Received 3 August 2020; revised 30 September 2020; accepted 7 October 2020.

Date of publication 16 October 2020;

date of current version 5 November 2020. The review of this paper was arranged by Associate Editor Valeriy Vyatkin.
Digital Object Identifier 10.1109/OJIES.2020.3031660

Quo Vadis Industry 4.0? Position, Trends,

and Challenges
PAULO LEITÃO 1 (Senior Member, IEEE), FLÁVIA PIRES 1 ,
STAMATIS KARNOUSKOS 2 (Senior Member, IEEE), AND ARMANDO WALTER COLOMBO 3 (Fellow, IEEE)
1
Research Centre in Digitalization and Intelligent Robotics (CeDRI), Instituto Politécnico de Bragança, Campus de Santa Apolónia, 5300-253 Bragança, Portugal
2
SAP, 69190 Walldorf, Germany
3
University of Applied Sciences Emden/Leer, 26723 Emden, Germany
CORRESPONDING AUTHOR: PAULO LEITÃO (e-mail: [email protected])
This work was supported by FCT – Fundacao para a Ciencia e Tecnologia within the Project Scope UIDB/05757/2020. The work of Flávia Pires was supported by
the Fundacao para a Ciência e Tecnologia (FCT), Portugal, for the Ph.D. under Grant SFRH/BD/143243/2019.

ABSTRACT Industry 4.0 vision and its mandated digital transformation are radically reshaping the way
business is carried out and the way overall industrial processes and collaborations are operating. In this
work, the objective is to analyze the current level of adoption of Industry 4.0, via the footprint available
in industrial and academic research works. The analysis performed reveals insights on how Industry 4.0
has impacted and is still influencing research and innovation in industrial systems, services, and business
approaches. It also reveals pertinent trends on key enabling features, technologies and challenges associated
with this 4th industrial revolution, mainly focusing on the pathways for wider industrial adoption of Industry
4.0-compliant technologies and solutions.

INDEX TERMS Industry 4.0, cyber-physical systems, digital transformation, industrial digitalization.

I. INTRODUCTION 4.0” in Italy, “Smart Industry” in the Netherlands, “Catapult”

Three industrial revolutions have radically reshaped the in- in the UK, “Alliance Industrie du Futur” in France, and “I40”
dustrial world, and most notably, the manufacturing sector. in Portugal. Other initiatives outside Europe are also promot-
The 1th industrial revolution was related to the introduction of ing similar visions, such as “Industrial Internet of Things”
mechanical production equipment driven by water and steam in the USA, “Made in China 2025” in China, and “Robot
power, the 2th one was based on industrial mass production Revolution Initiative” in Japan. The maturity level among the
organization and the use of electrical energy, and the 3th one countries shows significant differences even if one concen-
was based on the introduction of electronics and Information trates on a single region, e.g., Europe [5]. There are several
and Communication Technologies (ICT) in order to automate definitions of what Industry 4.0 entails [6]–[8], however, for
the production and machinery. Currently, we are witnessing this work, we do not focus on the subtle differences among
the 4th industrial revolution that is related to the smart au- the definitions or the country-specific innovation initiatives,
tomation based on the use of Cyber-Physical Systems (CPS) but commonly consider all of them under the umbrella term
in industry [1]–[4], which are complemented with the Internet “Industry 4.0”, as they share common ground towards a new
of Things (IoT) and Artificial Intelligence (AI) technologies, sophisticated infrastructure that will be based on Industrial
that are customized for industrial application. Cyber-Physical Systems (ICPS) that empowers industrial pro-
This digital transformation vision towards the factory of the cesses, services, systems, and applications.
future (FoF) is being promoted worldwide with governmen- The domain of the Industry 4.0 is complex and needs to
tal initiatives regarding strategies, research, and innovation be approached via a multi-angled view, as its implications
programs. Such efforts are included under the umbrella of pertain to business, technical, and socio-economic aspects [9].
country-specific initiatives such as “Industrie 4.0” in Ger- Significant economic benefits are expected from Industry 4.0,
many, “Industria Conectada 4.0” in Spain, “Piano Industria e.g., it is estimated that the digitization and Internet-based

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
298 VOLUME 1, 2020
networking of products and services generate approximately
110€ billion of additional revenues per year for the European
industry [10]. On the business side, an increase of 45–55% is
expected in the productivity of technical professions through
the automation of knowledge work, a reduction of 30–50%
of total machine downtime, a reduction of 20–50% in time to
market and a reduction of 10–40% of maintenance costs [11].
Furthermore, a recent survey of more than 2000 companies
from several industrial production sectors, claims an increase
in the level of digitization and networking in the industrial
companies by 2020 from 33% to 72% was predicted [10],
while others [12] point to an adoption rate of around 50%
of Industry 4.0 principles, concepts, and technologies until
FIGURE 1. Data extraction and data analysis scheme.
2035 for the Western European countries. Such expectations
are also of interest when they intersect with the adoption in
small and medium-sized enterprises [13], which for instance,
in Europe reflect the majority of companies. Because Industry analysis is performed, and insights are presented, including
4.0 is about value creation and not only optimizing existing trends in industrial research. In addition to the performed anal-
methods and processes, these examples point towards the sig- ysis, challenges are discussed that emerge from the analysis,
nificant influence that is expected to benefit from Industry 4.0, and that is expected to be in focus for the next years.
and which have implications, e.g., towards economic growth, The paper is organized as follows: After the introduction
innovation, and job creation. in Section I, the methodology used to gather the material and
In the Industry 4.0 context, several aspects come into perform the analysis is discussed in Section II. Based on the
play, including virtualization of products and services, new dataset retrieved from scientific repositories, the application
business models that are data-driven, and creation of col- of automated techniques and human analysis, Section III and
laborative ecosystems that bring business value [14], [15]. Section IV summarize the findings. Section V points out the
To realize this, autonomous and modular ICPS in produc- most critical challenges that emerge, while Section VI rounds
tion is needed, that can exhibit self-X features such as self- up the paper with conclusions and an overview of potential
organization, self-optimization, self-adaptation, etc., all of further directions.
which are vital to providing capabilities for decentralized and
autonomous decision-making, connectivity and interoperabil- II. METHODOLOGY
ity, modularity, service-orientation, optimized and real-time The aim of this work is to extract insights about the current
decision-making, virtualization, and human-machine integra- position, trends, and future challenges that pertain to Industry
tion. The realization of these capabilities implies the complete 4.0. To do so, an analysis of several articles that contained
digitization and networking of the industrial manufacturing specific Industry 4.0 relevant keywords has been performed.
sector, which will require the use of several enabling tech- Automated methods to search scientific repositories, and ap-
nologies, namely Big Data, IoT, Edge, and cloud computing plying ML techniques combined with human analysis, were
to support the data computational power and connectivity, Ma- employed.
chine Learning (ML) and advanced data analysis algorithms The dataset was acquired from well-known databases such
for the analytics and intelligence, Virtual Reality (VR) and as Scopus, WoS, and IEEE Xplore, which cover well all pub-
Augmented Reality (AR) for the human-machine interface, lications from major scientific publishers, and also include re-
and additive manufacturing and collaborative robotics for the sults from others, e.g. ACM and Springer, that are not directly
digital-to-physical conversion [16]. accessed. Such analyses utilizing these document repositories
Addressing the challenges of Industry 4.0 is a continuous is common in text-based analysis [17]–[21]. Specifically, doc-
process, which is well reflected in industrial research efforts uments were selected containing the case-insensitive as well
published in leading scientific journals, magazines, confer- as variations of keywords: ”industry 4.0” OR “industrie 4.0”
ences, workshops, technical reports, standards, etc. It is of in- OR “fourth industrial revolution” OR “I4.0”, as illustrated in
terest to see how Industry 4.0 aspects are reflected in literature, Figure 1. The query searches these terms in the title, abstract,
and attempt to identify the current level of adoption of Indus- and keywords of the papers. The time interval chosen for this
try 4.0 principles and Industry 4.0 key enabling technologies, search was between 2013 and 2020, resulting in a dataset
as well as its impact in the industry. This work attempts to containing 13.636 documents after the removal of duplicated
provide some insights towards these directions, i.e., see the entries. The size of the dataset is, therefore, larger than what
current adoption of Industry 4.0, its implications in industrial is analyzed in similar studies, e.g., [17]–[21].
and academy communities, and acquire insights on how it is The dataset analysis was conducted with various tools, in-
reshaping the industrial research and operational efforts in cluding custom Python scripts and utilization of the OpenRe-
the industry. A systematic keyword-based literature review fine tool for dataset cleanup, as well as the VOSViewer for

