IEEE AIKE 2019

International Conference on Artificial Intelligence and Knowledge Engineering

Doctoral Consortium and Young Investigator Training Program

Report on Research Visit

University of Brescia

Visitor: Rafael Escalfoni, PhD Student, Graduate Program in Informatics, UFRJ

Host Professor: Devis Bianchini, Department of Information Engineering, UNIBS
From June 6th to July 7th
Summary
1. Researchers' Background 3
About the visitor 3
About the host 3

2. General Activities 4

3. YITP Project Proposal 4

3.1 Introduction 4
3.2 Methodological Approach 5
3.3 Project Schedule 6

4. Project Execution 7
st
4.1 1 DRS Cycle - Problem Awareness 7
nd
4.2 2 DRS Cycle - Proposition 8
The Collective Knowledge of Experts 9
Group Storytelling 10
Data-driven Storytelling 11
Knowledge Management Layer to IDEAaS Framework 12

5. Conclusions and Future Works 13

6. References 13
1. Researchers' Background

About the visitor

Rafael Escalfoni is a Ph.D. student at the Graduate Program in Informatics, Federal University of Rio de
Janeiro - UFRJ, and teaching assistant at Federal Center for Technological Education of Rio de Janeiro
(CEFET/RJ). His work is related to Big Social Data, entrepreneurial ecosystems, social computing. He is
also involved in entrepreneurial activities of CEFET/RJ, such as the implementation of the new
technology incubator in the Nova Friburgo campus, and the development of Enactus project of
CEFET/RJ.

Research work: "Coral Framework - a big social data approach to boost startup ecosystems".
Abstract: The startup ecosystem is a creative environment that aims to promote entrepreneurship and
innovative business. In this complex social network, the activities derive from the ties of
interdependence and co-evolution between actors with different purposes, who share resources to
achieve common goals. A smart startup ecosystem demands an understanding of the interests and
capabilities of each participant as a way to identify convergences and possible partnerships that can
help in the development of new business. The Coral framework is an approach based on big social data
to assist in the analysis of existing relationships in startup ecosystems.

About the host

Professor Devis Bianchini is an associate professor in Computer Science at the Department of
Information Engineering, University of Brescia. His main research interests are focused on models and
tools for Web Information System (WIS) design, service-oriented architectures, resource discovery
enabled through techniques, tools and theoretical foundations of Semantic Web.

Research work: "A Relevance-based approach for Big Data Exploration"

Abstract: The collection, organization, and analysis of a large amount of data, Big Data, in different
application domains still require the involvement of domain experts for the identification of relevant data
only, without being overwhelmed by volume, velocity, and variety of collected data. According to the
“human-in-the-loop” vision, expert users inspect and explore data to take decisions in unexpected
situations, based on their long-term experience. We propose a novel framework for Big Data exploration,
relying on the combined use of different techniques: (i) an incremental clustering algorithm, to provide
summarised representation of collected data; (ii) data relevance evaluation techniques, to attract the
experts’ attention on relevant data only; (iii) a multi-dimensional organisation of summarised data, for
relevance-driven data exploration according to different analysis dimensions. The proposed framework
is the basis of a general approach to Big Data exploration. In particular, we demonstrate its effectiveness
and efficacy with extensive experimentation in the Industry 4.0 application domain, given the strategic
importance of Big Data exploration for Cyber-Physical Systems (CPS) in this context.
2. General Activities
In the first day of the visiting, professor Bianchini prepared a welcome meeting with all participants of his
research team. During the reception, they showed me the current work of the laboratory, and I could
make a short presentation of my study on big social data applied to entrepreneurial communities. Then,
we depicted the most important subjects related to the investigations: their objectives, conjectures, and
achievements.

After that, it was presented to me the university infrastructure and the place where I have been
working during the visit. I had access to the main paper of the research, and I shared my work to the
group. So, I spent two days reading the article and the most relevant references. I was preparing some
notes and searching for some convergence between our studies. During the week, I also had some
meetings with Ada Bagozi, a Ph.D. student of professor Bianchini, to ask questions and broaden my
comprehension about the group work.

