Towards Data-driven Production: Analysis of Data Models Describing
Machinery Jobs in OPC UA
Tonja Heinemann
a
, Marwin Gihr, Oliver Riedel
b
and Armin Lechler
c
University of Stuttgart, Institute for Control Engineering of Machine Tools and Manufacturing Units,
Seidenstraße 36, Stuttgart, Germany
Keywords:
OPC UA, Information Model, Standardization.
Abstract:
This work analyzes the Open Platform Communications Unified Architecture (OPC UA) specifications for flat
glass, plastics and rubber, machine vision, ISA-95 and machine tools regarding their job descriptions. Com-
mon contents of job models in the domain of machinery are deducted. Using a structured qualitative content
analysis, more than 70 functional elements used in OPC UA job models have been identified. While some
of these functional elements are modeled similarly in multiple domains, major differences are identified for
other functional elements. Especially those differences constitute impediments in the standardization of indus-
trial communication. The results of this work harmonize the contents and the modeling techniques regarding
machining jobs in OPC UA and provide a generally applicable method for the standardization of machine
communication throughout different domains. With this method for standardization, this work contributes
directly to the goal of OPC UA, to easily exchange data between platforms from multiple vendors.
1 INTRODUCTION
With the advent of Industry 4.0, cloud manufactur-
ing and lot size 1 in production, the data transfer of
machines between each other and between control-
ling systems like Manufacturing Execution Systems
(MES) raises in importance. A popular standard to
use for such communication today is Open Platform
Communications Unified Architecture (OPC UA).
OPC UA allows specifying data models for specific
use cases and specific domains. This opportunity is
used in the 36 Companion specifications (CS) pub-
lished today as well as in the specifications being cur-
rently developed (OPC Foundation, 2022c).
An important part of data transmission in the ma-
chinery domain is a description of jobs. A job is
considered the source of all activities, as well as data
container for all information and efforts necessary for
processing or originating from processing (Informa-
tionstechnik, 2016). Multiple of the existing CS de-
fine such jobs. These existing models have similar
intentions and overlap in content, but are different by
definition.
a
https://orcid.org/0000-0001-8601-7820
b
https://orcid.org/0000-0002-1883-6813
c
https://orcid.org/0000-0002-4073-1487
To solve such problems, especially for newly de-
veloped CS, but also in updates of the existing CS,
harmonization groups have formed (OPC Foundation,
2022b; VDMA e.V., 2022). These groups need to
know the contents of existing specifications to include
all the functionality that is already provided. Such an
overview is developed in this work.
This paper is structured as follows: In section 2,
the examined CS in this work are introduced along
with a brief description of the role of CS in OPC
UA. The content analysis used for the overview is de-
scribed in section 3. This method involves the defini-
tion of categories. These are introduced in section 4.
The resulting overview is presented in section 5, sec-
tion 6 gives context for the results. The last section 7
contains a discussion of the method used and the re-
sults generated.
2 OPC UA AND COMPANION
SPECIFICATIONS
The ISO-Standard OPC UA composes multiple exist-
ing paradigms in data transport for use in industrial
environments (IEC 62541-[1-14]:2020, 2020). These
include transport protocols, data formats, communi-
Heinemann, T., Gihr, M., Riedel, O. and Lechler, A.
Towards Data-driven Production: Analysis of Data Models Describing Machinery Jobs in OPC UA.
DOI: 10.5220/0011142900003271
In Proceedings of the 19th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2022), pages 729-736
ISBN: 978-989-758-585-2; ISSN: 2184-2809
Copyright
c
2022 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
729
Figure 1: Hierarchy of Information Models in OPC UA
(OPC Foundation, 2019).
cation paradigms such as server-client and publish-
subscribe and means of secure communication like
encryption algorithms. On top of this technological
foundation of data transport, OPC UA defines rules
for data representation and collaborating parties use
these rules to define data representations for differ-
ent domains (OPC Foundation, 2022c). These data
representations are called information models and are
hierarchically structured as shown in figure 1. The
main goal of information models is to describe the
structure of data and its intended usage on an OPC
UA interface. The “Core Information Models” con-
tain the most general aspects, intended to be reused
in all subsequent information models. “Companion
Information Models” are often defined by collaborat-
ing parties and contain data models for a specific use-
case or domain. The models regarded in this work are
all companion information models, described in so-
called companion specifications (CS) (40501-1, 2020;
40083, 2021; 40077, 2020; 40301, 2022; 10030,
2013; 10031-4, 2021; 30260, 2020; 40100-1, 2019).
