DIGITALIZATION AND DIGITAL TRANSFORMATION IN METAL FORMING: KEY TECHNOLOGIES, CHALLENGES AND CURRENT DEVELOPMENTS OF INDUSTRY 4.0 APPLICATIONS

Abstract

The conception of Industry 4.0 in 2011 marked the beginning of a new era of industrial manufacturing. Since then, data-driven decision making and smart networks supported by artificial intelligence have led to a continuous change in working environments in production. Nevertheless, there are many companies that have not fully exploit the potential of the fourth industrial revolution. To a considerable extent, this can be attributed to a low level of automation. This circumstance often results from financial or process-related restrictions and affects not only the production facilities but also the sensitive IT infrastructure. The lack of automation is therefore often compensated by expert knowledge and experience. The fear of job loss due to disruptive technologies is a further contribution to the significant delay in the digital transformation of such companies. Companies in the metal forming industry are particularly affected by this development. This paper describes the key technologies of digitalization and provides an outlook to possible solutions for specific challenges during the next years in the manufacturing and especially metal forming industry environment.

Full text

DIGITALIZATION AND DIGITAL TRANSFORMATION IN METAL
FORMING: KEY TECHNOLOGIES, CHALLENGES AND CURRENT
DEVELOPMENTS OF INDUSTRY 4.0 APPLICATIONS
Benjamin James Ralph1*, Martin Stockinger1
1Chair of Metalforming, University of Leoben, Franz Josef Straße 18, 8700 Leoben, Austria
ABSTRACT: The conception of Industry 4.0 in 2011 marked the beginning of a new era of industrial manufacturing.
Since then, data-driven decision making and smart networks supported by artificial intelligence have led to a continuous
change in working environments in production. Nevertheless, there are many companies that have not fully exploit the
potential of the fourth industrial revolution. To a considerable extent, this can be attributed to a low level of automation.
This circumstance often results from financial or process-related restrictions and affects not only the production
facilities but also the sensitive IT infrastructure. The lack of automation is therefore often compensated by expert
knowledge and experience. The fear of job loss due to disruptive technologies is a further contribution to the significant
delay in the digital transformation of such companies. Companies in the metal forming industry are particularly affected
by this development. This paper describes the key technologies of digitalization and provides an outlook to possible
solutions for specific challenges during the next years in the manufacturing and especially metal forming industry
environment.
KEYWORDS: Industry 4.0; Metal Forming; Smart Production; Digital Factory; Cyber Physical Production Systems;
Industrial Internet of Things; Big Data; Cloud Computing; Digital Twin; Finite Element Analysis;
1 INTRODUCTION
Since the German government introduced the term
“Industry 4.0” in 2011, a lot of scientific research
focused on the development and forms of its key
components [1,2]. In most of these works, several
problems arise with regard to the practical application of
Industry 4.0 (I 4.0) technologies in the metal forming
industry. In summary, there is a clear incoherence
between theoretical digital maturity levels and practical
work validated by case studies. Especially for smaller
batch sizes, apart from conventional mass production,
many recommended implementation approaches are not
feasible at the current state of the art [3]. In addition,
there are hardly any publications that significantly
consider the integration of complex numerical
simulations, which are widely used in forming
technology practice, into such I 4.0 strategies. This paper
gives an overview of current digitalization technologies
which are already used in forming technology or at least
have the potential to be implemented in the near future.
In addition, the potential challenges in this industry
sector are discussed.
2 KEY TECHNOLOGIES
The following subchapters are dedicated to core
technologies of the fourth industrial revolution that are
of the highest relevance from the perspective of current
scientific publications [1,2,3,4,5,6,7]. Despite the large
number of publications concerning these technologies,
there is still no uniform terminology [8]. For this reason,
the most recent scientific publications in sophisticated
journals are cited for the definition of the terms.
Although Additive Manufacturing (AM) is considered a
part of I 4.0 in most current publications [9], it will not
be discussed in detail in this paper. From the metal
forming point of view, additive manufacturing is a
special production technique, which was strongly driven
by I 4.0 technologies, but is not absolutely necessary in
order to completely digitalize forming technology
companies.
2.1 GENERIC INFRASTRUCTURE
To create a uniform operationalization of the I 4.0
concepts, various reference models were developed. One
of the most comprehensive and promising for further
adaptations is the Reference Architectural Model
Industrie 4.0 (RAMI 4.0) [10]. This model is an
extension of the Smart Grid Architecture Model
(SGAM) and takes into account the often complex
requirements of planning, implementation and operation
of a smart factory [11]. In addition to the consideration
of the hierarchical level and the layer level known from
classic automation technology, this model also considers
value stream and asset life cycle in the third dimension
[10]. Fig. 1 schematically shows the basic configuration
of this concept.

Figure 1: RAMI 4.0 concept [12]
An essential aspect of the digital transformation and thus
of I 4.0 is the vertical as well as the horizontal
integration along the supply chain. These are described
by the hierarchy levels (horizontal integration) and
layers (vertical integration). In the Life Cycle and Value
Stream dimension, both the initial implementation and
further optimization can be quickly visualized and
discussed, as required [12]. Depending on requirements,
this model can be adapted at will and is therefore
suitable for all types of production processes and
manufacturing companies [10,13,14].
2.2 CYBER PHYSICAL PRODUCTION SYSTEMS
(CPPS)
Cyber Physical Systems (CPS) were first mentioned in a
workshop of the American National Science Foundation
in 2006 [15]. The terminology of this concept has been
concretized several times in recent years. In general,
CPS refers to systems that acquire, store, analyze and
process data via Internet technologies (Internet of Things
(IoT)) and in the context of the integration of the real
physical and virtual world, including human machine
interaction. It is also seen as one of the essential building
blocks of the fourth industrial revolution [1,2]. A further
development and specially adapted extension for
production is the so-called CPPS. In addition to the
interaction of computer science, information and
communication technologies as well as human machine
interface known from the CPS concept, automation
technology is also taken into account more explicitly
[16]. For this purpose, instruments of classical
automation technology, e.g. sensors, actuators or
fieldbus interfaces are combined with modern
information technology hardware and software
[17,18,19]. The classic hierarchical structure of the well-
known automation pyramid is replaced by a
decentralized structure [20], starting above the field and
control level (Fig. 2).