In the following days, I focused on the development of a presentation of all works in progress on
my origin laboratory, the Lab Cores (Laboratory of Social Computing and Network Analysis). We had a
meeting at the end of the day to detail these studies and found some possible cooperation between
institutions. As a result, I started a document with my proposal of work to be delivered at the end of this
program and a bigger plan of cooperation, with various areas of collaboration. This partnership project
involves three prominent researchers1 from Brazilian institutions who were consulted and agreed with
the possibility of cooperation effort. Currently, they are working on the proposal with insights and
suggestions of projects.

3. YITP Project Proposal

In face to the reduced time to develop a research work of the YITP, we thought a project with short
cycles to produce some results at the end of the month. We have reached an agreement on the subject
of the YITP project in the area of knowledge management. The purpose is to recover tacit knowledge
from experts using data-driven storytelling about unexpected situations. For that, we are proposing a
new layer, or component, to IDEAaS framework in which a specific episode will be displayed to experts
using contextualized visualization models, and then they will collectively tell a story about the data,
according to their wisdom or experience.

3.1 Introduction
One of the most significant competitive differentials in organizations is based on the abilities of their
individuals and teams to create new assets (Omotayo 2015). Therefore, to promote and preserve the
knowledge, as well as provide efficient communication tools is crucial to innovative enterprises. In the
Cyber-Physical Production Systems (CPPS), there is high connectivity and cooperation between
humans, machines, and systems within production systems and with their environment as a whole. In

1 Professor Jonice Oliveira, Ph.D., UFRJ [https://dblp.org/pers/hd/o/Oliveira:Jonice]

Professor Claudio Miceli de Farias, Ph.D. UFRJ [https://dblp.org/pers/hd/f/Farias:Claudio_M=_de]
Professor Sergio Manuel Serra Cruz, Ph.D. UFRRJ
[https://dblp.org/pers/hd/c/Cruz:S=eacute=rgio_Manuel_Serra_da]
these smart factories, their smart devices and smart connected products can exhibit some behavior,
interpret information, and have some level of autonomy. All of these conditions require information
anywhere, anytime, for any user using any device, and be accessible in different formats ( Vernadat et al.
2018).

Considering that this workplace produces a large amount of heterogeneous data from different
sensors networks in a quickly changing environment, to provide human workers with more accurate and
relevant information is one of the great challenges of the area, especially to comprehend all facets
related to the information and recover all involved knowledge that could be helpful to get insights to
innovation and improvements.

To provide human workers with more accurate and relevant information is one of the significant
challenges of the area, especially to comprehend all facets related to the information and recover all
involved knowledge that could be helpful to get insights to innovation and improvements. In this way,
the IDEAaS (Interactive Data Exploration As a Service) framework provides a set of solutions based on
clustering algorithms to support the representation of collected data, evaluation mechanisms to display
only relevant data to experts, and multi-dimensional arrangements to aid data exploration according to
diverse analysis dimensions. It was designed to gather, structure, and analyze data through
summarisation techniques that provide information using data syntheses. Those syntheses are modeled
into a multi-dimensional model to aid the representation of unexpected behavior (Bagozi et al. 2019).

The proposal is to create a lessons-learned base using big data. For that, we can apply
visualization models to increase the experts' comprehension give them insights about the domain. This
method is so-called data-driven group storytelling: an approach based on a collaborative procedure to
aim the facets identification, and the pattern recognition grounded in the relates made by people evolved
in experiences observed.

3.2 Methodological Approach

The project has been conducted using the design science research (DRS) method. The DSR has
principles, practices, and procedures necessary to conduct scientific work, ensuring consistency with
previous literature, through defined steps for the research and a mental model to present and evaluate
the results achieved. The DRS execution adopted is a way to shorten the distance between industry and
academia. The cycle solve-problem supports specific situations that can be generalized in problem
classes (Aken et al. 2012), as shown in Figure 1.

