Int J Adv Manuf Technol

DOI 10.1007/s00170-017-0112-9

ORIGINAL ARTICLE

Additive manufacturing integration with topology optimization

methodology for innovative product design
Teresa Primo 1 & Maurizio Calabrese 1 & Antonio Del Prete 1 & Alfredo Anglani 1

Received: 13 October 2015 / Accepted: 27 January 2017

# Springer-Verlag London 2017

Abstract Additive manufacturing (AM) processes enable the been developed. An engineering intelligence tool has been
production of functional parts with complex geometries, used to post process the generated optimization results for
multi-materials, as well as individualized mass production. the three different approaches. Finally, the first three “best”
Another significant benefit of AM is the ability to produce structural solutions have been manufactured by 3D printing
optimized geometries with near perfect strength to weight ra- machine, with scaled dimensions, in order to evaluate the
tios. Weight plays a crucial role in many functional parts such printing time considering the geometry complexity for the
as parts used in automotive and aeronautic industries. Current chosen structural layout in order to have useful feedbacks on
topology optimization techniques, a powerful tool for weight Product/Process choices interaction.
reduction and product optimization in general, do not work
well for such kind of process since AM methods necessitate Keywords Additive manufacturing . Topology optimization .
applying own dedicated design rules. This paper reports a Key performance indicators . Lattice approach . Hybrid
product/process optimization study of a simple test case ge- approach
ometry (C-Clip), where structural optimization has been ap-
plied using an innovative approach based on the design of
lattice structure feasible thanks to additive process adoption. 1 Introduction
Moreover, it has been conducted a study to evaluate the pos-
sible advantages offered by the integration of the two previous Product weight has a significant impact not only on many
approaches in order to verify the required design specifica- production parameters but also on the component perfor-
tions. The aim of the work has been to evaluate the potentiality mance and its service life. The usage of the minimum quantity
offered by the integration of the two structural optimization as raw materials and the needed energy for fabricating engi-
approaches (topological and lattice structures design) to obtain neering components turn out to be the main issues in parts
innovative and highly performing structures. This activity rep- design rules [1]. Lighter products require less manufacturing
resents a necessary step for the definition and the subsequent and maintenance costs, on one hand, and cause inferior dam-
development of a methodology aimed to the creation of struc- ages to the environment due to their lower carbon consuming,
tures obtained with this combined design approach. In order to on the other. Furthermore, in some industrial contexts such as
define an objective evaluation of the component perfor- aerospace, the components weight plays a crucial role in suc-
mances, appropriate Key Performance Indicators (KPI) have cess of the industry from economical point of view. Reducing
a part weight during the design stage, however, cannot be
achieved straightforwardly and many other issues have to be
* Teresa Primo considered.
[email protected] Topology optimization, a mathematical tool used in con-
ceptual design stage, has been found to fulfill the “weight
1
Department of Engineering Innovation, Università del Salento,
reduction” aspect by optimal distributing of the material
Complesso Ecotekne—Edificio “Corpo O”—Via per Monteroni, throughout the component body in accordance with the given
73100 Lecce, Italy boundaries conditions.
 Int J Adv Manuf Technol

