The Overview of Digital Twins in Industry 4.0: Managing the Whole

Ecosystem

Cristina Rosaria Monsone

, Eunika Mercier-Laurent

and J´osvai J´anos

MMTDI, Sz´echenyi Istv´an University, Gy¨or, Hungary

Department of Computer Science, University of Reims, Champagne Ardenne, France

Department of Automobile Production Technology, Sz´echenyi Istv´an University, Gy¨or, Hungary

Keywords:

Digital Twin, Cyber Physical System, Internet of Things, Cloud, Artiﬁcial Intelligence, Industry 4.0.

Abstract:

Industry 4.0 aims in renewing processes using available technologies such as robots and other AI techniques

implemented in IoT, drones, digital twins and clouds. This metamorphose impacts the whole industry ecosys-

tems including people, information processing and business models. In this context, the accumulated knowl-

edge and know-how can be reused but has also to evolve. This paper focus on the role of digital twins in

transforming industrial ecosystems and discuss also the environmental impact.

1 INTRODUCTION

The industry sector has now embarked on a path of

profound transformation.

An important change of pace, which arises under

the pressure of an increasingly competitive economic

scenario enabled by technological levers such as Ar-

tiﬁcial Intelligence (AI) IoT and Cloud, accompanied

by a profound review of processes, culture and even

corporate business models (Madni et al., 2019). In

particular, with the advent of the Internet of Things

(IoT) and Future Factory, Digital Twin (DT) tech-

nology has become cost-effective to implement and

is gaining increasing acceptance in the Industrial In-

ternet of Things (IIoT) community, which tends to

focus on large, complex, capital-intensive equipment

(Moeuf et al., 2018).

At the same time, the aerospace and defense in-

dustry, which continues to invest in Industry 4.0, has

begun to invest in Digital Twin technology. There-

fore it is most important to state that Industry 4.0 is

not limited to the technical dimension of digitalizing

modern businesses (Felser et al., 2015), as it is rather

the complete new organizationand network coordina-

tion of value and supply chains (PlatformI40, 2018).

The role of the Digital Twin, a product avatar or

cyber-physicalequivalence,is to improve the business

performanceand costs in industry (Holler et al., 2016)

(Madni et al., 2019).

According to (Gartenr, 2018), by 2021 nearly half

the major industrial companies will be leveragingdig-

ital twin technology to facilitate the assessment of

system performanceand technical risks, while achiev-

ing approximately 10% improvementin system effec-

tiveness. In particular, the digital twin is at the top of

the peak of expectations in the hype cycle of ”Top

10 Strategic Technology Trends 2018” proposed by

(Gartenr, 2018).

The research have been reported that the research

company expects that by 2020 at least 50% of pro-

ducers with annual revenues exceeding 5 billion dol-

lars will have launched at least one initiative for dig-

ital twins for their products or assets. A trend that

will lead to tripling the number of companies that

use these solutions by 2022, considering also that the

companies that are implementing projects based on

the Internet of Things are already using or planning

to use ”digital twin”.

This paper presents our study of digital twins,

their applications and impacts.

2 INTERCONNECTION OF

DIGITAL TWIN AND IOT

The digital twin has long since established itself in in-

dustry, where it’s revolutionizing processes along the

entire value chain of a product, production process, or

performance, it enables the individual process stages

to be seamlessly linked,Figure 1.

Monsone, C., Mercier-Laurent, E. and János, J.

The Overview of Digital Twins in Industry 4.0: Managing the Whole Ecosystem.

DOI: 10.5220/0008348202710276

In Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2019), pages 271-276

ISBN: 978-989-758-382-7

271

Figure 1: Digital Twin.

As a virtual representation it was used in the

aerospace context within NASA’s Apollo program,

where two equal space shuttle , one - the physical de-

vice - in the space and the other - the virtual device- at

NASA’s Center, have worked in real time and in mir-

ror condition during the ﬂight (Boschert and Rosen,

2016). The possibility to work in real time and in dy-

namic conditions, creates a consistent improvement

in efﬁciency, minimizes failure rates, shortens devel-

opment cycles, and opens up new business opportu-

nities: in other words, it creates a lasting competitive

edge (Boschert and Rosen, 2016)(Holler et al., 2016).