The first step in the process is characterizing all aspects of the problematic situation. This
activity includes the outline of contextualization, gaps, main challenges, and trends. The problem
understanding will be useful to the development of artifacts, that will relate identified difficulties with their
reasonable solutions, based on the literature. The analysis and diagnosis phase searches for internal and
external specificities of the study object, aiming to get all aspects that surround the problem. Then, the
solution design supports all process of the artifact construction. The solution must be evaluated
concerning the circumstances. In the end, the learning and evaluation phase conducts to a reflection
about the whole cycle. The result evaluation supports the planning of the other iterations, aiding in the
researcher next actions.
 Figure 1 Problem-solving cycle (Aken et al. 2012)

3.3 Project Schedule

The activity plan was initially designed to have two DRS cycles, illustrated in Table 1. The first cycle is a
problem awareness step, in which we conducted a literature review and searched some possibilities of
new studies. The first cycle lasted two weeks and allowed the definition and scope delimitation for the
problematic situation.

The second cycle is a proposition step. According to the results obtained in the first cycle, we
chose one specific challenge in the problematic situation, and then it was proposed a solution. Our
taken issue was knowledge recovery in Industry 4.0. The suggested approach combines visualization
techniques to display context information and group storytelling to gather experts' testimonies.

Table 1 Activities Planning

Activities/Iterations Week 1 Week 2 Week 3 Week 4

Activity 1 x x

Activity 2 x

Activity 3 x

Activity 4 x

First DSR cycle

Activity 1 - Literature review
Expected artifacts: an outline overview of Industry 4.0, its trends and challenges.
Duration: 2 weeks.
Observations: As it was an adaptation phase, I dedicated two weeks instead of only one. The
biggest challenge was to acquaint better the concepts of Industry 4.0 and its related
technologies, once my original area is big social data.
Activity 2 - Exploring the knowledge recovering issue
Expected artifacts: appointments of related works and a draft of a proposal.
Duration: 1 week.
Observation: It is done. We collected some papers related to the contextualized visualization
that we plan to use to display CPS information to experts. We also searched for some articles
about knowledge-based tools in Industry 4.0.

Second DSR cycle

Activity 3 - Architecture proposition
Expected artifacts: a high-level scheme of our approach. A description of parts of the purposed
layer.
Duration: 1 week.
Observation: We defined a group storytelling approach to gather users' intuition. The project
was presented to professor Bianchini and it had good acceptance.

Activity 4 - Project requirements

Expected artifacts: a draft of an article depicting the parts of the architecture. Description of
some features to the group storytelling support system.
Duration: 1 week.
Observation: This cycle is planned to define the next steps in the research, including the remote
work that will be developed in cooperation of the University of Brescia and UFRJ.

4. Project Execution

4.1 1st DRS Cycle - Problem Awareness

In the first iteration, it was done a literature review about the concepts of Industry 4.0, Cyber-physical
systems (CPS), Big data, Expert-based Intuition, contextualized visualization, data-driven storytelling,
and group storytelling. The elements of Industry 4.0, CPS, and Big Data are parts of the problem
context. While expert-based intuition, contextualized, visualization, data-driven storytelling, and group
storytelling are the core of the designed solution.

There is a strong association between the search terms. Industry 4.0 encompasses several
technologies and associated paradigms that allowed high connectivity and cooperation between
humans, machines, and systems within production environments.

Among other technological advances that paved the era of Industry 4.0, the concept of "Internet
of Things" (IoT) and Cyber-physical systems (CPS) stand out. The IoT addresses the connectivity
existent in physical objects able to collect and transmit data. The advance in the network sector made it
possible for the interconnection of a large number of highly heterogeneous devices, from computers to
laptops, and smartphones. Nowadays, wireless networks are composed of sensor nodes, able to collect
physical world data (Nunes et al. 2018). The CPSs are responsible for monitoring physical processes,
creating a virtual copy of the physical world, and making decentralized decisions. The virtual
environment of CPSs produces a large amount of data, that results in research trends in the Big Data
area. One of these challenges is to make experts able to take decisions in unexpected situations based
on their long-term experience and data visualization (Lu 2017).