Generally, for implementing topology optimization tech- manufacture therefore provides an opportunity to manufacture
nique during the concept design stage, engineers need to de- components that more closely approximate the optimal geom-
termine the objective functions as well as the desired con- etry than traditional methods [14].
straints which have to be taken into account. Material proper- Additive manufacture can accommodate highly complex
ties, component essential geometric features, and loading con- geometries; for example, high-value jet engine [28], aerospace
ditions are normally considered as desired constrains. The [29], and automotive [30] components that are incompatible
topology optimizer will then attempt to optimize the material with traditional manufacturing processes. The unique capabil-
distribution in terms of possible layout within the design ity of additive manufacture to generate highly complex geom-
boundaries while meeting the design requirements [2]. etries provides an opportunity to manufacture components
Nowadays, computer-aided engineering (CAE) software and systems that are close to their theoretically optimal geom-
packages have started to employ this approach to assist de- etry. In particular, topologically optimal geometries often in-
signers. The result of topology optimization may include a re- clude highly complex truss networks and internal void struc-
designed model with complex structure and intermediate den- tures [10–12].
sities properly distributed in the assigned 3D design space. In this paper, a structural optimization study of a simply C-
Consequently, the feasibility of the design should be verified Clip component has been carried out considering the follow-
due to not taking into account the design for manufacturing ing approaches: topology optimization (TO), lattice approach
(DFM) requirements during topology optimization stage. (LA, based on additive process), and the hybrid approach
Generally, in traditional manufacturing processes, it is then (HA) as a “smart” mix of the previous two approaches in order
needed to review the optimized design and turn the complex to evaluate which the advantages are offered by their
areas into simple sections with fully dense (solid) or fully void integration.
(open) spaces to improve the product feasibility for the chosen To catch up the aim of a complete overview of the gener-
processing technology [3]. As result, it can be expected that ated data space, the authors have used an innovative approach
the fabricated part may perform in a different way for com- based upon the On-Line Analytical Processing (OLAP) data-
parison with the topology optimized one [4]. base and a properly defined engineering intelligence model.
Additive manufacturing has gained numerous attractions in Once simulations results are imported in OLAP database, it is
the recent years because of its unique offered advantages in- very simple to navigate throughout the data space and to cre-
cluding directing concept of DFM into manufacturing for de- ate clear graphs able to describe the effects of process param-
sign (MFD) approach in the part production. Additive eters on the monitored KPI. Finally, the three “best” solutions
manufacturing refers to manufacturing techniques that build for the studied component have been manufactured by a 3D
up three-dimensional components by adding material incre- printing machine in scaled dimensions.
mentally to produce the desired object layer-by-layer [3, 5].
The only limitation to build a part with these processes is just
the imagination during designing [6, 7]. 2 Methodology
Additive manufacture involves the progressive addition of
material to generate component geometry; this differs funda- Additive manufacturing is nowadays widely used in industrial
mentally from traditional, subtractive manufacture, whereby product development. The main advantage of the additive
material is progressively removed from an initial geometry as fabrication concept is the ability to create almost any possible
required [8]. Additive manufacture has been labeled as a dis- shape. This capability is governed by the built-up layer-by-
ruptive technology [9] due to the associated capacity to eco- layer process. There are several available technologies based
nomically manufacture at very low batch sizes and the capa- on this additive machining concept [31]: stereolithography,
bility to manufacture highly complex geometries [8–27]. This photo-masking, selective laser sintering, fused deposition
latter capability provides an opportunity to physically imple- modeling, 3D printing, etc. Researchers principally work on
ment topologically optimal geometries, which are often highly the influence of part orientation, slicing strategy, matching
complex, and therefore incompatible with traditional internal patterns to improve cost, product quality, built time,
manufacturing processes. etc. The additive manufacturing technologies are playing an
Topology optimization enables identification of the optimal increasingly important role in the production scenarios. In
structural connectivity for the following: a specific design particular, one of the main strength points is the production
scenario, boundary conditions, and available 3D design space. of small series components with innovative geometries not
Topology optimization is not based on a priori assumptions of feasible by traditional technologies. The additive manufactur-
material distribution, resulting in complex truss networks with ing tools able to support the designer during concept definition
high structural efficiency [10–12]. The geometric complexity have undergone a significant development.
of the topologically optimal design is typically incompatible In particular, in the last years, the efforts are focused on the
with traditional manufacturing methods [13]. Additive development of innovative tools based on the lattice structures
Int J Adv Manuf Technol

Table 1 KPIs thresholds

and the improvement of the topology optimization traditional – Phase 2: starting from the topology optimization
approach. (TO), different test cases have been developed
Therefore, the proposed study has been structured in two considering the integration of the two approaches
phases: described previously. The target of this methodol-
ogy has been to obtain an innovative concept that
– Phase 1: analysis of the capabilities of the two optimiza- exploits the advantages of the combined applica-
tion approaches (topological and lattice structures) for a tion of topology optimization and lattice
simplified geometry test case. approach.