It seems that the notion “Digital Twin” appears the

ﬁrst time in the year 2000 (Grievesand Vickers, 2017)

characterizes a digital twin concept by three main

components. “Physical products in real space, virtual

products in virtual space, and the connections of data

and information that ties the virtual and real prod-

ucts together”. Almost ten years later, it is provided

(Glaessgen and Stargel, 2012) a more ﬁne-grained

conceptualization (Holler et al., 2016).

“A digital twin is an integrated multiphysics, mul-

tiscale, probabilistic simulation of an as-built vehicle

or system that uses the best available physical models,

sensor updates, ﬂeet history to mirror the life of its

corresponding ﬂying twin ” (Glaessgen and Stargel,

2012)(p.7).

This multiphysics and sensored integration is

made thanks to the Cyber Physical System (CPS )

and Artiﬁcial Intelligence (AI) platform, which rad-

ically increased system where the re-conﬁgurability

and ﬂexibility allows a predictive planning and con-

trol in order to prevent and solve the potential failure

in a production or in a physical system (Yue et al.,

2015). The CPS allows to control and monitor the

industrial process via algorithms directly integrated

in the systems and users around them (Moeuf et al.,

2018).

Today, a digital twin consists of connected prod-

ucts, typically utilizing the IoT, and a digital thread.

The digital thread provides connectivity throughout

the system’s life-cycle and collects data from the

physical twin to update the models in the Digital

Twin.

In this context, the cloud technology provides

ﬂexible and elastic computing services for multiple

types of functions, this is one the innovative element

in the industry production (J´osvai et al., 2018).

Connected to the physical devices, it offers

broader connectivity and data storage/management

resources. Meanwhile, the cyber function mentioned

above, e.g. data processing, simulation, and optimisa-

tion, can also be executed in the cloud by its stronger

computing power. To maximise the system perfor-

mance, the real-time control tasks can be deployed

in the local controllers, while the resource-insensitive

tasks, like task planning, supervisory control, and mo-

bile worker assistance, can be migrated to the remote

computing clouds (Moeuf et al., 2018).

Indeed the interconnection between IoT and Dig-

ital Twin system is based on CPS, where the phys-

ical and software components are tightly coupled to

each other while the physical world (robots, monitors,

products, etc.) is reﬂected and controlled by the cyber

world models, monitoring data, software, task plans,

and so forth through Cloud (Lee et al., 2015).

In particular, the interactions between the cyber

and physical worlds can be further improved with

the help of the latest ICT achievements (Monostori,

2014).

It’s necessary to consider this aspect in the con-

text of cyber-physical systems which are “integration

of computation with physical processes, where em-

bedded computers and networks monitor and control

the physical processes, usually with feedback loops

where physical processes affect computations and

vice versa” (Lee, 2008)(p.1). Main research works

employing this nomenclature cope with visual com-

puting (Stork, 2015) (Holler et al., 2016). Digital

twin, by deﬁnition, requires a physical twin for data

acquisition and context-driven interaction. The vir-

tual system model in the digital twin can change in

real-time as the state of the physical system changes

(during operation).

3 LEVERAGE OF THE DIGITAL

TWIN IN INDUSTRY 4.0

Digital twins are becoming a business imperative,

covering the entire life-cycle of an asset or process

and forming the foundation for connected products

and services and allows analysis of data and monitor-

ing of systems to head off problems before they even

occur,preventdowntime, developing,ina cloud-based

system, new opportunitiesand even plan for the future

by using simulations, thinking of a digital twin as a

KMIS 2019 - 11th International Conference on Knowledge Management and Information Systems

272

bridge between the physical and digital world (Mus-

someli et al., 2018).

Simulating before doing helps understanding not

only the main processes about also interconnections

and impacts in aim optimizing design and minimizing

the footprint (Mercier-Laurent, 2013).

However, Industry 4.0 is a multi-faceted problem,

and it is unlikely that all aspects of it will be applica-

ble to all businesses.

Whilst the area of Intelligent Manufacturing is it-

self a multifaceted problem, the recurring element

that underpins much of this revolution is the col-

lection, utilization and understanding of data, or

the study of ‘Informatics’; almost all of the areas

linked with the intelligent manufacturing research

area rely on the capture and analysis of data in some

way,(Dworschak and Zaiser, 2014)(Shapira et al.,

2013).