Figure 2: CPPS – Scheme [22]
Due to the increasing spread of in memory technologies
on the Enterprise Resource Planning (ERP) and Business
Intelligence (BI) level as well as more sophisticated
Management Executions Systems (MES) in the metal
forming Industry, the use of this technology will also
increase in the future [21]. The challenge regarding the
high amount of different interfaces plays an important
role in this context. According to the current state of
research, OPC Unified Architecture (OPC UA) seems to
be the first standard communication protocol framework
for communication in the mostly heterogeneous machine
data environment. This format is open source, constantly
developing and compatible with almost all file formats.
Furthermore, adaptations can be made via a
corresponding Graphical User Interface (GUI) using
Python or C++ [23]. OPC UA is therefore seen as key
technology of Industrial Internet of Things (IIoT), which
can be seen as one of the main enabler of CPPS (2.3).
The inconsistency of scientific publications concerning I
4.0, HMI, CPS and CPPS can be illustrated by the
following Figure 3. While publications on the theory and
terminology of CPS almost exclusively see Human
Machine Interface (HMI) as an integral part, this topic is
often seen as a separate focus and therefore outstanding
add-on (3.2). Also the definition of CPPS is often
difficult to differentiate from CPS. In this paper, with
reference to the metal forming industry, CPPS is
nevertheless considered as a superordinate system,
which adds sophisticated HMI Technologies and
simulations to the CPS.
Figure 3:Commonly used definition of CPS [16]

Within the context of the metal forming industry, this
would imply that CPPS takes into account digital
shadows and digital twins in addition to HMI
technologies [24].
2.3 INDUSTRIAL INTERNET OF THINGS (IIOT)
According to the current state of research, the term IIoT
can be regarded as an evolution of the IoT concept [25].
Starting with the first significant mention in recognized
journals in 2013, there has been an exponential increase
in publications over the years, as demonstrated by Liao
et al. [26] until 2017, whereby articles from the Scopus,
IEEE Xplore and Science Direct databases were
collected in detail and analyzed using data denoising,
data confirmation, data enrichment and data
categorization. Despite this, there are still various
definitions of IoT in 2018 [26]. The underlying reasons
for this can often be attributed to different approaches
due to the interdisciplinary nature of digitalization. In
general, it can be postulated that the term IoT has
developed in most cases from the Fieldbus technology
known from automation technology. These process
automation protocols were and are still partly used today
for the implementation of the Supervisory Control and
Data Acquisition (SCADA) control level (Fig. 2, Process
Control Level), which then forwards agglomerated
information to the MES level above. This interface is
also one of the main drivers for the development of the
fourth industrial revolution. SCADA systems in most
cases include not only simple sensors and actuators but
also Programmable Logic Controllers (PLCs),
management consoles and Proportional-Integral-
Derivative (PID) controllers. These were passed on and
processed using the Fieldbus protocol format. The main
challenge in this case is the heterogeneity of these
protocols. The IEC-61158-1 and other known standards
include over 18 different families of fieldbus protocols,
e.g. modBus, ProfiNet, CANbus, EtherCAT and many
more. This heterogeneous protocol landscape leads to
extremely complex systems, especially for companies
with long system life and machine heterogeneity. Due to
the spread of the Internet in the industrial context, it was
also necessary to include Internet protocol standards in
this mostly already complex system. For the
manufacturing industry, the term IoT can therefore be
seen as a structured and standardized layer architecture
that provides standardized Internet protocols (IPv4 and
IPv6) as a superior instance for further processing while
simplifying complex SCADA systems as much as
possible. This abstraction is made possible by the
integration of gateways or data transformation using, for
example, IPv6 over Low Power Wireless Personal Area
Networks (6LoWPAN7) [27]. One of the most
promising recommended data protocol format for pure
machine communication in production is Message
Queuing Telemetry Transport (MQTT), as it provides
particularly efficient storage and thus reduces the
resulting amount of data. Another example is the
Extensible Messaging and Presence Protocol (XMPP),
which was designed specifically for HMI
communication [28]. Fig. 4 shows some further file
formats which are currently in use according to the
current state of literature on all levels of a fully
digitalized factory [29].
Figure 4: Smart Manufacturing systems and various data
formats [29]
2.3.1 Tracking Technologies
The basic prerequisite for the use of an IIoT environment
are so-called Smart Parts. These subsystems, which are
often also often referred to as Intelligent Parts, Products
or Machines, refer to sensors and/or actuators mounted
directly on the product or machine, which enable the
localisation of all entities involved in the production
process. While machines are usually integrated via
integrated controllers and gateways based on them, this
approach is often difficult to execute for semi-finished
products or starting materials. For this application, a
variety of different technologies are used, from
intelligent image recognition to Radio Frequency
Identification (RFID) [30], Bluetooth and WiFi
technologies [31], or (for mostly smaller quantities)
manual input into the system via HMI. It is important to
note that the full potential can hardly be exploited
without supply chain-wide tracking technologies. This
concept is also often referred to as Logistics 4.0 [32], but
in terms of the technologies used and the purpose, it can
be considered a subset of I 4.0 [33,34]. The full potential
of I 4.0 and the Smart Production concept can only be
achieved by fulfilling these technologies, so a separate
consideration and naming from the metal forming
industry´s point of view does not seem necessary.
2.3.2 IT Security
Cyberattacks have been common practice since the mass
suitability of the Internet. Parallel to the increasing use
of IIoT technology in manufacturing companies, attacks
on their IT infrastructure are also on the increase.
