Advanced Engineering Informatics xxx (2017) xxxxxx

Contents lists available at ScienceDirect

Advanced Engineering Informatics

journal homepage: www.elsevier.com/locate/aei

Knowledge-based design for assembly in agile manufacturing by using

Data Mining methods
R. Kretschmer b, A. Pfouga a, S. Rulhoff a, J. Stjepandic a,
a
PROSTEP AG, Head of Business Unit 3D Product Creation, Dolivostrabe 11, D-64293 Darmstadt, Germany
b
Miele & Cie. KG, Segment Professional Laundry, Director Segment Professional Laundry, Technology Lehrte, Industriestrae 3, 31275 Lehrte, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Decision making in early production planning phases is typically based on a rough estimation due to lack
Received 15 June 2016 of a comprehensive, reliable knowledge base. Virtual planning has been prevailed as a method used to
Received in revised form 16 December 2016 evaluate risks and costs before the concrete realization of production processes. The process of product
Accepted 22 December 2016
assembly, which yields a high share in total production costs, gets its particular importance. This paper
Available online xxxx
introduces a new approach and its initial implementation for knowledge-based design for assembly in
agile manufacturing by using data mining (DM) methods in the field of series production with high vari-
Keywords:
ance. The approach adopts the usage of bulk data with old, successful designs in order to extrapolate its
Design for assembly
Agile manufacturing
scope for assembly processes. Especially linked product and process data allow the innovative usage of
Digital factory DM methods in order to facilitate the front loading in the product development. The concept presents
Assembly an affordable assistance potential for development of new products variants along the product emer-
Process planning gence process (PEP). With this approach an early cost estimation of assembly processes in series produc-
Data mining tion can be conducted using advanced DM methods as shown in an industrial use case. Furthermore,
design and planning processes can be supported effectively.
2016 Elsevier Ltd. All rights reserved.

1. Introduction feasibility of series production must be assured with vague infor-

Today globally operating companies face additional challenges
due to turbulent fluctuations in market demand, increasing vari-
ability of products, shortened product lifecycles and corresponding
complexity of processes [1]. It results in higher flexibility expec-
tance in production system and economically reasonable dispatch
of new products in an existing production line. The pivotal link
between design and manufacturing lies in process planning. Pro-
cess planning deals with the selection of necessary manufacturing
processes and determination of their sequences to transform a
designers creation (namely the shaped part) into a physical com-
ponent economically and competitively. In the modern product
emergence process (PEP), production planning gains in importance
and has to be executed as parallel as possible to the product devel-
opment according to concurrent engineering principles [10]. In this
early product creation phase, a first step for planning processes is a
cost calculation for the industrialization of the product in existing
production lines regarding basic conditions [5]. The cost-effective
Corresponding author.
E-mail addresses: ralf.kretschmer@miele.de (R. Kretschmer), Alain.Pfouga@
PROSTEP.com (A. Pfouga), Stefan.rulhoff@prostep.com (S. Rulhoff), josip.stjepandic@
prostep.com (J. Stjepandic).
mation on product and given general conditions, e.g. shift model
[6]. This is a great challenge especially to planning cost-intensive
assembly of a product [41,53].
Front loading has been seen as an appropriate means to tackle
such challenges [43] and must be supported by adequate pro-
cesses, methods and tools. Subsequently, production planning
should start as early as possible in phase with product develop-
ment. The research and development project Prospective Deter-
mination of Assembly Work Content in Digital Manufacturing
(Pro Mondi) was initiated to link the early product design with
the early production planning using methods of data modeling
and data mining (DM) to generate information with focus on the
product assembly planning for new products in early production
planning phases. Aim of this project is to achieve accurate estima-
tion of expected assembly work and resulting costs in an early
stage of the product development. This should be realized in sync
with design through provision of additional support with assembly
knowledge for an underlying design. The approach, thus, contains
reuse of existing planning data in order to extrapolate assembly
processes. Especially linked product and process data allow the
innovative usage of Data Mining methods. As proof of concept this

http://dx.doi.org/10.1016/j.aei.2016.12.006
1474-0346/ 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
2 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx
approach has been evaluated with different manufacturing
companies.
2. Related work
2.1. Design for X
The literature offers besides Design for X also the term
Design to X, both of which are often grouped under the concept
Design for X, but have different meanings. Many definitions
highlight the fact that there are different objectives that come into
consideration in this work as well. The terms are interpreted
according to the definition of Vielhaber [62]:
1. According to Chen [14], Design for X describes the reactive,
methodological assistance in development of a product to a
property determined by a set of features which is necessary,
to make a follow-up process in the product lifecycle successful
(e.g., adaptability to manufacture).
2. Design to X describes the proactive, methodical assistance in
development of a product for a target property (e.g., cost-
effective products).
3. X-Oriented Design refers to the collection of all Design for X
and Design to X criteria.
4. The terms features and characteristics are understood
according to the view of the Characteristics-Properties Model-
ing (CPM)/Property-Driven Development (PDP) approaches
by Weber [66] in the definitions.
The main objectives (whether Design for X or Design to X) are
often repeatedly considered during the development process and
therefore decisions have to be brought about at a time. For exam-
ple, the criteria are taken into account in early stages of product
development, but may need to be focused again later during the
elaboration of the product shape and have to be adjusted [50].
Therefore, it can be stated in principle that the application of meth-
ods and tools of Design for X is a way to consider the numerous
restrictions in the development process [65]. Both analysis and
synthesis steps must be included. Weber explains the poor integra-
tion of Design for X-methods and tools in the design methodology
with the required characteristic by Design for X (e.g., material
strength or assemblability), which cannot be easily described by
terms such as functional structure or solution principle [67].
According to Bossmann [9], DfX methods can be subdivided in
five main groups (user, environment, functional reliability, costs,
and production). The latter is composed of the following sub-
groups: Machining, Reshaping, Primary forming, Component man-
ufacturing, and Assembly.
In the ideal case all guidelines are considered in designing a
new product, but that case is improbable. Therefore, the design
department has to be supported by other departments which pro-
vide their expertise in order to meet the Design for X
requirements.
There are several design for or design to criteria, which are
summarized with Design for X. Design for Disassembly and
Design for Recycling, for example, are covered by the same group,
which is headlined with Design for Environment. Boothroyd and
Alting concentrated on design for assembly [7]. In their concept
the reduction of the number of components is particularly
addressed to reduce assembly costs. A consistent implementation
in Swedish companies led to convincing results [61]. In 50% of
these companies a reduction of development time and develop-
ment costs by 33% was established. The assembly costs could be
reduced by up to 85% with the DFA approach by Boothryd and Alt-
ing [61]. Design for assembly and Design for Manufacturing are
Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
often summarized by the concept Design for Manufacturing and
Assembly (DFMA) [7]. The normal result of DFMA, as an integral
part of the design process, is simpler and more reliable products
that are less expensive to manufacture and assemble [37]. How-
ever, products designed in this way tend to have a smaller number
of complex components, making maintenance and upgrading diffi-
cult and expensive. The emphasis on reducing manufacturing costs
has, therefore, been at the detriment of in-service costs. This may
not be a particular problem for mass-produced (typically minimal
maintenance, low priced, short life span) products such as the
majority of domestic appliances [22]. Meeting manufacture crite-
ria, not only anticipated the manufacturing engineering auto-
body design activities reducing the time to market, but reduces
investments and structural cost of the plant are the findings of
the industrial study [36] [39]. These strategies have generated
resources with the mechanical presses selling and create a new
platform with fewer operations, on average, comparing similar
parts with previous designs [57]. Other approaches focus on the
complexity of design rules, especially in semantic assembly deci-
sion [15]. A major challenge of the complexity is a computational
issue associated with a very sophisticated and time-consuming
task with respect to semantic reasoning for ontology-based pro-
duct design. By using disparate attributes algorithm, computer-
aided systems become more easily able to understand and to dis-
cern joining types.
An excerpt of typical DFMA guidelines is provided in Fig. 1 [9].
An updated version comprises the broad, long-term experience
in DfMA by a set of generic guidelines [6].
2.2. Assembly planning
Assembly is the installation of geometrically defined objects by
joining, handling, calibration and control operations. The task of
assembly planning refers to the creation of necessary precondition
for economic installation procedure. This includes identification of
needs, targeted and efficient use of staff as well as resources.
Assembly is defined as joining or assembling manufactured items
into a functional unit. Within the installation planning, a distinc-
tion is made between operations and detailed planning. Function
of operations planning is to develop an assembly system and to
create a rough schedule [41,53].
The detailed planning is divided in concretization of the assem-
bly system and preparation of a detailed schedule. Table 1 shows
functions and planning phases of assembly planning as well as
its required input information and results to be achieved [9,11].
As product geometrical shape is main, but not only criterion for
assembly process, many relevant product properties are defined
in CAD systems [3,8].
In classic Assembly Planning the assembly sequence has to be
determined by a product analysis. Thereby, geometric relation-
ships between product structure components are examined in par-
ticular. For this purpose the product is dismantled into separate
parts. In some cases, the assembly sequence can be sufficiently
characterized by the joining sequence of the components [51].
For the determination of the assembly sequence Lotter and
Wiendahl evince that 40% of the time is to be invested in order
to determine the order of the components in the assembly while
performing the planning activities [41]. In recent years, some
expenditure shift could be registered due to an increased use of
computers. The largest effort is still located in determining the
assembly sequence, which is why this aspect focused in research
activities [18]. There are numerous approaches which can be
based, for example, on the disassembly sequence to determine an
appropriate assembly sequence [25,38,47].
One result of the assembly planning is the assembly graph,
which represents the possible assembly sequences in a network
 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx 3