VOLUME 1, 2020 299

LEITÃO ET AL.: QUO VADIS INDUSTRY 4.0? POSITION, TRENDS, AND CHALLENGES

the creation of network maps and visualization. Processing the

dataset included lemmatizing the words as well as determin-
ing the evolution of the type of publications, word frequency,
frequency of publications by country, a correlation between
technologies and domain applications, a correlation between
design principles and enabling technologies, and the evolution
of enabling technologies. This approach was used for all re-
sults presented in the paper, e.g. the analysis of technological
trends in Figure 12, performed for the periods 2006–2012 and
2013–2019, uses a subset of the dataset containing the specific
technology and timeframe analyzed. FIGURE 2. Overall evolution of Industry 4.0 publications.
There are several limitations pertaining to the methodology
of this work, including the dataset construction. The opera-
tionalized query used for the data extraction from the scientific
repositories considers only the major terms related to Industry
4.0 (as discussed in this section). However, these keywords are
prevalent in Europe (and to a large degree in associated with
the original Industrie 4.0 ideas and results in Germany), and
as such, other worldwide similar programs like the “Industrial
Internet of Things” in the US and the “Made in China 2025”
in China, maybe underrepresented in the dataset. As such, it is
seen as future work to expand the terms further and potentially
link them ontologically with satellite terms, in order to create
a more representative dataset that is more inclusive and has
sufficient representation of other worldwide initiatives.
Another issue is related to the method. Firstly we have con-
FIGURE 3. Detailed view of publication types related to Industry 4.0.
ducted a keyword-only text extraction from the repositories,
and for the search, only specific fields were searched, i.e., title,
abstract, and keywords. This decision was motivated in order
to narrow the search scope, and also it was considered that
if Industry 4.0 was a prevalent topic in the paper, this would
most probably be reflected as a term in one of those fields.
However, this is a limitation, as papers related to Industry 4.0
that do not have these terms in those areas or do not use the
well-known keywords are not considered.
Overall, while we do not claim that the dataset is complete,
nor that the utilized methods are without issues, they are
sufficient and suitable for the scope of this research, as the
derived dataset and its analysis enabled us to extract some
insights concerning Industry 4.0 and utilize them for further
critical discussions in this work. FIGURE 4. Geographical distribution of Industry 4.0 related publications.

III. CURRENT STATUS OF INDUSTRY 4.0 a projection based on the extrapolation performed in Fig-
A. ARTICLE DEMOGRAPHICS ure 2), one can observe the significant increase per publication
The initial analysis aims to understand the demographics of type. The results show continuous growth in the conference
the publications associated with Industry 4.0. Figure 2 illus- papers, but mostly a consolidated increase in journal papers,
trates the evolution of the number of publications referring which reflects and reinforces the relevance of Industry 4.0
to Industry 4.0 over the years. The growing interest of the topics of the scientific community in the last years. Such
research community in the Industry 4.0 topics is evident, es- evolution is expected since early results in conferences take
pecially after 2016. Considering the available up to now data some time to materialize in mature journal papers and also
for 2020, and the exponential trend so far, an estimation is for the community to generate interest in books dedicated to
also made for 2020 (represented by the green box), which is specialized issues relevant to Industry 4.0.
expected to expand further the interest in the domain. It is also of interest to investigate the countries involved
Taking a closer look, as shown in Figure 3, on the distribu- in the Industry 4.0 publications, as these are reflected in the
tion of publications per type (the green box in 2020 represents dataset. The geographic distribution illustrated in Figure 4,

300 VOLUME 1, 2020

 FIGURE 6. Design principles in Industry 4.0 related publications.

shows that Industry 4.0 is related to the application to in-

dustrial systems following the 4th industrial revolution. “Op-
timization”, “interoperability”, “simulation”, “IoT”, “cloud
computing”, “Big Data”, “additive manufacturing”, “virtual
reality” and “digital twin” terms also show the existence of
some important characteristics and technologies associated to
this new paradigm.

FIGURE 5. Co-occurrence keyword map for Industry 4.0 related B. INDUSTRY 4.0 DESIGN PRINCIPLES
publications.
Industry 4.0 is built upon several key design principles,
and among them, an integral role is attributed to the ICPS
paradigm as well as the Asset Administration Shell (AAS)
considers the affiliation of the authors of each article, using
that accompanies the assets along their life cycle [22]. Several
a composite score coming out from all authors listed in the
considerations about the design of Industry 4.0 systems are
publication, with linearly decreasing weights (where the first
reflected in the keywords of the articles.
author weighs more than the second, the second more than the
As shown in Figure 6, optimization, and interoperability,
third, and onwards). The results are normalized according to
followed by decentralization and real-time capability, are the
the maximum achieved value.
most relevant design principles. These results reflect the im-
As expected, Germany is the country that most contributes
portance that the authors have put towards developing decen-
to the publication of Industry 4.0 related articles. This does
tralized but interoperable digitized and networked solutions
not come as a surprise, as Germany very early kick-started its
in the scope of Industry 4.0. These include additional con-
Industrie 4.0 program, which laid detailed research and devel-
siderations, e.g., the realization of systems that can embed
opment (R&D) visions, actions, and targets for academia and
AI techniques to analyze collected data regarding the pro-
industry almost a decade ago. However, Figure 4 also provides
cess optimization, and on occasion, also addressing real-time
additional insights, as it clearly shows that other countries
constraints. The human-machine integration appears just after
within Europe use the same terminology, and from the rele-
these four key design principles, which reflect the importance
vant positions in the graph, one may also relate to the degree of
of the human role in the ICPS life cycle, given that humans can
penetration of Industry 4.0 concepts and ideas for these coun-
be the most flexible piece (asset) in the system. Other aspects
tries. In addition, Figure 4 shows that other countries beyond
are also evident such as service orientation and virtualization,
Europe use the same terminology, that there is cooperation
which is now at a mature level in the industry.
among countries on the domain, and that the phenomenon is
not geographically constrained. It is also possible to observe
that the Industry 4.0 topics are widely spread internationally, C. KEY ENABLING TECHNOLOGIES
mainly to those countries where the government R&D pro- Industry 4.0 is based on the development and application
grams are actively promoting the digital transformation, e.g., of service-oriented business performed using ICPS as a
Italy, United States, China, United Kingdom, and Spain. backbone platform, complemented with several key enabling
The frequency of terms used in the dataset is shown in communication and information technologies, namely IoT,
Figure 5, which covers the 2013-2020 period. The clusters of Big Data, Edge and Cloud Computing, Multi-agent Systems
the terms that have similarities (e.g., being used in the same (MAS), Data Analytics, Additive Manufacturing, VR/AR and
papers) are shown, while the thickness of the links among Collaborative Robotics, among others [10], [16]. These con-
them shows the level of their correlation. As it can be seen, cepts and associated technologies are well-reflected within the
the appearance of “cyber-physical systems”, “smart factory”, dataset, as shown in Figure 7. It has to be noted that some ar-
“industrial internet”, “digital transformation”, “digitalization” eas, e.g., ICPS, encompass diverse concepts and technologies
and “smart manufacturing” as the most common terms clearly themselves.