Already 2014, the manufacturing industry was the main
target of spear pishing attacks [35]. Another well-known
example of cyber physical attacks is the "WannaCry"
ransomware virus, which drove a large number of
automotive factories to a standstill in 2017 [36]. Such
cyberattacks range from theft and manipulation to the
deletion of sensitive production data. Intellectual
property is also affected [37]. Especially in European
and American metal forming industry, in which internal

know-how is in many cases an essential component of
competitiveness, such an attack can threaten the
existence of a company. It should also be noted that the
legal basis for successfully defending against data
transfer is often not sufficient to protect intellectual
property rights [38]. Sensitization and the involvement
of IT security experts can help companies to identify and
avoid potential dangers through well thought-out IT risk
management. Additionally, there are several frameworks
developed by experts, which support the development
and implementation of cybersecurity solutions, e.g. as
part of or under consideration of the RAMI 4.0 concept
[39].
2.4 DIGITAL TWINS
The terminology of Digital Twins, like most of the
technologies mentioned above, is not always clear
defined. Originally first mentioned in 2002 in the context
of Product Lifecycle Management (PLM), various fields
of application and interpretations have led to a diverse
development in the use of this term. In general, a Digital
Twin is defined as the virtual, digital equivalent of a
physical existing product [40]. However, this definition
is insufficient for practical applications in the metal
forming industry.
The first differentiation is based on the field of
application. In metal forming technology there are two
main application areas for this concept: i.) Digital twins
as a representation of the production process over parts
or the entire production chain, and ii.) Digital twins as a
representation of one or a manageable number of process
steps for the production of a semi-finished or finished
product. In ii.) the focus is mainly on numerical
simulation, e.g. Finite Element Analysis (FEA). The
mapping of the process chain according to definition i.)
is an important aspect in production plan optimization,
but does not differ significantly from the general
manufacturing industry. There are also a large number of
large-scale industrial solutions, often coupled or as an
integral part of modern MES systems. Definition ii.) is
of particular interest. In this case a Digital Twin can be
defined as a digital representation of processes, which
are often based on complex material science and process
engineering interactions [24]. For this reason, this article
will refer to definition ii.) when using Digital Twin
terminology.
2.4.1 Degree of Process Intervention
Another necessary differentiation is the distinction
between "actual" Digital Twins (DT), Digital Shadows
(DS) and Digital Models (DM). A Digital Model is a
digital copy of a real physical entity, but completely
without automatic data exchange between the virtual and
the real object [41], e.g. an FEA of an existing forming
process. This definition is mostly unambiguous in
current literature and therefore needs no further
concretization.
There are different definitions for DS in comparison. In
general, it can be postulated that at least one of the two
data connections between real and virtual objects is
automated [41]. However, there are definitions where
one of the two connections is explicitly defined as
manual and the other as automatic [42]. Due to the lack
of traceability and for reasons of generalizability, the
first definition will be chosen as the basis for further
considerations.
A Digital Twin is therefore a digital image of a real
physical entity, which automatically transfers data
bidirectional between both instances [41]. The
differences between these three definitions are
schematically visualized in Fig. 5.
Automatic Transfer
DM
Manual Transfer
DS
DT
Figure 5: Comparison of Data Connectivity: DM/DS/DT
A large number of scientific publications define DT as
virtual objects, which are in some way connected to a
real object. This approach is mainly used for case studies
and implementation approaches [41,42,43]. From this it
can be concluded that especially in the field of modeling
in manufacturing, theoretical progress and practical
implementation diverge substantially.
2.4.2 Modeling Techniques
Furthermore, a distinction must be made between the
origin of data used to model a DM, DS or DT. The
consideration of digital twins on the basis of real
material-physical laws is only one of two general
approaches in the metal forming industry. Another
approach, but one that is rather rarely used in the
development of companies in heavy industry, is the
creation of such a twin with strongly restricted or, in
extreme cases, without the inclusion of material physics.
Such approaches are based on stochastic methods and
find correlations and thus descriptive variables directly
from process and sensor data analysis. Multiple
digitalization solutions, which e.g. connect on the
SCADA level in production control, use a mixture of
both approaches: complex material-physical models are
coupled to the process and existing material models are
optimized under the supervision of experts with the
support of Artificial Intelligence (AI). In this case the
stochastic part of process modelling consists primarily in

the reduction of the complex real-physical models.
Nevertheless, it should be the ambition of the metal
forming industry to understand the underlying scientific
relationships even after abstraction. In the context of I
4.0 and Smart Production, the mixture of real-physical
laws and thus comprehensible calculations, called White
Box Modeling (WBM), and calculations that are no
longer comprehensible for a domain expert
(stochastically calculated modelling that is decoupled
from the physical problem, called Black Box Modeling
(BBM)) is referred to as Grey Box Modeling (GBM).
GBM approaches are becoming increasingly popular and
represent the tool of choice for many manufacturing
companies, in which also the greatest (near) future
potential with regard to the further spread of DS and DT
in production is represented [24,44]. Fig. 6 shows the
GBM approach schematically.
Figure 6: GBM approach for the metal forming industry
[24]
2.5 BIG DATA AND ANALYTICS
Data is the fundamental basis of all I 4.0 technologies. In
order for effective usage of data generated in a
production process, it must meet certain criteria. The big
data concept summarizes the problems that arise when
handling recorded data in an industrial context in defined
criteria. The number of these criteria again varies
depending on the scientific publication [45,46,47], but in
general production is based on three criteria, the 3 V of
Big Data (Fig. 6), which is lately scientifically
substantiated from M. Ghasemaghaei [48]. If these
criteria are fulfilled, the term Big Data is accurate.
Figure 7: IBM Big Data characteristics [49]
Volume refers to the continuously increasing amount of
data, whereby in most cases only from petabytes
(equivalent to 1000 terabytes) onward, a truly large
amount of data is assumed. Considering a fully
digitalized factory in which all machines and the entire
logistics communicate with each other using IIoT, CPPS,
MES and an ERP System, data volumes of this
magnitude can be reached even in medium-sized
companies [50].
Considering the volume, the variety of the data is also a
challenge. Data from different heterogeneous resources,
e.g. IIoT Gateways, Customer Relationship Management
(CRM) or Supplier Relationship Management (SRM)
Tools, in most cases lead to a data structure that is
difficult to handle [51]. In heavy industry in particular,
offline parallel structures are also a problem, as data is
often not entered into the digitalization system at all or
only insufficiently (e.g. missing database maintenance).
