Design of Machines and Structures, Vol. 11, No. 1 (2021), pp. 53–58.

https://doi.org/10.32972/dms.2021.007

STEPS OF GENERATIVE DESIGN IN INTEGRATED CAD SYSTEMS

KRISTÓF SZABÓ
University of Miskolc, Department of Machine Tools
3515 Miskolc-Egyetemváros
szabo.kristof@uni-miskolc.hu

Abstract: Due to the continuous development of various areas of the industry, such as mod-
ern production equipment, material technology, computer and software development, it is
possible to expand the range of conventional production technologies. These include additive
manufacturing technology, which provides a new opportunity to produce everyday products,
thereby satisfying market needs. Integrated CAD systems have occupied a place in the prod-
uct design and development process for decades, which has partially reformed classical de-
sign methods and its steps.

Keywords: product design methodology, topology optimisation, generative design

1. INTRODUCTION
A successful product meets the level of technical development of a given period and
fulfils the needs expressed by society. The aim of engineering design is to create a
suitable solution for a given problem, both from a technical and economic point of
view. Product design and development is an outstanding and special profession, as it
requires extensive experience, a unique vision and additional specific skills. Earlier it
has been accepted that the knowledge required for successful product design is a talent
that cannot be fully learnt, described, is not an exact science, and cannot be mecha-
nized. It was recognized in a short time that the quality of a product is greatly influ-
enced by the concept defined and selected in the design phase. Furthermore, a series
of decisions that arise during the design procedure play a key role in the product man-
ufacturing process, which can result in beneficial or disadvantageous changes. Based
on this philosophy, it can be said that in terms of the life cycle of a product, innovation
activities consume huge resources. Assuming that this type of activity can only be
properly performed by a competent person, design and development work proves to
be an expensive and long procedure. The increasing expectations dictated by the mar-
ket can be met as much as possible if a given product can be sold as soon as possible
and with the lowest financial cost. Accordingly, the design and construction tasks must
be transformed into tasks that can be performed by many, in which the individual
stages and steps can be well followed and performed [1].
54 Kristóf Szabó

2. MILESTONES OF DEVELOPMENT OF DESIGN METHODOLOGY

The development of various design methodological processes could be observed in the
last hundred years. Literature related to the field can be found mainly in Europe, but
there are researchers from all over the world whose work is related to this field of
science. The aim of the research is unchanged: the design process must be divided into
different stages, which can be clearly interpreted and followed in order to be applicable
for others. Kesselring published on evaluation procedures as early as 1937 and then
presented the basics of his convergent approximation procedure. Wögerbauer pro-
posed in 1943 that the entire design process should be divided into subtasks. The
founders of the Ilmenau school were Bischoff and Hansen. Hansen has been working
on the basics of design methodology since the 1950s, and, in 1965 he summarized the
theoretical aspects of his system. The founder of the Berlin school is Beitz, whose
work is closely linked to the founder of the Darmstadt School of Design, Pahl. In 1974,
Roth was among the firsts to realize that methodical design could be successfully au-
tomated using graphics computers and then developed an algorithmic design model.
In Hungary, the Budapest School of Design is worth mentioning, which deals with the
development and research of product design methodology and tools. The Hungarian
founder of this topic is Bercsey, who developed the Autogenetic Algorithm. It is im-
portant to mention the design school in Miskolc, which was founded by Terplán and
Tajnafői, and computer structure generation methods were created by Lipóth and
Takács [2], [3].

3. GENERATIVE DESIGN AND INTEGRATED CAD SYSTEMS

The generative design model is able to generate concepts using predefined require-
ments and constraints. The procedure, including shape and topological optimization,
was developed around the 1990s, but at that time could not lead to breakthrough
success.

Figure 1. The technological need of generative design

 Steps of Generative Design in integrated CAD Systems 55

The use of the programs was cumbersome, the capacity of the computers proved to
be insufficient, but the main drawback was that the result obtained could not be pro-
duced with the help of the traditional manufacturing technologies of the given era.
Over the next 20 years, the production of additives provided an opportunity to im-
plement 3D printing, and in the early 2000s it became clear that there was an oppor-
tunity for additive production of high-performance metallic components, which at-
tracted interest among integrated software manufacturers. Software supporting gen-
erative design appeared in the first half of the 2010s. Among the firsts
TrueSOLIDTM from Frustum can be mentioned, developed by Jesse Coors-Blank-
enship. The other big developer is AutoDesk, but recognizing the need for generative
design, more and more software development products have become available,
which are summarized in Table 1.
Table 1
Generative design softwares
Software developer Product
Frustum Generate
Generative design
nTopology Element
software
Paramatters CogniCAD
Altair OptiStruct
ANSYS ANSYS Mechanical
Tosca Structure, Tosca
CAE software sup- Dassault Systèmes
Fluid
porting generative de-
sign PAM-STAMP, Pro-
ESI Group
CAST, SYSTUS
MSC Software MSC Nastran Optimiza-
tion
Autodesk Fusion 360, Inventor
Dassault Systèmes TOSCA suite
Integrated systems Robert McNeel & Asso-
Rhino
with generative design ciates
module PTC Creo Simulate
Siemens NX, Solid Edge
Altair solidThinking Inspire