Table 2 LA: analyzed solutions

 Int J Adv Manuf Technol

Figure 1 shows the adopted workflow for the methodology design variables, objective function and last, but not the
developed in this work, where, the different steps of the two least manufacturing constraints. Of course, the given pos-
considered approaches, TO and LA, to obtain the respective final sible result of this first option has to be processed in order
design and their integration for HA evaluation, are reported. to obtain a feasible CAD model which is a needed input to
As can be detected by the represented workflow, the pro- verify the structure performances of the obtained solution.
posed method shares an initial step (project specification) in In this case, of unacceptable result for the given choice,
which the main key points of the problem are defined (loads, what has been just described can be executed again until
material, constraints, acceptable deformation). After that, the an appropriate solution is provided.
available choices are represented by: b) Thanks to the additive technologies, an alternative to the
abovementioned approach is represented by what has
a) A classical topology optimization approach able to pro- been called “Lattice Approach”. In this case, the defined
vide a new design solution in accordance with the given: design space is considered as the volume in which

Fig. 1 Methodology workflow

Int J Adv Manuf Technol

different options for: lattice structures, their main dimen- in what has been defined “Good Range”, while the “Weight”
sions, skin definitions, and skin thickness variability can KPI is close to the upper bound (22.9 kg, as reported Table 3) of
be considered. Also in this design approach, the obtained the given acceptable range (20.1–23.0 kg).
result will be a possible solution which has to be verified
and, if it will be necessary, subjected to design 2.1 Topology problem description
improvement.
Topologyoptimizationproblemcanbedefinedasthesearchforthe
As the workflow reports, authors believe that the best so- best material allocation or distribution in a given design space.
lution can be provided by an integrated approach of the two Topological optimization solves a specific structural problem:
above described procedures. n

