
software management techniques and methodologies,
”learning” in the sense of the acquisition of knowledge or skills
does not simply consist of immediate concrete experience as
the basis of observation and reflection [
        <xref ref-type="bibr" rid="ref13">13</xref>
        ]. To completely
”learn” and understand project management techniques,
reflective observation, abstract conceptualization and especially
active experimentation is required.
      </p>
      <p>
        Especially in an interdisciplinary context, different
people behave differently when it comes to meeting deadlines,
working in a team and showing initiative in group problem
solving. For example, while one student might only attend the
weekly team meetings and finishes his tasks solely at home,
another one needs the physical presence to be up to date with
the changes happening to the project. This self identification
only happens when truly experiencing the experiments inside
an interdisciplinary team [
        <xref ref-type="bibr" rid="ref14">14</xref>
        ], as the students only absorb
generalized content by the lectures and it’s exercise. The same
principle can be applied to intuitive-experimental thinking:
There are different types of thinking styles, for some
teachercentered teaching is effective, for the others not so much [
        <xref ref-type="bibr" rid="ref15">15</xref>
        ].
Only when a person with experimental-intuitive thinking style
jumps in at the deep end, and has not yet mastered the extent
of their study discipline, he encounters problems unforeseen
and trains his decision making and data interpretation abilities.
      </p>
      <p>As stated in Section II, experienced and unexperienced
students were put together in small sub-teams, which were
composed of both hardware and software components. An
important sub-team, for example, was the propulsion mechanism,
composed by both mechanical engineers and electronics. The
business team, on the other hand, was composed by business
students responsible for sponsoring, and web-developers
responsible for the website.</p>
      <p>
        By creating a practical learning environment similar to a
classroom, not only the students with little to no project
experience evolve individually, but also the lead of the
subteam in the role of a teacher [
        <xref ref-type="bibr" rid="ref16">16</xref>
        ]. The less experienced
students are provided with information about the requirements,
suitability and viability of possible technologies, expert
knowledge as well as lessons learned from prior mistakes. With
the guiding help of the experts the less experienced students
can save a lot of time when trying to meet the expected
requirements. Furthermore, the team members in the teaching
role have the chance to evolve by taking a leading position
and therefore sharpen soft-skill competencies such as time and
budget management competencies, as well as how to explain
technical aspects to an interdisciplinary audience. They also
get a better sense of self-efficacy [
        <xref ref-type="bibr" rid="ref17">17</xref>
        ] by observing their
competence in various disciplines while trying to convey ideas
to others.
      </p>
    </sec>
    <sec id="sec-7">
      <title>V. CONCLUSION</title>
      <p>In conclusion, in order for a multidisciplinary team to
function together, it is essential that each team member respects the
roles and issues of the others. By decomposing big systems
into smaller sub-problems, we were able to have each
subteam work on its own component, and always have continuous
feedback about the complete system. We have learned, that the
right composition of the team is necessary for it to function
properly. The ideal sub-team configuration contains not only
students from different faculties and backgrounds, but also
with different levels of experience, which can help balance
the learning experience for everyone involved.</p>
      <p>In addition, techniques used in agile methodologies provide
a good initial proposal of how to have these different teams
collaborate with each other, as well as the structure of the
emerged sub-systems in the system. Furthermore, it is of
importance to define the grades of coupling and cohesion
to guarantee a modular enough system design, while
keeping the developing time low. Together with prospective risk
management, the achievement of the project goals is thus best
secured. By encouraging teams to work together and giving
them goals that are dependent on complete collaboration, it is
possible to ensure that every participant is able to experience
new disciplines in a motivating manner.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <given-names>V.</given-names>
            <surname>Mahnic</surname>
          </string-name>
          , “
          <article-title>A capstone course on agile software development using scrum,” IEEE Transactions on Education</article-title>
          , vol.
          <volume>55</volume>
          , no.
          <issue>1</issue>
          , pp.
          <fpage>99</fpage>
          -
          <lpage>106</lpage>
          ,
          <year>2012</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <given-names>B.</given-names>
            <surname>Bruegge</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.</given-names>
            <surname>Krusche</surname>
          </string-name>
          , and L. Alperowitz, “
          <article-title>Software engineering project courses with industrial clients</article-title>
          ,
          <source>” Trans. Comput. Educ.</source>
          , vol.