4. STEPS OF GENERATIVE DESIGN IN INTEGRATED CAD SYSTEMS

Generative design is a design process in which an algorithm is used to optimize the
shape of a part for a given boundary condition. Designing the shape itself is not a
manual design task. The designer determines the functional boundary conditions of
the part, adds it into the software, which calculates the shape of the optimized part
according to the defined aspects during iteration processes [4], [8]. Limit states can
usually be divided into two groups, the calculation requires an initial geometry,
56 Kristóf Szabó

which must be constructed by traditional 3D modelling. This is quite similar to the

solution used in traditional FEM systems: it is necessary to determine which area of
the piece is subjected to which forces and which constraints [5–7]. Another possibil-
ity is to determine the volumes in which there can be no material because, for exam-
ple, some other component is moving there. If there is no starting workpiece, it
should be specified as a “volume part” that will be part of the finished part. The steps
in the generative design process that are valid and show similarity using all the inte-
grated CAD systems listed in Table 1 are detailed below.
After opening the given program, our first step is to define the design volume, for
which we have three options. The first way to do this is to define the geometries to
be retained, which remain an integral part of the yellow geometry. Specifying them
is a mandatory operation, and later these bodies and surfaces allow defining func-
tions, such as placing mortises. The second method of design space is to define so-
called interfering geometries, which can be used to specify those parts of space
where there can be no material. The geometry produced by the program can only be
located outside this space, but it can be applied in a similar way when the part is
limited in size. These volumes are optional during design. The third method is to
import a solid-state model whose shape features can be used to specify functions. In
this case, the outer surface of the original model does not limit the enclosing size of
the geometry produced during generative design by default.
After the precise definition of the design space, the second stage of the process
can follow, during which the fixing points and further constraints of our model can
be defined. It is possible to specify fixed points, but it is possible to unlock individual
planes and axes of rotation within it. We have the option of creating a hinge or pivot
point where radial, axial and tangential movement can be allowed. Furthermore, it is
allowed to define slip planes and friction surface pairs.
In the third stage of design, we get to defining the location and magnitude of the
loads. We have the ability to accommodate force, pressure, torque and distributed
load, the direction and magnitude of which can be changed indefinitely.
The fourth step is to decide on the design criteria and objectives. This can be
minimizing mass, maximizing stiffness, or developing minimal stress and its optimal
distribution. In this phase, a so-called safety factor can be set.
In the fifth step, it is possible to choose the production method, where the pro-
duction volume and the appropriate production technology can be selected. Optional
technologies include additive manufacturing, cutting processes such as milling, cut-
ting and casting. For each option, the minimum material thickness for the model and
the tools used in the particular technology, such as the geometric size of the milling
tool and the machining direction can be chosen. There is also the possibility that this
step will remain unselected, in which case the generation of models will be more
widely allowed.
In the sixth step of the process, the material has to be chosen from which the
product can be made. The selection can be made from the material catalogues of the
programs, but a new material with unique properties can also be defined. The mate-
rial properties of the items in the catalogue can be modified without any problem.
 Steps of Generative Design in integrated CAD Systems 57

Care must be taken to ensure that each manufacturing technology has a set of com-
patible materials.
After making these settings, a verification step becomes available that runs
through the data we enter and alerts the user in case of lack of data or poorly entered
conditions.
Once the check is done, the planning, i.e. the final calculation and generation
process, can be started. We have the opportunity to filter the obtained solutions by
categories and access the iteration results of the individual components.

Figure 2. Steps of generative design process

5. SUMMARY
The article reviews the development of the product design- and development field that
forms the basis of generative design, as well as its defining stages. Factors influencing
the spread of the generative design process and the development of the necessary tech-
nological processes are presented, and the article provides a short historical overview
of the topic of software supporting. Based on the software listed, the article describes
58 Kristóf Szabó

the steps required to use the method, which show a match for different programs. For
quick understanding and illustration, a flowchart for the operation of the method was
created, supplementing the possible iterations. By observing and following the steps
properly, we get successful solutions to the formulated task.

ACKNOWLEDGEMENT
The described article/presentation/study was carried out as part of the EFOP-3.6.1-
16-00011 Younger and Renewing University – Innovative Knowledge City – institu-
tional development of the University of Miskolc aiming at intelligent specialisation
project implemented in the framework of the Szechenyi 2020 program. The realiza-
tion of this project is supported by the European Union, co-financed by the European
Social Fund.

REFERENCES
[1] Takács, Gy. − Zsiga, Z. − Szabóné Makó, I. − Hegedűs, Gy.: Gyártóeszközök
módszeres tervezése. Nemzeti Tankönyvkiadó, Miskolc, 2011.
[2] Kamondi, L. − Sarka, F. − Takács, Á.: Fejlesztés-módszertani ismeretek.
Nemzeti Tankönyvkiadó, Miskolc, 2011.
[3] Takács, Á.: Computer Aided Concept Building. Solid State Phenomena, Vol.
261, 2017., pp. 204–207, DOI: 10.4028/www.scientific.net/SSP.261.402.
[4] Szabó, K. − Hegedűs, Gy.: A generatív tervezést támogató szoftverek rövid
áttekintése. Multidiszciplináris Tudományok, Vol. 10, No. 3, 2020, pp. 328–
337, DOI: 10.35925/j.multi.2020.3.39.
[5] Zuo, K. – Chen, L. – Zhang, Y. – Yang, J.: Study of key algorithms in topology
optimization. International Journal of Advanced Manufacturing Technology,
Vol. 32, No. 7–8, 2007, pp. 787–796, DOI: 10.1007/s00170-005-0387-0.
[6] Bendsøe, M.: Optimization of Structural Topology, Shape and Material.
Springer Verlag, Berlin, 1995., DOI: 10.1007/978-3-662-03115-5.
[7] Rozvany, G.: Aims, scope, methods, history and unified terminology of com-
puter-aided topology optimization in structural mechanics. Structural Multi-
disciplinary Optimization, pp. 90–108, 2001, DOI: 10.1007/s001580050174.
[8] Trautmann, L.: Product customization and generative design. Multidiszci-
plináris Tudományok, Vol. 11, No. 4, pp. 87–95., DOI:https://doi.org/10.35925
/.multi.2021.4.10