A strategic approach to improve the development of use-oriented knowledge-based engineering configurators (KBEC)

Abstract

The high complexity of today’s automation solutions often raises integration costs to an uneconomic level, particularly for small and medium-sized enterprises. Analyzing the total costs of automation solutions, engineering efforts account for the largest share. However, potentials for time and cost savings as well as quality improvements by reusing existing engineering knowledge are usually not exploited in industrial practice. In this context, knowledge-based configurators are the most popular expert systems and present an opportunity to automate the creation of customer-specific automation solutions. Especially for efficient knowledge reuse, constraint-based configurators seem suitable. However, existing methods for developing configurators focus on product configurators rather than on knowledge-based engineering configurators (KBEC). In addition, the necessary knowledge acquisition (KA) is still one of the major challenges in developing KBECs. Open fields of action include the definition of the optimal functional scope as well as the identification, prioritization, and selection of suitable knowledge sources. Another prerequisite represents the transparency of existing engineering processes and interests of all affected stakeholders. Therefore, this paper introduces a six-step approach enabling the development of use-oriented KBECs with the minimum required functional scope to reduce efforts for KA and thus overall development costs. Finally, the strategic approach is validated using the example of a KBEC for the concept planning of robot-based automation solutions.

Full text

ScienceDirect
Available online at www.sciencedirect.com
Available online at www.sciencedirect.com
ScienceDirect
Procedia CIRP 00 (2017) 000–000
www.elsevier.com/locate/procedia
2212-8271 © 2017 The Authors. Published by Elsevier B.V.
Peer-review under responsibility of the scientific committee of the 28th CIRP Design Conference 2018.
28th CIRP Design Conference, May 2018, Nantes, France
A new methodology to analyze the functional and physical architecture of
existing products for an assembly oriented product family identification
Paul Stief *, Jean-Yves Dantan, Alain Etienne, Ali Siadat
École Nationale Supérieure d’Arts et Métiers, Arts et Métiers ParisTech, LCFC EA 4495, 4 Rue Augustin Fresnel, Metz 57078, France
* Corresponding author. Tel.: +33 3 87 37 54 30; E-mail address: paul.stief@ensam.eu
Abstract
In today’s business environment, the trend towards more product variety and customization is unbroken. Due to this development, the need of
agile and reconfigurable production systems emerged to cope with various products and product families. To design and optimize production
systems as well as to choose the optimal product matches, product analysis methods are needed. Indeed, most of the known methods aim to
analyze a product or one product family on the physical level. Different product families, however, may differ largely in terms of the number and
nature of components. This fact impedes an efficient comparison and choice of appropriate product family combinations for the production
system. A new methodology is proposed to analyze existing products in view of their functional and physical architecture. The aim is to cluster
these products in new assembly oriented product families for the optimization of existing assembly lines and the creation of future reconfigurable
assembly systems. Based on Datum Flow Chain, the physical structure of the products is analyzed. Functional subassemblies are identified, and
a functional analysis is performed. Moreover, a hybrid functional and physical architecture graph (HyFPAG) is the output which depicts the
similarity between product families by providing design support to both, production system planners and product designers. An illustrative
example of a nail-clipper is used to explain the proposed methodology. An industrial case study on two product families of steering columns of
thyssenkrupp Presta France is then carried out to give a first industrial evaluation of the proposed approach.
© 2017 The Authors. Published by Elsevier B.V.
Peer-review under responsibility of the scientific committee of the 28th CIRP Design Conference 2018.
Keywords: Assembly; Design method; Family identification
1. Introduction
Due to the fast development in the domain of
communication and an ongoing trend of digitization and
digitalization, manufacturing enterprises are facing important
challenges in today’s market environments: a continuing
tendency towards reduction of product development times and
shortened product lifecycles. In addition, there is an increasing
demand of customization, being at the same time in a global
competition with competitors all over the world. This trend,
which is inducing the development from macro to micro
markets, results in diminished lot sizes due to augmenting
product varieties (high-volume to low-volume production) [1].
To cope with this augmenting variety as well as to be able to
identify possible optimization potentials in the existing
production system, it is important to have a precise knowledge
of the product range and characteristics manufactured and/or
assembled in this system. In this context, the main challenge in
modelling and analysis is now not only to cope with single
products, a limited product range or existing product families,
but also to be able to analyze and to compare products to define
new product families. It can be observed that classical existing
product families are regrouped in function of clients or features.
However, assembly oriented product families are hardly to find.
On the product family level, products differ mainly in two
main characteristics: (i) the number of components and (ii) the
type of components (e.g. mechanical, electrical, electronical).
Classical methodologies considering mainly single products
or solitary, already existing product families analyze the
product structure on a physical level (components level) which
causes difficulties regarding an efficient definition and
comparison of different product families. Addressing this
Procedia CIRP 96 (2020) 219–224
2212-8271 © 2020 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)
Peer-review under responsibility of the scientific committee of the 8th CIRP Global Web Conference – Flexible Mass Customisation
10.1016/j.procir.2021.01.078
© 2020 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)
Peer-review under responsibility of the scientic committee of the 8th CIRP Global Web Conference – Flexible Mass Customisation
2212-8271 © 2020 The Authors, Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer review under the responsibility of the scientific committee of the CIRPe 2020 Global Web Conference
CIRPe 2020 – 8th CIRP Global Web Conference – Flexible Mass Customization
A strategic approach to improve the development of use-oriented
knowledge-based engineering configurators (KBEC)
Eike Schäffera,*, Sara Shafieeb, Andreas Mayra, Jörg Frankea
aInstitute for Factory Automation and Production Systems, Friedrich-Alexander University Erlangen-Nuremberg, Egerlandstr. 7, 91058 Erlangen, Germany
bTechnical University of Denmark, Department of Mechanical Engineering, 2800 Kgs. Lyngby, Denmark
* Corresponding author. Tel.: +49 9131 85-28314; fax: +49 9131 302825; E-mail address: eike.schaeffer@faps.fau.de
Abstract
The high complexity of today's automation solutions often raises integration costs to an uneconomic level, particularly for small and medium-
sized enterprises. Analyzing the total costs of automation solutions, engineering efforts account for the largest share. However, potentials for time
and cost savings as well as quality improvements by reusing existing engineering knowledge are usually not exploited in industrial practice. In
this context, knowledge-based configurators are the most popularexpert systems and present an opportunity to automate the creation of customer-
specific automation solutions. Especially for efficient knowledge reuse, constraint-based configurators seem suitable. However, existing methods
for developing configurators focus on product configurators rather than on knowledge-based engineering configurators (KBEC). In addition, the
necessary knowledge acquisition (KA) is still one of the major challenges in developing KBECs. Open fields of action include the definition of
the optimal functional scope as well as the identification, prioritization, and selection of suitable knowledge sources. Another prerequisite repre-
sents the transparency of existing engineering processes and interests of all affected stakeholders. Therefore, this paper introduces a six-step
approach enabling the development of use-oriented KBECs with the minimum required functional scope to reduce efforts for KA and thus overall
development costs. Finally, the strategic approach is validated using the example of a KBEC for the concept planning of robot-based automation
solutions.
© 2020 The Authors, Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer review under the responsibility of the scientific committee of the CIRPe 2020 Global Web Conference
Keywords: configurator; knowledge-based; constraint-based; knowledge acquisition; KBEC; engineering; division of labor; planning; automation; robot
1. Introduction
Robot-based automation solutions (RAS) promise increases
in efficiency and the relief of workers in physically demanding
and dangerous tasks. Their field of application is constantly
growing due to technical innovations. In the area of machine
tending, robots can quickly and repetitively load and unload
automatic machines such as milling centers, resulting in higher
utilization, better amortization, and thus economic benefits. As
RAS are mostly individually engineered, costs are significant
and mostly unaffordable for small and medium-sized enter-
prises (SME). [1,2]
Knowledge-based configurators (KBC), often regarded as a
subgroup of expert systems [3], offer a technology to automate
knowledge work such as the aforementioned engineering of au-
tomation solutions [4]. So far, configurators are mostly used for
the customization of products such as automobiles or clothes
[5] rather than the engineering of automation solutions. More-
over, due to the various iterative planning phases of automation
solutions and the resulting uncertainty, the scope and functional
range of engineering configurators are less obvious and prede-
fined than for product configurators (PC), which can likewise
rely on a knowledge-based approach.
In general, information retrieval is considered as the first
critical step in the knowledge lifecycle [6]. As KBCs require an
explicit representation of knowledge, domain experts and
knowledge engineers have to invest a considerable amount of
time for acquiring and keeping knowledge up-to-date, also
known as knowledge acquisition (KA) bottleneck [7]. How-
ever, there is still a general lack of methods enabling the divi-
sion of labor in KA [8]. In addition, literature often neglects the
challenge of transparency in knowledge work and the processes
addressed. Hence, the relevant process requirements of the in-
volved stakeholders, their goals, and the available sources of
knowledge have to be collected in order to derive a selection of
minimum necessary knowledge functions, thereby avoiding
later over-engineering. This is particularly relevant for the idea
of knowledge-based engineering configurators (KBEC) since
most methods for KA originate from PC, where the functional
scope and focus is inherently given.