To this end the use of advanced analytic, suited

machine learning techniques and knowledge-based

AI is a key technology to develop to further these

other technologies; and the next step in this chain

lies in utilizing the vast reserves of data through data

mining and knowledgediscovery, to better understand

these manufacturing processes. The performance in-

dicators for a company which invest in new tech-

nologies are : lower costs, improved quality, im-

proved ﬂexibility,improved productivity (Raymond,

2005) (Moeuf et al., 2018).

The DT allows greater levelsat technological level

and management level. As presented on the following

sections, in particular the DT allows a high level of

operational efﬁciency: reduction of breakdowns, re-

duction of machine downtime, reduction of defects,

optimization of raw material procurement and logis-

tics, Figure 2.

Figure 2: Leverage of Digital Twin at management and

technological level.

In the same time, the management of the com-

pany wants to explore new business models, in which

the service component and ”recurring revenue” (re-

curring business) plays an increasingly important role

(Madni et al., 2019).

3.1 Leverage of DT at Technological

Level

Today, DT has a crucial role in the smart industry pro-

duction, with an important impact on leveraging also

the revenue stream of the industry and it provides an

edge-to-cloud architecture that is able to optimize op-

erational costs through the division between various

infrastructures.

Digital twin concepts span from aerospace (Kraft,

2016) to naval (Wuest et al., 2015), 2015) indus-

try. Up to the present day, most applications target

these industries.(Wuest et al., 2015) plan extensions

to luxury products and white goods in the future. In

that regard, identifying the relevant information for

the stakeholders represents an increasing challenge.

In line with these developments, it may be interest-

ing to examine applications in other manufacturing

industries in the engineer-to-order to make-to-stock-

planning spectrum (Holler et al., 2016).

This may be due to the current hype cycle of

where the technology stands at the time of writing,

as reported by (Gartenr, 2018). Digital twin is almost

at the peak of inﬂate expectations where we are seeing

some early adopters producing success stories, while

others may be failing or not starting to invest at all

(Eyre and Freeman, 2018).

The DT leverage many industrial sectors, like :

• Automotive: The convergence of physical and

their virtual products has the potential to address

many challenges which exists in the automotive

value chain today. The ‘digital twin’ in auto-

motive industry can enable convergence of exist-

ing gaps between physical and virtual versions of

product prototypes, shop ﬂoor and actual vehicle

on the road.

• Aerospace: Commercial plane’s thousandsof sen-

sors stream asset data to better system servicing

and operational status.

• Healthcare: Connected medical systems and tools

ensure product integrity and measure patient out-

comes.

• Manufacturing: Digital factory equipment and

machinery increase uptime and production yield,

while reducing repair and maintenance rates.

• Oil and Gas: Remote rig sends health data limit-

ing routine inspections and servicing.

The Overview of Digital Twins in Industry 4.0: Managing the Whole Ecosystem

273

• Rail: View of deployed locomotives and assets

health better optimize scheduling and reducing

servicing time.

• Utilities: Digital representation of systems on the

power grid improve demand response functions

and energy efﬁciency.

Of course, the value of the data is an indispens-

able condition in every smart manufacturing project,

since without the Digital Twin it is impossible to iden-

tify the KPIs and identify the areas in which to inter-

vene, it is equally true that its implementationrequires

method. It is important too that the entire value chain

is involved so that all aspects and all processes are

examined in a holistic and structured manner.

The big companies, like GE, Siemens, have de-

veloped some platforms for the utilisation of DT, that

was able to meet the expectations of the industrial

world by producingforecasts and results that could be

valid for any type of market a company was involved

with.

In particular developing the “Edge devices”.

The ”Edge devices” considers remote devices as

intelligent devices able to operate directly on the data

or on the resources to which it is close. For this rea-

son the devices that are used include the use of ma-

chine, which is able to guarantee them the use and or-

ganization of communication and authentication pro-

tocols. Being smart devices, they have the ability to

make forecasts directly without having to communi-

cate with the cloud; thanks to this they are able to in-

dependently detect anomalies and act directly on the

environment. Supported by the possibility of commu-

nicating with the cloud to obtain data on past history,

they are able to improve their forecasts over time.