Fig. 1. DFMA guidelines [9].

Table 1
Functions and planning phases of the assembly planning.

Task Input Information Output

Assembly system development
Operations planning
Assembly system refinement
Detailed planning
 Work system design  Assembly task  Assembly structure
 Developing system structures  Production program  Rough layout
 Capacity planning  Square measures  Capacity requirements
 Concatenation principle
 Structuring product  Parts list  Precedence graph
 Rough flow planning  Drawing  Rough flow structure
 Principle solution planning  Parts list  Detailed layout
 Workstations planning  Drawing  Workstations
 Concentration resource planning  Concentration resources
 Determining assembly content  Assembly structure  Assembly plan
 Creating assembly plan and assembly documents  Rough layout  Work instructions
 Spreadsheet

structure [19,20]. This network can be seen as a tree structure in
which individual parts are forming leaves, inner branches are seen
as sub-assemblies, which in turn are taken up by complex ramifi-
cations and, finally, end up in the root, the base element [28,29].
For this purpose feature models are used by the Feature Oriented
Domain Analysis (FODA) approach to represent the assembly itself
and its variance [54]. Similarly, a new approach for mining large
quantities of machining features, Computer-Aided Design (CAD)
models and manufacturing data is based on DM and uses
learning-logic classification techniques for mining 3D CAD data.
The proposed approach was evaluated in CAD model analysis,
specifically in classification tasks. The experimental results proved
that the method is effective in terms of classification accuracy and
can be used as an efficient data mining tool for CAD model analysis
and classification [2].
Low-fidelity CAD models can be used to predict assembly times,
thereby supporting earlier inclusion of design for assembly meth-
ods in the design process [45]. Expanding on previous work to
predict assembly times from detailed assembly models, low fide-
lity part models are used in a series of predictive performance
experiments. Results reveal that this tool can predict the assembly
time of a product to within 40% of the target as built time using a
high fidelity neural network and a low fidelity CAD model. The tool
is based on structural complexity, representing the assembly graph
as complexity vector of 29 metrics. A neural network is then used
to build a relationship between the complexity vector (input) and
the assembly time (output).
Various computer-aided planning approaches are compared
with an overweight of knowledge-based and STEP-based
approaches in recent years [71]. Advanced planning approaches
in assembly planning can be grouped in a number of categories,
i.e. feature-based technologies, knowledge-based systems, artifi-
cial neural networks, genetic algorithms, fuzzy set theory and
fuzzy logic, Petri nets, agent-based technology, Internet-based
technology, STEP-compliant CAPP and other emerging technolo-
gies [69].

Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
4 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx
A paper gives a framework and organization of the conceptual
assembly design system with help of knowledge-based engineer-
ing. It thereby takes assembly sequence, joint configuration, and
tolerance allocation in the auto-body assembly design process
planning into integrated consideration [40]. Heuristic and empiri-
cal knowledge play active roles in the generation of assembly
sequence, dimension chain, configuration of joint types, allocation
of tolerance limits and qualitative simulation which is knowledge
or rule reasoning. It is provided with various rule, criterion and
principle. Knowledge-based vehicle assembly design system can
improve assembly concept quality by means of qualitative simula-
tion to fulfill the concurrent integration between conceptual
design and detail design stages. Further example of a military sys-
tem is presented to demonstrate the feasibility of a three-stage
integrated approach with heuristic working rules [12,14].
Similarly, a planning system may include an assemble advisory
module, a knowledge-based module and a user interface to enables
design engineers to find and pick-up the best and fastest assembly
method [60].
2.3. Data mining
Concepts for data mining are projected to be useful to provide
historical insight and best-practices for an evolutionary assortment
of successful design solutions in the context of this paper. This
chapter is devoted to an introduction to this particular domain.
Data mining is a process of discovering valuable information
from observational data sets, which is an interdisciplinary field
bringing together techniques from databases, machine learning,
optimization theory, statistics, pattern recognition, and visualiza-
tion. Introduction of data mining concepts, implementation proce-
dures and application cases in engineering is described by [70,59].
The overall goal of the data mining process is to extract informa-
tion from a data set and transform it into an understandable struc-
ture for further use. One of the greatest expected benefits of DM
methods is the ability to link seemingly disparate disciplines, for
the purpose of developing and testing hypotheses that cannot be
approached within a single knowledge domain. Methods are
reviewed by which analysts can navigate through different data
resources (e.g. historical data) to create new, merged data sets
[17]. Significant factors are efficient knowledge utilization and
knowledge exchange on an interdisciplinary level. An empirical
study involving 206 Data Mining tools users was conducted to
evaluate the model using structural equation modeling. Results
Fig. 2. Typical context of data mining procedure [46].
Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
demonstrate that the proposed model explains 58% of the variance
[32] what indicates disparate usability of this approach. Three fac-
tors in the task-oriented dimension, output quality, response time,
and format, provide valuable points of reference for the designers
and developers of DM tools and relevant software [46]. Typical
context of the commercial DM procedure is shown in Fig. 2.
Phase 1 (Business Objectives) comprises the definition of a DM
project, including its initial concept, motivation, business objec-
tives, viability, estimated costs, and expected benefit (returns).
Phase 2 (Data Preparation) deals with possible sources of data
and information, different ways of conceptualization, defines the
required data quality, considers the topics of variable selection
and factor derivation, sampling and partitioning methods, which
is often done when the volume of data is too great to process.
Phase 3 (Data Analysis) provides a selection of the most common
types of data analysis for data mining. It includes the clustering
in combination with visualization techniques, transactional analy-
sis and time series analysis. Phase 4 (Modeling) begins with the
definition of a data model, followed by discussion of concepts such
as supervised and unsupervised learning, cross-validation. It con-
siders various techniques for modeling data, from AI (artificial
intelligence) approaches, such as neural networks and rule induc-
tion, to statistical techniques, such as regression. Finally, in Phase 5
(Deployment), the results of data mining can be fed into the
decision-making and operative processes of the business in differ-
ent ways (query/reporting, executive information systems and
expert systems).
A further article presents a data-miningaided optimal design
method that is able to find a competitive design solution with a rel-
atively low computational cost [33]. The method consists of four
components: (1) a uniform-coverage selection method, that
chooses design representatives from among a large number of
original design alternatives for a nonrectangular design space; (2)
feature functions, of which evaluation is computationally econom-
ical as the surrogate for the design objective function; (3) a cluster-
ing method, that generates a design library based on the evaluation
of feature functions instead of an objective function; and (4) a clas-
sification method to create the design selection rules, eventually
leading us to a competitive design.
New processes suitable to assemble the given new product shall
be designed based on this existing, historical data (product linked
to corresponding process). Successful designs from the past are
expected to be a good envelope for the solution space of the future
designs [27]. Automatic analysis with a specific DM method shall
 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx 5

Fig. 3. Design optimization with additional time data.

Fig. 4. Emergence of product assembly information PAI.

be used to create a first draft of the assembly process and estimate engineer can be supported at the selection of appropriate joining
the expected costs. Following production planning processes can elements. With this approach, an assembly knowledge based sup-
be supported by automatic proposals of adequate assembly pro- port of the designer in series production can be achieved using
cesses, which then can be customized [52]. Moreover, the design innovative DM methods.

Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
6 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx
2.4. Agile manufacturing
Agile manufacturing has been defined with respect to the agile
enterprise, products, workforce, capabilities and the environment
that gives impetus to the development of agile paradigm [72]. It
assimilates the full range of flexible production technologies, along
with the lessons learned from total quality management, just-in-
time production and lean production. Agility is the ability of a
business to grow in a competitive market of continuous and unan-
ticipated change, to respond quickly to rapidly changing markets
driven by customer-based valuing of products and services. In
other words agile organizations are able to compete on the basis
of time-compression. Finally, agile manufacturing is a system that
shifts quickly (speed and responsiveness) among product models
or between product lines (flexibility), ideally in real-time response
to customer demand (customer needs and wants).
3. Background and requirements
Worldwide available, Miele is the global premium brand of
domestic appliances and commercial machines in the field of laun-
dry care, dishwashing and disinfection [44]. A continuous stream
of innovations is the foundation of Mieles business success since
1899. In terms of quality, Miele appliances are considerably better
than those of the competition; otherwise they would not have
been able to compete successfully on such a fiercely competitive
market. To facilitate agile manufacturing, Miele is looking for
new approaches and methods for product and production plan-
ning. One of the main areas of interest is the assembly process
which is cost-intensive and, therefore, has high potential for
improvement by prospective planning. Consequently, fast search
in the existing databases promises good assistance for the product
designers. The main input for this research was the request for reli-
able information about the already successfully used connections
which shall be extracted from historical data.
In order to address the challenges of data mining, the integra-
tion of various planning tasks within the PEP, new concepts are
necessary. Though, as a part of integrated product and process
development there are different definitions for various phases
and aspects of planning activities along the PEP [68]. Regardless
of the specific definition of these phases and aspects; and based
on analysis, it is certain that great amount of their containing infor-
mation and knowledge are either utilized insufficiently and inef-
fectively or remain unused [26]. In this regard, the presented
Fig. 5. Workflow of the design optimization with additional time data.
Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
concept focuses on concurrent product design and production
assembly planning [42]. Product designer and production planer
are thereby required to be more efficiently assisted by various
applications to improve design or planning process through infor-
mation gathered by data mining [55]. As the assembly process con-
sumes a considerable share of production costs, its optimization
seems to be a promising strategy.
In this context, the following requirements for research ques-
tions arise:
1. The assembly planning should start as soon as possible (ideally,
front-loaded within product design).
2. To secure re-use of parts and processes, in particular of joining
elements, planning should be based on past empirical values,
the product assembly information (PAI).
3. Users should be supported by an intelligent assistant which
either pulls the empirical PAI by data mining in the historical
data or adds new PAI based e.g. on recent recommendations.
Because no commercial application is known supporting afore-
mentioned requirements, demonstrators need to be developed,
which in first step, focuses on the selection and design of appropri-
ate assembly connections.
3.1. Preparation and requirements
Fig. 3 illustrates the underlying approach to realize design opti-
mization enriched with fundamental data required from Manufac-
turing. It describes the overall product emergence process, which
combines process relevant attributes from design (bottom left)
and process patterns from manufacturing (bottom right). The inte-
grated product model is thereby enriched to an assembly-oriented
product model, having a good conjunction between these phases.
With this in place, a prospective estimation of the assembly work
and costs in series production can be achieved. Thus, design task
are effectively aligned with time requirements linked to assembly
planning process.
The emergence of Product Assembly Information is shown in
Fig. 4 and runs through the following steps:
Enrich CAD data with assembling information: Derived from
previously similar designed products, assembly information such
as time data about the actual design situation can be identified
and provided in order to support the designer. This additional
information can be used to enrich the CAD data to support the cur-
rent design and be updated in later assembly planning processes.
 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx
Suggesting assembly connections: An assisting option for the
designer is a suggestion list of similar previously designed assem-
bly connection variations. These lists give a quick overview of pos-
sible and already implemented connection types in the assembly.
Assembly process estimation: The focus is on the creation of
an assembly process for a new product. Based on existing product
and process data compilation of a first approximated assembly
process for a new product could be developed. From this, the pro-
duction planner can specify further details and thus determine a
first estimation regarding assembly time. Based on the assembly
time and associated calculation scheme, the planner can perform
the first cost estimation in a very early planning phase.
The information in production planning and design processes
including historical insights mutually enrich each other. Addition-
ally, intelligent interconnecting information from both areas cre-
ates added value. The newly obtained information supports the
workflow throughout the PEP. Therefore, as part of this concept
some requirements need to be met. Thus, the pre-conditions
attached to both systems as well as their respective processes have
to be fulfilled [56].
3.2. Attributes and data sources
Data Mining methods can be used for data clustering and clas-
sification, however criteria for comparison of data sets have to be
identified [30]. To determine the criteria for comparison, within
the scope of this project, a survey of users as well as an analysis
of various tools of the DM was performed. The objective of this
analysis was to identify attributes relevant for assembly processes
that could be assigned to products and parts in CAD [31], PDM and
production planning systems. In CAD systems, attributes assigned
to parts contain mainly geometric information including meta
information such as volume and weight. The PDM systems contain
organizational information, such as creator, version and revision as
well as the mentioned parts information form CAD [24]. In addition
Fig. 6. Assembly time estimation using PAI.
Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
7
to the conventional systems for design and stacking product parts
and assemblies, systems for process planning and time measure-
ment were also taken into account. They sustain a comprehensive
portfolio of information and therefore can be used to distinguish
different product parts and assemblies. The results of this analysis
are capsulated as an object oriented data model, further described
in the data model section.
The necessary enrichment of product and process data on the
fly (without holding the workflow) for the presented concept
requires additional efforts in design. This additional expenditure
also relates to assembly connections and includes the acquisition
of new information form the designers know how. The designer
usually defines assembly connections either implicitly through
formed locked joints by the shaping of the parts or explicitly by
connecting elements such as in screwed fasteners.
3.3. Data collection and availability
The designer of the assembly connection considers all these
information in the design but cannot store them in the CAD model
because the CAD tools for the most part are not able to define the
necessary attributes.
To overcome this problem as part of the concept presented in
this paper, the designer will be provided with an additional tool
in the CAD system. It can be used to create assembly connections
and gives additional information (PAI) and explicit design possibil-
ities. Thus, data will be collected in the source system, the CAD sys-
tem in particular. Since the defined objects are not part of PDM
systems, an extension is necessary in order to implement connec-
tions as objects and to store them after the transfer in the PDM sys-
tem persistently. In the further processing, the product
information will be linked to the planning processes. Unless the
storage of product data are in the same system as for production
planning, the information flow from the PDM system to the plan-
8 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx
Fig. 7. Data model overview.
ning system as well as the DM tool, for further analysis, has to be
ensured.
In current planning systems it is often possible to directly link
processes to products [49]. Thus, an allocation of to be assembled
product and the associated assembly processes can be realized.
In the assembly, however, parts are joined with other parts or
products. These assembly connections with their additional infor-
mation have no digital equivalent object yet. However by means
of an object such as the product assembly information, it is pos-
sible to store useful additional connection information, which
relates directly to the respective assembly connection. As part of
this concept, the combination of the products and processes does
not take place directly but through this object product assembly
information. The linking of product and process does not neces-
sarily need to occur at the part level.
4. Proposed concept
The concept presented in this paper describes an assisting
workflow to support the designer (Fig. 5). As part of a new or mod-
ified design, the designer creates new product data. A software
assistant supports the designer in creating assembly connections
and enriches the CAD model with product assembly information
for each connection. This product assembly information includes
additional connection information including e.g. the torque
screwed fasteners or the type and the form of a welded joint and
information about other connection types. In the ongoing design
process, the designer can trigger an evaluation regarding the
assembly connections in the model.
For this purpose, the characteristics of the CAD model are first
prepared and analyzed with Data Mining. The analysis focuses on
the product assembly information and their properties. The parts
associated with the product assembly information and their geo-
metric properties are also included in the analysis as additional
information set. Furthermore, an extended database is also pro-
vided and consists of product and process data of existing prod-
ucts, which are linked via the product assembly information. The
characteristics of the product assembly information of the new