          <volume>15</volume>
          , pp.
          <volume>17</volume>
          :
          <fpage>1</fpage>
          -
          <lpage>17</lpage>
          :
          <fpage>31</fpage>
          ,
          <string-name>
            <surname>Dec</surname>
          </string-name>
          .
          <year>2015</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <given-names>M. E.</given-names>
            <surname>Conway</surname>
          </string-name>
          , “
          <article-title>How do committees invent,” Datamation</article-title>
          , vol.
          <volume>14</volume>
          , no.
          <issue>4</issue>
          , pp.
          <fpage>28</fpage>
          -
          <lpage>31</lpage>
          ,
          <year>1968</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          [4]
          <string-name>
            <given-names>B.</given-names>
            <surname>Bruegge</surname>
          </string-name>
          and
          <string-name>
            <given-names>A. H.</given-names>
            <surname>Dutoit</surname>
          </string-name>
          ,
          <article-title>Object Oriented Software Engineering Using UML, Patterns, and</article-title>
          <string-name>
            <given-names>Java. Prentice</given-names>
            <surname>Hall</surname>
          </string-name>
          ,
          <year>2009</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [5]
          <string-name>
            <given-names>S.</given-names>
            <surname>Krusche</surname>
          </string-name>
          and
          <string-name>
            <given-names>L.</given-names>
            <surname>Alperowitz</surname>
          </string-name>
          , “
          <article-title>Introduction of continuous delivery in multi-customer project courses</article-title>
          ,” in
          <source>Companion Proceedings of the 36th International Conference on Software Engineering</source>
          , pp.
          <fpage>335</fpage>
          -
          <lpage>343</lpage>
          , ACM,
          <year>2014</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <given-names>C. U.</given-names>
            <surname>Smith</surname>
          </string-name>
          ,
          <string-name>
            <given-names>G. A.</given-names>
            <surname>Frank</surname>
          </string-name>
          , and
          <string-name>
            <given-names>J.</given-names>
            <surname>Cuardrado</surname>
          </string-name>
          , “
          <article-title>An architecture design and assessment system for software/hardware codesign</article-title>
          ,” in Design Automation,
          <year>1985</year>
          . 22nd Conference on, pp.
          <fpage>417</fpage>
          -
          <lpage>424</lpage>
          , IEEE,
          <year>1985</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [7]
          <string-name>
            <given-names>S.</given-names>
            <surname>Augustine</surname>
          </string-name>
          ,
          <string-name>
            <given-names>B.</given-names>
            <surname>Payne</surname>
          </string-name>
          ,
          <string-name>
            <given-names>F.</given-names>
            <surname>Sencindiver</surname>
          </string-name>
          , and
          <string-name>
            <given-names>S.</given-names>
            <surname>Woodcock</surname>
          </string-name>
          , “
          <article-title>Agile project management: steering from the edges,” Communications of the ACM</article-title>
          , vol.
          <volume>48</volume>
          , no.
          <issue>12</issue>
          , pp.
          <fpage>85</fpage>
          -
          <lpage>89</lpage>
          ,
          <year>2005</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [8]
          <string-name>
            <given-names>L.</given-names>
            <surname>Rapanotti</surname>
          </string-name>
          ,
          <string-name>
            <given-names>J. G.</given-names>
            <surname>Hall</surname>
          </string-name>
          ,
          <string-name>
            <given-names>M.</given-names>
            <surname>Jackson</surname>
          </string-name>
          , and
          <string-name>
            <given-names>B.</given-names>
            <surname>Nuseibeh</surname>
          </string-name>
          , “Architecturedriven problem decomposition,” in Requirements Engineering Conference,
          <year>2004</year>
          . Proceedings. 12th IEEE International, pp.
          <fpage>80</fpage>
          -
          <lpage>89</lpage>
          , IEEE,
          <year>2004</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <given-names>C.</given-names>
            <surname>Baldwin</surname>
          </string-name>
          and
          <string-name>
            <given-names>K.</given-names>
            <surname>Clark</surname>
          </string-name>
          , “
          <article-title>Modularity in the design of complex engineering systems,” Complex engineered systems</article-title>
          , pp.