f i ðxÞ−T i 2

The component material considered in this study has been the min ∑ ωi g j ðxÞ ≤ 0 j ¼ 1; m
i¼1 Ti
AISI 304 steel (Austenitic Cr-Ni stainless steel) with the follow-
ing mechanical properties: elastic module E = 190.000 MPa; In this equation, fi(x) are structural responses obtained from a
density ρ = 8.0 × 10 −9 ton/mm 3 , and yield strength classic finite element analysis (FEA), and gi(x) are the constraints
σs = 245 MPa. function. During the calculation, if g = 0 constraints are active, if
In order to ensure elastic operating conditions, a safety g < 0 constraints are considered as inactive, and if g > 0 con-
coefficient has been considered (Cf = 1.1) that determines a straints are defined as violated. The values Ti are the target values
maximum allowable stress equal to 222 MPa. for the particular response, and ωi is a weighting factor. In topol-
Moreover, the authors have considered two different types ogy optimization, design variables are element densities (SIMP
of key performance indicators in order to monitor the struc- method) [32, 33].
tural behavior of the component (product KPIs) and the part Under topology optimization, the material density of each
manufacturing feasibility (process KPI). element should take a value of either 0 or 1, defining the
The considered product KPIs are the following: maximum element as being either void or solid, respectively.
displacement (mm), maximum von Mises stress (MPa), and With the density method, the material density of each ele-
weight (kg). ment is directly used as the design variable, and varies con-
For the process KPI, it has been defined an appropriate tinuously between 0 and 1 (0= no material presence, 1= ma-
KPI: the needed printing time for each solution. The printing terial presence); these two possible status represent the state of
time depends of the 3D print technology used; therefore, in void and solid, respectively. The stiffness of the material is
order to monitor this aspect, it has been necessary to define a assumed to be linearly dependent on the density.
normalized time as KPI. This approach allows to monitor the Specific techniques need to be introduced to penalize inter-
part manufacture time considering the geometry complexity mediate densities and to force the final design to be represented
of each chosen solution. by densities of 0 or 1 for each element. The penalization tech-
The process KPI has been defined by the ratio between the nique used is the “power law representation of elasticity proper-
printing time of each considered component (tci) and the lower ties” and it can be expressed as follows:
printing time (tl) among all considered design solutions:
K ðr Þ ¼ ρp k
t ci
Printing time ¼ where K(r) is the penalized stiffness matrix, k is the real stiff-
tl
ness matrix, ρ is the density and p is the penalization factor
An intuitive approach has been used to define the KPIs which has to be superior to 1 [32, 33].
thresholds: “Good”, “Acceptable”, and “Unacceptable”. In this specific case, the authors have applied a topological
In Table 1, the three different thresholds for the defined KPIs optimization technology to obtain a structural shape in re-
are reported. The adopted colors in the legend are the following: sponse to the given loads and constraints. The typical resulting
red for the “Unacceptable” level, yellow for the “Acceptable” solutions are polymeshes that can be exported to computer-
quality, and green for the “Good” levels. The limits for each KPI aided design tools (CAD) as a source of inspiration for de-
and each interval have been selected having as reference a pos- signers. Figure 2a shows C-clip test case with the design and
sible design solution obtained as output as a simple topology non-design space: the first one represents the area where it is
optimization approach. A possible reference case is represented possible to explore new design solutions and the second is the
by the solution defined as the “TO” topology optimization ap- one which cannot be changed. In the “non- design” areas, the
proach. The product responses defined for this specific case boundary conditions have been applied (Fig. 2b). In particular,
have been considered to define each range for the considered in the “non-design” space 1 (Fig. 2a) the C-clip has been fixed,
KPIs. In particular, for this considered solution the measured: while the “non-design” space 2 (Fig. 2a) represents the areas
displacements, von Mises stress and printing time, are included in which the part has been loaded with a total force equal to
 Int J Adv Manuf Technol

Table 3 KPIs result analysis

15 kN on each side (Fig. 2b). The “design” space represents The TO problem has been defined in accordance with the
the optimization domain. following steps:

FORCE: 15 kN
300 mm

FORCE: 15 kN
Thickness= 50 mm

FIXED SUPPORT

400 mm
(a) (b)
Fig. 2 a C-clip test case–ID 1; b applied boundary conditions
Int J Adv Manuf Technol

Fig. 3 a Optimized design

obtained by TO; b, c C-CLIP
concept design by TO

& Design space definition and the relative design variables; 2.2 Innovative lattice approach (LA) description
& Output responses adopted for the optimization as objec-
tive and constraint functions of the given structural prob- Taking into account the advantages offered by the additive
lem. For the specific study displacements, stress, and mass manufacturing processes in the last years, different works
responses have been defined; have developed some innovative geometries having as dis-
& Constraints based on output responses defining for each tinctive feature the lattice structures [35, 36].
one of them the limit bounds. For the studied case: In this specific case, the authors have developed different
designs with the LA through the usage of a specific technol-
– a stress constraint equal to the yield stress versus safety ogy in order to optimize the quantity of material used.
coefficient ratio (equal to 1.1) has been defined; Thanks to which it is possible to define new geometries that
– a manufacturing constraint for minimum member size are lighter in weight but maintain the performance character-
control equal to 30 mm has been defined. istics being stiff in some areas and flexible in others.
These new geometries consist of: lattice structures and
& Objective function is the minimization (or the maximiza- boundary skins (Fig. 5).
tion) of the problem. For the specific study, the mass min- The capabilities of the tool are:
imization has been defined as objective function.
– Different lattice topologies (single shell and multi shell)
Figure 3 shows the optimized design obtained by TO in (Fig. 6);
terms of given shape by the optimizer (a) and the concept – Kernel function that allows to define different local den-
design defined (b, c) using a 3D hybrid modeling and render- sities for lattice and skins (Fig. 7);
ing environment based on the NURBS surfaces able to pro- – Different beams sections;
cess the shape results obtained from the optimizer. – Possibility to improve the quality of the model by means
For the concept design, a FEM analysis has been carried of the smoothing function.
out. In particular, the FE model, that consist of linear tetra
elements with an average size equal to 1 mm has been created
in the Altair Hypermesh® environment; and the linear struc- The authors, in this work, have used the capabilities of the
tural analysis has been performed with Altair Optistruct® lattice technology environment tool in order to reduce the
code. component weight ensuring the required stiffness.
In Fig. 4, it is possible to review the results in terms of A set of possible solutions have been studied considering
displacement (mm) and MPa for the studied case. different lattice typologies. In Table 2, are shown all the