Taking these challenges into account, this paper introduces
a strategic approach for improving the development of a use-
oriented KBEC through greater transparency and a clear focus
2212-8271 © 2020 The Authors, Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer review under the responsibility of the scientific committee of the CIRPe 2020 Global Web Conference
CIRPe 2020 – 8th CIRP Global Web Conference – Flexible Mass Customization
A strategic approach to improve the development of use-oriented
knowledge-based engineering configurators (KBEC)
Eike Schäffera,*, Sara Shafieeb, Andreas Mayra, Jörg Frankea
aInstitute for Factory Automation and Production Systems, Friedrich-Alexander University Erlangen-Nuremberg, Egerlandstr. 7, 91058 Erlangen, Germany
bTechnical University of Denmark, Department of Mechanical Engineering, 2800 Kgs. Lyngby, Denmark
* Corresponding author. Tel.: +49 9131 85-28314; fax: +49 9131 302825; E-mail address: eike.schaeffer@faps.fau.de
Abstract
The high complexity of today's automation solutions often raises integration costs to an uneconomic level, particularly for small and medium-
sized enterprises. Analyzing the total costs of automation solutions, engineering efforts account for the largest share. However, potentials for time
and cost savings as well as quality improvements by reusing existing engineering knowledge are usually not exploited in industrial practice. In
this context, knowledge-based configurators are the most popularexpert systems and present an opportunity to automate the creation of customer-
specific automation solutions. Especially for efficient knowledge reuse, constraint-based configurators seem suitable. However, existing methods
for developing configurators focus on product configurators rather than on knowledge-based engineering configurators (KBEC). In addition, the
necessary knowledge acquisition (KA) is still one of the major challenges in developing KBECs. Open fields of action include the definition of
the optimal functional scope as well as the identification, prioritization, and selection of suitable knowledge sources. Another prerequisite repre-
sents the transparency of existing engineering processes and interests of all affected stakeholders. Therefore, this paper introduces a six-step
approach enabling the development of use-oriented KBECs with the minimum required functional scope to reduce efforts for KA and thus overall
development costs. Finally, the strategic approach is validated using the example of a KBEC for the concept planning of robot-based automation
solutions.
© 2020 The Authors, Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer review under the responsibility of the scientific committee of the CIRPe 2020 Global Web Conference
Keywords: configurator; knowledge-based; constraint-based; knowledge acquisition; KBEC; engineering; division of labor; planning; automation; robot
1. Introduction
Robot-based automation solutions (RAS) promise increases
in efficiency and the relief of workers in physically demanding
and dangerous tasks. Their field of application is constantly
growing due to technical innovations. In the area of machine
tending, robots can quickly and repetitively load and unload
automatic machines such as milling centers, resulting in higher
utilization, better amortization, and thus economic benefits. As
RAS are mostly individually engineered, costs are significant
and mostly unaffordable for small and medium-sized enter-
prises (SME). [1,2]
Knowledge-based configurators (KBC), often regarded as a
subgroup of expert systems [3], offer a technology to automate
knowledge work such as the aforementioned engineering of au-
tomation solutions [4]. So far, configurators are mostly used for
the customization of products such as automobiles or clothes
[5] rather than the engineering of automation solutions. More-
over, due to the various iterative planning phases of automation
solutions and the resulting uncertainty, the scope and functional
range of engineering configurators are less obvious and prede-
fined than for product configurators (PC), which can likewise
rely on a knowledge-based approach.
In general, information retrieval is considered as the first
critical step in the knowledge lifecycle [6]. As KBCs require an
explicit representation of knowledge, domain experts and
knowledge engineers have to invest a considerable amount of
time for acquiring and keeping knowledge up-to-date, also
known as knowledge acquisition (KA) bottleneck [7]. How-
ever, there is still a general lack of methods enabling the divi-
sion of labor in KA [8]. In addition, literature often neglects the
challenge of transparency in knowledge work and the processes
addressed. Hence, the relevant process requirements of the in-
volved stakeholders, their goals, and the available sources of
knowledge have to be collected in order to derive a selection of
minimum necessary knowledge functions, thereby avoiding
later over-engineering. This is particularly relevant for the idea
of knowledge-based engineering configurators (KBEC) since
most methods for KA originate from PC, where the functional
scope and focus is inherently given.
Taking these challenges into account, this paper introduces
a strategic approach for improving the development of a use-
oriented KBEC through greater transparency and a clear focus
2212-8271 © 2020 The Authors, Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer review under the responsibility of the scientific committee of the CIRPe 2020 Global Web Conference
CIRPe 2020 – 8th CIRP Global Web Conference – Flexible Mass Customization
A strategic approach to improve the development of use-oriented
knowledge-based engineering configurators (KBEC)
Eike Schäffera,*, Sara Shafieeb, Andreas Mayra, Jörg Frankea
aInstitute for Factory Automation and Production Systems, Friedrich-Alexander University Erlangen-Nuremberg, Egerlandstr. 7, 91058 Erlangen, Germany
bTechnical University of Denmark, Department of Mechanical Engineering, 2800 Kgs. Lyngby, Denmark
* Corresponding author. Tel.: +49 9131 85-28314; fax: +49 9131 302825; E-mail address: eike.schaeffer@faps.fau.de
Abstract
The high complexity of today's automation solutions often raises integration costs to an uneconomic level, particularly for small and medium-
sized enterprises. Analyzing the total costs of automation solutions, engineering efforts account for the largest share. However, potentials for time
and cost savings as well as quality improvements by reusing existing engineering knowledge are usually not exploited in industrial practice. In
this context, knowledge-based configurators are the most popular expert systems and present an opportunity to automate the creation of customer-
specific automation solutions. Especially for efficient knowledge reuse, constraint-based configurators seem suitable. However, existing methods
for developing configurators focus on product configurators rather than on knowledge-based engineering configurators (KBEC). In addition, the
necessary knowledge acquisition (KA) is still one of the major challenges in developing KBECs. Open fields of action include the definition of
the optimal functional scope as well as the identification, prioritization, and selection of suitable knowledge sources. Another prerequisite repre-
sents the transparency of existing engineering processes and interests of all affected stakeholders. Therefore, this paper introduces a six-step
approach enabling the development of use-oriented KBECs with the minimum required functional scope to reduce efforts for KA and thus overall
development costs. Finally, the strategic approach is validated using the example of a KBEC for the concept planning of robot-based automation
solutions.
© 2020 The Authors, Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer review under the responsibility of the scientific committee of the CIRPe 2020 Global Web Conference
Keywords: configurator; knowledge-based; constraint-based; knowledge acquisition; KBEC; engineering; division of labor; planning; automation; robot
1. Introduction
Robot-based automation solutions (RAS) promise increases
in efficiency and the relief of workers in physically demanding
and dangerous tasks. Their field of application is constantly
growing due to technical innovations. In the area of machine
tending, robots can quickly and repetitively load and unload
automatic machines such as milling centers, resulting in higher
utilization, better amortization, and thus economic benefits. As
RAS are mostly individually engineered, costs are significant
and mostly unaffordable for small and medium-sized enter-
prises (SME). [1,2]
Knowledge-based configurators (KBC), often regarded as a
subgroup of expert systems [3], offer a technology to automate
knowledge work such as the aforementioned engineering of au-
tomation solutions [4]. So far, configurators are mostly used for
the customization of products such as automobiles or clothes
[5] rather than the engineering of automation solutions. More-
over, due to the various iterative planning phases of automation
solutions and the resulting uncertainty, the scope and functional
range of engineering configurators are less obvious and prede-
fined than for product configurators (PC), which can likewise
rely on a knowledge-based approach.
In general, information retrieval is considered as the first
critical step in the knowledge lifecycle [6]. As KBCs require an
explicit representation of knowledge, domain experts and
knowledge engineers have to invest a considerable amount of
time for acquiring and keeping knowledge up-to-date, also
known as knowledge acquisition (KA) bottleneck [7]. How-
ever, there is still a general lack of methods enabling the divi-
sion of labor in KA [8]. In addition, literature often neglects the
challenge of transparency in knowledge work and the processes
addressed. Hence, the relevant process requirements of the in-
volved stakeholders, their goals, and the available sources of
knowledge have to be collected in order to derive a selection of
minimum necessary knowledge functions, thereby avoiding
later over-engineering. This is particularly relevant for the idea
of knowledge-based engineering configurators (KBEC) since
most methods for KA originate from PC, where the functional
scope and focus is inherently given.
Taking these challenges into account, this paper introduces
a strategic approach for improving the development of a use-
oriented KBEC through greater transparency and a clear focus
2212-8271 © 2020 The Authors, Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer review under the responsibility of the scientific committee of the CIRPe 2020 Global Web Conference
CIRPe 2020 – 8th CIRP Global Web Conference – Flexible Mass Customization
A strategic approach to improve the development of use-oriented
knowledge-based engineering configurators (KBEC)
Eike Schäffera,*, Sara Shafieeb, Andreas Mayra, Jörg Frankea
aInstitute for Factory Automation and Production Systems, Friedrich-Alexander University Erlangen-Nuremberg, Egerlandstr. 7, 91058 Erlangen, Germany
bTechnical University of Denmark, Department of Mechanical Engineering, 2800 Kgs. Lyngby, Denmark
* Corresponding author. Tel.: +49 9131 85-28314; fax: +49 9131 302825; E-mail address: eike.schaeffer@faps.fau.de
Abstract
The high complexity of today's automation solutions often raises integration costs to an uneconomic level, particularly for small and medium-
sized enterprises. Analyzing the total costs of automation solutions, engineering efforts account for the largest share. However, potentials for time
and cost savings as well as quality improvements by reusing existing engineering knowledge are usually not exploited in industrial practice. In
this context, knowledge-based configurators are the most popularexpert systems and present an opportunity to automate the creation of customer-
specific automation solutions. Especially for efficient knowledge reuse, constraint-based configurators seem suitable. However, existing methods
for developing configurators focus on product configurators rather than on knowledge-based engineering configurators (KBEC). In addition, the
necessary knowledge acquisition (KA) is still one of the major challenges in developing KBECs. Open fields of action include the definition of
the optimal functional scope as well as the identification, prioritization, and selection of suitable knowledge sources. Another prerequisite repre-
sents the transparency of existing engineering processes and interests of all affected stakeholders. Therefore, this paper introduces a six-step
approach enabling the development of use-oriented KBECs with the minimum required functional scope to reduce efforts for KA and thus overall
development costs. Finally, the strategic approach is validated using the example of a KBEC for the concept planning of robot-based automation
solutions.