product are compared with the properties of the product assembly
information of the existing product in the extended database. Then,
the most similar product assembly information is determined from
the existing products. This analysis can be restricted by a class of
the connection types (screwed, weld, rivet) or deliberately left
open to widen the solution space and to provide the designer with
information about other assembly connections.
A limitation on the particular type of connection yields as a
result of the closest realized assembly connection of the same kind.
Depending on the properties of the parts, other mounting connec-
tions can also be found and proposed to the designer in a proposal
list.
The presented application for the support of the design process
uses the product assembly information identified in the analysis of
Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
the PDM database to determine the respective associated and
related sub processes (Fig. 6) where eBOM and mBOM describe
the engineering and manufacturing BOM respectively. Each pro-
duct assembly information record represents an assembly connec-
tion. By multiple connections within the assemblies, multiple sub-
processes for the assembly can be determined. These processes
contain the time data relevant for the new product design. Corre-
sponding time information and, if requested, along with an alter-
native proposal list is transferred and displayed in the CAD
System. This assembly time information of the existing product
represents a first approximated assembly time for the new prod-
uct. So the designer is provided with this additional information
regarding the assembly time and with an enterprise specific factor
the corresponding cost of the current design solution. In the final
step, the designer is able to optimize the product iteratively on
the basis of anticipated assembly time and costs for each design.
4.1. Data model
Based on determined assembly characteristics, a range of attri-
butes is derived to classify the assembly of the parts. Fig. 7 shows
an excerpt of the generated data model for the data mining
analysis.
The ProductAssemblyInformation (PAI) is the central element in
this data scheme and represents the assembly of the product parts.
References for time analysis, assembly requirements, designed
parts or products as well as a wide range of meta data including
the assembly department and other information are lodged. This
element is supplemented with attributes of different connection
types (Fig. 8).
The second fundamental object in the data model is the Item. It
contains references to existing sub-assembly units, geometrical
characteristics as well as ProductAssemblyInformation. Each Item
refers to the ProductAssemblyInformation, which also refers to fur-
ther used Items. This construct is chosen to enable Data Mining
methods to determine exact similarities between new parts and/
or products and other existing parts. Furthermore, it makes com-
parison parts and products the new and existing ones, in any order
and combination interchangeably possible.
In the first approach the attributes for screw connections are
clustered and evaluated regarding the relevance for assembly oper-
ation. The identified attributes are classified in the categories fas-
teners, installation/assembly situation, tools/equipment,
installation regulations and additional assembly elements. These
attributes are represented in the data model in different object cat-
egories (here illustrated by color1, c.f. Fig. 8). The evaluation regard-
ing the influence on the assembly time provides a first indication for
the relevance in the data mining analysis. Production planner
weighted the attributes with low medium and high in order to deter-
1
For interpretation of color in Fig. 8, the reader is referred to the web version of
this article.
 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx
Fig. 8. Structure of Product Assembly Informations (PAI) for weighting.
mine a first selection of attributes relevant for the data mining.
Which attributes really are significant for the similarity of assembly
connection have to be determined in a data mining analysis with a
large quantity of product data. The columns design and work plan-
ning shall indicate in which department the attributes can be col-
lected either by IT system (column System) or experienced
users (column Know-How).
Not all of these attributes can be identified in the mechanical
design in the CAD system. Some can be determined in production
planning workshops in order to optimize the current design. Expe-
rienced designers and production planner can pre-allocate some
parameters with estimated values, which can be reviewed later.
Other parameter and the corresponding value data can be
extracted out of other systems e.g. the attributes of standard parts.
Supplementary connection types can be added to the data
model. To provide required information for time analysis, a stan-
dardized data model is applied. In this regard, ADiFa projects
Application-specific data models, so called ADiFa Application
Protocols, were used. It offer the integration of processes and data
for different DM systems [49].
The application data model of the ADiFa Application Protocol is
specified in an UML class diagram illustrated roughly in Fig. 9. It
reflects additional requirements of the user, needed to be defined
in the application process model.
The core objects in the application data model are the operation
definition, the time module definition and the time module occur-
rence. In Fig. 9 they are highlighted in blue. The operation object
includes hierarchical structure of processes and activities with
the definition of successor and child operations. The object time
module definition is designed for the encapsulation of all time
management data. It is a general definition of every describing
attribute and time value of an operation. The time module defini-
tion can contain numerous time values, each with the classification
Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
9
time type and unit. The application model supports the distinction
of simple time values or time ranges.
The time module occurrence represents the specific incident of
an operation. For example if in an assembly operation several
screws have to be tightened, in the time module definition there
are all time types with value and units for the single operation of
one screw and in the time module occurrence is the factor for
the number of screws used in this specific process.
Additionally the object operation definition includes a reference
container. This object connects the operation to any resources and
parts used in the operation. With this structure the application
model represents the important triangle of product, process
and resource. This application model is then translated in the final
data exchange format in XML. For this mapping the static and
dynamic data exchange has to be considered.
The implementation concept of the ADiFa protocol provides
capabilities for static and dynamic data exchange (see Fig. 10).
The static data exchange, which is in fact file-based, occurs in form
of XML files, which are compiled and read from the interface pro-
cessors. The dynamic data exchange transmits the same data
though an SOA (Service-oriented Architecture) concept by using
web service technologies. In those web services the requests and
the corresponding data from the application interpreted model
are defined in order to be exchanged between the programs in
engineering and manufacturing domains. Applications involved
thereby will act both as a server to provide data to another
requesting application as well as a client to retrieve data from
other systems.
4.2. Data mapping and data mining
After aggregating and appending data subsets from different
sources and systems, it is necessary to remove redundant ones
10 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx
Fig. 9. Key objects of the ADiFa Application protocol.
[48] for the DM process. The next step is converting and porting
data in the presented data model. Depending on data source the
conversion is either fully automated or partially automated with
further manual adjustment. Often, value and scale of different
attributes are heterogeneous. In these cases, a normalization of rat-
ings prevents the undesired high or low impact of certain attri-
butes on the results and evaluation process. In this regard, a [0,
1] linear normalization has been used. Additionally, a further attri-
bute prioritizing via weighting can be necessary to define the
importance of each attribute for the evaluation [58]. An automated
learning of the weights via machine learning methods depends on
the existing data sets and their quality. Otherwise, they are deter-
mined based on expert knowledge or a combination of both meth-
ods. To prevent further expansion of scope and the complexity of
existing problem, expert knowledge was applied to determine
the attribute weights. This support was provided by an expert from
the vendor of the DM tool. Furthermore, it is possible to have more
than a single weight vector. This approach is useful, if there are
various object types or parts, which have different prioritization
for their attributes [73]. To identify the objects with most similar
product assembly information for a new object, the classification
algorithm k-nearest neighbour (kNN) [21] with Euclidean distance
as evaluation function is used. From the identified objects, a list is
generated and the most related one can be manually chosen, which
passes its assembly process data to new object. To assure the reli-
ability of the presented method and prevent over fitting, a cross
validation [34] is used.
Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
The implementation of the presented approach is challenging
due to high requirement for interconnection and overall quality
of the existing data in different source system. In particular the
pure number of realized and existing assembly connections and,
thus, necessary instances of a product assembly information as
well as the quality of the data regarding their attributes are
important.
At this point of time, the fulfillment of these high requirements
has to be verified. In particular, the quality of the linkage of pro-
duct data with the corresponding assembly processes poses a chal-
lenge. The selection of the properties and attributes for the DM
analysis also has to be determined based on production data to
ensure the fidelity of generated results (Fig. 11). In this scope a spe-
cial focus is on characteristics of parts and of the connection itself.
In conformity with the presented objectives and concept, an appli-
cation of the methodology is described as follows.
Suggesting assembly connections and enrichment of CAD data:
The designer with expertise in assembly process creates a new
module with already known and new assembly connections in
the CAD system. He designs assembly connections and comple-
ments these with properties in context of a new module. Via an
automated DM process, he is provided with various data about
assembly connections. Moreover, for each assembly connection a
list of alternative or ever realized connections can be created.
Depending on the product properties, the five most similar product
assembly configurations are made available to the designer as a
prepared proposal list, which is generated through cluster analysis
 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx 11

