Available online at www.sciencedirect.

com

Procedia CIRP 7 (2013) 431 – 436

Forty Sixth CIRP Conference on Manufacturing Systems 2013

3D Design Support for Rapid Virtual Prototyping of Manufacturing

Systems
István Németha,*, János Püspökia, Csaba Haraszkóa, Gyula Mátyásia, Tibor Nagya,
Christopher Freemanb, Robin W. Scottb, James S. Baldwinb
a
Department of Manufacturing Science and Technology, Budapest University of Technology and Economics, Egry J. u. 1, Budapest 1111, Hungary
b
Advanced Manufacturing Research Centre with Boeing, University of Sheffield, Advanced Manufacturing Park, Wallis Way, Catcliffe, Rotherham
S60 5TZ, United Kingdom
* Corresponding author. Tel.: +36-1-463-2512; fax: +36-1-463-3176; E-mail address: [email protected].

Abstract

The paper presents several 3D layout configurator software tools and a library of manufacturing system components to rapidly
build up the 3D models of manufacturing systems; some analysis tools to analyze the created 3D models; and some methods to
rapidly create 3D models from point cloud of scanned manufacturing facilities. Within the configurators the user can define the
following data as input: production process (e.g. type and order of operations), type of production equipment, type of material
handling systems (e.g. robots, conveyors), type and sub-type of layout, and some spatial constraints (e.g. distance between
machines, spatial arrangement). The configurators are able to create the 3D model of the manufacturing systems based on the
selection of an appropriate layout template. The layout templates are defined by the species identified with a cladistic classification
system or by the sub-species where each sub-species inherits the characteristics of its species. Based on the selected layout template
and the defined production process the appropriate elements of the manufacturing system can be selected by the user from a
predefined component library. Such library includes the 3D models and technical data of workstations, machine tools, material
handling equipment, etc. Two analysis tools have been developed to evaluate the created 3D models: a collision elimination tool
and a robot programming tool. The paper also presents two solutions to rapidly create 3D models from point cloud of scanned
manufacturing facilities; as well as an application to automatically create a simplified quasi-volumetric point cloud model that can
be used for collision detection.

2013The
© 2013 TheAuthors.
Authors.Published
Published
by by Elsevier
Elsevier B.V.B.V.
Selectionand
Selection and/or peer-review
peer-review underunder responsibility
responsibility of Professor
of Professor Pedro
Pedro Filipe doFilipe
CarmodoCunha
Carmo Cunha

Keywords: manufacturing systems; cladistic classification; layout design; layout planning; layout configurators; point cloud

– similarly to the term ‘design’ used in design science

1. Introduction [2] – the synthesis, analysis and optimization of the
selection and arrangement of manufacturing system
The European joint research project COPERNICO equipment mainly focusing on the layout of machine
was initiated with the aim of developing methodologies, tools, workstations and material handling in 3D
tools and a cooperation environment for rapid design and environments.
prototyping of manufacturing facilities [1]. This paper
presents the results of the project achieved in the 2. Cladistic classification of manufacturing systems –
development of 3D design support tools for rapid virtual layout species and layout templates
prototyping of manufacturing systems. In particular,
these results are configurator software tools, A cladistics based classification of manufacturing
manufacturing system component databases, component systems has been developed within the COPERNICO
selection tools, and techniques to rapidly create 3D project. Cladistics is an evolutionary classification
models from scanned point cloud data. In the context of scheme that describes the attributes of existing entities as
this paper the term ‘layout design’, as an activity, means well as their ancestral characteristics. Cladistic

2212-8271 © 2013 The Authors. Published by Elsevier B.V.

Selection and peer-review under responsibility of Professor Pedro Filipe do Carmo Cunha
doi:10.1016/j.procir.2013.06.011
432 István Németh et al. / Procedia CIRP 7 (2013) 431 – 436

relationships are commonly understood as the

evolutionary branching sequences using taxonomies. It
is assumed that characters can be expressed in discrete
states and arrayed in a presumed evolutionary order from
primitive to derived advanced. The application of
cladistics to manufacturing systems has opened up a new
way to classify and link evolutionary changes to main
manufacturing changes of the past and present. [3]
Using the cladistic approach, the evolutionary
relationships between forty-six candidate species of
manufacturing systems were hypothesized, described Fig. 1. 3D models of conveyor elements
and presented diagrammatically (as it was in the 4th
generation cladogram of the COPERNICO project). The Fig. 1 shows the 3D models of the elements of four
manufacturing species were then organized in a conveyor types.
hierarchical classification using genera, families and For the population of the component library machine
orders under the class of discrete manufacturing. configurators developed in a previous research project
Some of the species and their layout patterns have called NEXT [4] have been used. These machine
been selected as layout templates for rapid systematic configurators are able to create the 3D models of some
layout design. Then further layout sub-species (or sub- manufacturing equipment [5].
types) of each of the selected species have been Most of the populated components have fixed
identified. Configurator software tools have been geometrical model but some of the components have
developed to create the 3D models of several sub-species parametric 3D model. For instance, the 3D models of the
of the following layout species: configured conveyor assemblies can be built up from
Unidirectional robot line; parametric 3D models of the conveyor elements (Fig. 1).
Robot centered FMS;
Rotary indexer table. 4. Component selection