Each item can also be combined with a position. This

is because in certain work areas it is normal to sup-

pose that products are in motion and theywant to keep

track of them. As mentioned before, the great advan-

tage of Digital Twin concerns the optimization at pro-

cess level and also at operational costs level through

the division between various infrastructures, achieved

important investment by worldwide companies.

3.2 Leverage of DT at Management

Level

With the growing importance of devices utilized in

the smart industry, especially for Digital Twin, it is

becoming increasingly important, within a company,

the opportunity for employees, who are now able to

perform more tasks using their help.

The employees are more connected than ever, able

to improve his own efﬁciency, reduce costs for the

company and increase customer satisfaction.

The digital technology also needs high-level ex-

perts that have a “suited” knowledge regarding this

new methodology about the control and productivity

of the company.

It’s recognize a global level that speciﬁc invest-

ments needs the use of speciﬁc expertise in a view of

a more dynamic and ﬂexible organisation’s strategy

(Moeuf et al., 2018).

(Moeuf et al., 2018) have adapted the works of

(Porter and Heppelmann, 2014), initially proposed for

measuring the capacity of new smart connected prod-

ucts and services, to establish a list of four distinct

managerial capacities aligned with the concept of In-

dustry 4.0.

Summarizing, in this list 4 main aspects are con-

sidered:

• Monitoring: the role of decision- making (Ve-

landia et al., 2016)is based on the analysis of his-

torical data provided by various connected objects

(Wang et al., 2015) in smart manufacturing. In

this way any change occurred during the produc-

tion process is reported as warming and the anal-

ysis of these data allows a further elaboration to

improve the production performance,

• Control: all data that have a different behavior re-

spect to the planned performance, allows to deﬁne

algorithms with the speciﬁc role of detection of

alert situation (Aruv¨ali et al., 2014),

• Optimisation: it’s possible to optimise in real time

the production process, (Saenz de Ugarte et al.,

2011) combining the analysis of data and the uti-

lization of speciﬁc algorithm for detection of alert

situation,

• Autonomy: Thanks to the above mentioned oper-

ations, it’s possible to reach a new levels of au-

tonomy in production system (Khalid et al., 2016)

with a great advantage – with AI - to develop sys-

tems capable of learning autonomouslyand adapt-

ing themselves from their own behaviour(Bagheri

et al., 2015)

In this way, a better level of production process is

offered by the utilization of DT through data acquired

by sensors in real time, and the consequently modeli-

sation, reached often by AI, fostering the entire value

chain of Industry4.0 also a knowledge management

level.

In fact, Digital twin focus on three key assets: the

technologicalasset, represented for example by all the

initiatives of Industrial Internet of Things, the product

life cycle, the management of the value chain. The

latter is probably the aspect on which attention is less

often concentrated. And yet, it is precisely in optimiz-

ing the entire value chain that the potential of smart

KMIS 2019 - 11th International Conference on Knowledge Management and Information Systems

274

manufacturingprojects emerges. In particular, the DT

leverage the needs of ”digital skilled” employees:

• professions related to the processing and analysis

of information (big data, business intelligence);

• professionalism in Artiﬁcial intelligent ﬁeld (ma-

chine learning, deep learning);

• professionals specialized in the automation of

productive and automated processes.

Dealing with a smart manufacturing project with a

keen eye on the impact on the entire value chain

means breaking away from a short-term approach and

acquiring a strategic and long-term vision thanks to

new levels of collaboration with partners and suppli-

ers. A well managed value chain trough Digital Twin

makes it possible to integrate processes, reduce in-

ventories, improve service levels, with an overall im-

provement in both products and customer satisfaction

levels.