For individual information models, often defined by a
single company, there is the information model group
of “Vendor Specific Extensions”.
The CS are defined by a so-called joint working
group consisting of OPC Foundation members of the
respective domain. This working group defines the
applications and use cases for the CS, develops the
information model and edits the CS documentation.
Resulting CS do include concepts, that are similar in
principle, but handled differently in various CS. Such
a concept is the representation of production jobs.
Based on the description of all CS published by the
OPC Foundation (OPC Foundation, 2022a), the CS
for Machine Tools, Plastics and Rubber Machinery,
Flat Glass, OPEN-SCS, ISA-95 and Machine Vision
contain an information model describing jobs (40501-
1, 2020; 40083, 2021; 40077, 2020; 40301, 2022;
10030, 2013; 10031-4, 2021; 30260, 2020; 40100-1,
2019).
In OPC 40501, the CS for machine tools, the job
model focuses on times spent per program, part and
job for a communication for systems like MES (Man-
ufacturing Execution Systems). The individual enti-
ties, programs, parts and jobs, have key indicators like
a unique identifier. However, more complex proper-
ties like identification of material lots used in parts or
a detailed representation of subprogram structures is
not included. The machine tools model is represent-
ing the job without providing an interface to control
it. (40501-1, 2020)
Two of the specifications for plastics and rubber
machinery contain information about jobs, namely
OPC 40083 and OPC 40077. In combination, they
provide a communication interface between machines
and MES systems. The job description contains the
planned jobs and related information as well as a man-
agement interface for production data like programs.
Using this interface, some aspects of production like
enabling and disabling automatic runs, can be con-
trolled by an OPC UA client. (40083, 2021; 40077,
2020)
The model of OPC 40301 aims to provide a com-
munication interface between MES or ERP (Enter-
prise Resource Planning) and glass processing ma-
chines. It represents jobs, instructions used for pro-
duction and the material used in production. In addi-
tion to representing the state of jobs, the model also
allows to manage jobs by e.g. adding, deleting, sus-
pending and releasing jobs. (40301, 2022)
The specifications OPC 10030 and 10031-4 are
mappings of the ISA-95 standard defined by the ISA
(International Society of Automation) to communi-
cate between MES and diverse manufacturing soft-
ware systems. OPC 10030 contains models for man-
agement of material, personnel and components. In
OPC 10031-4, these models are extended by means
to control machine jobs. The jobs are connected to
the related material, equipment, physical assets and
personnel. In addition, the interface allows to control
jobs. (10030, 2013; 10031-4, 2021)
The specification of OPC 30260, representing the
Open Serialization Communication Standard (OPEN-
SCS) in OPC UA, utilizes the ISA-95 standard and is
thus implicitly represented in this work, but not ex-
plicitly analyzed (30260, 2020).
In OPC 40100-1, communication among machine
vision systems and between vision systems and con-
trollers or MES is specified. Machine vision systems
are used e.g. to gain information about production
quality and to identify products. In the OPC UA
model, jobs are displayed as “recipes”, and can be
modified over the interface.
Even though the different models exist and are
known to the specification groups, no comparison or
comprehensive overview has been developed so far.
With such a comparison, the individual concepts and
aspects of each model can be compared and be used
as groundwork for a harmonized model containing all
ICINCO 2022 - 19th International Conference on Informatics in Control, Automation and Robotics
730
Figure 2: Qualitative Content Analysis for OPC UA CS.
aspects. For such a harmonized model, all informa-
tion needed for production and thus also the informa-
tion contained in the job model, needs to be included.
In addition, the tasks in job management that are cur-
rently contained in the individual specifications need
to be portrayed. The analysis shown in this work
aims to find the information contained in job mod-
els as well as the tasks in job management necessary
to transmit between communication partners as indi-
cated by the existing OPC UA job models.
3 METHOD
To identify the information contained in existing OPC
UA job models as well as the tasks in job management
displayed in these OPC UA job models, a structured
qualitative content analysis is performed.
This kind of analysis strives to generate a detailed
understanding of the models in the specifications. For
this analysis, first a set of main categories is defined.
The first set of categories is defined a-priori by means
of a literature review. In Figure 2 this is depicted as
step 1. Individual elements of the OPC UA models
are then categorized according to the category set; in
the end all elements need to be included in a category.