The third challenge is to provide the required data at the
required velocity. This problem seems to be the biggest
challenge especially in the forming technology practice,
considering e.g. DS or DT approaches. Also with regard
to innovation this point seems to be the most critical one
[52].
A variety of infrastructures have also been developed for
Big Data applications. In general, the necessary
architecture can be divided into six different layers. For
each of these layers, different software and database
developments should be preferred.
Among the most widely used is the Apache Hadoop
framework developed in Java. This is a scalable tool for
centrally operating software applications and is based on
Google's MapReduce algorithm, which enables the
efficient clustering of computing processes for
processing large amounts of data. Fig. 8 shows an
example of the process from tapping data from coupled
machines to visualization of the analyzed data via HMI.
Figure 8: Example of a big data architecture framework
[53]

The data from e.g. IIoT Gateways or directly from the
machine controller is sent to the Hadoop Distributed File
System (HDFS). The second step is a real-time
monitoring of the file systems using e.g. Apache Flume.
These are then collected in real-time and sent to (in this
example) Apache Kafka in agglomerated form. Apache
Kafka then acts as an interface for loading and exporting
data streams to preferred third party systems. In the
following layer, e.g. Yarn or Apache SPARK can be
used for package management of JavaScript and Node.js.
This layer is also responsible for the actual data analysis.
Besides e.g. streaming functionality they also include
machine learning, deep learning and graph processing
tools. Applications can be written directly with Java,
Scala, Python, R or SQL. The fact that SPARK and
Hadoop (HDFS, Yarn) applications have partially
overlapping functions must be considered. On one hand,
SPARK does not have its own file management and is
therefore relying on HFDS or similar technology (e.g.
Cassandra, HBase). On the other hand, SPARK can
perform most calculations in memory, which leads to an
outperformance of traditional Hadoop MapReduce
platforms in most cases [54]. Nevertheless, which
architecture is chosen depends on the preference of those
responsible for implementation and the specific use case.
The results in this use case are then stored in
ElasticSearch as NoSQL. Kibana is a browser-based,
open-source analysis platform, and a practical example
for a program in the final big data layer that allows to
search and visualize the data stored in e. g. ElasticSearch
[53,54].
2.6 CLOUD COMPUTING
The definition of cloud computing varies depending on
considered application area. The National Institute of
Standard and Technology (NIST) refers in a very general
way to the following definition to the cloud computing
terminology, as “a model for enabling convenient, on-
demand network access to a shared pool configurable
computing resources (e.g. networks, servers, storage,
application, and services) that can be rapidly
provisioned and released with minimal management
effort or service provider interaction” [55]. According to
Mell and Grants [55], the common architecture of a
cloud solution consists of five essential characteristics,
three service models and four deployment models (Fig.
9). It is considered outsourcing of data storage, usually
accompanied by corresponding Service Level
Agreements (SLAs).
On demand self-service is the definition for the
possibility of customers of a cloud service to
automatically access required resources, e.g. storage
capacities, without additional human interaction. Broad
network access enables customers to access all agreed
resources with a variety of heterogeneous client
platforms (e.g. smartphones, tablets, workstations). The
resources provided by the provider are used by a large
number of customers, who have limited possibilities to
localize the physical data storage (resource pooling).
Rapid elasticity refers to the ability of the provider to
adapt its provided resources highly flexibly to the
requirements of customers (e.g. upscaling). Another part
of the essential characteristics is the possibility for the
user to monitor and analyze his provided data streams
and to request reports (measured service). As usual in
many sub-areas of IT services, pay-per-use models are
usually used for this purpose (e.g. for extended reporting
possibilities, data volume (per time unit)).
In order to be able to use the advantages of such a
service offer as effectively as possible, some providers
also offer to adopt a corresponding big data architecture
(SaaS, PaaS). The advantages, especially for smaller
manufacturing companies, lie in the reduced in house
expert know-how and required infrastructure (IaaS)
required in this area [55, 56].
Private clouds are exclusively used by one organization,
divided into e. g. business units, managed by internal IT,
external organizations or a combination of both.
Community clouds are provided for a defined
community that share common interests, e.g. security
requirements. It can be operated and owned by one or
more organizations in the community, a third party, or a
combination of both. Public clouds are infrastructures
which are provided for open use by GOs, NGOs or
academic organizations. Hybrid clouds consist of two or
more different cloud infrastructures (public, community,
private) that remain unique entities, but are connected by
a single standardized framework technology. This
enables a fast and relatively simple exchange of data
between individual entities involved [55].
Figure 9: Cloud computing terminology after NIST [55]
3 POTENTIAL CHALLENGES
The metal forming industry in general is characterized
by a high degree of heterogeneity. It includes a variety of
production processes, materials, machine systems, but
also organizational structures and sizes. In addition,
there is currently no Austria-wide umbrella organization
which represents companies assigned to this industrial
sector as a unit. The heterogeneity in all these cases also

results in a reduced number of publications that address
the requirements within this technical discipline.
Digitalization and digital transformation are topics that
have been in the focus of a large number of interest
groups in recent years. Nevertheless, a large number of
developed concepts and case studies are not applicable to
a significant part of the metal forming industry. For this
reason, this chapter deals with what the author believes
to be the greatest challenges in metal forming technology
in relation to digitalization and digital transformation,
supported by current scientific publications. Despite the
technical component, also industrial-economical and
partly also legal components are considered.
3.1 RETROFITTING
Retrofitting in the context of the fourth industrial
revolution is defined as the upgrading of machine
systems to make them viable for I 4.0 applications. In
scientific publications, this process is usually based on
the planning, implementation and validation of suitable
infrastructure, communication and applications. In
general, the goal of these procedures is to turn machines
with older technologies into fully functional CPPS. This
enables them to connect to existing IIoT networks and
big data applications. Considering the high degree of
heterogeneity, especially in metal forming industry, this
seems to be one of the main challenges in implementing
a smart production unit (fully digitalized production).