© 2020 The Authors, Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer review under the responsibility of the scientific committee of the CIRPe 2020 Global Web Conference
Keywords: configurator; knowledge-based; constraint-based; knowledge acquisition; KBEC; engineering; division of labor; planning; automation; robot
1. Introduction
Robot-based automation solutions (RAS) promise increases
in efficiency and the relief of workers in physically demanding
and dangerous tasks. Their field of application is constantly
growing due to technical innovations. In the area of machine
tending, robots can quickly and repetitively load and unload
automatic machines such as milling centers, resulting in higher
utilization, better amortization, and thus economic benefits. As
RAS are mostly individually engineered, costs are significant
and mostly unaffordable for small and medium-sized enter-
prises (SME). [1,2]
Knowledge-based configurators (KBC), often regarded as a
subgroup of expert systems [3], offer a technology to automate
knowledge work such as the aforementioned engineering of au-
tomation solutions [4]. So far, configurators are mostly used for
the customization of products such as automobiles or clothes
[5] rather than the engineering of automation solutions. More-
over, due to the various iterative planning phases of automation
solutions and the resulting uncertainty, the scope and functional
range of engineering configurators are less obvious and prede-
fined than for product configurators (PC), which can likewise
rely on a knowledge-based approach.
In general, information retrieval is considered as the first
critical step in the knowledge lifecycle [6]. As KBCs require an
explicit representation of knowledge, domain experts and
knowledge engineers have to invest a considerable amount of
time for acquiring and keeping knowledge up-to-date, also
known as knowledge acquisition (KA) bottleneck [7]. How-
ever, there is still a general lack of methods enabling the divi-
sion of labor in KA [8]. In addition, literature often neglects the
challenge of transparency in knowledge work and the processes
addressed. Hence, the relevant process requirements of the in-
volved stakeholders, their goals, and the available sources of
knowledge have to be collected in order to derive a selection of
minimum necessary knowledge functions, thereby avoiding
later over-engineering. This is particularly relevant for the idea
of knowledge-based engineering configurators (KBEC) since
most methods for KA originate from PC, where the functional
scope and focus is inherently given.
Taking these challenges into account, this paper introduces
a strategic approach for improving the development of a use-
oriented KBEC through greater transparency and a clear focus
2212-8271 © 2020 The Authors, Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer review under the responsibility of the scientific committee of the CIRPe 2020 Global Web Conference
CIRPe 2020 – 8th CIRP Global Web Conference – Flexible Mass Customization
A strategic approach to improve the development of use-oriented
knowledge-based engineering configurators (KBEC)
Eike Schäffera,*, Sara Shafieeb, Andreas Mayra, Jörg Frankea
aInstitute for Factory Automation and Production Systems, Friedrich-Alexander University Erlangen-Nuremberg, Egerlandstr. 7, 91058 Erlangen, Germany
bTechnical University of Denmark, Department of Mechanical Engineering, 2800 Kgs. Lyngby, Denmark
* Corresponding author. Tel.: +49 9131 85-28314; fax: +49 9131 302825; E-mail address: eike.schaeffer@faps.fau.de
Abstract
The high complexity of today's automation solutions often raises integration costs to an uneconomic level, particularly for small and medium-
sized enterprises. Analyzing the total costs of automation solutions, engineering efforts account for the largest share. However, potentials for time
and cost savings as well as quality improvements by reusing existing engineering knowledge are usually not exploited in industrial practice. In
this context, knowledge-based configurators are the most popularexpert systems and present an opportunity to automate the creation of customer-
specific automation solutions. Especially for efficient knowledge reuse, constraint-based configurators seem suitable. However, existing methods
for developing configurators focus on product configurators rather than on knowledge-based engineering configurators (KBEC). In addition, the
necessary knowledge acquisition (KA) is still one of the major challenges in developing KBECs. Open fields of action include the definition of
the optimal functional scope as well as the identification, prioritization, and selection of suitable knowledge sources. Another prerequisite repre-
sents the transparency of existing engineering processes and interests of all affected stakeholders. Therefore, this paper introduces a six-step
approach enabling the development of use-oriented KBECs with the minimum required functional scope to reduce efforts for KA and thus overall
development costs. Finally, the strategic approach is validated using the example of a KBEC for the concept planning of robot-based automation
solutions.
© 2020 The Authors, Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer review under the responsibility of the scientific committee of the CIRPe 2020 Global Web Conference
Keywords: configurator; knowledge-based; constraint-based; knowledge acquisition; KBEC; engineering; division of labor; planning; automation; robot
1. Introduction
Robot-based automation solutions (RAS) promise increases
in efficiency and the relief of workers in physically demanding
and dangerous tasks. Their field of application is constantly
growing due to technical innovations. In the area of machine
tending, robots can quickly and repetitively load and unload
automatic machines such as milling centers, resulting in higher
utilization, better amortization, and thus economic benefits. As
RAS are mostly individually engineered, costs are significant
and mostly unaffordable for small and medium-sized enter-
prises (SME). [1,2]
Knowledge-based configurators (KBC), often regarded as a
subgroup of expert systems [3], offer a technology to automate
knowledge work such as the aforementioned engineering of au-
tomation solutions [4]. So far, configurators are mostly used for
the customization of products such as automobiles or clothes
[5] rather than the engineering of automation solutions. More-
over, due to the various iterative planning phases of automation
solutions and the resulting uncertainty, the scope and functional
range of engineering configurators are less obvious and prede-
fined than for product configurators (PC), which can likewise
rely on a knowledge-based approach.
In general, information retrieval is considered as the first
critical step in the knowledge lifecycle [6]. As KBCs require an
explicit representation of knowledge, domain experts and
knowledge engineers have to invest a considerable amount of
time for acquiring and keeping knowledge up-to-date, also
known as knowledge acquisition (KA) bottleneck [7]. How-
ever, there is still a general lack of methods enabling the divi-
sion of labor in KA [8]. In addition, literature often neglects the
challenge of transparency in knowledge work and the processes
addressed. Hence, the relevant process requirements of the in-
volved stakeholders, their goals, and the available sources of
knowledge have to be collected in order to derive a selection of
minimum necessary knowledge functions, thereby avoiding
later over-engineering. This is particularly relevant for the idea
of knowledge-based engineering configurators (KBEC) since
most methods for KA originate from PC, where the functional
scope and focus is inherently given.
Taking these challenges into account, this paper introduces
a strategic approach for improving the development of a use-
oriented KBEC through greater transparency and a clear focus

220 Eike Schäffer et al. / Procedia CIRP 96 (2020) 219–224
2 Eike Schäffer et al. / Procedia CIRP 00 (2019) 000–000
on the minimum required functional scope. Section 2 first ex-
plains the relevant basics and clarifies the need for action. Sub-
sequently, the research approach (section 3) as well as the de-
veloped six-step method (section 4) are presented. After the
validation example in section 5, section 6 concludes with a
summary and an outlook on future activities.
2. Relevant basics, related work and need for action
Starting from the product emergence process (PEP), the fol-
lowing subsections introduce the different types of configura-
tors, some existing methods for their development as well as
fundamental problems associated with KA.
2.1. Lack of tools for concept planning within PEP
The PEP is divided into three phases: product development,
production planning, and production [9]. Within the PEP, soft-
ware tools support the employees involved and aim at reducing
the total effort required. In the product development phase,
tools for Computer-Aided Design (CAD), Product Data Man-
agement (PDM) as well as Product Lifecycle Management
(PLM) are well-established [10]. In production, Production
Planning and Control Systems, Manufacturing Execution Sys-
tems (MES), and Enterprise Resource Planning (ERP) systems
are used [10]. While there is a variety of software tools for de-
tailed planning of production systems, such as kinematics sim-
ulation and virtual commissioning, there are only a few ap-
proaches, such as in [11], for the early planning phases.
Within requirements definition and concept planning, a high
degree of uncertainty is common. This in turn means that the
level of detail and accuracy is significantly lower than in later
phases. After all, these early phases of production planning are
more about finding a general solution concept rather than con-
cretizing all specific details. Since classical engineering ap-
proaches are time-consuming and cost-intensive [12], the idea
of KBECs is moving into focus as they allow the structured
reuse of existing engineering knowledge elements.
2.2. PC vs. KBEC
In literature, the term engineering configurator is frequently
used but actually implies a PC in most cases as in [13] or [14].
In general, a PC gives automatic access to different existing
product variants [5]. For PCs, the main value proposition of the
product is usually clear and does not change significantly
through configuration, unless its performance or additional
functionalities are altered. Accordingly, most of the existing
configuration tools and methods have been developed for PCs
[15] providing a modular, interchangeable product structure
[16]. When planning a RAS, however, the individual compo-
nents are supplied by different manufacturers, mostly without
any comprehensive standardization. As a result, KBECs be-
come complex due to an indefinite number of manufacturers
with different standards and product variants [13,16], which
have to be configured to one consistent solution.