The data science has been playing an important role in providing meaningful value to decision-
making from big data (Xu et al. 2018). Its techniques are developed and employed in a cyber-physical
production network to analyze massive amounts of data generated from IoT applications and existing
ICT systems. On the other hand, there is a boundary to the comprehension provided by statistical and
computational models. As a way to explore the potentiality of this big data asset, we are proposing a
knowledge management approach that intends to offer insights from teams' experience.

The collaboration is an essential strategy for dealing with problems that are too complex to
handle individually. This is the case of knowledge recovery, which presents sufficiently serious difficulties
to justify the adoption of collaborative work strategies. By joining diverse people with distinct
backgrounds, skills, and perspectives to interact, discuss, and reflect together it could create synergies
that allow increased capacity for reflection and a rise in the possibilities of mutual learning (Escalfoni et
al. 2011).

To conduct our work, we formulated the following research question:

How to support the knowledge management in unexpected situations

occurred in Industry 4.0 context?

Our main conjecture:

The knowledge recalling from previous episodes constitutes an essential

strategy of work understanding, as well as influencing its practices and how it
is done.

Objectives:

● To analyze the knowledge management applications in Industry 4.0;

● To contribute with new knowledge management approaches in data-
driven contexts.

4.2 2nd DRS Cycle - Proposition

Industry 4.0 highlights the prominent role of knowledge to the competitiveness and sustainability of the
business. It occurs because a large part of the value of the products derives from creativity and
knowledge assets due to accelerated technological development (Nunes et al. 2018). Thus, the
appropriate management of this asset can make all the difference to organizations (Dias et al. 2018).

Even in the face of all technological progress of CPSs, there is a research trend in Industry 4.0
involving the resilience of smart factories (Xu 2017). The resilience of the smart factory is a property that
defines the system tolerance to disruptions (GTAI 2014). These efforts can demand interdisciplinary
research to aid the development of a resilient industrial ecosystem (Xu et al. 2018).

Therefore, we proposed a knowledge management layer to the IDEAaS framework to broaden

the comprehension of previous events based on the experts' background. Our proposal utilizes a
contextualized visualization model to collect testimonies of specialists to assemble a collective story
applying group storytelling technique. In other contexts, the building of lessons-learned base from past
episodes has been considered a helpful strategy for understanding the work, its practices, and how it is
carried out (Escalfoni et al. 2011; Vivacqua and Borges 2012).

The Collective Knowledge of Experts

Experts knowledge represents the intuition and wisdom applied to deal with unexpected situations,
adapt to uncertainty, providing resilience to systems, and work with resource constraints or gaps to their
achievements (Dias et al. 2018). The importance of experts knowledge refers to the collection of patterns
in the specialists' mind that allows taking a better decision than other less experienced workers.
Moreover, when specialists leave their work teams, it causes significant losses for organizations (Neiva
and Borges 2019).

Thus, as a way to improve organizational processes and products and to save operating costs,
is crucial implements effective knowledge sharing mechanisms. There are two forms of knowledge: tacit
and explicit, which the explicit knowledge represents the organizational asset, codified by documents,
texts, databases, reports. While the tacit knowledge consists of the inherited abilities, developed by
individual experiences and used to act (Dias et al. 2018). The tacit knowledge is hard to be formalized,
and consequently, difficult to transfer. Mostly because this kind of knowledge is embedded in and
dependent on many factors, such as culture, identity, routines, policies, as well as individual employees
(Yih-Tong and Scott 2005).

So, the capture and maintenance of tacit knowledge represent a significant challenge because it
is not logical and strictly documented. Carminatti et al. (2006) recommend transforming tacit into explicit
knowledge through the recalling of episodes that occurred in the past and which has been witnessed by
a group of people. This group can produce parts of a story to try to reconstitute the event.