Some work in this field deals with the creation of
standardized frameworks, which should allow a
structured approach for retrofitting. However, in the case
of forming technology, these frameworks are either too
general to directly initiate necessary steps (e.g. no
suitable recommendation of defined interfaces,
hardware, software (SW)) [57] or too specific (only
suitable for a defined use case [58]). Based on renowned
publications between 2014 and 2018 [59,60,61], Lins et
al. [57] summarized the requirements for successful
conversion of older systems to full CPPS, by retrofitting
in a partially digitalized environment, in 13 points (Table
1, modified by the author). For the first-time
digitalization in a non-industry 4.0 environment, a
similar procedure can be followed once at least IIoT
technology has been implemented. Recommendations
for a first-time introduction of IIoT are the provision of
high speed internet coverage and the server architecture
required for local data processing. Sufficient resources
for sustainable IT security should also be considered.
Big data applications and mostly embedded artificial
intelligence systems as well as the associated data
expenditure must be taken into account in the planning
stage. The storage location of the data volume should
also be defined in the planning phase (locally or via
cloud (2.6)). An essential point is also the choice of IT
standards used (e.g. MQTT, OPC UA, MES, general
layer architecture). Appropriately trained personnel must
be considered if not available in the own company (for
planning, implementation and also ongoing operation
and maintenance).
Table 1: Retrofit approach for the development of a
CPPS for an existing I 4.0 environment [57]
Infrastructure
1
Identification and visualization of requirements
and potential improvements, for each sub-process
and machinery to retrofit
2
Adding of suitable IIoT Devices directly to the
selected machinery (e.g. smart sensors)
3
Adding associated, but not directly connected IIoT
Devices to the machinery (e.g. IIoT gateways)
Communication
4
Identification and visualization of existing
communication technologies and protocols with I
4.0 standard, if applicable (e.g. MQTT protocols)
5
Integration of existing not integrated
communication technologies in an I 4.0 network
(e. g. through IIoT gateways )
6
Integration of communication management,
avoiding usage of a network manager for different
communication types
7
Support for IIoT networks (e.g. I 4.0 specialists)
8
Implementation of real-time communication
between all production levels (most important
between SCADA and MES level)
Application
9
Identification and visualization of existing,
implemented software and applications and
required software/applications for running I 4.0
applications on the system to retrofit
10
Adding of interfaces for the connection of not I
4.0 SW and applications to I 4.0 SW and
applications (e.g. connections to cloud)
11
Integration of existing not I 4.0 ready SW to I 4.0
applications which are part of existing CPPS (e.g.
OPC UA interfaces from SPS)
12
Implementation of monitoring applications for the
supervising of all generated data of the
retrofitting system in conjunction with added IIoT
devices (HMI via various GUIs)
13
Implementation of remote access technology for
the users of the CPPS
3.2 HUMAN MACHINE INTERACTION
(INDUSTRY 5.0)
HMI technologies are in the focus of current research.
Some recognized researchers in the field of digitalization
also refer to this focus as Industry 5.0 [62]. In this case it
is not only about the purely technical component, but
primarily about interaction challenges in human-robot
collaboration. In the metal forming industry, the author
believes that the interaction between complex process-
optimizing algorithms and employees involved is a
particular challenge in this context, which must be

considered when implementing an I 4.0 solution in the
manufacturing process [63,64].
3.3 DIGITAL TWIN INTEGRATION
In the metal forming industry, the simulation of
processes and the resulting material behavior is of high
importance. Important process parameters (e.g. material
flow, temperature range, force required) as well as the
resulting material characteristics (e.g. strength, residual
stress, temperature resistance) can be supported by the
use of FEA to replace costly and uneconomical practical
tests. Furthermore, theoretical process planning and
process optimization can be investigated and validated
quickly and cost-effectively. Due to the advancing
digitalization and the use of I 4.0 technologies, it is also
possible to use FEA in the form of digital twins for in
situ process optimization. The biggest challenges are the
efficient programming and abstraction of corresponding
FEAs to make such a process synchronous calculation
possible. Furthermore, the automated integration of FE
programs into the IT architecture of a digital factory is
only possible by means of standardized interfaces.
However, the difficulty of programming such interfaces
can vary greatly depending on the supplier [24].
3.4 VERTICAL INTEGRATION
From a purely production engineering perspective,
vertical integration offers one of the greatest potentials
for optimizing production processes. Successfully
considered already in the planning phase, it can support
greenfield approaches (completely new planning of
digital factories) to make economic sense in the first
place. When it comes to the digitalization of existing
factories (brownfield approach), a transparent vertical,
but also horizontal integration can quickly become
complex. This is caused, among other aspects, by
historically grown IT and machine infrastructure, but
also by organizational structures. For this reason, in
many cases a multitude of measures are necessary to
successfully implement a transparent integration. In
addition to the selection of suitable personnel, suitable
hardware and software, the accompanying use of change
management techniques is also unavoidable in most
cases.
3.5 DATA SECURITY AND LEGAL ASPECTS
The use of big data and artificial intelligence
applications in the metal forming industry may be
outsourced to external providers, if there is a lack of
human resources. Many of these providers also include
in their scope of services the physical storage of data to
be analyzed from the production process. In addition to
the advantages of lower administrative and personnel
costs, there are numerous risks associated with this
approach. In many cases the physical storage of data
takes place in specially designed server farms, which are
often operated in countries that are subject to
fundamentally different legal obligations. For this
reason, a legal examination of contracts with third party
providers regarding data security and also data
ownership rights must be taken into account when
implementing such applications (2.5, 2.6).
3.6 INTERDISCIPLINARITY AND EDUCATION
The implementation of a holistic digitalization solution
generally requires a high level of interdisciplinary
expertise. In addition to traditional engineering
knowledge of forming processes, basic knowledge of
network technology, programming languages,
production logistics and industrial economics is required,
even when outsourcing to third parties. Smart Production
concepts also require a change in the organizational
structure. According to a personal interview of the
author with a well-known consulting company in heavy
industry, the greatest challenges in the implementation of
Industry 4.0 technologies are not the provision and use
of modern technologies, but rather the inadequate
coordination of those responsible for production and the
internal IT department. These departments often have
too limited resources to plan, implement and maintain a
digitalization solution. It is therefore recommended to
involve the responsible persons as early as possible.