4 CONCLUSIONS

Considering the role of the Digital Twin in the context

of Industry 4.0, it seems clear that any strategy that

leads the company to move in a logic of smart manu-

facturing cannot be limited to a reﬂection on cost sav-

ings on improving productivity, also in terms of Re-

turn on Investment (ROI). In particular it’s possible to

deﬁne three main aspect. The ﬁrst aspect is the role of

the digital twin of the product. It is created as early as

the deﬁnition and design stage of a planned product,

allowing the engineers to simulate and validate prod-

uct properties depending in advance. The second as-

pect is the digital twin of production. It involves every

aspect, from the machines and plant controllers to en-

tire production lines in the virtual environment using

real time data. This simulation process, as mentioned

in the previous sections, can be used to optimize pro-

duction in advance - with PLC code or AI. As a result,

sources of error or failure can be identiﬁed and pre-

vented beforeactual operation begins. This savestime

and lays the groundworkforcustomized mass produc-

tion, because even highly complex production routes

can be calculated, tested, and programmed with min-

imal cost and effort in a very short time.

Last but no list, the aspect related to the digital

twin of performance. It is constantly fed with oper-

ational data from products or the production plant.

This allows information like status data from ma-

chines and energy consumption data from manufac-

turing systems to be constantly monitored. In turn,

this makes it possible to perform predictive main-

tenance to prevent downtime and optimize energy

consumption. This generates a completely closed

decision-making loop for the continuous optimization

process. The conclusion is that the Digital twins are

not only a trend, but play important role in industrial

innovation. They allow simulation based not only on

collected data but also exploring existing knowledge

and know-how to understand, improve, take into con-

sideration the ”weak signals” and contextual knowl-

edge as well.

Renewing industry is not only about introducing

trendy technology, but also about understanding the

related ecosystems including people, business and en-

vironment.

Digital twins are useful for training and simulation

but ﬁnal goal is to create a sustainable synergy be-

tween technology and humans allowing to explore the

best the both capacities. Enabling this transparency

across the organizational value chain through digital

and physical world convergence is becoming increas-

ingly necessary as products shift to ‘as-a-service’

models. This digital transformation and ﬁnal link of

the digital thread is capable through the advent and

adoption of the digital twin.

It’ also true that some recent disasters demon-

strated that fully automated systems are not 100% re-

liable, especially in case of missing data during the

operation and when the device is commercialised with

undiscovered bugs during prototyping phase. The

main challenge of this work is to study the efﬁciency

and beneﬁts of I4 when combining data and knowl-

edge and integrating human into automated systems.

It is also to propose the systems able to combine the

best of machine and human capacities. We focus ﬁrst

on digital twins considered as a component of the

whole I4 and eco-innovation systems. Other compo-

nents of I4 will be added in the future work, applying

incremental approach.

Our future research will focus on this collabora-

tion.

REFERENCES

Aruv¨ali, T., Maass, W., and Otto, T. (2014). Digital object

memory based monitoring solutions in manufacturing

processes. Procedia Engineering, 69:449–458.

Bagheri, B., Yang, S., Kao, H.-A., and Lee, J. (2015).

Cyber-physical systems architecture for self-aware

machines in industry 4.0 environment. IFAC-

PapersOnLine, 48(3):1622–1627.

Boschert, S. and Rosen, R. (2016). Digital twin—the sim-

ulation aspect. In Mechatronic Futures, pages 59–74.

Springer.

Dworschak, B. and Zaiser, H. (2014). Competences for

The Overview of Digital Twins in Industry 4.0: Managing the Whole Ecosystem

275

cyber-physical systems in manufacturing–ﬁrst ﬁnd-

ings and scenarios. Procedia CIRP, 25:345–350.

Eyre, J. and Freeman, C. Immersive applications of indus-

trial digital twins. The Industrial Track of EuroVR

2018.

Felser, W., Kirsch, A., Kletti, J., Wießler, J., and Meuser,

D. (2015). Industrie 4.0 kompakt. NetSkill Solutions,

K¨oln.

Gartenr, I. (2018). Smart insights. retrieved from

gartner hype cycle 2018 – most emerg-

ing technologies are 5-10 years away.

https://www.smartinsights.com/managing-digital-

marketing/managing-marketing-technology/gartner-

hype-cycle-2018-most-emerging-technologies-are-5-

10-years-away/.

Glaessgen, E. and Stargel, D. (2012). The digital twin

paradigm for future nasa and us air force vehi-

cles. In 53rd AIAA/ASME/ASCE/AHS/ASC Struc-

tures, Structural Dynamics and Materials Conference

20th AIAA/ASME/AHS Adaptive Structures Confer-

ence 14th AIAA, page 1818.