The collective approach to recovering knowledge is necessary because rarely the recalling of
relevant information due to a single person who witnessed the events or participated in them. The
combination of the different point of views of the work with varying levels of expertise makes the
collective knowledge wealthier than the sum of individual pieces of knowledge because it rises
connections that are not perceived when the testimonies are presented isolated (Vivacqua and Borges
2012).

However, the collaborative effort of knowledge gathering has some barriers that do not exist in
the individual dynamics. These difficulties have social or cultural reasons, such as resistance to share,
relationship problems, conflicts, among others. For that, the report of an experience can have four
versions, as illustrated in Figure 2 (Carminatti et al. 2006):

1. The preserved version in the people's mind who witnessed or participated in the episode as a
whole or some event of the episode;
2. The reported version by these people, in other words, the externalization of their tacit
knowledge;
3. The known version by these people, the set of knowledge that the participants have.
4. The real description of the events, all events that form the episode. Usually, this version does not
exist because of missing information.
 Figure 2 The Four Versions of an Episode (Carminatti et al. 2006)

Group Storytelling
Several studies state that the group storytelling approach usually produces a knowledge recalling
process richer than individual dynamics (Carminatti et al. 2006, Escalfoni et al. 2011, Vivacqua and
Borges 2012). The technique consists of the collaboration of a group to recover a story, synchronously
or asynchronously, in the same or a different place. If the storytellers have the opportunity to discuss
with the others, the knowledge will be enriched with the different point of views of the members.

Because each involved in the storytelling process can contribute to distinct ways, it is natural
that his or her point of view is different from other people. Therefore, it is crucial to define a structure to
tell a story to aid the group assimilates it more easily. The story can then be divided into fragments,
where the narrator does not need to remember all events. In this way, the storyteller does not have to tell
the whole story at once; s(he) can do this gradually, as s(he) remembers (Carminatti et al. 2006).

In the semantics pre-defined by Carminatti et al. (2006), the description fragments present
information about people, places, artifacts, and activities performed. The fact fragments describe
accurate information related to an event that can not be answered. The statement fragments are those
that tell details of what happened, according to Figure 3.
 Figure 3 The Structure of a Story in the Group Storytelling (Carminati et al. 2006)

There is no consensus about the utilization of Group Storytelling technique. The method
depends on the scenario and the intended objectives. Thus, to improve knowledge recovery, some
studies recommend some distinct phases: pre-dynamic, dynamic, and post-dynamic (Escalfoni et al.
2011, Vivacqua and Borges 2012). The first phase defines the participants, identifies available
environmental data, and sets up the system to collect narratives. During the dynamic, the storytellers
can expose and discuss their memories collectively. Finally, the post-dynamic encompasses the
activities to build a story timeline, identify significant actions, select and join relevant information in
lessons learned database.

Data-driven Storytelling
The creation of stories supported by facts extracted from data analysis has been grown in many ways
recently. The potentiality of the data-driven stories has increased in order the advances in data science
and visualization areas. The data science helps to identify patterns in a large amount of produced data in
our highly connected society while the visualization area makes the data understandable to the general
public. The actions towards data-driven stories can have a significant potential to inform and sharing
knowledge (Riche et al. 2018).

Nowadays, most of the research efforts of visual analytics and big data are focused on the
integration of artificial intelligence methods with visual representation. Those studies basically search to
improve exploratory analysis and data mining mechanisms through machine learning and statistical
models (Thomas and Cook 2005). Regarding the visualization field, the consolidation of development
technologies, such as JavaScripts Web Technologies, enables a broader public to create data
visualizations. Therefore, data-driven research challenges should focus on comprehending the power of
these tools to communicate information.