VOLUME 1, 2020 301

LEITÃO ET AL.: QUO VADIS INDUSTRY 4.0? POSITION, TRENDS, AND CHALLENGES

in distributed systems exchanging data and is aligned with the

international research and innovation programs, e.g., the up-
coming 100€ billion research and innovation program frame-
work “Horizon Europe”, as well as multi-annual roadmaps
and strategic research and innovation agendas of industry plat-
forms [26]. 5G is expected to be the prevalent key enabling
communication technology that will boost even more the use
of IoT, Edge, and link to Cloud technologies, and will also
impact other areas including, for example, the collaborative
robots (as shown in Figure 7).
Figure 9 shows clearly the evolution of these enabling tech-
nologies, and as it can be seen, some have a very steep slope
FIGURE 7. Enabling technologies in Industry 4.0. (e.g., AI and IoT), which indicates the recent rapid increase in
interest within the research community. It is no surprise that
in continuation of the digital transformation envisioned within
As seen in Figure 7, IoT, AI, and Big Data are the key Industry 4.0, the application of IoT as well as AI in industry,
enabling technologies most cited in the Industry 4.0 related are seen as key enabling technologies. Both technologies can
articles, meaning that at this stage, the data collection, storage, be utilized to enhance existing processes or enable the real-
and analysis assumes crucial importance in the digital trans- ization of new ones. Also, others such as Big Data related to
formation process. These technologies can be seen as the pil- Analytics, Digital Twins, and Simulation are well reflected,
lars and the most representative of Industry 4.0, aligned with and they maintain a continuous and consolidated growing,
the key design principles reflected in Figure 6. The growth again indicating a strong interest in them.
of the importance of technologies regarding simulation and By correlating the design principles and enabling technolo-
digital twin evidences their importance in the implementation gies, as shown in Figure 10, some interesting observations can
of the next stage of the digital transformation, namely the be made. Industrial IoT, simulation, AI, MAS, and Big Data
design, implementation, operation, and maintenance of ICPS have a strong presence in the optimization design principle,
and the AAS [23], as defined by the Reference Architectural but also industrial IoT, AI, and Big Data are closely related
Model for Industrie 4.0 (RAMI 4.0) [22]. This relationship to the engineering of the Industry 4.0-systems with real-time
is captured in Figure 8, and provides a coherent way for capability. It is also noticed that blockchain, 3D printing,
the integration and interaction of Industry 4.0- and Industrial cybersecurity, and 5G have not yet significant presence in
Internet-compliant components via AAS within RAMI 4.0. the key design principles, which may be explained by their
[24]. relatively early stage of development, which has not led to
A main concrete objective for research and particularly for their wider adoption yet.
innovation activities, related to the implementation of Industry
4.0-compliant solutions, is the formal technical implemen- D. INDUSTRY 4.0 APPLICATION DOMAINS
tation of the AAS containing the layers 2–6 of the vertical Industry 4.0 comprises several domains such as industrial
dimension (digitalization and networking of assets) within manufacturing, logistics, maintenance, energy, health, agri-
the RAMI 4.0 specification. For achieving such objective, culture, and retail. In order to acquire new insights, we have
the following challenges can be highlighted: (i) The AAS carried out a bigram analysis in order to create the correlation
specifying the entire life cycle (following the IEC 62890) of the design principles and enabling technologies with the
of Assets positioned within the IEC62264 (ISA’95) and/or different domains, and the results are shown in Figure 11.
IEC 61512 (ISA’88), i.e., products, equipment, machinery, It is observed that production and manufacturing are the
and production systems; (ii) The AAS enabling the interoper- two domains that dominate, as most of the technologies and
able exchange of information between value creation partners design principles relate to them. This is also understandable,
within an I4.0-network; (iii) The AAS specifying and imple- since these foster the birthplace of Industry 4.0 efforts in
menting (preparing and structuring) information in a package Germany. Logistics, maintenance, health, and supply chain
for being exchanged between I4.0-components and systems appear as growing applications with some sizable footprint;
(digitalized assets); (iv) Moreover, designed available for non- however, retail and agriculture are less represented.
intelligent and intelligent Assets and containing digitalized
data and information, correctly modeled, the AAS becomes IV. TRENDS IN INDUSTRY 4.0
a real implementation of the “digital twins” of the digital- Another type of analysis conducted aimed to analyze the tech-
ized assets, facilitating cross-vendor interoperability of assets nological trends associated with Industry 4.0 in more recent
through their networking by an I4.0-conform communication years. For this purpose, two consecutive periods were taken
[23], [25]. into consideration: the first period from 2006–2012, which
Cybersecurity, blockchain, and 5G are also noticed in Fig- represents the early era where several Industry 4.0 related
ure 7, reflecting the fundamental importance of these issues technologies were developed (although not explicitly linked

302 VOLUME 1, 2020

FIGURE 8. Asset Administration Shell in the RAMI 4.0

FIGURE 9. Evolution of the key enabling technologies.

FIGURE 11. Correlation between technologies and domain applications.

FIGURE 10. Correlation between design principles and enabling

technologies. FIGURE 12. Comparison of the adoption of enabling technologies for the
2006–2012 (early) and 2013–2019 (contemporary) periods.

to Industry 4.0), and the period 2013–2019, which represents

the more contemporary one. as it is marked by the presentation of the final report of the
Figure 12 provides a comparative view of the two periods, working group on Industrie 4.0 in the Hanover Fair.
before and after the official introduction of the Industry 4.0. The analysis of the achieved results shows that the tech-
2013 was selected as a pivotal point for this segmentation, nologies nowadays associated with Industry 4.0 were already