As can be seen in Fig. 1, with a KBEC, new solutions can
be created and validated, whereas a PC, such as in [13] or [5],
provides existing product knowledge. Therefore, in this paper,
the idea of “real” KBECs is pursued and differentiated from
ordinary PCs.
Fig. 1. Differentiation of PCs and the introduced concept of KBECs
according to our synthesis out of [3,4,13,17–19]
2.3. KA and development methods for KBC
As a subgroup of expert systems [3], KBCs can be devel-
oped table-, statement-, rule- or constraint-based [20]. Among
all these alternatives, a constraint-based approach offers the
greatest advantages in terms of transferability, maintainability,
and scalability [21,22] for KBECs, as constraints provide a
general functional description rather than case-specific models.
Research has already revealed numerous generic methods for
developing KBCs. According to [20], for example, the devel-
opment is divided into four steps: 1) problem characterization,
2) shell development, 3) knowledge base building, and
4) knowledge base maintenance. The KA, e.g. based on expert
interviews, can again be subdivided as follows [23]: 1) themati-
zation, 2) drafting, 3) interviewing, 4) transmitting, 5) analyz-
ing, 6) verifying, and 7) reporting. For scoping configuration
projects and managing the knowledge they require, Shafiee et
al. [24] propose a four-step framework.
The abovementioned and most other methods in literature
mainly refer to the development of PCs. However, for KBECs,
all the required knowledge cannot usually be taken from prod-
uct development and interviewing development employees.
Therefore, it is necessary to elaborate on the KA process and
possible knowledge sources in more detail for KBECs.
In recent research, Schäffer et al. [25] have already proposed
a method for the division of labor in KA, especially suited for
KBEC. For unifying the underlying semantics, they introduce
a procedure for the collaborative development of an ontology,
validated based on AutomationML [26].
2.4. Problems and need for action within the KA processes
Before the actual KA can take place, however, sufficient
transparency about the addressed processes, stakeholders, and
knowledge sources is required. Data, information, and
knowledge are valuable resources that need to be planned, or-
ganized, and utilized [6]. Theories of the mechanism-design
theory [27,28] can be used to understand the problem with dif-
ferent interest groups occurring in knowledge work. Mecha-
nism-design theory is a branch of game theory that defines gen-
eral rules and incentives for games (e.g. business situation) to
achieve a desired overall result (e.g. a social welfare or effi-
ciency level), even if the players (e.g. companies, employees)
are pursuing exclusively own interests (e.g. profit maximiza-
tion, job security). Apparently, it is not in the interest of indi-
viduals to make all information and intentions transparent [27].
Product configurator
(Reproduction/ closed world assumption)
+ Defined functional scope
of the product (less complex)
+ Machine components uniquely selected
during product developm ent
+ Expert knowledge for the configurator
can be acquired from product
development
+ Known or defined number of
variants (2-n )
Knowledge-based engineering configurator
(Creation, validation/open world assumption)
- Due to the complexity of the engineering process,
functional scope of configurator not obvious to
determine
- Special machine construction and therefore
dynamic selection of machine components
- No central place where expert knowledge
can be acquired
- No standardized pl ant structures
(according to the state of research)
Higher uncertainty as there are
more possible s olutions and no
defined product structure
If a uniform product structure can be defined and variants derived, a product configurator is conceivable.
Eike Schäffer et al./ Procedia CIRP 00 (2019) 000–000 3
According to our experience from the research project RO-
BOTOP, established stakeholders (e.g. system integrators) pos-
sess the expertise (e.g. engineering know-how) but are not will-
ing to share it in order to protect their intellectual property and
thus competitive advantage. Sufficient transparency (e.g.
through available literature) as well as a regulation (through
which the general interest is attained by pursuing one's own in-
terests) can provide a solution to this general problem [27,28].
Even if unlimited knowledge access exists, we didn’t find
suitable and practical strategies within the literature to handle
extensive amounts of knowledge sources for enabling an effi-
cient KA process. Thus, it is necessary to elaborate on the cur-
rent KA process and possible knowledge sources in more detail
for KBECs. Without better KA, costs for knowledge acquisi-
tion and harmonization will successively increase due to local
instead of global optimization, as mentioned in [29] in the con-
text of process management. These tasks are referred as man-
agement functions in a company; but mostly executive manag-
ers are left to their individual knowledge approaches [24].
3. Applied research method
In this study, we observed and condensed several practical
cases, developing KBECs within system integration for RAS.
The introduced six-step method and the knowledge sources in
Table 2 are accomplished, further enhanced, and validated
based on exemplary KA processes based on several student
projects, our own knowledge work, and the research project
ROBOTOP. This type of empirical inquiry investigates a con-
temporary phenomenon within its real-life context [30]. Case
study research enables deep observation of the phenomenon
under investigation, and for a given set of available resources,
fewer cases allow for deeper observation [31].
4. Six-step method to improve the development of KBECs
In the following, a six-step method is introduced, which in-
tends to increase the transparency of relevant business pro-
cesses and stakeholders’ interests. Based on this, the minimal
necessary scope of knowledge functions can be derived and
suitable knowledge sources be identified. Since this method ad-
dresses the root challenges of knowledge work, it must be ap-
plied even before the actual KBEC development. Table 1 sum-
marizes the individual steps and their respective goals.
Table 1. Six-step method for improving the development of KBECs
Step Goal
1) Collection and synthesis of
business processes and outputs
Transparency of the market or com-
pany demand and current standard
procedures
2) Collection of the affected
stakeholders and their goals
Transparency of process-relevant
stakeholders and their requirements
3) Definition of the minimal
necessary scope of functions
Overall focus on the first minimal
viable configurator
4) Multidimensional classifica-
tion of needed knowledge
Detailed focus on the required
knowledge
5) Collection of available
knowledge sources
Transparency of available
knowledge sources and their focus
6) Structured knowledge acquisi-
tion and implementation
Operative collection of constraints to
build the KBEC
4.1. Collection and synthesis of business processes and
outputs
According to our experience, missing business process and
output transparency are considered as one of the primary chal-
lenges when creating innovative business models based on new
technologies or improvement potentials. The granularity de-
pends on the respective target. Depending on the availability of
knowledge, collection and synthesis of business processes can
be achieved using literature, consulting, or cooperative ven-
tures. For the specification of business processes, established
graphical representations like BPMN or event-driven process
chain (EPC) can be used. On this basis, new business possibil-
ities for KBECs can be identified based on the current market
environment and unsatisfied customers.
4.2. Collection of the affected stakeholders and their goals
A central challenge in the design of software solutions and
expert systems [24], especially KBECs, is the identification of
affected stakeholders in order to include and scope their rele-
vant requirements and to enable a use-oriented software design.
This is particularly important as various stakeholders represent
several process-relevant viewpoints and essential activities
aligned with a complete project or single business processes.
Optimal solutions can be achieved easier if the involved stake-
holders and their business needs, performance, and even per-
sonal goals are transparent. For example, most salespeople op-
timize their commission and if such goals are not transparent,
this can cause the failing of IT solutions or KBECs, due to the
lack of acceptance.
4.3. Definition of the minimal necessary scope of functions
Based on step 2, a highly promising business case or sub-
process within is selected. Within the business processes and
involved stakeholders, the minimum required functional scope
of the KBEC and therefore the necessary amount and depth of
knowledge are defined. This step is relatively straight forward,
based on the consolidation of information from steps 1 and 2.
4.4. Multidimensional classification of needed knowledge
For a focused KA of constraints, a multidimensional classi-
fication matrix as shown in Fig. 3 is needed. The matrix helps
to precisely classify the kind of constraints necessary for KA
and the optimal level of detail for the KBEC. This step consid-
ers the context in order to make targeted trade-offs in terms of
focus, depth, and number of necessary constraints.
4.5. Collection of available knowledge sources
Depending on the requirements and the minimum required
functional scope, the actual knowledge source pre-selection
can be performed. For transparency reasons, an overview of the
available knowledge sources as given in Table 2 should be cre-
ated in order to prioritize and select the most suitable sources
for the given context. In addition, Table 2 enables a cross-pro-
ject reuse of already available knowledge work by the explicit

Eike Schäffer et al. / Procedia CIRP 96 (2020) 219–224 221
2 Eike Schäffer et al. / Procedia CIRP 00 (2019) 000–000
on the minimum required functional scope. Section 2 first ex-
plains the relevant basics and clarifies the need for action. Sub-
sequently, the research approach (section 3) as well as the de-
veloped six-step method (section 4) are presented. After the
validation example in section 5, section 6 concludes with a
summary and an outlook on future activities.
2. Relevant basics, related work and need for action
Starting from the product emergence process (PEP), the fol-
lowing subsections introduce the different types of configura-
tors, some existing methods for their development as well as
fundamental problems associated with KA.
2.1. Lack of tools for concept planning within PEP
The PEP is divided into three phases: product development,
production planning, and production [9]. Within the PEP, soft-
ware tools support the employees involved and aim at reducing
the total effort required. In the product development phase,
tools for Computer-Aided Design (CAD), Product Data Man-
agement (PDM) as well as Product Lifecycle Management
(PLM) are well-established [10]. In production, Production
Planning and Control Systems, Manufacturing Execution Sys-
tems (MES), and Enterprise Resource Planning (ERP) systems
are used [10]. While there is a variety of software tools for de-
tailed planning of production systems, such as kinematics sim-
ulation and virtual commissioning, there are only a few ap-
proaches, such as in [11], for the early planning phases.