          <fpage>175</fpage>
          -
          <lpage>205</lpage>
          ,
          <year>2006</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [10]
          <string-name>
            <given-names>H.</given-names>
            <surname>Dhama</surname>
          </string-name>
          , “
          <article-title>Quantitative models of cohesion and coupling in software</article-title>
          ,
          <source>” Journal of Systems and Software</source>
          , vol.
          <volume>29</volume>
          , no.
          <issue>1</issue>
          , pp.
          <fpage>65</fpage>
          -
          <lpage>74</lpage>
          ,
          <year>1995</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          [11]
          <string-name>
            <given-names>N.</given-names>
            <surname>Grebennikov</surname>
          </string-name>
          and
          <string-name>
            <given-names>A.</given-names>
            <surname>Kireev</surname>
          </string-name>
          , “
          <article-title>Electromagnetic suspension used for high-speed vacuum transport</article-title>
          ,”
          <source>International Journal of Applied Engineering Research</source>
          , vol.
          <volume>12</volume>
          , no.
          <issue>12</issue>
          , pp.
          <fpage>3293</fpage>
          -
          <lpage>3297</lpage>
          ,
          <year>2017</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          [12]
          <string-name>
            <given-names>A.</given-names>
            <surname>Jaafari</surname>
          </string-name>
          , “
          <article-title>Management of risks, uncertainties and opportunities on projects: time for a fundamental shift</article-title>
          ,”
          <source>International journal of project management</source>
          , vol.
          <volume>19</volume>
          , no.
          <issue>2</issue>
          , pp.
          <fpage>89</fpage>
          -
          <lpage>101</lpage>
          ,
          <year>2001</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          [13]
          <string-name>
            <given-names>D. A.</given-names>
            <surname>Kolb</surname>
          </string-name>
          , “
          <article-title>Management and the learning process,” California management review</article-title>
          , vol.
          <volume>18</volume>
          , no.
          <issue>3</issue>
          , pp.
          <fpage>21</fpage>
          -
          <lpage>31</lpage>
          ,
          <year>1976</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          [14]
          <string-name>
            <given-names>L.</given-names>
            <surname>Jaccheri</surname>
          </string-name>
          and G. Sindre, “
          <article-title>Software engineering students meet interdisciplinary project work and art</article-title>
          ,” in Information Visualization,
          <year>2007</year>
          . IV'
          <volume>07</volume>
          . 11th International Conference, pp.
          <fpage>925</fpage>
          -
          <lpage>934</lpage>
          , IEEE,
          <year>2007</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref15">
        <mixed-citation>
          [15]
          <string-name>
            <given-names>S.</given-names>
            <surname>Epstein</surname>
          </string-name>
          ,
          <string-name>
            <given-names>R.</given-names>
            <surname>Pacini</surname>
          </string-name>
          ,
          <string-name>
            <given-names>V.</given-names>
            <surname>Denes-Raj</surname>
          </string-name>
          ,
          <article-title>and</article-title>
          <string-name>
            <given-names>H.</given-names>
            <surname>Heier</surname>
          </string-name>
          , “
          <article-title>Individual differences in intuitive-experiential and analytical-rational thinking styles</article-title>
          .,
          <source>” Journal of personality and social psychology</source>
          , vol.
          <volume>71</volume>
          , no.
          <issue>2</issue>
          , p.
          <fpage>390</fpage>
          ,
          <year>1996</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref16">
        <mixed-citation>
          [16]
          <string-name>
            <given-names>J.</given-names>
            <surname>Harland</surname>
          </string-name>
          and
          <string-name>
            <given-names>K.</given-names>
            <surname>Kinder</surname>
          </string-name>
          , “
          <article-title>Teachers' continuing professional development: framing a model of outcomes,”</article-title>
          <source>British Journal of In-service Education</source>
          , vol.
          <volume>23</volume>
          , no.
          <issue>1</issue>
          , pp.
          <fpage>71</fpage>
          -
          <lpage>84</lpage>
          ,
          <year>1997</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref17">
        <mixed-citation>
          [17]
          <string-name>
            <given-names>A.</given-names>
            <surname>Bandura</surname>
          </string-name>
          , “
          <article-title>Perceived self-efficacy in cognitive development and functioning,” Educational psychologist</article-title>
          , vol.
          <volume>28</volume>
          , no.
          <issue>2</issue>
          , pp.
          <fpage>117</fpage>
          -
          <lpage>148</lpage>
          ,
          <year>1993</year>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>