Grieves, M. and Vickers, J. (2017). Digital twin: Mitigat-

ing unpredictable, undesirable emergent behavior in

complex systems. In Transdisciplinary perspectives

on complex systems, pages 85–113. Springer.

Holler, M., Uebernickel, F., and Brenner, W. (2016). Digital

twin concepts in manufacturing industries-a literature

review and avenues for further research.

J´osvai, J., Pfeiffer, A., Sz´ant´o, N., and Monek, G. (2018).

The role of digital twin in a cyber-physical production

environment with prescriptive learning.

Khalid, A., Kirisci, P., Ghrairi, Z., Thoben, K.-D., and Pan-

nek, J. (2016). A methodology to develop collabo-

rative robotic cyber physical systems for production

environments. Logistics Research, 9(1):23.

Lee, E. A. (2008). Cyber physical systems: Design chal-

lenges. In 2008 11th IEEE International Symposium

on Object and Component-Oriented Real-Time Dis-

tributed Computing (ISORC), pages 363–369. IEEE.

Lee, J., Ardakani, H. D., Yang, S., and Bagheri, B. (2015).

Industrial big data analytics and cyber-physical sys-

tems for future maintenance & service innovation.

Procedia Cirp, 38:3–7.

Madni, A. M., Madni, C. C., and Lucero, S. D. (2019).

Leveraging digital twin technology in model-based

systems engineering. Systems, 7(1):7.

Mercier-Laurent, E. (2013). Innovation ecosystems. John

Wiley & Sons.

Moeuf, A., Pellerin, R., Lamouri, S., Tamayo-Giraldo, S.,

and Barbaray, R. (2018). The industrial management

of smes in the era of industry 4.0. International Jour-

nal of Production Research, 56(3):1118–1136.

Monostori, L. (2014). Cyber-physical production systems:

Roots, expectations and r&d challenges. Procedia

Cirp, 17:9–13.

Mussomeli, A., Meeker, B., Shepley, S., and Schatsky, D.

(2018). Expecting digital twins. Deloitte Insight 2018.

PlatformI40 (2018). Aspects of the research roadmap

in application scenarios. https://www.plattform-

i40.de/I40/Redaktion/EN/Downloads.

Porter, M. E. and Heppelmann, J. E. (2014). How

smart, connected products are transforming competi-

tion. Harvard business review, 92(11):64–88.

Raymond, L. (2005). Operations management and ad-

vanced manufacturing technologies in smes: a con-

tingency approach. Journal of Manufacturing Tech-

nology Management, 16(8):936–955.

Saenz de Ugarte, B., Pellerin, R., and Artiba, A. (2011). An

improved genetic algorithm approach for on-line op-

timisation problems. Production Planning & Control,

22(8):742–753.

Shapira, P. P., Wessner, C. W., Council, N. R., et al. (2013).

21st century manufacturing: the role of the manu-

facturing extension partnership program. National

Academies Press.

Stork, A. (2015). Visual computing challenges of advanced

manufacturing and industrie 4.0 [guest editors’ intro-

duction]. IEEE computer graphics and applications,

35(2):21–25.

Velandia, D. M. S., Kaur, N., Whittow, W. G., Conway,

P. P., and West, A. A. (2016). Towards industrial

internet of things: Crankshaft monitoring, traceabil-

ity and tracking using rﬁd. Robotics and Computer-

Integrated Manufacturing, 41:66–77.

Wang, L., T¨orngren, M., and Onori, M. (2015). Current

status and advancement of cyber-physical systems in

manufacturing. Journal of Manufacturing Systems,

37:517–527.

Wuest, T., Hribernik, K., and Thoben, K.-D. (2015). Ac-

cessing servitisation potential of plm data by applying

the product avatar concept. Production Planning &

Control, 26(14-15):1198–1218.

Yue, X., Cai, H., Yan, H., Zou, C., and Zhou, K.

(2015). Cloud-assisted industrial cyber-physical sys-

tems: An insight. Microprocessors and Microsystems,

39(8):1262–1270.

KMIS 2019 - 11th International Conference on Knowledge Management and Information Systems

276