Within requirements definition and concept planning, a high
degree of uncertainty is common. This in turn means that the
level of detail and accuracy is significantly lower than in later
phases. After all, these early phases of production planning are
more about finding a general solution concept rather than con-
cretizing all specific details. Since classical engineering ap-
proaches are time-consuming and cost-intensive [12], the idea
of KBECs is moving into focus as they allow the structured
reuse of existing engineering knowledge elements.
2.2. PC vs. KBEC
In literature, the term engineering configurator is frequently
used but actually implies a PC in most cases as in [13] or [14].
In general, a PC gives automatic access to different existing
product variants [5]. For PCs, the main value proposition of the
product is usually clear and does not change significantly
through configuration, unless its performance or additional
functionalities are altered. Accordingly, most of the existing
configuration tools and methods have been developed for PCs
[15] providing a modular, interchangeable product structure
[16]. When planning a RAS, however, the individual compo-
nents are supplied by different manufacturers, mostly without
any comprehensive standardization. As a result, KBECs be-
come complex due to an indefinite number of manufacturers
with different standards and product variants [13,16], which
have to be configured to one consistent solution.
As can be seen in Fig. 1, with a KBEC, new solutions can
be created and validated, whereas a PC, such as in [13] or [5],
provides existing product knowledge. Therefore, in this paper,
the idea of “real” KBECs is pursued and differentiated from
ordinary PCs.
Fig. 1. Differentiation of PCs and the introduced concept of KBECs
according to our synthesis out of [3,4,13,17–19]
2.3. KA and development methods for KBC
As a subgroup of expert systems [3], KBCs can be devel-
oped table-, statement-, rule- or constraint-based [20]. Among
all these alternatives, a constraint-based approach offers the
greatest advantages in terms of transferability, maintainability,
and scalability [21,22] for KBECs, as constraints provide a
general functional description rather than case-specific models.
Research has already revealed numerous generic methods for
developing KBCs. According to [20], for example, the devel-
opment is divided into four steps: 1) problem characterization,
2) shell development, 3) knowledge base building, and
4) knowledge base maintenance. The KA, e.g. based on expert
interviews, can again be subdivided as follows [23]: 1) themati-
zation, 2) drafting, 3) interviewing, 4) transmitting, 5) analyz-
ing, 6) verifying, and 7) reporting. For scoping configuration
projects and managing the knowledge they require, Shafiee et
al. [24] propose a four-step framework.
The abovementioned and most other methods in literature
mainly refer to the development of PCs. However, for KBECs,
all the required knowledge cannot usually be taken from prod-
uct development and interviewing development employees.
Therefore, it is necessary to elaborate on the KA process and
possible knowledge sources in more detail for KBECs.
In recent research, Schäffer et al. [25] have already proposed
a method for the division of labor in KA, especially suited for
KBEC. For unifying the underlying semantics, they introduce
a procedure for the collaborative development of an ontology,
validated based on AutomationML [26].
2.4. Problems and need for action within the KA processes
Before the actual KA can take place, however, sufficient
transparency about the addressed processes, stakeholders, and
knowledge sources is required. Data, information, and
knowledge are valuable resources that need to be planned, or-
ganized, and utilized [6]. Theories of the mechanism-design
theory [27,28] can be used to understand the problem with dif-
ferent interest groups occurring in knowledge work. Mecha-
nism-design theory is a branch of game theory that defines gen-
eral rules and incentives for games (e.g. business situation) to
achieve a desired overall result (e.g. a social welfare or effi-
ciency level), even if the players (e.g. companies, employees)
are pursuing exclusively own interests (e.g. profit maximiza-
tion, job security). Apparently, it is not in the interest of indi-
viduals to make all information and intentions transparent [27].
Product configurator
(Reproduction/ closed world assumption)
+ Defined functional scope
of the product (less complex)
+ Machine components uniquely selected
during product developm ent
+ Expert knowledge for the configurator
can be acquired from product
development
+ Known or defined number of
variants (2-n )
Knowledge-based engineering configurator
(Creation, validation/open world assumption)
- Due to the complexity of the engineering process,
functional scope of configurator not obvious to
determine
- Special machine construction and therefore
dynamic selection of machine components
- No central place where expert knowledge
can be acquired
- No standardized pl ant structures
(according to the state of research)
Higher uncertainty as there are
more possible s olutions and no
defined product structure
If a uniform product structure can be defined and variants derived, a product configurator is conceivable.
Eike Schäffer et al./ Procedia CIRP 00 (2019) 000–000 3
According to our experience from the research project RO-
BOTOP, established stakeholders (e.g. system integrators) pos-
sess the expertise (e.g. engineering know-how) but are not will-
ing to share it in order to protect their intellectual property and
thus competitive advantage. Sufficient transparency (e.g.
through available literature) as well as a regulation (through
which the general interest is attained by pursuing one's own in-
terests) can provide a solution to this general problem [27,28].
Even if unlimited knowledge access exists, we didn’t find
suitable and practical strategies within the literature to handle
extensive amounts of knowledge sources for enabling an effi-
cient KA process. Thus, it is necessary to elaborate on the cur-
rent KA process and possible knowledge sources in more detail
for KBECs. Without better KA, costs for knowledge acquisi-
tion and harmonization will successively increase due to local
instead of global optimization, as mentioned in [29] in the con-
text of process management. These tasks are referred as man-
agement functions in a company; but mostly executive manag-
ers are left to their individual knowledge approaches [24].
3. Applied research method
In this study, we observed and condensed several practical
cases, developing KBECs within system integration for RAS.
The introduced six-step method and the knowledge sources in
Table 2 are accomplished, further enhanced, and validated
based on exemplary KA processes based on several student
projects, our own knowledge work, and the research project
ROBOTOP. This type of empirical inquiry investigates a con-
temporary phenomenon within its real-life context [30]. Case
study research enables deep observation of the phenomenon
under investigation, and for a given set of available resources,
fewer cases allow for deeper observation [31].
4. Six-step method to improve the development of KBECs
In the following, a six-step method is introduced, which in-
tends to increase the transparency of relevant business pro-
cesses and stakeholders’ interests. Based on this, the minimal
necessary scope of knowledge functions can be derived and
suitable knowledge sources be identified. Since this method ad-
dresses the root challenges of knowledge work, it must be ap-
plied even before the actual KBEC development. Table 1 sum-
marizes the individual steps and their respective goals.
Table 1. Six-step method for improving the development of KBECs
Step Goal
1) Collection and synthesis of
business processes and outputs
Transparency of the market or com-
pany demand and current standard
procedures
2) Collection of the affected
stakeholders and their goals
Transparency of process-relevant
stakeholders and their requirements
3) Definition of the minimal
necessary scope of functions
Overall focus on the first minimal
viable configurator
4) Multidimensional classifica-
tion of needed knowledge
Detailed focus on the required
knowledge
5) Collection of available
knowledge sources
Transparency of available
knowledge sources and their focus
6) Structured knowledge acquisi-
tion and implementation
Operative collection of constraints to
build the KBEC
4.1. Collection and synthesis of business processes and
outputs
According to our experience, missing business process and
output transparency are considered as one of the primary chal-
lenges when creating innovative business models based on new
technologies or improvement potentials. The granularity de-
pends on the respective target. Depending on the availability of
knowledge, collection and synthesis of business processes can
be achieved using literature, consulting, or cooperative ven-
tures. For the specification of business processes, established
graphical representations like BPMN or event-driven process
chain (EPC) can be used. On this basis, new business possibil-
ities for KBECs can be identified based on the current market
environment and unsatisfied customers.
4.2. Collection of the affected stakeholders and their goals
A central challenge in the design of software solutions and
expert systems [24], especially KBECs, is the identification of
affected stakeholders in order to include and scope their rele-
vant requirements and to enable a use-oriented software design.
This is particularly important as various stakeholders represent
several process-relevant viewpoints and essential activities
aligned with a complete project or single business processes.
Optimal solutions can be achieved easier if the involved stake-
holders and their business needs, performance, and even per-
sonal goals are transparent. For example, most salespeople op-
timize their commission and if such goals are not transparent,
this can cause the failing of IT solutions or KBECs, due to the
lack of acceptance.
4.3. Definition of the minimal necessary scope of functions
Based on step 2, a highly promising business case or sub-
process within is selected. Within the business processes and
involved stakeholders, the minimum required functional scope
of the KBEC and therefore the necessary amount and depth of
knowledge are defined. This step is relatively straight forward,
based on the consolidation of information from steps 1 and 2.
4.4. Multidimensional classification of needed knowledge
For a focused KA of constraints, a multidimensional classi-
fication matrix as shown in Fig. 3 is needed. The matrix helps
to precisely classify the kind of constraints necessary for KA
and the optimal level of detail for the KBEC. This step consid-
ers the context in order to make targeted trade-offs in terms of
focus, depth, and number of necessary constraints.