Fig. 4 Test TO case study: a displacement (mm); b MPa

 Int J Adv Manuf Technol

Figure 8 reports the results in terms of displacement (mm)

and MPa for the LA21.

2.3 Hybrid approach description

Starting from the topology solution (TO), different test cases

have been developed considering the integration of the two
approaches previously described (Fig. 9). The target of this
methodology has been to obtain an innovative concept that
exploits the advantages of TO and LA.
Fig. 5 Lattice structure The integration of the two approaches has allowed the def-
inition of three hybrid concepts which actually represent three
analyzed designs well represented in six groups. In particular, sub-phases of the integration that are described in Fig. 10.
the groups can be described in the following way:
– STEP 1: In this step, it has been conducted a preliminary
– Lattice: 1 and 2 are two different solutions analyzed to activity aimed at the evaluation of the structural perfor-
test the performances of the lattice structure with constant mances of a solution that involves the insertion of uni-
density. These components have been developed in the form lattice structures within the topologically optimized
preliminary phase; profile. The aim has been to investigate the maximum
– Lattice: 3, 4, 5, and 6 are the solutions analyzed to mon- weight reduction and to estimate the structural behavior.
itor the advantages offered by the kernel function for The output of this step has been the concept designed
some lattice typologies. named HA01.
– STEP 2: This step has regarded the development of a
The study has consisted of the following main steps: hybrid concept exploiting the capabilities offered by the
generation tool lattice structures. In particular, locally, the
1. Defining the possible test cases of interest; skin’s thickness (skins kernel) and reticular structures di-
2. Monitoring the weight for each one; ameters (lattice kernel) have been considered variable
3. Performing a FEM analysis in order to verify the perfor- taking into account of the boundary conditions applied
mance in terms of maximum displacement and MPa. on the test case. The objective has been to evaluate the
influence, in terms of weight and structural performances,
The FEM analysis have been performed by means the of the local alterations of the geometrical characteristics
Altair Hypermesh® environment for the pre-processing phase of the considered test case. The output of this step has
and the Optistruct® solver for the structural analysis. The type been the concept designed named HA02.
of element used is linear tetra with an average size equal to – STEP 3: This step has consisted in the following sub-
1 mm. phases:

Fig. 6 a Single shell SINGLE SHELL MULTI SHELL

JUNCTION

JUNCTION

All beams are as 1 shell and

there is a proper junction Each beam is a separate shell
generated
(a) Single Shell (b) Multi Shell
Int J Adv Manuf Technol