Fig. 10. Process for implementation of the ADiFa Application protocol.

Fig. 11. Formation of product clusters and process agglomerations.

of existing product data. These configurations can be used directly,
later or in an extended context of the product thus enhancing the
CAD model. If the analysis is dispensed with filtering of associated
connections of the product assembly information, the designer can
also be provided with other not associated connections as
alternatives.
Estimation of assembly process and information: The produc-
tion planner drafts an initial assembly process for a new assembly
at an early stage of product development. Similar to the use case of
the designer, for known assembly connections that are imple-
mented in the new product as well as in the old product data,
the right product assembly information and, thus, the assembly
processes are found. For new unknown connections, the most sim-
ilar product assembly information and related assembly processes
from the database are determined and duplicated. Each of the
founded product assembly information represents a single connec-
tion and the linked process represents precisely the assembly work
instructions for this connection. The sum of the individual connec-
tions for the new product is its first assembly process. Thereby, an
initial draft of an assembly process of the new module can be gen-
erated. Founded individual connections, individual process, as well
as the overall process can be used in different ways to assist

Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
12 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx
Fig. 12. Assessment schema for validation.
designer and production planner. Planner and designer addition-
ally get a first estimation for expected assembly time and costs
in the automated process. In addition, the production planner
can increase the quality of the process by manual intervention.
On the one hand, he adapts the product assembly information cre-
ated by the designer before the DM analysis. On the other hand, he
can complete the product assembly information with practical
knowledge. At this, he has an impact on the input of the DM anal-
ysis and increases the quality of the result thereby. Furthermore,
the designer has a first draft for the assembly process at ones dis-
posal and a first estimated assembly time in the current CAD sys-
tem. By a company-specific factor, the designer receives also
information about the cost of the connection in the assembly. By
verifying this information, the designer can evaluate and compare
the alternatives for different connections.
5. Use case evaluation
The general assessment schema is based on ISO/IEC 9126 for
software quality as shown in Fig. 12. The validation criteria are
subdivided in 4 weighted categories (methodological, organiza-
tional, technical and data validation), which comprise a plethora
of singular criteria (e.g. precision). Each evaluation assigns one of
3 performance levels in 2 stages. The total sum yields the overall
evaluation score for dedicated use case.
5.1. Building of product clusters
Currently, several individual Use Cases are considered for vali-
dation purposes. First, the formation of the product cluster was
Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
considered. To estimate a suitable number for the product cluster
structural data from part lists of assemblies were used. Then, the
attributes such as size, material and weight which were shown
previously were used in the first approach. In this case, the calcu-
lation of the component dimensions of the bounding box proved to
be sufficiently accurate. This way, 5 product clusters with respec-
tive reference to the component category could be formed. With
regard to the associated assembly processes, a first result was that
components with similar design characteristics also need similar
assembly processes. This seemingly trivial statement however val-
idates an important requirement for the applicability of existing
assembly processes for current products to future assembly pro-
cesses for new-developed products.
5.2. Building of process clusters
The clustering for the processes is based on time analysis in
Methods Time Measurement (MTM). Basic principle is the determi-
nation of target times by combining the time measurement units.
Depending on the containing time blocks, 7-process cluster with
explicit reference to the component category were formed. Addi-
tional accuracy of clustering can be achieved with different simi-
larity searches within the process data. The different similarity
searches are used:
1. Similarity search via process parameter: Single time blocks such
as pick and place are composed of individual movements.
Each single time block has attributes like distance to pick.
These attributes are used in the Data Mining method K-Means
to divide similar assembly processes into clusters.
 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx
Fig. 13. Similarity search for process data.
2. Similarity search via description text: The similarity of descrip-
tion texted is evaluated based on the text mining. In particular,
key words as switch panel base or steering obtain a high
weighting.
3. Similarity search via sequences: Structural characteristics of
individual assembly processes are considered. In addition, the
question of how many same sequences of individual time
blocks are used in a parent sequence is analyzed. The more
identical or similar sequences of time blocks, the more similar
the considered assembly processes.
Fig. 13 presents the automated similarity search with the tool
Data Miner. In different process blocks, data are read, processed
and analyzed with different similarity searches.
6. Conclusion and outlook
Through utilization of Data Mining tools, the efficient design of
assembly connection, the quality of planning results and planning
processes can be increased, while time and cost reduction can
simultaneously be achieved. With this approach working sched-
ules as planning results are based on field-tested assembly pro-
cesses and contain the implicit knowledge used in similar
assembly planning processes. The automatic generation of an
adapted assembly process enables fast customization to a concrete
setting at the shop floor. In this regard, the presented approach
contributes an important added value to production design and
planning through usage of knowledge in existing systems. Thus,
reduction of planning time, increasing availability of information
in product design as well as making the cooperation between the
designer and product planning teams easier are the consequences.
The feasibility of this concept with productive data is in evaluation.
New approaches with clustered data to improve data quality are in
Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
13
an assessment. However, to produce reliable outcomes, the pro-
duct data have to fulfil high requirements in regard to connection
elements. Concurrently, the necessary data model and some tool
sets are provided to make the data integration easier. In the future,
further development of tool sets and methods could help to reduce
the high initial effort for adjustment of the data even more. Besides
the evaluation of the results based on product data, it is important
to investigate the behavior and results of the methodology for new
and innovative assembly technologies. Furthermore, for analyzing
more complex data sets as well as obtaining better results, it is
important to develop and refine the concept and to apply further
Data Mining methods.
Acknowledgement
The research project Prospective Determination of Assembly
Work Content in Digital Manufacturing (ProMondi) is supported
by the German Federal Ministry of Education and Research (BMBF)
within the Framework Concept Research for Tomorrows Produc-
tion (funding number 02PJ1110) and managed by the Project
Management Agency Karlsruhe (PTKA). The authors are responsi-
ble for the contents of this publication.
References
[1] S. Alguezaui, R. Filieri, A knowledge-based view of the extending enterprise for
enhancing a collaborative innovation advantage, Int. J. Agile Syst. Manage. 7 (2)
(2014) 116131.
[2] H. Al-Mubaid, E.S. Abouel Nasr, A.K. Kamrani, Using data mining in the
manufacturing systems for CAD model analysis and classification, Int. J. Agile
Syst. Manage. 3 (1/2) (2008) 147162.
[3] A. Armillotta, G. Moroni, M. Rasella, Computer-aided assembly planning for the
diemaking industry, Robot. Comput.-Integr. Manuf. 22 (2006) 409419.
[5] H. Bley, C. Franke, Integration of product design and assembly planning in the
digital factory, Ann. CIRP 53 (1) (2004) 2530.
14 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx
[6] G. Boothroyd, Assembly Automation and Product Design, third ed., Taylor &
Francis Group, Boca Raton, 2011.
[7] G. Boothroyd, W. Knight, P. Dewhurst, Product Design for Manufacture and
Assembly, second ed., Marcel Dekker, New York, 1994.
[8] G. Bornet dit Vorgeat, P. Pu, R. Clavel, A. Csabai, F. Sprumont, P. Xiroochakis, M.-
T. Ivorra, MicroCE: Computer-Aided Support for DFMA Conceptual Design
Phase, CiteseerX: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.
16.1358, Accessed: Sep 15, 2015.
[9] M. Bossmann, Feature-basierte Produkt- und Prozessmodelle in der
integrierten Produktentstehung (Ph.D. thesis), University of Saarland, 2007.
[10] U. Bracht, T. Masurat, The Digital Factory between vision and reality, Comput.
Ind. 56 (4) (2005) 325333.
[11] A. Bryan, J. Ko, S.J. Hu, Y. Koren, Co-evolution of product families and assembly
systems, Ann. CIRP 56/1/2007 (2007) 4144.
[12] R.-S. Chen, K.-Y. Lu, P.-H. Tai, Optimizing assembly planning through a three-
stage integrated approach, Int. J. Prod. Econ. 88 (2004) 243256.