4.5. Collection of available knowledge sources
Depending on the requirements and the minimum required
functional scope, the actual knowledge source pre-selection
can be performed. For transparency reasons, an overview of the
available knowledge sources as given in Table 2 should be cre-
ated in order to prioritize and select the most suitable sources
for the given context. In addition, Table 2 enables a cross-pro-
ject reuse of already available knowledge work by the explicit

222 Eike Schäffer et al. / Procedia CIRP 96 (2020) 219–224
4 Eike Schäffer et al. / Procedia CIRP 00 (2019) 000–000
compilation of knowledge sources in one table. Thus,
knowledge engineers do not have to rework the same ideas on
appropriate sources of knowledge in each project. Without such
a tabular overview, employees usually tend to use the most fa-
miliar knowledge sources, as they mostly do not have time or
interest to research all available knowledge sources.
4.6. Structured knowledge acquisition and implementation
The tabular knowledge source overview and pre-selection
enables a focused acquisition of knowledge in the final step.
The operative division of labor and KA process can be accom-
plished according to the method introduced in [25]. In the latter
work, a so-called 150% topology for RAS is introduced, bring-
ing the structures of an engineering configurator closer to those
of a PC. Besides, a constraint matrix for a targeted division of
labor in KA is used. The 150% topology provides generic top-
ological classes for engineering elements such as robot arm,
gripper, or other robot peripheries, similar to a modular product
structure. Each element can be assigned to different experts for
knowledge formulations via a two-dimensional assignment
matrix. For example, a constraint-based KBC framework such
as Tacton can be used for the implementation of the knowledge.
5. Validation of the method
As already mentioned in section 3, the six-step approach
was derived and validated based on various practical cases,
mainly in the context of the research project ROBOTOP. In the
latter, a web-based KBEC for supporting the concept planning
of RAS was developed [25,32]. Data collection was performed
using multiple data sources via non-systematic direct observa-
tion, literature review, archived document analysis, and semi-
structured interviews. Additionally, all available documents
generated during the different KA processes and projects were
collected and studied.
1) Collection and synthesis of business processes and out-
puts: In the context of the research project ROBOTOP, the
business processes analyses and required process granularity
could be achieved by literature and document synthesis, in
combination with expert interviews. Fig. 2 shows the resulting
engineering and integration process of production systems,
which is sequenced into six phases starting with the clarifica-
tion of the requirements of the customer respectively the man-
ufacturing company up to the operating phase. Due to its gen-
erality, Fig. 2 offers a reference process for almost all KBECs.
In the next step, the affected stakeholders can be identified
based on the business processes and outputs.
Fig. 2. Business processes and output documents along the system integration
according to our synthesis out of [18,19,33,34] and expert interviews
2) Collection of the affected stakeholders and their goals:
From a system integrator’s perspective in the context of a
KBEC for RAS, the technical sales department maximizes sold
systems and their commission, project management seeks to
meet time as well as cost targets, and engineering ensures the
functional performance. From the company’s manufacturing
perspective, production planning and purchasing aim to mini-
mize the costs while simultaneously claiming for the required
function, quality, and flexibility.
3) Definition of the minimal necessary scope of func-
tions: In the case of the exemplary KBEC for RAS, the pro-
posal to rough planning phases are covered (see Fig. 2). The
first relevant output documents are requirement specification
and quotation proposal. Thus, for the minimal necessary scope,
a so-called Best Practice selection, a subsequent adaptation
configuration, an extraction of a requirement specification and
a quotation proposal have to be configured and generated
[32,35].
4) Multidimensional classification of needed knowledge:
Depending on the objective, it can be important to focus the
development of a KBEC based on the different relevant dimen-
sions as seen in Fig. 3 in order to avoid over-engineering and
unnecessary development expenses. In some use cases, a broad
overview is desired, where the validity of the knowledge is not
quite as relevant. In other cases, a very detailed, complete, and
precise configuration is required. Fig. 3 shows our concept for
a multidimensional requirements assessment matrix along the
following dimensions: Validity of the knowledge, process
phase, and universality. Within the example use case, we fo-
cused on the gap between classical sales systems and engineer-
ing tool. In doing so, the KBEC for RAS represents a technical
niche, specifying requirements, generating proposals up to the
first technical specification for rough planning.
Fig. 3. Three-dimensional requirements assessment for needed knowledge
5) Collection of available knowledge sources: Based on
the projects and authors experience with KA, an exemplary list
of knowledge sources is summarized in Table 2. This overview
should be elaborated, classified, and evaluated in detail in the
course of further research and industrial practice. In addition,
the authors emphasize the importance of available knowledge
overviews to enable employees to make the most informed de-
cisions in KA. Within the use case, some of the knowledge
sources in Table 2, in particular 1.1 to 1.3, 2.2, 2.3, 2.5, 2.6,
2.8, 2.9, and 3.1 to 3.3, were beneficial for the planning of RAS.
6) Structured knowledge acquisition and implementa-
tion: For the actual KA based on constraints, the aforemen-
tioned method from
[25] was applied. Therefore the elements
of the 150% topology were used as modules within the assign-
ment matrix to distribute the KA tasks to different people in-
volved in the project.
Output documents
Order
Time
Feasibility
check
Machine
concept
Detailed
design and
development
Assembly,
launch
test and
acceptance
Training and
optimization
Customer (Input :
requirements) Engineering department (Output: system integration)
Requirement
specifications
Technical specifications System
documentation
Production
Sales department
Information
about the
production
system
Quotation
proposal
System
integrator
Eike Schäffer et al. / Procedia CIRP 00 (2019) 000–000 5
Table 2. Exemplary knowledge sources including their description and qualitative evaluation based on our experience in the field of knowledge work
Category No. Knowledge source Level of detail and focus Quality Costs Machine
executable
1) In-house
knowledge
(developed
knowledge
within the
company;
not accessible to
others)
1.1
Process documentation
(e.g. wiki) and documents
(e.g. Word, Excel)
Application-specific and context-dependent
documents and presentations Low to high Low No to depends
1.2 Templates (e.g. Word, Excel) Task- or function-specific templates Low to high Low Yes
1.3 Expert interviews Subjective experience High Medium No
1.4 Operative employee interviews Experience of usage and maintenance Low to high Low No
1.5 IT system or software tool for
knowledge automation
From validation (e.g. simulation) to
knowledge automation (e.g. configurator) Low to high Low to
high Depends to yes
1.6 Dataset (existing exported data) Objective and partly transferable Low to high Medium Depends
1.7 New experiments Objective and problem-specific Low to high Very high Depends
2) Inexpensive
or simply
acquirable
external
knowledge
(publicly
available
knowledge or
relatively
in-expensively
purchasable for
anyone)
2.1 Encyclopedia General overview (e.g. wikipedia.com) Medium Free Depends
2.2 Expert internet forums Problem-specific questions and discussions Low to high Free No
2.3 Non-scientific publication Superficial, populist, or marketing-oriented Low Free No
2.4 White paper First concept exploration Medium Free No
2.5 Text- and picture-based
technical book or handbook
State of the art within practical application
(written text, general information)
Medium to
high
Free to
Medium
No
2.6 Table- and equation-based
technical book or handbook
Logical rules or constraints, problem-
specific equations, algorithms, and tables
Medium to
high
Free to
Medium Depends to yes
2.7 Scientific publication (e.g.
conference papers, journals)
Very detailed scientific novel methods or
concepts High Medium
to free No
2.8 Standard and norm Highly condensed knowledge Very high Low No to depends
2.9 Implemented standard and norm Implemented data schemas (e.g. eCl@ss) High Medium Depends to yes
2.10 Free software tool or library Implemented solution for a specific problem Low to high Free Yes
2.11 Public statistics Market statistics (e.g. from statistical
platforms like statista.com) Medium Low Depends
2.12 Law General orientation or directive High Free No
3) Elaborately
available exter-
nally acquirable
knowledge (ex-
ternal project,
service, business
or service con-
tract, joint ven-
ture, company
acquisition)
3.1 Exhibition and presentation Company-specific solution or novelty Medium Low No
3.2 Expert interview Specific and subjective High Medium No
3.3 Customer or
supplier
Level of knowledge, requirements and ex-
pectations from end-user Medium Low to
high No
3.4 Training or seminar Topic-specific event Low to high High No
3.5 Commercial software tool Implemented solution for a specific problem Low to high High Yes
3.6 Consulting Comparison or general approaches High High No
3.7 Research institutes
(e.g. university collaboration)
Future topics (e.g. new concept or
technology) High Low to
high No
3.8 Joint venture Access to external processes and experience Low to high High No to depends
Particularly for persons with little experience in the field
of KA, a focused and consistent procedure could be ob-
served. In contrast to earlier approaches, the general needed
training period and time for KA were reduced while produc-
ing better results in the form of more constraints relevant to
the problem specification.
In addition, a 3D visualization was developed for the
KBEC to further improve its usability (Fig. 4). The 3D-
KBEC was implemented using the configuration platform
Tacton and the 3D-CAD tool Inventor. The general architec-
ture of the configurator builds upon a Best Practice based ad-
aptation respectively change configuration. [25,32,36]
Fig. 4. Prototypical implementation of the 3D-KBEC for RAS; Best
Practice selection (left) and adaptation configuration (right)
In general, it was shown that the KA process could be op-
timized and accelerated by the proposed strategic six-step ap-
proach. Using the method, an overall transparency and un-
derstanding of which functionalities are relevant was
achieved. Certain aspects such as the overview of knowledge
sources collected in Table 2 can be re-used for future KBEC
projects.