KERN
NEL FUNCT
TION SECTION OF THE
E BEAMS SMOOT
THING FU
UNCTION

Fig. 7 Tool capabilities

Comparison in terms of volume and material distribution approaches (topological and lattice structures) for obtaining
between the TO output and engineered design TO. The innovative and highly performing structures. This activity rep-
goal has been to identify the added volume of the resents a necessary step for the definition and the subsequent
engineered solution respect to topology output. development of a methodology aimed to the creation of hybrid
Insertion of lattice structures in technological interest structures.
zones and testing of the defined solution. This study has allowed the authors to understand which
the main necessary technological aspects are to develop a
The output of this step has been the concept named HA03 hybrid concept achieving the engineering requirements.
(Fig. 11). After this work, taking into account the results obtained
The general objective has been to evaluate the effec- in terms of feasibility of hybrid structures and structural
tiveness in the integration of the two approaches for the performance, the authors intend to develop a methodology
setup of highly performing innovative geometries in able to consider shape variables to perturb the engineered
terms of weight reduction and mechanical characteristics topology results in order to increase the volume without
in order to match the advantages that each method offers, increasing the final design weight thanks to the adoption
or rather: of the trabecular design.
The Fig. 12 shows how the lattice capabilities could
– Simple process in the case of topological optimization increase the stiffness performance and reduce the weight
tools to support the creation of high-performance respect to the conventional topological design. In particu-
engineered profiles; lar, these capabilities are summarized in the following
– Components’ performance improvement through the in- considerations:
troduction of the trabecular structures in the case of the
lattice tools adoption. – The engineered topology concept contains some ribs with
full density (VF1) and a given stiffness (KF1);
The aim of the work has been to evaluate the potentiality – These lattice capabilities allow to reduce the volume/
offered by the integration of the two structural optimization weight of the rib (VL1 < VF1; WL1 < WF1), but this

Fig. 8 Test LA21 innovative LA case study: a displacement (mm); b MPa

 Int J Adv Manuf Technol

Fig. 9 Hybrid approach

solution could bring to obtain a reduction of the stiffness The Table 3 shows the results for the all most interesting
performance (KL1 < KF1); studied solutions. The other solutions do not satisfy the de-
– The perturbation of the rib dimensions could allow to fined KPIs.
increase the performance in the following aspects: The analysis results can be summarized in the following
observations:
Stiffness: the rib could present a stiffness greater or equal
than the stiffness related the topological rib (KL2 ≥ KF1); – Topology optimization test case: this solution represents a
Volume/weight: the lattice rib could present a volume good compromise in terms of product KPIs and process
lower than the volume related the topological rib KPI. In particular, for this last aspect, the printing time is
(VL2 < VF1; WL2 < WF1). the lowest.
– LA (LA19, LA20, LA21 test cases)

3 Engineering intelligence model and data analysis & Product KPIs: the best solution is represented by
LA21 thanks to which it is possible to reduce the
In order to achieve the aim of a complete overview of the maximum stress (200 MPa) and the weight (20.5 kg).
generated data, the authors have used an innovative approach & Process KPI: the best solution is represented by LA20
based upon the OLAP databases [34]. thanks to which it is possible to obtain a “Good”
Once optimization results have been imported in OLAP da- printing time result. The LA21, instead, requires a
tabase, an appropriate engineering entelligence (EI) model has p r i n t i n g t i m e t h a t i t ha s b e e n c o n s i d e r e d
been developed in order to allow the navigation throughout the “Acceptable”.
data space and to create clear plots of the wanted responses.
These are able to evaluate the effects of the defined design var- – Hybrid approach (HA02, HA03 test cases): this approach
iables on the monitored KPI: maximum displacement, MPa, C- represents a merge of the advantages offered by the TO
Clip weight (product KPIs), and printing time (process KPI) for and lattice methodology integration. In particular, the
each optimization approach. goal of this method it has been to obtain the following:
The OLAP technique has been allowed to monitor
the solutions considering quantitative and qualitative re- & A significant weight reduction of the weight
sults. In particular, in the present work, Altair® (ensuring the structural performances) by means
HiQube® has been used to monitor all the optimization the possibility to apply the lattice structures in
results. some areas;