VOLUME 1, 2020 303

LEITÃO ET AL.: QUO VADIS INDUSTRY 4.0? POSITION, TRENDS, AND CHALLENGES

developed to a degree before the official introduction of In- capabilities offered by the Industry 4.0-compliant solu-
dustry 4.0 paradigm. This comes as no surprise since there has tions. This convincing process starts not only with publica-
been observed a continuous development of ground-breaking tions about research and innovation results but particularly
concepts and technologies, which have set the baseline for the with clear real demonstrations of implemented Industry 4.0-
Industry 4.0 vision. In the contemporary period (2013–now), compliant solutions in lighthouse use cases that show tangible
all these enabling technologies have registered a significant in- benefits for the adoption of the new technologies, models, and
crease of attention by the research community, with particular ways to conduct business. Currently, in Europe, there are sev-
emphasis on IoT, Big Data, Blockchain, and 3D printing. eral research and innovation actions that are targeted towards
It may, therefore, be possible to consider that the presen- established industrial research agendas such as the ECS SRA
tation of the Industry 4.0 vision, as well as the inclusion of [26] and aim at demonstrating results that are on the high end
selected promising and suitable technologies in Industry 4.0 of the European Commission’s Technology Readiness Level
context, also acted as a booster to the further use and devel- (TRL) [28]. However, even with European-wide initiatives,
opment of these enabling technologies for ICPS. In a sense, there are significant disparities among European countries [5].
from 2013 onwards, we have a significant focus on this new
era of enabling technologies and concepts that ride the vision
V. FUTURE CHALLENGES
set by Industry 4.0 and which have significantly contributed
Several challenges/barriers related to Industry 4.0 [7], [8],
to the digital transformation in the industry, as this is reflected
[20], [21] and more specifically ICPS [2]–[4] have emerged.
in innovation and industrial research.
Combining such challenges, with the results of this analysis
Considering the analysis of the key enabling technologies
but also the essential information issued in the Multi-Annual
discussed in Section III, as well as key roadmaps set out by
Plans (MASPs) and Strategic Research and Innovation Agen-
industry associations, we can acquire additional insights on
das (SRIAs) of Industrial Associations and Platforms like
the convergence of visions as well as potential influencing
[26], [27], [29], enables the calibration of such insights. Ta-
technologies and design principles that that may impact the
ble 1 summarizes some of the most critical issues to imple-
next decade. Several roadmaps exist at the national and in-
ment Industry 4.0 and points out the main actions required
ternational levels, and an analysis of them is seen as future
to address each one of these challenges. The findings below
work. In Europe, the best representative is the Multi-Annual
are in line with similar systematic literature reviews [17],
Plan (MASP) [27] and the Strategic Research and Innovation
[19]–[21].
Agenda (SRIA) [26] stemming from three European industry
associations, i.e., AENEAS, ARTEMIS-IA, and EPoSS, that
represent Electronic Components & Systems (ECS) stake- A. EDUCATION, NEW SKILLS, AND JOB PROFILES
holders. The specific SRIA clearly delineates the challenges The success of Industry 4.0 strongly depends on the skills and
and strategic priorities related to Industry 4.0-compliant solu- competences that the workforce can have along the different
tions and Industrial Cyber-Physical Systems for the European dimensions of the new vision, and mainly related to the devel-
industry in order to become globally competitive and have a opment and implementation of key enabling technologies, as
beneficial societal and economic impact. As such, the SRIA well as the development and use of the new business process
[26] captures well the game-changing directions towards a that the new digitization and networking paradigm proposes.
number of key domains such as transport & smart mobility, Expertise and new skills are now crucial to implement this
health & well-being, energy, digital industry, digital life, sys- vision and enable the acceptance of Industry 4.0, including
tems & components, connectivity & interoperability, safety– finding qualified personnel [30]. It should be noted that several
security–reliability, computing & storage, process technology, job profiles such as information and digitization/digitalization
equipment, materials and manufacturing for ECS, etc. engineer, big data analyst, data scientist, machine learning en-
Another way to acquire additional insights beyond the in- gineer, and cloud services experts, did not exist ten years ago
dustry roadmaps is to see specific efforts that are in the core [31]. This is also attested in a recent analysis of Industry 4.0
of Industry 4.0, such as the RAMI 4.0 [22] and AAS [23]. The related job profiles [32]. Enabling the workforce to acquire the
analyzed dataset showed a significant increase in references to required knowledge [33] and skills to respond to challenges
RAMI4.0 aspects in the literature after 2016, the year also the and expectations of the industry when implementing the In-
RAMI 4.0 was standardized in DIN [22], which denotes the dustry 4.0, as well as migrating legacy industrial systems into
growing interest in it. Industry 4.0-compliant systems are seen as vital. Such know-
Industry 4.0 is about digital transformation, which, in how requires multidisciplinary knowledge [33] and skills,
conjunction with digitization and digitalization, aims to mi- which need to be acquired via the educational systems [34],
grate industry to a new era of ecosystem-wide interactions as well as complementary training and lifelong learning for
and increased optimization of its business processes via the employees. Especially considering the effects of COVID-19
usage of ICT. These processes imply changes and adjust- in business, the trend for online learning and training via
ments for all actors along the industry value stream and Massive open online courses (MOOCs) [35] is reinforced,
value chain, and associated business process [15]. For such which fits well with other complementary activities such as
change to happen, the stakeholders must be buy-in to the new VR for employee training [36].

304 VOLUME 1, 2020

TABLE 1. Challenges and Actions for Implementing Industry 4.0

The new generation of industrial engineers that will work demands. This is challenging, but universities that are at the
on aspects related to Industry 4.0 areas need beyond inte- forefront of research, and have an early glimpse of things to
grating multidisciplinary and cross-domain knowledge to also come, should adjust in a timely manner their offered programs
focus more on understanding the complex dynamic systems and specializations to cover these needs. The matching of pre-
and system of systems (SoS) perspective [37]. They will have dicted industry needs, technology trends, and a plan on how
to be able to navigate and utilize new paradigms and concepts to address them in order to be able to sufficiently cover new
such as SoS Engineering, data and information modelling, job profiles within Industry 4.0-compliant ecosystems [39] is
digitization/digitalization methods and tools, engineering dig- seen as a key to thriving Industry 4.0 utilization [38].
ital twins, simulation of systems and infrastructures, seman-
tics, interoperability, as well as emergent technologies such B. INTEROPERABILITY, STANDARDIZATION
as Industrial IoT, Big Data and advanced data analytics, digi- AND MIGRATION
tized lifecycle engineering. Additionally, in the ICPS context, Interoperability is one central issue for a successful imple-
engineers are not anymore dealing only with the physical mentation of Industry 4.0 [40], being related to the capabil-
counterpart, i.e., hardware, but increasingly with the cyber ity of different systems and services to be able to connect
counterpart of complex industrial systems, which requires an and interact transparently. In the ICPS context, this need for
integrative learning process to understand their interplay, how interoperability emerges from the difficulties of integrating
to link systems in their [15] and what the impacts might be. data from various sources and services that can understand
Lifelong learning is seen as crucial here, since due to the rapid each other’s semantics, across the different layers of RAMI
advances in hardware but mostly in software, the acquired 4.0. Additionally, the migration process conducting to the
knowledge quickly becomes obsolete and needs to be kept integration of legacy systems, both at Operational Technology
pertinent. (OT) as well as Information technology (IT) and between
The job profiles requested for Industry 4.0 related tasks them, assumes crucial importance for the operation of such
show a mixture of soft skills and diverse technical expertise systems, e.g., PLC, RC, CNC, databases, SCADA (Supervi-
in new technologies [32]. Education aims to prepare the pro- sory Control and Data Acquisition), and MES (Manufacturing
fessionals for today’s jobs rather than for ones of the future Execution System) systems [41].
[38] since it is difficult to determine what will be tomorrow’s The achievement of this demand, as well as in general for
jobs. However, the early identification of new job profiles tar- implementing RAMI 4.0-compliant solutions, requires the use
geting future needs is vital to train professionals for emerging of existing industry standards [42], as well as their potential

VOLUME 1, 2020 305

LEITÃO ET AL.: QUO VADIS INDUSTRY 4.0? POSITION, TRENDS, AND CHALLENGES

addressing interoperability, standardization, and migration

dilemmas.