6. Conclusion and outlook on future research activities
In general, the design of and the KA process for imple-
menting a KBEC can only be performed effectively through
a strategic understanding of the minimum required functional
scope as well as the available and optimal knowledge
sources. On the way to knowledge process automation in en-
gineering, some basic principles still need to be researched
as shown in the state of the art section. Therefore, this paper
suggests a six-step approach for the strategic design of
KBECs (step 1 to 5) before the operative KA and implemen-
tation (step 6). For the first phases of KBEC development, a
more detailed specification of the requirements as well as the
analysis of the involved stakeholders are necessary.

Eike Schäffer et al. / Procedia CIRP 96 (2020) 219–224 223
4 Eike Schäffer et al. / Procedia CIRP 00 (2019) 000–000
compilation of knowledge sources in one table. Thus,
knowledge engineers do not have to rework the same ideas on
appropriate sources of knowledge in each project. Without such
a tabular overview, employees usually tend to use the most fa-
miliar knowledge sources, as they mostly do not have time or
interest to research all available knowledge sources.
4.6. Structured knowledge acquisition and implementation
The tabular knowledge source overview and pre-selection
enables a focused acquisition of knowledge in the final step.
The operative division of labor and KA process can be accom-
plished according to the method introduced in [25]. In the latter
work, a so-called 150% topology for RAS is introduced, bring-
ing the structures of an engineering configurator closer to those
of a PC. Besides, a constraint matrix for a targeted division of
labor in KA is used. The 150% topology provides generic top-
ological classes for engineering elements such as robot arm,
gripper, or other robot peripheries, similar to a modular product
structure. Each element can be assigned to different experts for
knowledge formulations via a two-dimensional assignment
matrix. For example, a constraint-based KBC framework such
as Tacton can be used for the implementation of the knowledge.
5. Validation of the method
As already mentioned in section 3, the six-step approach
was derived and validated based on various practical cases,
mainly in the context of the research project ROBOTOP. In the
latter, a web-based KBEC for supporting the concept planning
of RAS was developed [25,32]. Data collection was performed
using multiple data sources via non-systematic direct observa-
tion, literature review, archived document analysis, and semi-
structured interviews. Additionally, all available documents
generated during the different KA processes and projects were
collected and studied.
1) Collection and synthesis of business processes and out-
puts: In the context of the research project ROBOTOP, the
business processes analyses and required process granularity
could be achieved by literature and document synthesis, in
combination with expert interviews. Fig. 2 shows the resulting
engineering and integration process of production systems,
which is sequenced into six phases starting with the clarifica-
tion of the requirements of the customer respectively the man-
ufacturing company up to the operating phase. Due to its gen-
erality, Fig. 2 offers a reference process for almost all KBECs.
In the next step, the affected stakeholders can be identified
based on the business processes and outputs.
Fig. 2. Business processes and output documents along the system integration
according to our synthesis out of [18,19,33,34] and expert interviews
2) Collection of the affected stakeholders and their goals:
From a system integrator’s perspective in the context of a
KBEC for RAS, the technical sales department maximizes sold
systems and their commission, project management seeks to
meet time as well as cost targets, and engineering ensures the
functional performance. From the company’s manufacturing
perspective, production planning and purchasing aim to mini-
mize the costs while simultaneously claiming for the required
function, quality, and flexibility.
3) Definition of the minimal necessary scope of func-
tions: In the case of the exemplary KBEC for RAS, the pro-
posal to rough planning phases are covered (see Fig. 2). The
first relevant output documents are requirement specification
and quotation proposal. Thus, for the minimal necessary scope,
a so-called Best Practice selection, a subsequent adaptation
configuration, an extraction of a requirement specification and
a quotation proposal have to be configured and generated
[32,35].
4) Multidimensional classification of needed knowledge:
Depending on the objective, it can be important to focus the
development of a KBEC based on the different relevant dimen-
sions as seen in Fig. 3 in order to avoid over-engineering and
unnecessary development expenses. In some use cases, a broad
overview is desired, where the validity of the knowledge is not
quite as relevant. In other cases, a very detailed, complete, and
precise configuration is required. Fig. 3 shows our concept for
a multidimensional requirements assessment matrix along the
following dimensions: Validity of the knowledge, process
phase, and universality. Within the example use case, we fo-
cused on the gap between classical sales systems and engineer-
ing tool. In doing so, the KBEC for RAS represents a technical
niche, specifying requirements, generating proposals up to the
first technical specification for rough planning.
Fig. 3. Three-dimensional requirements assessment for needed knowledge
5) Collection of available knowledge sources: Based on
the projects and authors experience with KA, an exemplary list
of knowledge sources is summarized in Table 2. This overview
should be elaborated, classified, and evaluated in detail in the
course of further research and industrial practice. In addition,
the authors emphasize the importance of available knowledge
overviews to enable employees to make the most informed de-
cisions in KA. Within the use case, some of the knowledge
sources in Table 2, in particular 1.1 to 1.3, 2.2, 2.3, 2.5, 2.6,
2.8, 2.9, and 3.1 to 3.3, were beneficial for the planning of RAS.
6) Structured knowledge acquisition and implementa-
tion: For the actual KA based on constraints, the aforemen-
tioned method from
[25] was applied. Therefore the elements
of the 150% topology were used as modules within the assign-
ment matrix to distribute the KA tasks to different people in-
volved in the project.
Output documents
Order
Time
Feasibility
check
Machine
concept
Detailed
design and
development
Assembly,
launch
test and
acceptance
Training and
optimization
Customer (Input :
requirements) Engineering department (Output: system integration)
Requirement
specifications
Technical specifications System
documentation
Production
Sales department
Information
about the
production
system
Quotation
proposal
System
integrator
Eike Schäffer et al. / Procedia CIRP 00 (2019) 000–000 5
Table 2. Exemplary knowledge sources including their description and qualitative evaluation based on our experience in the field of knowledge work
Category No. Knowledge source Level of detail and focus Quality Costs Machine
executable
1) In-house
knowledge
(developed
knowledge
within the
company;
not accessible to
others)
1.1
Process documentation
(e.g. wiki) and documents
(e.g. Word, Excel)
Application-specific and context-dependent
documents and presentations Low to high Low No to depends
1.2 Templates (e.g. Word, Excel) Task- or function-specific templates Low to high Low Yes
1.3 Expert interviews Subjective experience High Medium No
1.4 Operative employee interviews Experience of usage and maintenance Low to high Low No
1.5 IT system or software tool for
knowledge automation
From validation (e.g. simulation) to
knowledge automation (e.g. configurator) Low to high Low to
high Depends to yes
1.6 Dataset (existing exported data) Objective and partly transferable Low to high Medium Depends
1.7 New experiments Objective and problem-specific Low to high Very high Depends
2) Inexpensive
or simply
acquirable
external
knowledge
(publicly
available
knowledge or
relatively
in-expensively
purchasable for
anyone)
2.1 Encyclopedia General overview (e.g. wikipedia.com) Medium Free Depends
2.2 Expert internet forums Problem-specific questions and discussions Low to high Free No
2.3 Non-scientific publication Superficial, populist, or marketing-oriented Low Free No
2.4 White paper First concept exploration Medium Free No
2.5 Text- and picture-based
technical book or handbook
State of the art within practical application
(written text, general information)
Medium to
high
Free to
Medium
No
2.6 Table- and equation-based
technical book or handbook
Logical rules or constraints, problem-
specific equations, algorithms, and tables
Medium to
high
Free to
Medium Depends to yes
2.7 Scientific publication (e.g.
conference papers, journals)
Very detailed scientific novel methods or
concepts High Medium
to free No
2.8 Standard and norm Highly condensed knowledge Very high Low No to depends
2.9 Implemented standard and norm Implemented data schemas (e.g. eCl@ss) High Medium Depends to yes
2.10 Free software tool or library Implemented solution for a specific problem Low to high Free Yes
2.11 Public statistics Market statistics (e.g. from statistical
platforms like statista.com) Medium Low Depends
2.12 Law General orientation or directive High Free No
3) Elaborately
available exter-
nally acquirable
knowledge (ex-
ternal project,
service, business
or service con-
tract, joint ven-
ture, company
acquisition)
3.1 Exhibition and presentation Company-specific solution or novelty Medium Low No
3.2 Expert interview Specific and subjective High Medium No
3.3 Customer or
supplier
Level of knowledge, requirements and ex-
pectations from end-user Medium Low to
high No
3.4 Training or seminar Topic-specific event Low to high High No
3.5 Commercial software tool Implemented solution for a specific problem Low to high High Yes
3.6 Consulting Comparison or general approaches High High No
3.7 Research institutes
(e.g. university collaboration)
Future topics (e.g. new concept or
technology) High Low to
high No
3.8 Joint venture Access to external processes and experience Low to high High No to depends
Particularly for persons with little experience in the field
of KA, a focused and consistent procedure could be ob-
served. In contrast to earlier approaches, the general needed
training period and time for KA were reduced while produc-
ing better results in the form of more constraints relevant to
the problem specification.
In addition, a 3D visualization was developed for the
KBEC to further improve its usability (Fig. 4). The 3D-
KBEC was implemented using the configuration platform
Tacton and the 3D-CAD tool Inventor. The general architec-
ture of the configurator builds upon a Best Practice based ad-
aptation respectively change configuration. [25,32,36]
Fig. 4. Prototypical implementation of the 3D-KBEC for RAS; Best
Practice selection (left) and adaptation configuration (right)
In general, it was shown that the KA process could be op-
timized and accelerated by the proposed strategic six-step ap-
proach. Using the method, an overall transparency and un-
derstanding of which functionalities are relevant was
achieved. Certain aspects such as the overview of knowledge
sources collected in Table 2 can be re-used for future KBEC
projects.