Fig. 10 Hybrid approach: output

of the steps
Int J Adv Manuf Technol

Fig. 11 STEP 3, overview

& A significant printing time reduction respect to 4 Test case manufacturing

the pure lattice design thanks to the presence of
different areas designed with TO. In order to monitor the printing time (process KPI), the six
solutions represented in the Table 3 in scaled dimensions
Anyway, the judgment of what can be considered the “best (1:5), have been manufactured by high precision 3D printer
solution” cannot be defined as an absolute reference. The rea- Solidscape 3Z STUDIO (non-structural demonstrators). In
son of this assumption stands in fact that each possible final Fig. 13, the most interesting solutions (LA21, TO, HA03 test
application context of the studied case defines the ranking of case) are shown.
the most important aspects to be considered for the right The 3D printer receives digital input from three-dimensional
choice. The HA03 test case can be considered the best solu- data files (STL, SLC) and creates solid, three-dimensional parts
tion because it guaranties to obtain reasonable results for each through an additive, layer-by-layer process with a layer thick-
KPI. However, if the most important aspect is the “mass re- ness (mm) from 0.00625 to 0.0762 and a resolution of (dpi)
duction” without any concern about printing time and costs 5000 × 5000 × 8000 XYZ.
(like it is in industrial application represented by aerospace The used printer combines drop-on-demand (“DoD”) ther-
industry), than HA02 looks like the more promising one. moplastic ink-jetting technology and high precision milling of
For what has just been said, the best “design concept” can each layer.
be accomplished only once; the design options take into ac- Figure 13 shows the most interesting solution analyzed for
count the final application context. For this reason, the hybrid the C-Clip scaled demonstrators obtained by 3D printer
approach represents a powerful tool for designers which can Solidscape 3Z STUDIO, respectively are topology, lattice,
explore many different solutions for any given proposed en- and hybrid approach. It is evident of the high precision and
gineering problem. detail refinement.

Fig. 12 Hybrid approach: future

enhancements
 Int J Adv Manuf Technol

Fig. 13 C-Clip non-structural

demonstrators obtained by: a
topology TO; b LA LA21; c
hybrid approach HA03

The work plane of this 3D printer is 150 × 150 × 50 mm KPI is the printing time. For each KPI, the authors have con-
(XYZ). Therefore, in order to manufacture the chosen com- sidered appropriate thresholds of product acceptability.
ponents, it has been necessary to scale the dimensions (1:5). The obtained results have been compared by means of an
The machine capabilities allow to have extremely precision in engineering intelligence tool that allows to observe the sce-
terms of minimum lattice section diameter obtained (about nario with a multi-dimensional database (OLAP technique).
0.2 mm). The objective of the authors has been not to find the best
solution, but to propose a novel approach hybrid, which al-
lows to have more degrees of freedom in the design definition
5 Conclusions and further developments and where the possible and the best solution depends from the
final application.
In this work, the authors have considered a simple test case (C- Further researches will be conducted to study and manu-
Clip) in order to evaluate the capabilities offered by the fol- facture an industrial test case with the methodology analyzed
lowing approaches: topology optimization, the innovative lat- in the present work. In order to verify the component perfor-
tice approach, and the hybrid approach. In all cases, the goal mances, a numerical and experimental comparison will be
has been to minimize the mass of the component and guaran- performed.
tee the performances in terms of maximum displacement, The study has been conducted on a simple test case in order
maximum von Mises stress, and the maximum printing time to evaluate the potentialities of the methodology. The authors,
allowable. downstream of the work, will apply the methodology on cases
The first approach has been based on the topology optimi- of industrial interest. In particular, the studies are focused on
zation; therefore, the weight reduction has been obtained re- the structural optimization of structural brackets and, in addi-
moving the material that is not necessary to satisfy the defined tion, the authors are involved in a specifically related research
design constraints. project.
The lattice approach is an innovative way to design and
manufacture the component by means the 3D printing. In this
scenario, it is possible to define innovative geometries in-
spired by the lightness that cannot be produced by the tradi-
tional subtractive processes. In this work, the authors have
References
considered a set of possible solutions using different lattice
1. Becker R, Grzesiak A, Henning A (2005) Rethink assembly design.
structures and exploiting the capabilities offered by the kernel Assem Autom 25:262–266
functions in order to increase the local density. 2. Almeida HDA, da Silva Bártolo PJ (2010) Virtual topological op-
Afterwards, the authors have developed an hybrid method timisation of scaffolds for rapid prototyping. Medical Engineering
in order to exploit the advantages offered by the combination & Physics 32:775–782
of the topology and the lattice approaches. In particular, con- 3. Sundararajan VG (2010) Topology optimization for additive
manufacturing of customized meso-structures using homogeniza-
sidering the topology results, in order to obtain a further tion and parametric smoothing functions, Master of Science in
weight reduction and to contain the printing time, different Engineering, Texas at Austin
lattice structures have been applied in some areas of the con- 4. Rezaie R, Badrossamay M, Ghaie A, Moosavi H (2013) Topology
sidered solution. optimization for fused deposition modeling process. Procedia CIRP
6:521–526
In order to assess the performance, two types of key per-
5. Chen Y (2007) 3D texture mapping for rapid manufacturing.
formance indicators (KPIs) have been defined: product KPIs Computer-Aided Design & Applications 4:761–771
and process KPI. The product KPI are: maximum displace- 6. Boothroyd G and Marinwscu I (2002) Product design for manufac-
ment, maximum von Mises stress, and weight. The process ture and assembly Marcel Dekker
Int J Adv Manuf Technol