C. VIRTUALIZATION AND DIGITAL TWINS

The application of the Digital Twin [46], which is a crucial
technology in the Industry 4.0 context, will increase the effi-
ciency of current automation systems and companies in 10%
[47]. The Digital Twin is one of the top ten strategic technol-
ogy trends and is expected that by 2023 the global market on
FIGURE 13. Evolution of the cumulative references to standards. this technology will grow by 38% annually, representing $16
billion [48]. Currently, the majority of Digital Twin research
works focuses on manufacturing [46], but also other areas
adaptation to encompass the Industry 4.0 envisioned capa- linked to Industry 4.0 are addressed.
bilities. Particularly concerning connectivity [43] and inter- In order to achieve the mentioned numbers, several actions
operability related issues, standards addressing data models, should be taken. The integration of industrial IoT technologies
e.g., AutomationML, B2MML, and FIWARE, and ICT, e.g., is essential for performing the data collection to feed the Dig-
OPC-UA, REST and MQTT, need to be properly realized in ital Twin. Moreover, the specification [23], development [49],
the enterprise architectures. It is also required the adoption and utilization of the AAS, as a concrete implementation of
of some technological choices in the design process: the use the Digital Twins within an Industry 4.0-compliant solution, is
of service-orientation to encapsulate functionalities as ser- one essential challenge to be approached in the next couple of
vices and loosely-coupled approaches actively contribute to years [50]. The main challenges behind the digital twin tech-
the interoperability and scalability of the system. Industry nology include the interplay with Big Data [51], simulation
4.0-compliant solutions mean working with a set of well- [52], standardization [53], and also the security and privacy
defined standards and norms, starting with those that consti- of the networks created for networking digitized data, virtual
tute the background dimensions of the RAMI 4.0. Analyzing data, and information models [54].
the dataset of Industry 4.0 papers, it is possible to verify a The intelligence behind the decision support is another im-
consistent increase in the adoption of these industry standards portant challenge to be tackled. In this case, the simple use of
over the years, as shown in Figure 13. Among these standards, ML techniques for data analysis may not always be sufficient
the most frequently mentioned are the IEC 62541 – OPC-UA, to achieve the best results, being necessary to be combined
IEC 61131 – PLC OPEN XML, VDMA 24582 – Condition with simulation capabilities. However, there are recent studies
monitoring, IEC 61804 – Function Blocks for process con- that show several limitations in the use of pure AI or sim-
trol, and ISO/IEC 20140 – Energy Efficiency. Note that the ulations approaches for producing knowledge for decision-
numbers provided for the year 2020 (in green) are a projection support [55]. The main problem with a mixed approach is
based on the analyzed data. the difficulty in the integration of the simulation models with
An essential action is the identification of gaps and the AI techniques. Improving the modeling and simulation in
initiation of actions to define new standards or complement digital twins is critical [54], and a possible way to do this
existing ones with the missing functionality. When the new is to integrate MAS applications to enhance the capabilities
enabling technologies, as well as the new requirements im- of generating different scenarios based on the interaction of
posed by such systems, are taken into account, the creation of multiple intelligent entities [56].
a unified approach that builds on existing efforts has the po- Last but not least, beyond security [54], the trust of the user
tential to boost Industry 4.0 while also making the migration is one of the most critical factors since this can dictate the
process towards it easier. The establishment of best practices, acceptance or rejection of the Digital Twin system [57]. For
especially those linked with credible migration strategies, are this purpose, the development of trust mechanisms, supported
needed to support the industry transition from their traditional with AI techniques, assumes an important role in developing
rigid and centralized production systems towards the new de- fully functional and trustworthy systems [58].
centralized, autonomous, and intelligent ICPS infrastructure.
This migration should be an evolutionary process based on the
continuous, phased, and smooth transition of existing systems D. HPC AND PROCESSING INFRASTRUCTURES
and not a disruptive process, to preserve investments that are A crucial challenge to boost the implementation of Industry
already done in brownfield infrastructure [41], [44]. However, 4.0 is the need to have available not only a sophisticated
greenfield investments to new infrastructures offer a unique communication infrastructure [43] but also computational in-
opportunity to avoid the complexities of migration and focus frastructures that store the collected data and run the AI and
on cutting-edge technology, which, however, bears the po- advanced data analytics algorithms, ranging from the edge and
tentially higher risk and cost. Overall cost, performance, and fog levels to High Processing Computing (HPC) infrastruc-
sustainability can serve as useful indicators [45] towards tures for the cloud computing level [7].

306 VOLUME 1, 2020

 In the last years, we have seen the rise of powerful edge F. CYBERSECURITY
devices that provide the computational power to run AI algo- Cybersecurity is an integral path of Industry 4.0 efforts [70],
rithms at the edge, near to the data sources and in real-time [71], especially considering that several systems are utilized
[59]. Since algorithms running on the edge can provide a in critical infrastructures that can be targeted even for po-
fast response to condition change and also reduce the size of litical reasons, as the operationalization of Stuxnet [72] has
data to be sent to the cloud by implementing pre-processing shown. As envisioned in Industry 4.0, the majority of the
methods, and AI algorithms running in the cloud can pro- developed systems, applications, and services go beyond their
vide optimization, prediction, and planning functions. Using utilization in controlled environments, e.g., factories, and may
distributed computational layers, it is imperative to balance expand towards consumer systems and interaction with them.
the intelligence and data analytics correctly among edge, fog, Traditional research focusing on the pillars of information
and cloud layers, while respecting security, privacy and other security, i.e., addressing confidentiality, availability, and in-
federated data concerns. tegrity, is expected to continue. This implies, for instance,
Combined with these computational infra-structures, other the development of new algorithms for encryption, authenti-
key enabling technologies can boost the implementation of the cation/authorization, attack detection, etc., that increasingly
Industry 4.0 principles. As an example, photonics is seen as takes into consideration the unique characteristics, needs, and
one fundamental technology in digitalization, as assumed by emerging trends on infrastructures for ICPS [70]. In parallel,
the European Commission in its digitalization strategy, that additional evolutionary steps are expected that takes the state
could facilitate a change in data-center architectures [60]. Of of the art research and development and advances it around
equally high importance are 5G and its M2M capabilities that ICPS and their ecosystem. As an example, intrusion detection
will provide the means to boost the communication between systems (IDS) for ICPS that are part of critical infrastructures
the distributed entities in ICPS systems [61]. may evolve an adapt themselves in real-time, considering the
behavior and roles of an ICPS within its operational context.
New cybersecurity challenges posed in Industry 4.0 are
E. HUMAN INTEGRATION also expected to be a significant focus area. These challenges
The human integration in ICPS [62] plays a crucial role in the pertain to security, trust, and privacy management in non-
era of the digital transformation, notably because humans are controlled and highly dynamic environments, at ICPS as well
seen as the most flexible driver in an automated system, due to as a system level. Systems of ICPS are expected to interact
their flexibility and capability to solve unexpected situations. and even cooperate autonomously, e.g., autonomous vehicles
However, the full exploitation of the potential of humans in self-regulating traffic at smart cities, etc. Also, due to the
ICPS requires the use of human-centric design approaches, envisioned high utilization of ML in such systems, aspects
combined with technologies for the symbiotic integration of about dependability, trust, and explainability for ML-driven
the human activities in ICPS [63], [64], e.g., using new user- decision making, especially for critical systems, are emerging.
friendly interfaces and VR/AR technologies. Finally, as other technologies evolve, synergies can be created,
Two main models for human activities are usually referred and their interaction with cybersecurity is eminent. Typical
[63]: Human-in-the-Loop (HitL), related to tasks in which the such examples include blockchain and quantum computing,
worker (usually an operator) is directly participating in the both of which have the potential to reshape cybersecurity for
production process and its control loop, and Human-in-the- ICPS. Managing cybersecurity in such diverse and highly so-
Mesh (HitM), related to tasks in which the worker (usually a phisticated infrastructures is expected to be challenging [71],
manager) is participating in the production planning process and therefore it should be addressed from a holistic point of
and its control loop. HitL tasks are more focused on the ex- view that includes technical and non-technical aspects [66].
ecution of the production processes and have an immediate
impact on the quality produced products, while HitM tasks
have a long-term impact on production efficiency and require G. BUSINESS ORIENTATION AND ETHICS
sophisticated methods for planning and simulation based on A crucial issue to be taken into account in the development
data analytics and AI. In the scope of Industry 4.0, equal of such smart ICPS systems is to look and think beyond
importance should be given to both. the technology, taking special attention to business, social-
The integration of humans and the collaboration among organizational, economic, and ethical aspects. Beyond tech-
Industry 4.0 components (that include humans) has several nology, financial and knowledge constraints are found to be
dimensions that include safety [65], security [66], but also key challenges for small and medium-sized enterprises that
ethical considerations [67]. To better understand the role of constitute the majority of Europe’s businesses [13] and simi-
humans in Industry 4.0, it is necessary to perform an im- lar results are also evident in other studies [73]. In addition,
plication and impact research study, from a socio-technical other aspects such as resistance by employees or middle-
angle [68]. In the scope of Industry 4.0 related challenges management need to be considered as they hinder the adoption
and opportunities, new engineering approaches will need to be of Industry 4.0 [74].
developed that are human-centric and focus on the interactions As any other technology or product, a very initial step is to
of humans with the other Industry 4.0 components [69]. convince and empower the stakeholders about the benefits of