This is shown in step 2 in Figure 2. The category
overview in the sketch contains all elements with their
sources and additional information. By analyzing the
elements in each category, finding similarities and dif-
ferences, sub-categories can be deducted based on the
evaluated contents. In figure 2 this is represented by
step 3. The result is an overview of functional ele-
ments in CS, that can be used as requirements for the
generation of new specifications or for harmonization
of existing CS. In this step, the previous categories
can be modified if needed. If so, the elements in the
specification are assigned to the new categories again.
For this reason, the process is drawn as a circle in fig-
ure 2. In the end, all delimitable elements are assigned
to at least one category. (Kuckartz, 2012)
Table 1: Categories for Job Models Defined in OPC UA CS.
Main category Sub category
Data for job
preparation
Ensure capability
Data for job execution -
Data during job
execution
Production state
Production mode
Execution information
Parameter monitoring
Data after job
execution
-
Resource capability
description
Equipment description
Production data set
description
Configuration data set
description
Variant parameter
description
Personnel description
Material description
Job management -
Production
management
Machine control
Production mode
Device control
Production parameters
Result management -
Safety Functional reliability
Security Authenticity
Availability
4 CATEGORY SYSTEM FOR
EXAMINED JOB MODELS
Table 1 shows the categories for the information and
task management elements contained in a job. The
data for job preparation concerns checking the capa-
bilities and capacities of machines needed in advance
to schedule and start production activities. As soon
as a job is scheduled, the information required by ma-
chines to execute the job is sent to the machine.
During job execution, the machine transmits in-
formation for systems to monitor the production.
This category is subdivided in information about
the state and the operation mode of the machine, as
well as in information about the job execution itself
and the monitoring of job-related parameters.
After job execution, results and information for
product tracing are transmitted.
Towards Data-driven Production: Analysis of Data Models Describing Machinery Jobs in OPC UA
731
The capabilities of machines are not only needed
in advance to start a job but also to manage it during
execution. Furthermore, they need to be adapted in-
dependent of jobs, like information about tools avail-
able on the machine. For this reason, a category for
resource capability description is considered. Subcat-
egories include the description of equipment used in
production, like tools and fixtures, personnel required
for the job, and material used in production. The pro-
duction data set description contains programs and in-
structions for production. The description of configu-
ration data sets concerns settings that can be made on
machines and production equipment. And the variant
parameter description contains all information that
changes in different variants of similar products.
The category job management contains elements,
e.g. used to add or delete jobs on machines.
All information concerning functions like starting
or stopping the job is gathered in the category pro-
duction management. The respective subcategories
further specify if actions concern the machine con-
trol, change the production mode, control additional
devices apart from the machine itself or changes of
parameters like controller settings or tolerances.
If the communication partner can control how and
how long results are available, especially independent
of active jobs, the respective elements are categorized
as result management.
For the operation of manufacturing equipment,
safety plays a crucial role. As to communication part-
ners, ensuring the functional reliability by informing
about errors and invalid system states is identified in
the OPC UA specifications.
In terms of security, OPC UA itself provides op-
tions for authentication of communication partners, as
well as methods to encrypt the data transfer. For this
reason, these aspects don’t need to be handled explic-
itly in companion specifications. Nevertheless, in the
examined specifications, aspects related to authentic-
ity and availability and thus to security are included.
5 RESULTS
Within the categories and sub-categories, the ele-
ments are evaluated and grouped using the following
common functional elements. More than 70 func-
tional elements for job models are found. The cate-
gories and the functional elements therein are shown
in table 2. In the rightmost five columns, the exis-
tence of a functional element in the specifications is
shown with an “x”. The following text will use the
abbreviations for the specifications used in the table.
In the category “Data for job preparation”, PAR,
G and V are represented. The models provide lists
of configurations and available job data like machine
programs.
Concerning the category “Data for job execution”,
MT is the only specification without elements in this
category. All other specifications’ models contain
most of the functional elements identified. For the
functional element “Production data set”, different
modeling approaches are used to link a data set to a
job. PAR, I95 and V transmit the data set ID to link
to the job, while G requires to transmit the data set
along with the job. In similar fashion, material data
is referenced by ID in PAR, I95 and V while for G
all types of material are provided wile creating a job.