Depending on the size of the company, more and more
companies are also using a separate instance for the
digital transformation of production. The most
prominent example of this approach is the role of the
Chief Digital Officer (CDO). In most cases, the CDO
does not replace the IT manager, but is deployed in
parallel to him specifically for innovations in the area of
the fourth industrial revolution. The combination of top-
down commitment through the implementation of such
functions in the organizational structure, and the
promotion of bottom-up commitment through the use of
change management at all hierarchical levels,
significantly increases the probability of a company's
successful digital transformation. In order to be able to
meet the increasing demand for qualified specialists in
the future from today's perspective, appropriate training
and further training measures must be developed. Figure
10 shows an example of the knowledge required in the
field of digital and advanced analytics (DnA) for a
company in heavy industry [65].
Figure 10: Required Skills for Data Analytics in heavy
industry [65].

4 CONCLUSION AND OUTLOOK
The challenges listed in Chapter 3 represent major
obstacles to the success and competitiveness of the
Austrian metal forming industry. In addition, it must
always be taken into account that even the simplest form
of digitization can only bring an advantage if the
preceding process is mastered and capable. Traditional
concepts of Lean Management must therefore always be
considered as an initial instance. It should also be
ensured in advance that manual and automated data,
which are digitized or fed into a digitalization system,
are complete and valid.
Digitalization and digital transformation are research
priorities of a large number of academic and private
research institutions. For this reason, several research
priorities regarding digitalization were set at the Chair of
Metalforming. Since 2019, in cooperation with the
company ibaAG, the Chair of Metalforming has been
working on the networking of the most important in-
house forming technology aggregates [66]. Furthermore,
the possibilities of integrating an FEA-based digital twin
into a Smart Production Lab are being investigated in the
context of a dissertation. This is done in cooperation of
the University of Leoben and the FH Joanneum,
University of Applied Sciences. In 2020, the Chair of
Metalforming will also develop a digital shadow at
SCADA level using Simatic S7 1200 control and
Simufact FEA. Furthermore, a quantitative survey of the
digital maturity level of the metal forming industry in
Austria will be carried out by 2021. Within the
framework of this survey, potential correlations between
economic success and the degree of digitalization will
also be determined. In the field of academic education,
digitalization in the context of the metal forming
industry will also be a major focus in 2021. In this
context, the digitalization projects at the chair, for
practical illustration of theory, will also be included.
These projects should make an important contribution to
preserve and expand the competitiveness of the Austrian
metal forming industry.
LITERATURE
[1] Alcácer V., Cruz-Machado V.: Scanning the
Industry 4.0: A Literature Review on Technologies
for Manufacturing Systems, Engineering Science
and Technology, an International Journal, 22 (2019)
899-919.
[2] Ghobakhloo M.: The future of manufacturing
industry: a strategic roadmap toward Industry 4.0,
Journal of Manufacturing Technology Management,
29 (2018) 910-936.
[3] Mittal S., Khan M. A., Romero D., Wuest T.: A
critical review of smart manufacturing & Industry
4.0 maturity models: Implications for small and
medium-sized enterprises (SMEs), Journal of
Manufacturing Systems, 49 (2018) 194-214.
[4] Osterrieder P., Budde L., Friedli T.: The smart
factory as a key construct of industry 4.0: A
systematic literature review, International Journal of
Production Economics, in press (2020), doi:
https://doi.org/10.1016/j.ijpe.2019.08.011.
[5] Hagenah H., Schulte R., Vogel M., Hermann J.,
Scharrer H., Lechner M., Merklein M.: 4.0 in metal
forming – questions and challenges, Procedia CIRP,
79 (2019) 649-654.
[6] Culot G., Nassimbeni G., Orzes G., Sartor M.:
Behind the definition of industry 4.0: Analysis and
open questions, International Journal of Production
Economics, accepted paper (2020), doi:
https://doi.org/10.1016/j.ijpe.2020.107617.
[7] Kerin M., Pham D. T.: A review of emerging
industry 4.0 technologies in remanufacturing,
Journal of Cleaner Production, 237 (2019) 1-16.
[8] da Silva E. H. D. R., Shinohara A. C., de Lima E. P.,
Angelis J., Machado C. G.: Reviewing Digital
Manufacturing concept in the Industry 4.0
paradigm, Procedia CIRP, 81 (2019) 240-245.
[9] Dilberoglu U. M., Gharehpapagh B., Yaman U.,
Dolen M.: The Role of Additive Manufacturing in
the Era of Industry 4.0, Procedia Manufacturing, 11
(2017) 545-554.
[10] Wang Y., Towara T., Anderl R.: Topological
Approach for Mapping Technologies in Reference
Architectural Model Industrie 4.0 (RAMI 4.0),
Proceedings of the World Congress on Engineering
and Computer Science (WCECS), Vol II (2017),
San Francisco, USA.
[11] Pisching M. A., Pessoa M. A. O., Junqueira F., dos
Santos Filho D. J., Miyagi P. E.: An architecture
based on RAMI 4.0 to discover equipment to
process operations required by products, Computers
& Industrial Engineering, 125 (2018) 574-591.
[12] “Referenzarchitekturmodell Industrie 4.0 (RAMI
4.0)”, DIN SPEC 91345 (2016).
[13] Mourtzis D., Gargallis A., Zogopoulos V.:
Modelling of Customer Oriented Applications in
Product Lifecycle using RAMI 4.0, Procedia
Manufacturing, 28 (2019) 31-36.
[14] Schaefer I., Cleophas L., Felderer M.: Model-driven
Requirements Engineering Using RAMI 4.0 Based
Visualizations, Workshops at Modellierung 2018,
Adequacy of Modeling Methods (AQEMO) (2018)
21-30.
[15] Lee E. A., Cyber-Physical-Systems – Are
Computing Foundations Adequate?, Position Paper
NSF Workshop Cyber Physical Systems (2006),
Austin (Texas), USA.
[16] Brettel M., Friederichsen N., Keller M., Rosenberg
N.: How virtualization, decentralization and network
building change the manufacturing landscape: An
Industry 4.0 Perspective, International Journal of
Science, Engineering and Technology, 8 (2014) 37-
44.