VOLUME 1, 2020 307

LEITÃO ET AL.: QUO VADIS INDUSTRY 4.0? POSITION, TRENDS, AND CHALLENGES

Industry 4.0 vision and technologies, identifying where they principles for the near future implementation of Industry 4.0
can help to improve the performance, competitiveness, and compliant systems.
resilience of their systems. As such, Industry 4.0 is inherently The results presented are in line with other similar studies
linked with the circular economy [75]. The application of [17]–[21], while the critical discourse on the challenges re-
digital technologies and the digital transformation of existing vealed additional insights and directions. These challenges are
ICPS and business systems only makes sense if the expected mainly based on the fact that digitized and adequately mod-
return has a positive business impact, e.g., new market oppor- elled data/information is at the disposal for being analyzed
tunities, or may contribute to the improvement of the com- and used to support all functions associated with the different
pany’s competitiveness. This issue is highlighted in the layers phases of the life cycle of the physical and cyber-part of a
of RAMI 4.0 [22], where the top layer, called “Business”, is RAMI 4.0-compliant solutions. Apart from the technological
related to business and organizational processes of the asset or challenges, the business, social-organizational and ethical as-
system to be digitalized [76]. Besides defining how to digitally pects are also pivotal and require to be treated at par with the
transform assets to Industry 4.0 components (via the AAS), it technological ones, if Industry 4.0 concepts and technologies
is crucial also to establish why this asset should be digitalized, are to be accepted.
i.e., which desired business functions should be offered as a
service, and what assets should be digitalized to provide the REFERENCES
desired functions [34]. [1] E. A. Lee, “Cyber physical systems: Design challenges,” in Proc. 11th
Another dimension relevant for modern ICPS, uniquely IEEE Int. Symp. Object Component-Oriented Real-Time Distrib. Com-
put., May 2008, pp. 363–369.
since these are integrated increasingly into modern society [2] P. Leitão, A. W. Colombo, and S. Karnouskos, “Industrial automation
and interact with users, is that of ethical design and decision based on Cyber-Physical Systems technologies: Prototype implementa-
making [67]. As ICPS expects to be fueled by AI, they re- tions and challenges,” Comput. Ind., vol. 81, pp. 11–25, Sep. 2016.
[3] A. W. Colombo, S. Karnouskos, O. Kaynak, Y. Shi, and S. Yin, “Indus-
quire special attention to the boundaries of the autonomous trial cyberphysical systems: A backbone of the fourth industrial revolu-
decisions regarding the ethical frameworks under which they tion,” IEEE Ind. Electron. Mag., vol. 11, no. 1, pp. 6–16, Mar. 2017.
make decisions, as well as the ethical implication of the de- [4] S. Karnouskos, P. Leitao, L. Ribeiro, and A. W. Colombo, “Industrial
agents as a key enabler for realizing industrial cyber-physical systems:
cisions and their effects on humans and society at large [77]. Multiagent systems entering Industry 4.0,” IEEE Ind. Electron. Mag.,
As we increasingly rely on AI and autonomous ICPS, we do vol. 14, no. 3, pp. 18–32, Sep. 2020.
place trust that they will make the best possible decisions; [5] I. Castelo-Branco, F. Cruz-Jesus, and T. Oliveira, “Assessing Industry
4.0 readiness in manufacturing: Evidence for the European Union,”
however, the criteria, ethical frameworks, cultural consider- Comput. Ind., vol. 107, pp. 22–32, May 2019.
ations, different legal and regulatory aspects, transparency [6] G. Beier, A. Ullrich, S. Niehoff, M. Reißig, and M. Habich, “Indus-
and explainability of AI decision, etc., come on top of other try 4.0: How it is defined from a sociotechnical perspective and how
much sustainability it includes—A literature review,” J. Cleaner Prod.,
challenges that need to be resolved [78] when creating ethical vol. 259, Jun. 2020, Art. no. 120856.
machines though is not without risk [79]. [7] A. Raj, G. Dwivedi, A. Sharma, A. B. L. de Sousa Jabbour, and S.
Rajak, “Barriers to the adoption of Industry 4.0 technologies in the
manufacturing sector: An inter-country comparative perspective,” Int.
VI. CONCLUSIONS J. Prod. Econ., vol. 224, Jun. 2020, Art. no. 107546.
The 4th industrial revolution is reshaping the existing indus- [8] G. Culot, G. Nassimbeni, G. Orzes, and M. Sartor, “Behind the defini-
trial production systems through a digital transformation pro- tion of Industry 4.0: Analysis and open questions,” Int. J. Prod. Econ.,
vol. 226, Aug. 2020, Art. no. 107617.
cess, based on the ICPS complemented with IoT and AI to [9] S. Bag, S. Gupta, and S. Kumar, “Industry 4.0 adoption and 10r advance
collect and extract value and knowledge from the massive manufacturing capabilities for sustainable development,” Int. J. Prod.
amount of available data. The Industry 4.0 initiative is the Econ., vol. 231, Jan. 2021, Art. no. 107844.
[10] R. Geissbauer, J. Vedso, and S. Schrauf, “Industry 4.0:
most well-known face of this revolution, attracting a growing Building the digital enterprise,” PWC, London, U.K., Tech.
interest by industry and scientific communities. Rep., 2016, Accessed: Oct. 29, 2020. [Online]. Available:
This paper has analyzed and discussed the current state of https://www.pwc.com/gx/en/industries/industries-4.0/landing-
page/industry-4.0-buildingyour-digital-enterprise-april-2016.pdf
implementation of Industry 4.0, as this is reflected in a rich [11] C. Baur and D. Wee, “Manufacturing’s next act,” 2015. [Online]. Ava-
dataset collected from several literature repositories. Particu- ilable: https://www.mckinsey.com/business-functions/operations/our-
larly this work has presented insights about the trends in the insights/manufacturings-next-act
[12] M. Blanchet and T. Rinn, “The Industrie 4.0 transition
adoption of the design principles and enabling technologies, quantified,” Roland Berger GmbH, Munich, Germany, Tech.
as well as towards the identification of future challenges re- Rep., 2016. Accessed: Oct. 29, 2020. [Online]. Available:
lated to Industry 4.0. The results show that the Industry 4.0 https://www.rolandberger.com/publications/publication_pdf/
roland_berger_industry_40_20160609.pdf
initiative was highly correlated to the increased interest in [13] T. Masood and P. Sonntag, “Industry 4.0: Adoption challenges and
specific technologies and concepts related to ICPS, including benefits for SMEs,” Comput. Ind., vol. 121, Oct. 2020, Art. no. 103261.
industrial IoT, Big Data, data analytics, additive manufac- [14] H. Hirsch-Kreinsen et al., “Key themes of Industrie 4.0: Research and
development needs for successful implementation of Industrie 4.0,”
turing, digitization and networking, service-oriented business Research Council of the Plattform Industrie 4.0, Tech. Rep., 2019,
processes. In particular, ICPS, industrial IoT, and industrial Accessed: Oct. 29, 2020. [Online]. Available: https://en.acatech.de/
AI are those that mostly impact these new systems in the publication/key-themes-of-industrie-4-0/
[15] G. B. Benitez, N. F. Ayala, and A. G. Frank, “Industry 4.0 innovation
future. Additionally, decentralization, interoperability, real- ecosystems: An evolutionary perspective on value cocreation,” Int. J.
time capability, and optimization are the most critical design Prod. Econ., vol. 228, Oct. 2020, Art. no. 107735.