3. Component library Three component selection tools have been developed

that are able to filter the components populated in the
A library has been populated with manufacturing library presented above: one graphical user interface
equipment components that is used by the layout design (GUI) has been developed for filtering and selecting
tools. The components included in the library have two robots, another GUI for conveyors, and another general
sorts of data: 3D models and properties. The properties GUI for filtering and selecting workstations. These
of the components contain the name of the components, selection tools filter the components using their property
the file name of the 3D models and many other technical data. On the GUIs as many filters can be added as the
and non-technical data. Such data are stored in a user wants using comparing rules, such as ‘>’, ‘<’ or ‘=’
Microsoft Access database. The selection of the in AND relation. For example, the user can look for all
components (see Section 4) is based on the properties the press machines where the length and width are
stored in the Access data tables. smaller than the values specified on the GUI.
The following types of components have been These filtering and selection tools can be easily used
populated so far: press machines, milling machines, for simple problems where the user can search for and
turning machines, grinding machines, wire and sinking manually pick the needed components. For more
electro discharge machines, rotary indexer tables and sophisticated and automated selection problems,
stations, conveyor elements, robots, bending machines, ontologies are being developed to aid the selection of
cutting machines, drilling machines, plastic injection suitable resources which meet specific requirements.
machines, and welding workstations.
Fig. 4 below shows some instances of 3D models of 5. 3D layout configurators
press machines. The properties of the press machines are
the following: name of the component, name of the 3D Five 3D configurators have been developed to
model, name of the manufacturer, press force, number of support the rapid creation of the 3D models of
press heads, max. distance between bad and ram, stroke, manufacturing facilities. These software tools have been
ram length, ram width, power, speed, rapid speed, developed under the Visual Components 3DCreate 3D
length, width, height, weight, existence of autofeeder, modeling and simulation system using the C#
type of operation, MTBF, MTTR, and price. programming language and COM interface.
 István Németh et al. / Procedia CIRP 7 (2013) 431 – 436 433

Right sided

Both sided

Fig. 2. GUI of unidirectional robot line layout configurator for the layout sub-type segmented L

Fig. 3 depicts the automatically created 3D model of

5.1. Unidirectional robot line configurator the right curved segmented L unidirectional robot line
layout configured with the settings of the GUI presented
The left side of Fig. 2 presents the configurator GUI in Fig. 3.
developed for the species ‘Unidirectional robot line’.
With the help of this configurator various layout sub- 5.2. Robot centered layout configurator
types can be selected, and then the 3D model can be
created after configuring some layout parameters. With A configurator together with a special GUI has been
this GUI 15 unidirectional robot line layout sub-types (or developed for the layout species ‘Robot centered FMS’.
templates) can be configured: straight line, left or right With the help of this configurator various layout sub-
curved segmented L, left or right curved segmented U, types can be selected, and then the 3D model can be
where all can have left sided, right sided or both sided created after configuring some layout parameters.
robot arrangements. For example, on the GUI presented Several layout sub-types have been defined for the type
in Fig. 2 the left sided segmented L layout sub-type is ‘Robot centered FMS’, such as line, double line,
selected, while the right side of Fig. 2 shows the rectangular, circular, L-form, U-form, and matrix. The
pictogram of the right sided and both sided robot GUI of this configurator enables the selection of the
arrangements as well. robot and stations by calling the GUIs described in
On the GUI of this configurator the user can define Section 4. Fig. 4 shows a 3D model created
the layout sub-type, the number of, the geometric automatically using this configurator. The layout sub-
dimensions of and the distances between the stations, type of this robot centered system is ‘Line’ and the
and he can select robots using the selection tools selected robot is a general portal robot.
presented in Section 4. Having the robots selected the
3D model of the manufacturing system can be created by
pressing the ‘Create’ button on the GUI.