[14] W.-C. Chen, Y.-Y. Hsu, L.-F. Hsieh, P.-H. Tai, A systematic optimization
approach for assembly sequence planning using Taguchi method, DOE, and
BPNN, Expert Syst. Appl. 37 (2010) 716726.
[15] K. Choi, K.-Y. Kim, H.-J. Yang, Disparate attributes algorithm for semantic
assembly design rule management, Adv. Eng. Inform. 27 (2013) 5165.
[17] W.W. Chu, Data Mining and Knowledge Discovery for Big Data, Methodologies,
Challenge and Opportunities, Springer-Verlag, Berlin, Heidelberg, 2014.
[18] A. Coralo, A. Margherita, G. Pascali, Digital mock-up to optimize the assembly
of a ship fuel system, J. Modell. Simula. Syst. 1 (2010) 412.
[19] F. Demoly, X.-T. Yan, B. Eynard, L. Rivest, S. Gomes, An assembly oriented
design framework for product structure engineering and assembly sequence
planning, Robot. Comput.-Integr. Manuf. 27 (2011) 3346.
[20] F. Demoly, A. Matsokis, D. Kirisits, A mereotopological product relationship
description approach for assembly oriented design, Robot. Comput.-Integr.
Manuf. 28 (2012) 681693.
[21] S. Dhanabal, S. Chandramathi, Review of various k-nearest neighbor query
processing techniques, Int. J. Comput. Appl. 31 (7) (2011) 1422.
[22] K.L. Edwards, Towards more strategic product design for manufacture and
assembly: priorities for concurrent engineering, Mater. Des. 23 (2002) 651
656.
[24] M. Eigner, R. Stelzer, Product Lifecycle Management Ein Leitfaden Fr
Product Development Und Life Cycle Management, Springer, Berlin,
Heidelberg, 2009.
[25] T.-H. Eng, Z.K. Ling, W. Olson, C. Mclean, Feature-based assembly modeling
and sequence generation, Comput. Ind. Eng. 3 (1999) 1733.
[26] O. Erohin, P. Kuhlang, J. Schallow, J. Deuse, Intelligent utilisation of digital
databases for assembly time determination in early phases of product
emergence, Procedia CIRP 45th CIRP Conference on Manufacturing
Systems vol. 3 (2012) 424429.
[27] L. Graening, B. Sendhoff, Shape mining: a holistic data mining approach for
engineering design, Adv. Eng. Inform. 28 (2014) 166185.
[28] S. Gupta, Computer-aided generation of modularized conceptual designs with
assembly and variety considerations (M.Sc. thesis), The Pennsylvania State
University, 2008.
[29] S. Gupta, G.E. Okudan, Computer-aided generation of modularized conceptual
designs with assembly and variety considerations, J. Eng. Des. 19 (6) (2009)
533551.
[30] J. Han, M. Kamber, J. Pei, Data Mining: Concepts and Techniques, third ed.,
Morgan Kaufmann Publishers, Waltham, 2012.
[31] J. Hartung, J. Schallow, S. Rulhoff, Moderne Produktionsplanung Integration
in der Produktentstehung, ProduktDaten J. 19 (1) (2012) 2021.
[32] T.C.-K. Huang, C.-C. Liu, D.-C. Chang, An empirical investigation of factors
influencing the adoption of data mining tools, Int. J. Inf. Manage. 32 (2012)
257270.
[33] P. Kim, Y. Ding, Optimal engineering system design guided by data-mining
methods, Technometrics 47 (3) (2005) 336348, August.
[34] R. Kohavi, A study of cross-validation and bootstrap for accuracy estimation
and model selection, Proceedings of the 14th International Joint Conference on
Artificial intelligence, (IJCAI95), vol. 2, Morgan Kaufmann Publishers Inc., San
Francisco, 1995, pp. 11371143.
[36] F.Z. Krumenauer, C.T. Matayoshi, I.B. da Silva, M.S. Filho, G.F. Batalha,
Concurrent engineering and DFMA approaches on the development of
automotive panels and doors, J. Achievements Mater. Manuf. Eng. 31 (2)
(2008) 690698.
[37] T.C. Kuo, S.H. Huang, H.C. Zhang, Design for manufacture and design for X:
concepts, applications, and perspectives, Comput. Ind. Eng. 41 (2001) 241
260.
[38] A.J.D. Lambert, Optimal disassembly sequence generation for combined
material recycling and part reuse, in: Proceedings of the 1999 IEEE
International Symposium on Assembly and Task Planning, 1999, pp. 146151.
[39] M.C. Leu, H.A. ElMaraghy, A.Y.C. Nee, S.K. Ong, M. Lanzetta, M. Putz, W. Zhu, A.
Bernard, CAD model based virtual assembly simulation, planning and training,
CIRP Ann. Manuf. Technol. 62 (2013) 799822.
[40] Y.-B. Li, G.-L. Chen, X.-M. Lai, S. Jin, Y.-F. Xing, Knowledge-based vehicle body
conceptual assembly design, Proc. I Mech E Part D J. Automob. Eng. 222 (2)
(2008) 221234.
[41] B. Lotter, H.-P. Wiendahl, Montage in Der Industriellen Produktion, Ein
Handbuch Fr Die Praxis, 2. Auflage, Springer-Verlag, Berlin-Heidelberg, 2013.
Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006
[42] E.B. Magrab, S.K. Gupta, F.P. McCluskey, P.A. Sandborn, Integrated Product and
Process Design and Development: The Product Realization Process, second ed.,
Taylor & Francis Group, Boca Raton, 2010.
[43] A. McLay, Re-reengineering the dream: agility as competitive adaptability, Int.
J. Agile Syst. Manage. 7 (2) (2014) 101115.
[44] Miele, <https://www.miele.com/de/com/philosophie-2095.htm>, Accessed
May 16th, 2016.
[45] E.Z. Namouz, J. Summers, Complexity connectivity metrics predicting
assembly times with low fidelity assembly CAD models, in: M. Abramovici,
R. Stark (Eds.), Smart Product Engineering, Springer-Verlag, Berlin-Heidelberg,
2013, pp. 777786.
[46] D. Nettleton, Commercial Data Mining: Processing, Analysis and Modeling for
Predictive Analytics Projects, Morgan Kaufmann, Waltham, 2014.
[47] B.A. Nicholds, J. Mo, S. Bridger, Determining an action plan for manufacturing
system improvement: a case study, Int. J. Agile Syst. Manage. 7 (1) (2014) 1
25.
[48] L. Ohno-Machado, H.S. Fraser, A. hrn, Improving machine learning
performance by removing redundant cases in medical data sets, Proc. AMIA
Fall Symposium (1998) 523527.
[49] D. Petzelt, J. Schallow, J. Deuse, S. Rulhoff, Anwendungsspezifische
Datenmodelle in der Digitalen Fabrik, ProduktDaten J. 16 (1) (2009) 4548.
[50] J. Ponn, U. Lindemann, Konzeptentwicklung und Gestaltung technischer
Produkte Systematisch von Anforderungen zu Konzepten und
Gestaltlsungen, Springer, Berlin, Heidelberg, 2011.
[51] M. Putz, A. Richter, M. Pfeifer, Adaptive planning and optimization of joining
and assembling sequences using parallel acting working units, CIRP Ann.-
Manuf. Technol. 59 (2010) 5760.
[52] R.V. Rao, Advanced Modeling and Optimization of Manufacturing Processes.
International Research and Development, Springer-Verlag, London, 2011.
[53] B. Rekiek, A. Delchambre, Assembly Line Design The Balancing of Mixed-
Model Hybrid Assembly Lines with Genetic Algorithms, Springer-Verlag,
London, 2006.
[54] G. Rock, K. Theis, P. Wischnewski, Variability management, in: J. Stjepandic
et al. (Eds.), Concurrent Engineering in the 21st Century: Foundations,
Developments and Challenges, Springer International Publishing,
Switzerland, 2015, pp. 491519.
[55] S. Rulhoff, R. Jalali Sousanabady, J. Deuse, C. Emmer, Concept and data model
for assembly work content determination, in: J. Zh (Ed.), Enabling
Manufacturing Competitiveness and Economic Sustainability Proceedings of
5th CIRP Conference on Changeable, Agile, Reconfigurable and Virtual
Production (CARV2013), Springer, Mnchen, Germany, Berlin, Heidelberg,
New York, 2012, pp. 353360.
[56] J. Schallow, K. Magenheimer, J. Deuse, G. Reinhart, Application protocols for
standardising of processes and data in digital manufacturing, in: H.A.
ElMaraghy (Ed.), Enabling Manufacturing Competitiveness and Economic
Sustainability Proceedings of 4th CIRP Conference on Changeable, Agile,
Reconfigurable and Virtual Production (CARV2011), Springer, Montreal,
Canada, Berlin, Heidelberg, New York, 2011, pp. 648653.
[57] Robert B. Stone, Daniel A. McAdams, Varghese J. Kayyalethekkel, A product
architecture-based conceptual DFA technique, Des. Stud. 25 (2004) 301325.
[58] B. Strug, G. Slusarczyk, Reasoning about designs through frequent patterns
mining, Adv. Eng. Inform. 23 (2009) 361369.
[59] D. Talia, P. Trunfio, Service-Oriented Distributed Knowledge Discovery, CRC
Press, Boca Raton, 2013.
[60] C.F. Tan, L.S. Wahidin, S.N. Khalil, An architecture framework of design for
assemble expert system, in: Proc. of the Intl. Conf. on Advances, in: Civil,
Structural And Mechanical Engineering CSM, 2014.
[61] L. Trygg, Concurrent engineering practices in selected swedish companies: a
movement or an activity on the few?, J Prod. Innov. Manage 10 (1993) 403
415.
[62] M. Vielhaber, Design to knowledge - a root design principle, in: P. Scharff, P.
Kurtz (Eds.), Proceedings of the 56th International Scientific Colloquium
(IWK), TU Ilmenau, Ilmenau, 2011.
[65] S. Wartzack, Predictive Engineering Assistenzsystem zur multikriteriellen
Analyse alternativer Produktkonzepte, Forschritt-Berichte VDI, Reihe 1, Band
336, Dsseldorf, VDI-Verlag, 2001.
[66] C. Weber, CPM/PDD an extended theoretical approach to modelling products
and product development processes, in: H. Bley, et al. (eds.), Proceedings of the
2nd German-Israeli Symposium on Advances in Methods and Systems for
Development of Products and Processes, Stuttgart, Fraunhofer-IRB-Verlag,
2005.
[67] C. Weber, Produkte und Produktentwicklungsprozesse mit Hilfe von
Merkmalen und Eigenschaften eine kritische Zwischenbilanz, in: D. Krause,
K. Paetzold, S. Wartzack (eds.), Beitrge zum 23, Symposium DfX, Hamburg,
TuTech, 2012.
[68] D.E. Whitney, Mechanical Assemblies: Their Design, Manufacture, and Role in
Product Development, Oxford University Press, Oxford New York, 2004.
[69] X. Xu, L. Wang, S.T. Newman, Computer-aided process planningA critical
review of recent developments and future trends, Int. J. Comput. Integr. Manuf.
24 (1) (2011) 131.
[70] Y. Yin, I. Kaku, J. Tang, J.M. Zhu, Data Mining: Concepts, Methods and
Applications in Management and Engineering Design, Springer-Verlag,
London, 2011.
[71] Y. Yusof, K. Latif, Survey on computer-aided process planning, Int. J. Adv.
Manuf. Technol. 75 (2014) 7789.
 R. Kretschmer et al. / Advanced Engineering Informatics xxx (2017) xxxxxx 15