308 VOLUME 1, 2020

[16] H. Bauer, C. Baur, G. Camplone et al., “Industry 4.0: How to Navigate [34] A. W. Colombo, G. Veltink, M. L. Caliusco, and J. Roa, “Learn-
Digitization of the Manufacturing Sector,” McKinsey, New York, NY, ing industrial cyber-physical systems and Industry 4.0-compliant so-
USA: Tech. Rep., 2015. lutions,” in Proc. 3rd IEEE Int. Conf. Ind. Cyber-Phys. Syst., 2020,
[17] Y. Lu, “Industry 4.0: A survey on technologies, applications and pp. 219–224.
open research issues,” J. Ind. Inf. Integr., vol. 6, pp. 1–10, Jun. [35] S. Karnouskos, “Massive open online courses (MOOCs) as an enabler
2017. for competent employees and innovation in industry,” Comput. Ind.,
[18] M. Mariani and M. Borghi, “Industry 4.0: A bibliometric review of vol. 91, pp. 1–10, Oct. 2017.
its managerial intellectual structure and potential evolution in the ser- [36] P. Zawadzki, K. Zywicki, P. Bun, and F. Gorski, “Employee training
vice industries,” Technological Forecasting Social Change, vol. 149, in an intelligent factory using virtual reality,” IEEE Access, vol. 8,
Dec. 2019, Art. no. 119752. pp. 135 110–135 117, 2020.
[19] F. Galati and B. Bigliardi, “Industry 4.0: Emerging themes and future [37] A. W. Colombo, S. Karnouskos, and T. Bangemann, “A system of
research avenues using a text mining approach,” Comput. Ind., vol. 109, systems view on collaborative industrial automation,” in Proc. IEEE
pp. 100–113, Aug. 2019. Int. Conf. Ind. Technol., Feb. 2013, pp. 1968–1975.
[20] F. S. T. da Silva, C. A. da Costa, C. D. P. Crovato, and R. da Rosa [38] S. Fareri, G. Fantoni, F. Chiarello, E. Coli, and A. Binda, “Estimating
Righi, “Looking at energy through the lens of Industry 4.0: A systematic Industry 4.0 impact on job profiles and skills using text mining,” Com-
literature review of concerns and challenges,” Comput. & Ind. Eng., put. Ind., vol. 118, Jun. 2020, Art. no. 103222.
vol. 143, May 2020, Art. no. 106426. [39] L. Sakurada, C. Geraldes, F. Fernandes, J. Pontes, and P. Leitao, “Anal-
[21] L. B. Furstenau et al., “Link between sustainability and Industry ysis of new job profiles for the factory ofthe future,” in Proc. 10th
4.0: Trends, challenges and new perspectives,” IEEE Access, vol. 8, Eur. Workshop Service Oriented, Holonic Multi-Agent Manuf. Syst. Ind.
pp. 140079–140096, 2020. Future (SOHOMA20), 2020.
[22] DIN SPEC 91345: Reference Architecture Model Industrie 4.0 [40] Industrie 4.0 Platform, “2030 Vision for Industrie 4.0: Shaping Dig-
(RAMI4.0), Deutsches Institut fr Normung (DIN) Std., Berlin, Ger- ital Ecosystems Globally,” 2019. [Online]. Available: https://www.
many, 2016. [Online]. Available: https://www.din.de/en/wdc-beuth: plattform-i40.de/PI40/Redaktion/EN/Standardartikel/vision.html
din21:250940128 [41] A. Calà, A. Cachada, F. Pires, J. Barbosa, J. Wermann, and A. W.
[23] Platform-I4.0, “Details of the Asset Administration Shell Part 1 - Colombo, “Migration strategy toward innovative, digitalized, and har-
The exchange of information between partners in the value chain of monized production systems,” in Digitalized and Harmonized Indus-
Industrie 4.0,” Federal Ministry for Economic Affairs and Energy, trial Production Systems. Boca Raton, FL, USA: CRC Press, Oct. 2019,
Berlin, Germany, Tech. Rep., 2020, Accessed: October 29, 2020. pp. 151–176.
[Online]. Available: https://www.plattform-i40.de/PI40/Redaktion/EN/ [42] Fraunhofer, “Industry 4.0 standards website,” 2016. Accessed: Oct. 29,
Downloads/Publikation/Details-of-the-Asset-Administration-Shell- 2020. [Online]. Available: https://i40.semantic-interoperability.org
Part1.pdf [43] G. Aceto, V. Persico, and A. Pescape, “A survey on information and
[24] B. Boss et al., “Digital twin and asset administration shell concepts communication technologies for Industry 4.0: State-of-the-art, tax-
and application in the industrial internet and industrie 4.0,” Plattform onomies, perspectives, and challenges,” IEEE Commun. Surv. Tut.,
Industrie 4.0 and Industrial internet Consortium, Tech. Rep., 2020. vol. 21, no. 4, pp. 3467–3501, Oct.–Dec. 2019.
Accessed: Oct. 29, 2020. [Online]. Available: https://www.plattform- [44] T. Lins and R. A. R. Oliveira, “Cyber-physical production systems
i40.de/PI40/Redaktion/EN/Downloads/Publikation/Digital-Twin-and- retrofitting in context of Industry 4.0,” Comput. Ind. Eng., vol. 139,
Asset-Administration-Shell-Concepts.pdf Jan. 2020, Art. no. 106193.
[25] A. W. Colombo, “Engineering ISoCPS. Digitalizing and Network- [45] D. Gürdür and F. Asplund, “A systematic review to merge discourses:
ing an Industrial Eco-System using the DIN SPEC 91345 / IEC Interoperability, integration and cyber-physical systems,” J. Ind. Inf.
PAS 63088 RAMI4.0. Distinguished Lecture IEEE Systems Coun- Integr., vol. 9, pp. 14–23, Mar. 2018.
cil,” 2020. [Online]. Available: https://ieeesystemscouncil.org/pages/ [46] A. Fuller, Z. Fan, C. Day, and C. Barlow, “Digital twin: Enabling
distinguished-lecturers technologies, challenges and open research,” IEEE Access, vol. 8,
[26] ECS, “Strategic research agenda for electronic components and pp. 108 952–108 971, 2020.
ecosystems,” European industry associations: AENEAS, ARTEMISIA [47] C. Pettey, “Prepare for the Impact of Digital Twins,” 2017. [On-
and EPoSS, Tech. Rep., 2020. Accessed: Oct. 29, 2020. [Online]. line]. Available: https://www.gartner.com/smarterwithgartner/prepare-
Available: https://ecscollaborationtool.eu/publication/download/ecs- for-the-impact-of-digital-twins/
sra-2020.pdf [48] K. Panetta, “Top 10 Strategic Technology Trends for 2018,” 2017. [On-
[27] ECSEL, “Multi annual strategic plan (MASP),” European line]. Available: https://www.gartner.com/smarterwithgartner/gartner-
industry as sociations: AENEAS, ARTEMIS-IA and EPoSS, top-10-strategic-technology-trends-for-2018/
Tech. Rep., 2020. Accessed: Oct. 29, 2020. [Online]. Available: [49] J. Fuchs, J. Schmidt, J. Franke, K. Rehman, M. Sauer, and S.
https://ec.europa.eu/research/participants/data/ref/h2020/other/legal/ Karnouskos, “I4.0-compliant integration of assets utilizing the asset
jtis/ecsel-multi-stratplan-2020_en.pdf administration shell,” in Proc. 24th IEEE Int. Conf. Emerg. Technol.
[28] European Commission, “Technology readiness levels (TRL),” 2014. Factory Autom., Sep. 2019.
Accessed: Oct. 29, 2020. [Online]. Available: https://ec.europa.eu/ [50] X. Ye and S. H. Hong, “Toward Industry 4.0 components: Insights into
research/participants/data/ref/h2020/wp/2014_2015/annexes/h2020- and implementation of asset administration shells,” IEEE Ind. Electron.
wp1415-annex-g-trl_en.pdf Mag., vol. 13, no. 1, pp. 13–25, Mar. 2019.
[29] EU-Robotics-Platform, “Roadmap and multi annual plan (MASP),” [51] Q. Qi and F. Tao, “Digital twin and big data towards smart manufac-
2020. [Online]. Available: https://www.eu-robotics.net/sparc/about/ turing and Industry 4.0: 360 degree comparison,” IEEE Access, vol. 6,
roadmap/ pp. 3585–3593, 2018.
[30] S. Karadayi-Usta, “An interpretive structural analysis for Industry [52] M. Schluse, M. Priggemeyer, L. Atorf, and J. Rossmann, “Experi-
4.0 adoption challenges,” IEEE Trans. Eng. Manag., vol. 67, no. 3, mentable digital twins—Streamlining simulation-based systems engi-
pp. 973–978, Aug. 2020. neering for Industry 4.0,” IEEE Trans. Ind. Inform., vol. 14, no. 4,
[31] T. A. Leopold, V. S. Ratcheva, and S. Zahidi, “The future of jobs report,” pp. 1722–1731, Apr. 2018.
World Economic Forum, Cologny, Switzerland, Tech. Rep., 2018, Ac- [53] Y. Lu, C. Liu, K. I.-K. Wang, H. Huang, and X. Xu, “Digital twin-driven
cessed: Oct. 29, 2020. [Online]. Available: https://www3.weforum.org/ smart manufacturing: Connotation, reference model, applications and
docs/WEF_Future_of_Jobs_2018.pdf research issues,” Robot. Comput.-Integr. Manuf., vol. 61, Feb. 2020, Art.
[32] M. Pejic-Bach, T. Bertoncel, M. Meško, and Ž. Krstić, “Text mining no. 101837.
of Industry 4.0 job advertisements,” Int. J. Inform. Manag., vol. 50, [54] A. Rasheed, O. San, and T. Kvamsdal, “Digital twin: Values, chal-
pp. 416–431, Feb. 2020. lenges and enablers from a modeling perspective,” IEEE Access, vol. 8,
[33] M. F. Manesh, M. M. Pellegrini, G. Marzi, and M. Dabic, “Knowledge pp. 21 980–22 012, 2020.
management in the fourth industrial revolution: Mapping the literature [55] F. M. Kasie, G. Bright, and A. Walker, “Decision support systems
and scoping future avenues,” IEEE Trans. Eng. Manag., to be published, in manufacturing: A survey and future trends,” J. Modelling Manag.,
doi: 10.1109/TEM.2019.2963489. vol. 12, no. 3, pp. 432–454, Jul. 2017.