I95 only makes a difference between resources and
resource classes. A resource class represents the ma-
terial properties in an abstract fashion (e.g. size) while
resources are directly linked to real material (e.g. with
lot number).
The functional elements in “Data during job exe-
cution” are apparent in all specifications. They follow
similar OPC UA modeling principles in all specifica-
tions. The same can be said for the three specifica-
tions implementing “Data after job execution”.
Each specification contains functional elements of
the category “Resource capability description”. How-
ever, MT and G only implement few functional ele-
ments while PAR, I95 and V implement the majority.
Concerning similarities and differences in modeling,
one sub category is taken as an example. As a sub
category, the material description occurs in all spec-
ifications but MT. In I95, the material is represented
in an array structure containing material IDs. In PAR
and V, a list containing entries that directly represent
the material is used. G, on the other hand, uses an
array containing references to the OPC UA represen-
tation of the material.
The category “Job management” again is present
in all specifications. The MT specification however
only implements one functional element, “Display
job plans”. Concerning job plans, the representations
differ: MT and G use a list, while I95 and PAR con-
tain the priority and intended start time of jobs. While
the list in MT and G represents the production order
of jobs, I95 uses an array for this information while
PAR displays the current and the upcoming job. For
the functional element “Receive job”, however, PAR,
G, I95 and V use OPC UA methods, and thus similar
modeling concepts.
ICINCO 2022 - 19th International Conference on Informatics in Control, Automation and Robotics
732
Table 2: Overview of functional elements and their occurrence in CS.
PAR: OPC 40083 and OPC 40077 (Plastics and Rubber)
G: OPC 40301 (Flat Glass)
I95: OPC 10030 and OPC 10031-4 (ISA-95)
V: OPC 40100-1 (Machine Vision)
MT: OPC 40501 (Machine Tools)
Main category Sub category Functional element PAR G I95 V MT
Data for job prepara-
tion
Ensure capability Filter and transmit capabilities x x
Prerequisites x
Add capabilities x x
Request capabilities x
Data for job execu-
tion
Job identification x x x x
Job description x x
Production data set x x x x
Variant parameters x x x
Job meta data x x x
Material data x x x x
Equipment data x x
Personnel data x
Instructions x
Data during job exe-
cution
Production state Informative x x x x x
Interaction needed (Server) x x x x x
Interaction needed (Client) x
Production mode Informative x x x
Execution information Job identification x x
Job description x
Process duration x x x x
Products x x x
Resources used x x
Parameter monitoring x x x
Data after job execu-
tion
Product data x x x
Resource capability
description
Equipment description Identify usable equipment x x x
Specify usable equipment x x x
Show inactive equipment x
Production data set de-
scription
Receive production data sets x x
Identify production data sets x x x
Specify production data sets x x x
Manage production data sets x x
Material relation x
Show inactive data sets x
Configuration data set
description
Receive configuration data
sets
x
Identify configuration data
sets
x
Specify configuration data sets x
Towards Data-driven Production: Analysis of Data Models Describing Machinery Jobs in OPC UA
733
Table 2: Overview of functional elements and their occurrence in CS (cont.).
PAR: OPC 40083 and OPC 40077 (Plastics and Rubber)
G: OPC 40301 (Flat Glass)
I95: OPC 10030 and OPC 10031-4 (ISA-95)
V: OPC 40100-1 (Machine Vision)
MT: OPC 40501 (Machine Tools)
Main category Sub category Functional element PAR G I95 V MT
Manage configuration data
sets
x
Show inactive data sets x
Variant parameter de-
scription
Identify variant parameters x x
Specify variant parameters x x
Personnel description Identify personnel x x x
Specify personnel x x x
Inform about inactive person-
nel
x
Material description Identify material x x x x
Specify material x x
Manage material x x
Job management Receive job plans x x x
Display job plans x x x x
Release job x
Receive job x x x x
Request job x x
Job management meta data x
Production manage-
ment
Machine control Start processing x x x
Stop processing x x x
Request sample product x
Production mode Set automatic mode x
Set simulation mode x
Set supervised mode x
Prevent production mode x
Device control Forward production data sets x
Inform about devices x x
Control device x
Production parameters Production settings x
Supervision settings x
Result management Provide results x
Identify results x
Specify results x
Manage results x
Safety Functional reliability Information x
Prevent undefined system
states
x
Lock processing x x
Security Authenticity Server settings x x x
Availability Limit number of clients x x
Optimize computing resources x
ICINCO 2022 - 19th International Conference on Informatics in Control, Automation and Robotics
734
Category “Production management” contains
multiple diverse functional elements that are largely
supported by PAR and V, G and I95 support individ-
ual functional elements. They are largely similar in
design if multiple specifications support a functional
element, the only exception is the sub category “Pro-
duction mode”. While V models the machine state us-
ing the concept of a state machine (displaying states
and possible transitions between them), PAR only dis-
plays possible states using an enumeration.