[72] Y.Y. Yusuf, M. Sarhadi, A. Gunasekaran, Agile manufacturing: the drivers,
concepts and attributes, Int. J. Prod. Econ. 62 (1999) 3343.
[73] D. Zhang, P.L. Yu, P.Z. Wang, State-dependent weights in multicriteria value [23]
functions, J. Optim. Theory Appl. 74 (1) (1992) 121.
[35]
Further reading
[4] R.C. Beckett, Functional system maps as boundary objects in complex system [63]
development, Int. J. Agile Syst. Manage. 8 (1) (2015) 5369.
[13] W.-C. Chen, P.-H. Tai, W.-J. Deng, L.-F. Hsieh, A three-stage integrated
approach for assembly sequence planning using neural networks, Expert Syst.
Appl. 34 (2008) 17771786. [64]
[16] M.-C. Chiu, G. Okudan, Evolution of design for X tools applicable to design
stages: a literature review, in: Proceedings of the ASME 2010 International
Design Engineering Technical Conferences & Computers and Information in
Engineering Conference IDETC/CIE 2010, August 1518, 2010, Montreal,
Quebec, Canada.
K. Ehrlenspiel, Integrierte Produktentwicklung Denkablufe,
Methodeneinsatz, Zusammenarbeit, Hanser, Mnchen, Wien, 2009.
R. Kretschmer, S. Rulhoff, J. Stjepandic, Design for assembly in series
production by using data mining methods, in: J. Cha et al. (Eds.), Moving
Integated Product Development to Service Clods in Global Economy, IOS
Press, Amsterdam, 2014, pp. 617626.
R. Wallis, J. Stjepandic, S. Rulhoff, F. Stromberger, J. Deuse, Intelligent
utilization of digital manufacturing data in modern product emergence
processes, in: J. Cha et al. (Eds.), Moving Integated Product Development to
Service Clods in Global Economy, IOS Press, Amsterdam, 2014, pp. 617626.
H. Wang, Y. Rong, D. Xiang, Mechanical assembly planning using ant colony
optimization, Comput. Aided Des. (2014).

Please cite this article in press as: R. Kretschmer et al., Knowledge-based design for assembly in agile manufacturing by using Data Mining methods, Adv.
Eng. Informat. (2017), http://dx.doi.org/10.1016/j.aei.2016.12.006