Fig. 3. Example of the 3D model of a right curved segmented L

unidirectional robot line layout Fig. 4. Example of the 3D model of a robot centered FMS layout:
layout sub-type is ‘Line’; 5 press machines served by a portal robot
434 István Németh et al. / Procedia CIRP 7 (2013) 431 – 436

In the specific case presented in Fig. 4 the dimensions

of the portal robot are calculated taking the dimensions
of the selected stations and the distances specified on the
configurator GUI into account. Then the parametric 3D
model of the portal robot is adjusted to the required size.
In other cases, where the components are fixed in size
(i.e. not parametric), with the help of this configurator Fig. 6. Example of basic conveyors layouts: rectangular and parallel
paths
the user can quickly validate if, for example, the
dimension range of the selected robot matches the
5.4. Rotary indexer configurator
dimensions of the selected stations. This is one of the
strengths of the 3D configurators presented in the paper.
The rotary indexer configurator helps the rapid
creation of rotary indexer systems. A GUI has been
5.3. Conveyor configurator
developed where the user can define the workpiece
(mass, dimensions), the number of stations, the working
The conveyor configurator has been developed to
directions (top, bottom, radial), and the type of the
support the fast definition and 3D model creation of
primary and secondary material handling system
conveyor assemblies. Fig. 5 presents the GUI of this
(human, robot, auto-feeder). Based on these
configurator. With this GUI the user can select the type
configuration data the 3D model of the rotary indexer
of the conveyor added in the component library, define
can be automatically created.
the major dimensions (width and height), and he can
build his conveyor assembly from either basic units or
5.5. Factory building configurator
predefined basic layouts. Currently, there are four basic
units available: straight, curve, divert Y, and merging Y.
The factory building configurator has been developed
Moreover, there are eight predefined basic layouts
in order to support the rapid creation of walls around a
available: line, circular, rectangular, parallel paths,
manufacturing system or within a factory. With this
lateral branching, U-form, L-form, and S-form. Each
configurator the area to be surrounded by walls can be
basic unit and basic layout has some driving parameters
directly defined by the user or the configurator can
to be specified. For example, Fig. 5 shows the U-form
automatically recognize the overall area of a
basic layout selected on the GUI, and Fig. 6 shows other
manufacturing system (or subsystem) that has been
two basic layouts and their driving parameters.
created before. With the configurator the places and
sizes of windows and doors can be easily specified and
created as well. The configurator can also create pillars
in the factory, where the user can specify the number of
rows and columns and the cross-section of the pillars.

6. 3D analysis tools

Two analysis tools have been developed to evaluate

and improve the 3D design of the layouts: a collision
elimination tool and a robot programming tool.

6.1. Collision elimination tool

The collision elimination tool has been developed to

check the geometric collisions and support their
elimination. Fig. 7 presents the GUI developed for the
tool. On the GUI the user can select the 3D objects to be
included in the analysis and then the collisions can be
identified (it is a standard feature of Visual
Components). An application has been developed that
attempts to eliminate the collisions by moving the
selected objects along the specified directions using
some embedded methods. These methods can be
configured on the GUI.
Fig. 5. GUI of the conveyor configurator
 István Németh et al. / Procedia CIRP 7 (2013) 431 – 436 435

Fig. 8. Point cloud (raw data); courtesy of Footprint Sheffield, UK

7.1. User assisted creation of CAD models from big

point cloud

Existing software technologies have been used to

create CAD models from point cloud.
In the first method, the point cloud was imported into
Autodesk AutoCAD Architecture as a reference. This
allowed the cloud to be used as a base from which a
Fig. 7. GUI of collision elimination tool CAD mesh could be generated. A series of slices were
taken through the cloud helping to identify the location
6.2. Robot programming tool of key features. Once the key features had been
generated, the cloud was removed leaving the standalone
The main function of the robot programming tool is CAD mesh. The final data was exported as a .dwg and
to help the user evaluate if a selected robot reaches the viewed in stereo through Autodesk Navisworks (Fig. 9).
selected stations to be served or how the robot moves Navisworks was chosen because it enabled both the
around the stations. On the GUI developed for this tool mesh and point cloud to be viewed in stereo
the user can specify the stations and their order to be simultaneously, allowing the viewer to compare the two
reached, the approaching points, the orientation data sets. The resulting mesh can be used for more
methods, the insertion methods, and the interpolation powerful 3D visualization.
modes. Then the robot program can be automatically In the second method the Rapidform XOR 3 software
created. For this some points have been predefined was used which enables the recognition of some features
within the 3D components populated in the system (e.g. plane, cylinder) from point cloud. After recognizing
library. The generated motion programs can be virtually some features the contour of surfaces can be detected by
executed by the robot within Visual Components. the intersection of the recognized features. Then the
During the simulation the robot motions can be visually closed surface models can be created using the well-
evaluated, or eventually, even collision detection can known solid body modeling techniques (e.g. sweeping).
also be used. An error message appears if the robot Fig. 10(c) shows the surface model created within
cannot reach a particular station. In this case the layout Rapidform from the point cloud of a machine.
has to be changed for which the configurators presented
above may provide a valuable support.