The category “Result management” is only in-
cluded in V.
The categories safety and security show sparse oc-
currence in the specifications. In the safety category,
the functional elements contain model elements re-
porting safety relevant system states, preventing un-
defined system states and locking operations on the
interface. The G specification allows locking the pro-
duction of a job, while PAR provides the ability to
prohibit changes in data sets. Concerning security, the
specifications contain diverse measures, like requiring
a specific minimum strength of encryption algorithms
for data transmission, limiting the number of simulta-
neous service users and providing the option to delete
unused data and thus optimize computer resource us-
age.
6 ANALYSIS
When looking at the functional elements contained in
the different specifications in table 2, the PAR specifi-
cation clearly contains the most functional elements.
Compared to G, MT and V, the PAR specification is
older, with its first release in 2016 and updates to the
specification since. In addition, both PAR and I95 are
based on previous existing standards while G, MT and
V have developed the model initially for OPC UA.
The V specification stands out, being the only
specification including configuration data set descrip-
tion and result management while including fewer of
the functional elements in data during job execution
and job management than the other specifications.
This might be attributed to the difference in the type
of device: While the injection molding machines of
PAR, the glass processing machines and the machine
tools all generally use input materials to produce out-
put materials or parts, the vision systems don’t touch
the items they are processing. Moreover, a process
on a vision device is often much shorter and involves
fewer individual steps, e.g. a program for a milling
machine. In addition, parameters like temperature
and forces, that are often monitored during different
production processes, do not occur in the same con-
text for machine vision.
The G, I95 and MT specifications cover many of
the same functional elements, with the MT specifi-
cation implementing the fewest functional elements.
One reason for this is the MT specification not defin-
ing interaction possibilities on the OPC UA interface
- neither can jobs be managed (added, deleted, edited
etc.) nor can the production be managed (start, stop,
set modes, ...).
Especially in the categories “Data for job execu-
tion”, “Data during Job Execution” and “Job manage-
ment”, all specifications show similarities. Both in
the functional elements they implement, but largely
also in the way, these functional elements are im-
plemented. In cases where the specifications handle
functional elements differently, the same information
is conveyed, be it the style of reference between job
and production data set or the order of execution, if
multiple jobs are present.
Major differences between the specifications arise
in “Data for job preparation”, “Resource capability
description”, “Production Management” and “Result
Management”. In ”Data for job preparation”, V fo-
cuses on recipes, on descriptions of process steps.
PAR focuses on production data sets. G informs about
prerequisites like the allowed file format accepted in
file transfer. As the examples are this sparse, this cat-
egory seems to be less focused on than the categories
more closely related with job execution.
The whole sub category “Configuration data set
description” only appears in V. This might be due
to the different domain of vision systems and sub-
sequently to the greater significance of configuration
data. However, configuration data may still be of
value for the other domains. Similar assumptions are
true for the category “Result management”. For the V
specification, the process results are the main process
outcome, as the product is in the other four specifi-
cation domains. And the “Products” functional ele-
ment (Data during job execution), the “Product data”
functional element (Data after job execution) and the
“Material description” subcategory (Resource capa-
bility description) are implemented widely among the
specifications.
When regarding safety and security, neither CS
contains comprehensive measures. This seems odd
given the importance of those two domains in ma-
chinery and industry 4.0. However, the safety in ma-
chinery is usually ensured at controller level. So re-
gardless if a start/stop command originates from the
machine panel or the OPC UA interface, the con-
troller checks if that command can be safely executed.
For this reason, the safety functions themselves don’t
need to be included in the interface.
Towards Data-driven Production: Analysis of Data Models Describing Machinery Jobs in OPC UA
735
In case of security, the CS contain few and diverse
measures. This aspect is to be regarded with the ar-
chitecture of OPC UA in mind. Most CS don’t spec-
ify the security algorithms to be used, similar to them
not specifying the transmission protocols to be used.