VOLUME 1, 2020 309

LEITÃO ET AL.: QUO VADIS INDUSTRY 4.0? POSITION, TRENDS, AND CHALLENGES

[56] P. Leitao, N. Rodrigues, and J. Barbosa, “What-if game simulation in [67] IEEE, “Ethically aligned design: A vision for prioritizing human well-
agent-based strategic production planners,” in Proc. IEEE 20th Conf. being with autonomous and intelligent systems,” The IEEE Global
Emerg. Technol. Factory Autom., Sep. 2015, pp. 1–8. Initiative on Ethics of Autonomous and Intelligent Systems, 2017. [On-
[57] P. Madhavan and D. A. Wiegmann, “Similarities and differences be- line]. Available: https://standards.ieee.org/content/dam/ieee-standards/
tween human–human and human–automation trust: An integrative re- standards/web/documents/other/ead_v2.pdf
view,” Theoretical Issues Ergonom. Sci., vol. 8, no. 4, pp. 277–301, [68] B. A. Kadir, O. Broberg, and C. S. da Conceição, “Current research and
Jul. 2007. future perspectives on human factors and ergonomics in Industry 4.0,”
[58] F. Pires, A. Cachada, J. Barbosa, A. P. Moreira, and P. Leitao, Comput. Ind. Eng., vol. 137, Nov. 2019, Art. no. 106004.
“Digital twin in Industry 4.0: Technologies, applications and chal- [69] M. P. Taylor, P. Boxall, J. J. Chen, X. Xu, A. Liew, and A. Adeniji,
lenges,” in Proc. IEEE 17th Int. Conf. Ind. Inform. (INDIN), Jul. 2019, “Operator 4.0 or maker 1.0? exploring the implications of industrie
pp. 721–726. 4.0 for innovation, safety and quality of work in small economies and
[59] Y. Shi, K. Yang, T. Jiang, J. Zhang, and K. B. Letaief, “Communication- enterprises,” Comput. & Ind. Eng., vol. 139, Jan. 2020, Art. no. 105486.
efficient edge AI: Algorithms and systems,” IEEE Commun. Surv. Tu- [70] K. Tange, M. D. Donno, X. Fafoutis, and N. Dragoni, “A system-
tor., to be published. atic survey of industrial internet of things security: Requirements
[60] European Commission, “Digital futures final report ‘A journey into and fog computing opportunities,” IEEE Commun. Surv. Tut., doi:
2050 visions and policy challenges,’ DG CONNECT, European 10.1109/COMST.2020.3011208.
Commission, Tech. Rep., 2016. Accessed: Oct. 29, 2020. [Online]. [71] A. Corallo, M. Lazoi, and M. Lezzi, “Cybersecurity in the context of
Available: https://ec.europa.eu/futurium/en/system/files/ged/futurium_ Industry 4.0: A structured classification of critical assets and business
scientific_report_v10revcl_v2_0.pdf impacts,” Comput. Ind., vol. 114, Jan. 2020, Art. no. 103165.
[61] C. Zunino, A. Valenzano, R. Obermaisser, and S. Petersen, “Factory [72] S. Karnouskos, “Stuxnet worm impact on industrial cyber-physical sys-
communications at the dawn of the fourth industrial revolution,” Com- tem security,” in Proc. 37th Annu. Conf. IEEE Ind. Electronics Soc.,
put. Standards Interfaces, vol. 71, Aug. 2020, Art. no. 103433. Nov. 2011, pp. 4490–4494.
[62] F. Lamnabhi-Lagarrigue et al., “Systems & control for the future of [73] A. G. Khanzode, P. Sarma, S. K. Mangla, and H. Yuan, “Modeling
humanity, research agenda: Current and future roles, impact and grand the Industry 4.0 adoption for sustainable production in micro, small
challenges,” Annu. Reviews Control, vol. 43, pp. 1–64, 2017. & medium enterprises,” J. Cleaner Prod., vol. 279, Jan. 2021, Art.
[63] P. Fantini, P. Leitõ, J. Barbosa, and M. Taisch, “Symbiotic integration no. 123489.
of human activities in cyber-physical systems,” IFAC-PapersOnLine, [74] D. Horváth and R. Z. Szabó, “Driving forces and barriers of In-
vol. 52, no. 19, pp. 133–138, 2019. dustry 4.0: Do multinational and small and medium-sized companies
[64] A. Bueno, M. G. Filho, and A. G. Frank, “Smart production planning have equal opportunities?,” Technological Forecasting Social Change,
and control in the Industry 4.0 context: A systematic literature review,” vol. 146, pp. 119–132, Sep. 2019.
Comput. & Ind. Eng., vol. 149, Nov. 2020, Art. no. 106774. [75] S. Rajput and S. P. Singh, “Connecting circular economy and Industry
[65] S. Robla-Gomez, V. M. Becerra, J. R. Llata, E. Gonzalez-Sarabia, C. 4.0,” Int. J. Inf. Manag., vol. 49, pp. 98–113, Dec. 2019.
Torre-Ferrero, and J. Perez-Oria, “Working together: A review on safe [76] M. Hankel and B. Rexroth, “The Reference Architectural Model Indus-
human-robot collaboration in industrial environments,” IEEE Access, trie 4.0 (RAMI 4.0),” ZWEI: Die Elektroindustrie, vol. 1, pp. 1–2, 2015.
vol. 5, pp. 26 754–26 773, 2017. [77] G. Hurlburt, “Toward applied cyberethics,” Computer, vol. 51, no. 9,
[66] R. Sharpe, K. van Lopik, A. Neal, P. Goodall, P. P. Conway, and pp. 80–84, Sep. 2018.
A. A. West, “An industrial evaluation of an Industry 4.0 refer- [78] S. Karnouskos, “Self-driving car acceptance and the role of ethics,”
ence architecture demonstrating the need for the inclusion of se- IEEE Trans. Eng. Manag., vol. 67, no. 2, pp. 252–265, May 2020.
curity and human components,” Comput. Ind., vol. 108, pp. 37–44, [79] S. Cave, R. Nyrup, K. Vold, and A. Weller, “Motivations and risks of
Jun. 2019. machine ethics,” Proc. IEEE, vol. 107, no. 3, pp. 562–574, Mar. 2019.

310 VOLUME 1, 2020