[17] Lee J., Bagheri B., Kao H.-A.: A Cyber-Physical-
Systems architecture for Industry 4.0-based
manufacturing systems, Manufacturing Letters, 3
(2015) 18-23.

[18] Monostori L.: Cyber-physical production systems:
Roots, expectations and R&D challenges, Procedia
CIRP, 17 (2014) 9-13.
[19] N.N. Cyber-Physical-Systems: Driving force for
innovation in mobility, health, energy and
production, acatech Position Paper, 48 (2011).
[20] Cardin O.: Classification of cyber-physical
production systems applications: Proposition of an
analysis framework, Computers in Industry, 104
(2019) 11-21.
[21] Große Kamphake A.: Digitization in Controlling,
Springer Gabler Wiesbaden, (2020) 3-25, doi:
https://doi.org/10.1007/978-3-658-28741-2_2.
[22] VDI/VDE-Gesellschaft Mess und
Automatisierungstechnik (GMA), Cyber-physical
systems: Chancen nutzen aus Sicht der Automation,
Thesen und Handlungsfelder, (2013).
[23] Bano H. F., Garcia C. A., Cabrera A A., Castellanos
E. X., Naranjo J. E., Garcia M. V.: Low-Cost
Embedded System for Shop Floor Communications
and Control Based on OPC-UA, Advances in
Emerging Trends and Technologies (ICAETT) 2
(2019), doi: https://doi.org/10.1007/978-3-030-
32033-1_1.
[24] Ralph B.: Kategorisierung und Auswahl geeigneter
numerischer Simulationssoftware für praktische
Problemstellungen in der umformtechnischen
Industrie im Kontext der vierten industriellen
Revolution, FH Joanneum University of Applied
Science, Master Thesis (2019), Institute of Industrial
Management, Kapfenberg, Austria.
[25] Boyes H., Hallaq B., Cunningham J., Watson T.:
The industrial internet of things (IIoT): An analysis
framework, Computers in Industry, 101 (2018) 1-12.
[26] Liao Y., de Freitas Rocha Loures E., Deschamps F.:
Industrial Internet of Things: A Systematic
Literature Review and Insights, IEEE Internet of
Things Journal, 5 (2018) 4515-4525.
[27] Cheruvu S., Kumar A., Wheeler D.M.: IoT
Frameworks and Complexity, Demystifying Internet
of Things Security (2020),Apress, Berkeley
(California), USA, doi: https://doi.org/10.1007/978-
1-4842-2896-8_2.
[28] Ben-Daya M., Hassini E., Bahroun Z.: Internet of
things and supply chain management: a literature
review, International Journal of Production
Research, 57 (2017) 4719-4742.
[29] Cui Y., Kara S., Chan K. C.: Manufacturing big data
ecosystem: A systematic literature review, Robotics
and Computer Integrated Manufacturing, 62 (2020)
1-20.
[30] Zhekun L., Gadh R., Prabhu B. S.: Applications of
RFID technology and smart parts in manufacturing,
Proceedings of DETC, 4 (2004) 1-7.
[31] Zhao J., Olivieri H., Seppänen O., Peltokorpi A.,
Badihi B., Lundström P.: Data analysis on applying
real time tracking in production and control of
construction, 2017 IEEE International Conference
on Industrial Engineering and Engineering
Management (IEEM), (2017) 573-577, Singapore,
doi: 10.1109/IEEM.2017.8289956.
[32] Strandhagen J. O., Vallandingham L. R., Fragapane
G., Strandhagen J. W., Stangeland A. B. H., Sharma
N.: Logistics 4.0 and emerging sustainable business
models, Advances in Manufacturing, 5 (2017) 359-
369.
[33] Hofman E., Rüsch M.: Industry 4.0 and the current
status as well as future prospects on logistics,
Computers in Industry, 89 (2017) 23-34.
[34] Barreto L., Amaral A., Pereira T.: Industry 4.0
implications in logistics: an overview, Procedia
Manufacturing, 13 (2017) 1245-1252.
[35] N. N.: Internet Security Threat Report 2015, 20
(2015), Symantec Corporation.
[36] Liptak A.: Renault Shut Down Several French
Factories after Cyberattack 2017,
https://www.theverge.com/2017/5/14/15637472/ren
ault-nissan-shut-down-french-uk-factories-
wannacry-cyberattack (accessed January 13, 2020).
[37] Elhabashy A. E., Wells L. J., Camelio J. A.: Cyber-
Physical Security Research Efforts in Manufacturing
– A Literature Review, Procedia Manufacturing, 34
(2019) 921-931.
[38] Albrecht F.: Informations- und
Kommunikationsrecht: Lehrbuch für das gesamte
IT-Recht, Kohlhammer Verlag (2018), ISBN:
9783170314054.
[39] Flatt H., Schriegel S., Jasperneite J., Trsek H.,
Adamczyk H.: Analysis of the Cyber-Security of
industry 4.0 technologies based on RAMI 4.0 and
identification of requirements, 2016 IEEE 21st
Conference on Emerging Technologies and Factory
Automation (ETFA), (2016) 1-4, Berlin, Germany.
[40] Grieves M.: Digital Twin: Manufacturing
Excellence Through Virtual Factory Replication,
2015,
https://www.researchgate.net/publication/27521104
7_Digital_Twin_Manufacturing_Excellence_throug
h_Virtual_Factory_Replication , (accessed January
13, 2020).
[41] Kritzinger W., Karner M., Traar G., Henjes J., Sihn
W.: Digital Twin in manufacturing: A categorical
literature review, IFAC PapersOnLine, 51 (2018)
1016-1022.
[42] Zhang H., Ma L., Sun J., Lin H., Thürer M.: Digital
Twin in Services and Industrial Product Service
Systems: Review and Analysis, Procedia CIRP, 83
(2019) 57-60.
[43] Redelinghuys A. J. H., Basson A. H., Kruger K.: A
six-layer architecture for the digital twin: a
manufacturing case study implementation, Journal
of Intelligent Manufacturing, (2019), doi:
10.1007/s10845-019-01516-6.