7. Creating 3D models from point cloud

Within the COPERNICO project several techniques

have been investigated on data processing of scanned 3D
factory objects. As a case study the point cloud of the
workshop of Footprint Sheffield, one of the industrial
partners of the project, was created by a laser scanning
method. This point cloud contains more than 10 million
colored points (Fig. 8) that should be further processed Fig. 9. CAD mesh model of the roof, building, and floor of a workshop
for efficient 3D visualization and layout design. (machines are removed)
436 István Németh et al. / Procedia CIRP 7 (2013) 431 – 436

manufacturing systems. The 3D modeling techniques

presented in the paper offer much functionality that
current factory design tools provide. The 3D
configurators support the user in the fast configuration
and creation of the 3D models of manufacturing systems
based on well-defined layout templates that are
identified and defined by a cladistics based classification
system. The analysis tools support the user to evaluate
and improve the 3D design of the layouts: they help
detect or eliminate collisions and analyze whether the
Fig. 10. (a) Photo of the original machine; (b) Point cloud of the selected components match each other or fit into a
machine; (c) Surface model of the machine created in Rapidform
system arrangement. The populated factory components
library serves as a collection of essential building blocks;
The creation of the CAD models presented above
it can always be extended. For the extension the point
needed some user interaction but the creation of the
cloud based modeling techniques may provide valuable
models took less time than starting the 3D modeling
support to rapidly created 3D models of existing objects.
from scratch.
The presented 3D modeling techniques indicate the
direction in which factory design tools should advance in
7.2. Automatic creation of quasi-volumetric point cloud
the future. In this way, the factory design tools may
models from big point cloud
reach a similar functionality that current CAD systems
provide for product design.
Another aim was to automatically create quasi-
volumetric CAD models from the point cloud.
Therefore, an algorithm and a GUI for its configuration
Acknowledgements
have been developed within Visual Components that
reduces the number of points in a customized manner.
The authors acknowledge the support of the European
With its help small spheres or cubes are created in the
Commission through the 7th Framework Programme
centers of the filtered points. In this manner a quasi-
under NMP-2008-3.4-1 Rapid Design and Virtual
volumetric point cloud model is created which can be
Prototyping of Factories (COPERNICO; contract
used for e.g. collision detection during layout design.
number 229025-2). The first and third authors
The left side of Fig. 11(b) shows the solid body model of
acknowledge the support of the project “Talent care and
the machine presented in Fig. 11(a). The creation of this
cultivation in the scientific workshops of BME” that is
solid body model could take some time. The right side of
supported by the grant TÁMOP-4.2.2.B-10/1-2010-
Fig. 11(b) shows the quasi-volumetric point cloud model
0009.
made from small spheres created automatically within a
very short time.
References

[1] COPERNICO Project, 2010. Description of Work, Project title:

“Cooperation Environment for Rapid Design, Prototyping and
New Integration Concepts for Factory of the Future”. EU FP7
integrated project, contract number: NMP2-LA-2010-229025;
Project duration: 01. 05. 2010 – 30. 04. 2014.
[2] Németh, I., 2003. “A CAD Tool for the Preliminary Design of 3-
Axis Machine Tools: Synthesis, Analysis and Optimisation”. PhD
thesis, Katholieke Universiteit Leuven, ISBN 90-5682-387-6, p.
340.
[3] Baldwin, J. S., Rose-Anderssen, C., Ridgway, K., 2012. Linnaean
and Cladistic Classifications of Manufacturing Systems, in
Fig. 11. (a) Photo of the original machine; (b) Solid body model (left) “Enabling Manufacturing Competitiveness and Economic
and quasi-volumetric point model (right) of the machine Sustainability”, ElMaraghy, H., Editor. Springer-Verlag, Berlin
Heidelberg, pp. 22-34.
[4] Németh, I., Püspöki, J., 2010. “Development of a Manufacturing
8. Conclusions Equipment Configurator and an NC Simulator”, Proceedings of
the 43rd CIRP Conference on Manufacturing Systems, Vienna,
Several methods, databases and software tools have Austria, 26-28 May 2010, pp. 375-382.
been developed in the framework of the COPERNICO [5] NEXT Project, 2009. Final reports. Project title: “Next Generation
research project which support the rapid 3D design of Production Systems”. EU FP6 integrated project, contract number:
NMP2-CT-2005-011815.