This leaves all possible security algorithms described
in OPC UA Part 2 as possible choices for implementa-
tion along with the CS model (OPC, 2018). Similarly,
the number of maximum clients usually depends on
the hardware resources of OPC UA server products.
The handling of server resources is therefore often not
part of the CS.
7 DISCUSSION
Using a structured qualitative content analysis on
OPC UA CS yields a more profound understanding
of functional elements focused on in the group of CS.
This kind of understanding achieves the goal set in
this work: to identify the information contained in
existing OPC UA job models as well as the tasks in
job management displayed in these models. How-
ever, such a structured content analysis may produce
different results based on the initial chosen set of cat-
egories. A different set of categories used in this case
would have led to a different structure in the resulting
overview. However, it is likely that the goal of this
work would still have been achieved.
As this work only regarded existing specifications,
the job model may not be complete, including all
aspects possible. The result also does not take job
models in other representations than OPC UA into
account. Still, some models are not OPC UA spe-
cific, but rather implementations of previously exist-
ing standards in OPC UA (40083, 2021; 40077, 2020;
10030, 2013; 10031-4, 2021). The benefit of limiting
the analysis to OPC UA models is a higher compara-
bility of modeling techniques, as all models regarded
in this work have to follow the same rules.
As a result of the content analysis, the overview
presented in table 2 has been created. This overview
can now be used as a basis in extending or harmoniz-
ing the above models. The category system developed
in the analysis could even be used itself to structure
data on OPC UA interfaces. Additionally, the cate-
gory system serves as a basis to identify the elements
that are important to model. This prevents working
groups from overlooking aspects, that other specifica-
tions already contain. The additional documents gen-
erated in the analysis also serve as a guideline how
these elements may be modeled.
ACKNOWLEDGEMENT
The authors thank the German Federal Ministry of
Economic Affairs and Climate Action (BMWK) for
supporting the project ”SDM4FZI” under the funding
programme ”Zukunftsinvestitionen in der Fahrzeug-
industrie”.
REFERENCES
10030, O. (2013). Opc 10030: Opc ua for isa-95 common
object model. Standard, OPC Foundation.
10031-4, O. (2021). Opc 10031-4: Opc ua for isa-95 part 4:
Job control. Standard, OPC Foundation.
30260, O. (2020). Opc 30260: Opc ua for opcn-scs. Stan-
dard, OPC Foundation.
40077, O. (2020). Opc 40077: Opc ua interfaces for plas-
tics and rubber machinery - data exchange between in-
jection moulding machines and mes. Standard, OPC
Foundation.
40083, O. (2021). Opc 40083: Opc ua for plastics and rub-
ber machinery - general type defninitions. Standard,
OPC Foundation.
40100-1, O. (2019). Opc 40100-1: Opc ua for machine
vision part 1. Standard, OPC Foundation.
40301, O. (2022). Opc 40301: Opc ua for flat glass. Stan-
dard, OPC Foundation.
40501-1, O. (2020). Opc 40501: Opc ua for machine tools.
Standard, OPC Foundation.
IEC 62541-[1-14]:2020 (2020). Opc unified architecture.
Standard, INTERNATIONAL ELECTROTECHNI-
CAL COMMISSION, Geneva, CH.
Informationstechnik, V.-F. (2016). 5600-1: Manufacturing
execution systems (mes).
Kuckartz, U. (2012). Qualitative inhaltsanalyse. Beltz Ju-
venta.
OPC, U. (2018). Specification: Part 2–security model, ver-
sion 1.04.
OPC Foundation (2022a). Markets & collaboration - opc
foundation. https://opcfoundation.org/markets-collab
oration/.
OPC Foundation (2022b). Opc-f working groups. https:
//opcfoundation.org/about/working-groups/opcf-wg/.
OPC Foundation (2022c). Opc ua information models - opc
foundation. https://opcfoundation.org/developer-tool
s/specifications-opc-ua-information-models.
OPC Foundation (26.09.2019). Unified architecture - opc
foundation. https://opcfoundation.org/about/opc-tech
nologies/opc-ua/.
VDMA e.V. (2022). Opc ua for machinery. https://opcfou
ndation.org/markets-collaboration/opc-ua-for-machi
nery/.
ICINCO 2022 - 19th International Conference on Informatics in Control, Automation and Robotics
736