[44] Bortz M.: Grey-Box-Modelle – Integration von
Anwendungswissen in Lernverfahren, ITWM
Fraunhofer,
https://www.itwm.fraunhofer.de/de/abteilungen/opt/
maschinelles-lernen-hybride-modelle/grey-box-

modelle-fraunhofer-projekt.html (accessed January
13, 2020).
[45] Demchenko Y., De Laat C., Membrey P.: Defining
architecture components of the Big Data Ecosystem,
2014 International Conference on Collaboration
Technologies and Systems (CTS), IEEE (2014) 104-
112.
[46] Chen J., Chen Y., Du X., Li C., Lu J., Zhao S., Zhou
X.: Big data challenge: a data management
perspective, Frontiers of Computer Science, 7
(2013) 157-164.
[47] Khan M. A., Uddin M. F., Gupta N.: Seven V´s of
Big Data understand Big Data to extract value,
Proceedings of the 2014 Zone 1 Conference of the
American Society for Engineering Education,
(2014) 1-5, Bridgeport (Connecticut), USA.
[48] Ghasemaghaei M.: Understanding the impact of big
data on firm performance: The necessity of
conceptually differentiating among big data
characteristics, International Journal of Information
Management, in press (2020), doi:
https://doi.org/10.1016/j.ijinfomgt.2019.102055.
[49] Zikopoulos P., Eaton C.: Understanding Big Data:
Analytics for Enterprise Class Hadoop and
Streaming Data, 1st ed. McGraw-Hill Osborne
Media (IBM) (2011), ISBN: 9780071790543.
[50] N. N.: Infinidat Announces “Scale to Win” Storage
Vision for the Multi-Petabyte Data Center to drive
Customers’ Competitive (2019),
https://www.bloomberg.com/press-releases/2019-
05-07/infinidat-announces-scale-to-win-storage-
vision-for-the-multi-petabyte-data-center-to-drive-
customers-competitive (accessed January 14, 2020).
[51] Zicaro R. V., Rosselli M., Ivanov T., Korfiatis N.,
Tolle K., Niemann R., Reichenbach C.: Setting up a
Big Data Project: Challenges, Opportunities,
Technologies and Optimization, Big Data
Optimization: Recent Developments and
Challenges, 6th Chapter (2015), doi: 10.1007/978-3-
319-30265-2_2.
[52] Ghasemaghaei M., Calic G.: Assessing the impact
of big data on firm innovation performance: Big
data is not always better data, Journal of Business
Research, 108 (2020) 147-162.
[53] Ye Y., Wang M., Yao S., Jiang J. N., Liu Q.: Big
data processing framework for manufacturing,
Procedia CIRP, 83 (2019) 661-664.
[54] Islam M. T., Srirama S. N., Karunasekera S., Buyya
R.: Cost-efficient Dynamic Scheduling of Big Data
Applications in Apache Spark on Cloud, The
Journal of Systems & Software, in press (2020), doi:
https://doi.org/10.1016/j.jss.2019.110515.
[55] Mell P., Grance T.: The NIST definition of cloud
computing, NIST Special Publication 800-145,
(2011) 1-7.
[56] Sheikh A., Munro M., Budgen D.: Systematic
Literature Review (SLR) of resource scheduling and
security in cloud computing, International journal of
advanced computer science and applications, 10
(2019) 35-44.
[57] Lins T., Oliveira R. A. R.: Cyber-physical
production systems retrofitting in context of
industry 4.0, Computers & Industrial Engineering,
139 (2020) 106193, doi:
https://doi.org/10.1016/j.cie.2019.106193.
[58] Baker J., et al.: Requirements Engineering for
Retrofittable Subsea Equipment, 2016 IEEE 24th
International Requirements Engineering Conference
(RE), (2016) 226-235, Beijing, China.
[59] Jazdi N.: Cyber physical systems in the context of
industry 4.0, 2014 IEEE international conference on
automation, quality and testing, robotics, (2014) 1-4,
doi: 10.1109/AQTR.2014.6857843.
[60] Hermann M., Pentek T., Otto B.: Design Principles
for Industry 4.0 Scenarios, 2016 49th Hawaii
International Conference on System Sciences
(HICSS), (2016) 3928-3937, doi:
10.1109/HICSS.2016.488.
[61] Lins T., Oliveira R. A. R., Correia L. H. A., Sá Silva
J.: Industry 4.0 Retrofitting, 2018 VIII Brazilian
Symposium on Computing Systems Engineering
(SBESC), (2018) 8-15, Salvador, Brazil, doi:
10.1109/SBESC.2018.00011.
[62] Skobelev P. O., Borovik S. Yu.: On the way from
Industry 4.0 to industry 5.0: from digital
manufacturing to digital society, Industry 4.0, 6
(2017) 307-311.
[63] Özdemir V., Hekim N.: Birth of Industry 5.0:
Making Sense of Big Data with Artificial
Intelligence, “The Internet of Things” and Next-
Generation Technology Policy, OMICS: A Journal
of Integrative Biology, 22 (2018) 65-76.
[64] Zhou J., Li P., Zhou Y., Wang B., Zang J., Meng L.:
Toward New-Generation Intelligent Manufacturing,
Engineering, 4 (2018) 11-20.
[65] Korbel M., Sim S., van Niel J.: Enabling a digital
and analytics transformation in heavy-industry
manufacturing, McKinsey & Company Operations
Practice, (2019) 1-8,
https://www.mckinsey.com/~/media/McKinsey/Busi
ness%20Functions/Operations/Our%20Insights/Ena
bling%20a%20digital%20and%20analytics%20tran
sformation%20in%20heavy%20industry%20manufa
cturing/Enabling-a-digital-and-analytics-
transformation-in-heavy-industry-
manufacturing.ashx (accessed January 18, 2020),
reproduced with permission.
[66] Ralph B. J., Schwarz A., Stockinger M.: An
Implementation Approach for an Academic
Learning Factory for the Metalforming Industry
with Special Focus on Digital Twins and Finite
Element Analysis, Procedia Manufacturing,
accepted (2020).