Our society lives a data-driven era in which enormous amounts of data are generated. Thus,
analyzing this massive and complex dataset is essential to make important discoveries and
communicate them. In this way, the data-driven storytelling techniques can be useful in the growing data
scenario through interactive visualization systems to support big data storytelling. It can aid our
cognitive ability to understand the data, make reliable and informed decisions (Riche et al. 2018).

Knowledge Management Layer to IDEAaS Framework

The IDEAaS framework provides a complete exploration tool for heterogeneous data from CPS
environments. We are proposing a knowledge management layer to IDEAaS framework, represented in
Figure 4. Our component intends to apply contextualized visualization to display information about an
observed episode to experts. Then, they will participate in group storytelling dynamic to recalling all
related knowledge. We believe that this component can be useful to support experts' insights, avoiding
an undesirable situation, comprehending peculiarities of processes and products, mitigating risk
situation, and providing more resilience to systems.

Figure 4 The Proposed Knowledge Management Layer

One of the biggest problems for knowledge workers today is access to the right information at
the right time. In this way, our approach predicts that the information will be displayed according to the
context. For this, the component will have a repository of visualization models to be chosen according to
a set of user characteristics. As a result, the content can be effectively learned, assimilated, and used
(Godehardt 2009).

The process to recall knowledge from an episode is based on the following steps:
 1. Select an unexpected situation: it encompasses all preparation activities, such as the database
configuration; identification and invitation of participants to the dynamic; definition of duration and
objectives of the dynamic.
2. Define visualization model: it applies a strategy to choose the best contextualized visualization
model to display information to experts
3. Group storytelling: it gathers fragments of stories of the information. It builds the complete story,
proposing a discussion about the episode.
4. Post-dynamic: It identifies the relevant knowledge, adding the lessons learned on the knowledge
management repository.

5. Conclusions and Future Works

The new era of Industry 4.0 encompasses several technologies and associated paradigms that allowed
high connectivity and cooperation between humans, machines, and systems within production
environments. In these integrated environments, or Cyber-Physical Production Systems (CPPS), the
smart devices and smart connected products can exhibit some behavior, interpret information, and have
some level of autonomy. It attends the agile and dynamic requirements of production and improves the
effectiveness and efficiency of the entire industry (Lu 2017).

Despite all technological advances, there are some Industry 4.0 challenges related to the
experts' knowledge, mainly associated with the resilience of smart factories. In the face of this issue, we
proposed a knowledge management layer to the IDEAaS framework to broaden the comprehension of
previous events based on the specialists' background.

Our component is data-driven storytelling based that uses collected testimonies from specialists
to assemble a collective story, assembling a lessons-learned base. This method has been considered a
helpful strategy for understanding the work, its practices, and how it is carried out (Escalfoni et al. 2011;
Vivacqua and Borges 2012).

This report describes work in progress using design science research as a scientific method.
Currently, we have developed two whole cycles, producing an initial literature review and an emerging
proposal of a component to consolidated research - the IDEAaS framework. It is necessary to deepen
the studies, especially on contextualized visualization, identifying applicable models. In further cycles, it
is also necessary to elicit the requirements to the entire module, focus on the data-driven storytelling
approach as a way to support the comprehension of big data.

6. References
AKEN, J. E. V.; BERENDS, H.; BIJ, H. V. D. Problem solving in organizations. Cambridge: University Press
Cambridge, 2012.

BAGOZI, A., BIANCHINI, D., ANTONELLIS, V., GARDA, M., and MARINI, A.: A Relevance-based approach for Big
Data Exploration. Future Generation Computer Systems (2019).

CARMINATTI, N., BORGES, M. R. S., and GOMES, J. O.: Analyzing Approaches to Collective Knowledge Recall.
Computing and Informatics, vol. 25, 2006, pp. 547-570.
[http://www.cai.sk/ojs/index.php/cai/article/viewArticle/358].