6. Conclusion and outlook on future research activities
In general, the design of and the KA process for imple-
menting a KBEC can only be performed effectively through
a strategic understanding of the minimum required functional
scope as well as the available and optimal knowledge
sources. On the way to knowledge process automation in en-
gineering, some basic principles still need to be researched
as shown in the state of the art section. Therefore, this paper
suggests a six-step approach for the strategic design of
KBECs (step 1 to 5) before the operative KA and implemen-
tation (step 6). For the first phases of KBEC development, a
more detailed specification of the requirements as well as the
analysis of the involved stakeholders are necessary.

224 Eike Schäffer et al. / Procedia CIRP 96 (2020) 219–224
6 Eike Schäffer et al. / Procedia CIRP 00 (2019) 000–000
Within the six-step method, a matrix for the focused ac-
quisition of constraint knowledge within the dimensions va-
lidity of knowledge, process phase, and universality is pro-
posed. In addition, a first overview table of available
knowledge sources for further KA is composed in Table 2.
In doing so, constraint knowledge re-use across different in-
dependent configurators or even configuration microservices
[2,37,38] is conceivable.
For further optimization of the knowledge processes, it is
nevertheless necessary to conduct a more detailed require-
ments assessment and to integrate more concrete mecha-
nisms and process management approaches into knowledge
work. Moreover, various aspects of user-centred design, e.g.
[39], need to be studied and integrated into KBECs [40].
Acknowledgements
The research project ROBOTOP (01MA17009E) is
funded by the German Federal Ministry of Economic Affairs
and Energy (BMWi). It is part of the technology program
“PAiCE – Digitale Technologien für die Wirtschaft”.
References
[1] Reinhart G, Magaña Flores A, Zwicker C. Industrieroboter: Planung,
Integration, Trends. Vogel Communications; 2018.
[2] Schäffer E, Pownuk T, Walberer J, Fischer A, Schulz J-P,
Kleinschnitz M et al. System architecture and conception of a
standardized robot configurator based on microservices.
Tagungsband des 3. Kongresses Montage Handhabung
Industrieroboter 2018:159–66.
[3] Koch M. Wissensbasierte Unterstützung der Angebotsbearbeitung in
der Investitionsgüterindustrie. Zugl.: Erlangen-Nürnberg, Univ.,
Diss.,. München: Hanser; 1995.
[4] Valavanis KP, Kokkinaki AI, Tzafestas SG. Knowledge-based
(expert) systems in engineering applications: A survey. J Intell Robot
Syst 1994(2):113–45.
[5] Blažek P, Kolb M, Streichsbier C, Honetz S. The Evolutionary
Process of Product Confi gurators. In: Bellemare J, Carrier S, Nielsen
K, Piller FT, editors. Managing Complexity: Proceedings of the 8th
World Conference on Mass Customization, Personalization, and Co-
Creation. Springer International Publishing; 2017, p. 161–172.
[6] Bodendorf F. Daten- und Wissensmanagement. 2nd ed. Berlin,
Heidelberg, New York: Springer; 2006.
[7] Mandl M, Felfernig A, Teppan E, Schubert M. Consumer decision
making in knowledge-based recommendation. J Intell Inf Syst
2011;37(1):1–22.
[8] Stumptner M. An overview of knowledge-based configuration. AI
Communications 1997;10(2):111.
[9] Feldmann K, Schmuck T, Brossog M, Dreyer J.
Beschreibungsmodell zur Planung von Produktionssystemen:
Entwicklung eines Beschreibungsmodells für Produkte, Prozesse
und Ressourcen zur rechnergestützten Planung
produktionstechnischer Systeme. wt-online 2008;98(3):156–62.
[10] Vajna S, Bley H, Hehenberger P, Weber C, Zeman K. CAx für
Ingenieure: Eine praxisbezogene Einführung. 2nd ed. Berlin,
Heidelberg: Springer; 2009.
[11] Helbig TB. Methode zur Verbesserung der domänenübergreifenden
Zusammenarbeit während des Engineering-Prozesses im
Sondermaschinenbau. 1st ed. Aachen: Shaker; 2017.
[12] Reinema C, Pompe A, Nyhuis P. Agiles Projektmanagement. ZWF
2013;108(3):113–7.
[13] Felfernig A, Hotz L, Bagley C, Tiihonen J. Knowledge-based
configuration: From research to business cases. Waltham, MA:
Morgan Kaufmann; 2014.
[14] Gönnheimer P, Kimmig A, Ehrmann C, Schlechtendahl J, Güth J,
Fleischer J. Concept for the Configuration of Turnkey Production
Systems. Procedia CIRP 2019;86:234–8.
[15] Haug A, Shafiee S, Hvam L. The causes of product configuration
project failure. Computers in Industry 2019;108:121–31.
[16] hosh A, Myrodia A, Mortensen NH, Hvam L. Reusing components
across multiple configurators. Proceedings of the 21st Configuration
Workshop 2019(Vol. 2467):53–60.
[17] Hayes-Roth F (ed.). Building expert systems. London: Addison-
Wesley; 1983.
[18] Rimpau C, Reinhart G. Knowledge-based risk evaluation during the
offer calculation of customised products. Prod. Eng. Res. Devel.
2010;4(5):515–24.
[19] Klein PW. Methode zum Engineering von Produktionsanlagen durch
Wiederverwendung von Modulen [Dissertation]. Siegen; 2015.
[20] Puppe F. Einführung in Expertensysteme. Springer; 1991.
[21] Hofstedt P. Constraints. In: Görz G, Schneeberger J, Schmid U,
editors. Handbuch der künstlichen Intelligenz, 5th ed. München:
Oldenbourg; 2014, p. 205–233.
[22] Fleischanderl G, Friedrich GE, Haselböck A, Schreiner H, Stumptner
M. Configuring large systems using generative constraint
satisfaction. IEEE Intelligent Systems 1998;13(4):59–68.
[23] Kvale S. InterViews: An introduction to qualitative research
interviewing. Thousand Oaks u.a: Sage; 2004.
[24] Shafiee S, Kristjansdottir K, Hvam L, Forza C. How to scope
configuration projects and manage the knowledge they require.
Journal of Knowledge Management 2018;22(5):982–1014.
[25] Schäffer E, Fröhlig S, Mayr A, Franke J. A method for collaborative
knowledge acquisition and modeling enabling the development of a
knowledge-based configurator of robot-based automation solutions.
Procedia CIRP 2019;86:92–7.
[26] Schäffer E, Penczek L, Mayr A, Bakakeu J, Franke J, Kuhlenkötter
B. Digitalisierung im Engineering: Ein Ansatz für ein
Vorgehensmodell zur durchgehenden, arbeitsteiligen Modellierung
am Beispiel von AutomationML. I40M 2019;2019(1):61–6.
[27] Milgrom P. Putting auction theory to work. Cambridge: Cambridge
University Press; 2010.
[28] Hehenkamp B. Die Grundlagen der Mechanismus-Design-Theorie.
Wirtschaftsdienst 2007;87(11):768–72.
[29] Becker J, Kahn D. Der Prozess im Fokus. In: Prozessmanagement.
Springer; 2012, p. 3–16.
[30] Yin RK. Case study research: Design and methods. 3rd ed. Thousand
Oaks, Calif: SAGE Publications; 2007.
[31] Voss C, Tsikriktsis N, Frohlich M. Case research in operations
management. Int Jrnl of Op & Prod Mnagemnt 2002;22(2):195–219.
[32] Schäffer E, Thi K, Franke J. Bedarf und Konzeptvorstellung des
Forschungsprojekts ROBOTOP für einen webbasierten Konfigurator
für den Massenmarkt 2020.
[33] VDI/VDE 3694 - Lastenheft/ Pflichtenheft für den Einsatz von
Automatisierungssystemen; 2014.
[34] Franke J, Götz J. Vernetzung für effiziente Anlagenplanung.
chemie&more 2014(2):11–4.
[35] Schäffer E, Schulz J-P, Franke J. Robotiklösungen im
Baukastenprinzip für den Mittelstand; (2019); In: bigdata-insider;
Available from: www.bigdata-insider.de/robotikloesungen-im-
baukastenprinzip-fuer-den-mittelstand-a-882401.
[36] Schäffer E, Mayr A, Fuchs J, Sjarov M, Vorndran J, Franke J.
Microservice-based architecture for engineering tools enabling a
collaborative multi-user configuration of robot-based automation
solutions. Procedia CIRP 2019;86:86–91.
[37] Schäffer E, Leibinger H, Stamm A, Brossog M, Franke J.
Configuration based process and knowledge management by
structuring the software landscape of global operating industrial
enterprises with Microservices. Procedia Manufacturing
2018;24:86–93.
[38] Schäffer E, Mayr A, Huber T, Höflinger T, Einecke M, Franke J.
Gradual tool-based optimization of engineering processes aiming at
a knowledge-based configuration of robot-based automation
solutions. Procedia CIRP 2019;81:736–41.
[39] Mao J-Y, Vredenburg K, Smith PW, Carey T. The state of user-
centered design practice. COMMUNICATIONS OF THE ACM
2005(Vol. 48, No. 3):105–9.
[40] Schäffer E, Shafiee S, Frühwald T, Franke. A development approach
towards user-centered front-ends for knowledge-based engineering
configurators: a study within planning of robot-based automation
solutions. In: Proceedings of the 22st ConfWS'20; 2020, "not yet
published".