DIAS, A., FRANÇA, J. B. S., NEIVA, F. W., BORGES, M. R. S., VIVACQUA, A. S.: Towards the Identification of
Cognitive Barriers to Knowledge Transfer in Teamwork. In Proceedings of the 2018 IEEE 22nd International
 Conference on Computer Supported Cooperative Work in Design (CSCWD 2018), pp. 731-736, 2018,
[https://doi.org/10.1109/cscwd.2018.8465277]

ESCALFONI, R., BRAGANHOLO, V., BORGES, M.: A Method for Capturing Innovation Features using Group
Storytelling. Expert Systems with Applications 2011, vol. 38, pp. 1148-1154.
[http://dx.doi.org/10.1016/j.eswa.2010.05.008]

GODEHARDT, E.: A Framework for Contextualized Visualization Supporting Informal Learning. iJAC, vol. 2(3), Aug.
2009. [https://doi.org/10.3991/ijac.v2i3.996]

GTAI (Germany Trade & Invest): Industries 4.0 - Smart Manufacturing for the Future. Berlin, 2014.

KOUBÂA., A, ANDERSSON, B.: A vision of cyber-physical internet. in Proc. of the Workshop of Real-Time
Networks (RTN 2009), Satellite Workshop to (ECRTS 2009), 2009.

LU, Y. Industry 4.0: A survey on technologies, applications and open research issues. Journal of Industrial
Information Integration 6 (2017) 1–10. [http://doi.org/10.1016/j.jii.2017.04.005].

NEIVA, F. W., BORGES, M. R. S.: Towards Expertise-based Intuition Sharing. In: Proceedings of the 17th European
Conference on Computer-Supported Cooperative Work: The International Venue on Practice-centred
Computing and the Design of Cooperation Technologies - Demos and Posters, Reports of the European
Society for Socially Embedded Technologies (ISSN 2510-2591), 2019.
[https://doi.org/10.18420/ecscw2019_p11]

NUNES, D., SILVA, J. S., BOAVIDA, F.: A Practical Introduction to Human-in-the-Loop Cyber-Physical Systems.
John Wiley & Sons, Incorporated, 2018.

OMOTAYO, F. O., "Knowledge Management as an important tool in Organisational Management: A Review of

Literature". Library Philosophy and Practice. 1238, 2015. [http://digitalcommons.unl.edu/libphilprac/1238]

RICHIE, N. H., HURTER, C., DIAKOPOULOS, N., CARPENDALE, S.: Data-Driven Storytelling. New York: Taylor and
Francis Group. (ISBN 978-1-138-48225-8), 2018.

RUDTSCH, V., GAUSEMEIER, J., GESING, J., MITTAG, T., PETER, S.: Pattern-based Business Model
Development for Cyber-Physical Production Systems. In Proceedings of the 8th International Conference on
Digital Enterprise Technology - DET 2014. Procedia CIRP 25 (2014) 313-319.

THOMAS, J. J., and COOK, K. A., eds. Illuminating the Path: The Research and Development Agenda for Visual
Analytics. Los Alamitos, CA: IEEE CS Press, 2005.

VERNADAT, F. B., CHAN, F. T. S., MOLINA, A., NOF, S. Y., PANETTO, H.: Information systems and knowledge
management in industrial engineering: recent advances and new perspectives, International Journal of
Production Research, vol. 56 (8), pp. 2707-2713, 2018. [https://doi.org/10.1080/00207543.2018.1454615]

VIVACQUA, A. S., and BORGES, M. R. S.: Taking advantage of collective knowledge in emergency response
systems. Journal of Network and Computer Applications, vol. 35, pp. 189-198, 2012.
[https://doi.org/10.1016/j.jnca.2011.03.002]
XU, L. D., XU, E. L., and Li, L.: Industry 4.0: State of the Art and Future Trends. International Journal of Production
Research, vol 56(8), pp. 2941-2962, 2018, [https://doi.org/10.1080/00207543.2018.1444806]

YIH-TONG SUN, P., and SCOTT, J. L.: An investigation of barriers to knowledge transfer. Journal of Knowledge
Management, vol 9 (2), 7590, 2005.