7. Godazandeha E, Badrossamay M, Rezaei R, and Tavoosi M (2011) 21. Michell AGM (1904) LVIII. The limits of economy of material in
Evaluating fabrication of rapid manufacturing textiles by applying frame-structures. London Edinburgh Dublin Philos Mag J Sci
CAD/CAE/AM, presented at the 3rd International Conference on 8(47):589–597
Manufacturing Engineering ICME2011, Iran, Tehran 22. Eschenauer HA, Olhoff N (2001) Topology optimization of contin-
8. Zhai Y, Lados DA, LaGoy JL (2014) Additive manufacturing: mak- uum structures: a review. Appl Mech Rev 54(4):331–390
ing imagination the major limitation. JOM 66(5):808–816 23. Bendsoe M, Kikuchi N (1988) Generating optimal topologies in
9. Hopkinson N, Hague R, Dickens P (ed) (2006) Rapid manufactur- structural design using a homogenization method. Comput
ing: an industrial revolution for the digital age. John Wiley & Sons Methods Appl Mech Eng 71(2):197–224
10. Bendsoe M, Sigmund O (2003) Topology optimization: theory, 24. Suzuki K, Kikuchi N (1991) A homogenization method for shape
methods and applications. Springer, 370 pages and topology optimization. Comput Methods Appl Mech Eng
11. Bendsoe M (1995) Optimization of structural topology, shape and 93(3):291–318
material. Springer 25. Bendsoe MP (1989) Optimal shape design as a material distribution
12. Seepersad CC (2004) A robust topological preliminary design ex- problem. Struct Optim 1(4):193–202
ploration method with materials design applications. Doctoral dis- 26. Zhou M, Rozvany GIN (1991) The COC algorithm, Part II: topo-
sertation. Georgia Institute of Technology logical, geometrical and generalized shape optimization. Comput
13. Chang KH, Tang PS (2001) Integration of design and manufactur- Methods Appl Mech Eng 89(1):309–336
ing for structural shape optimization. Adv Eng Softw 32(7):555– 27. Rozvany GIN (2001) Aims, scope, methods, history and unified
567 terminology of computer-aided topology optimization in structural
14. Leary M, Merli L, Torti F, Mazur M, Brandt M (2014) Optimal mechanics. Struct Multidiscipl Optim 21(2):90–108
topology for additive manufacture: a method for enabling additive 28. (2012) Components for rocket engines. In: Ti 2011—Proceedings
manufacture of support-free optimal structures. Mater Des 63:678– of the 12th world conference on titanium, vol. 3, Beijing, China, p.
690 1715–8.
15. Leary M, Babaee M, Brandt M, Subic A (2013) Feasible build 29. Brandt M, Sun S, Leary M, Feih S, Elambasseril J, Liu Q (2013)
orientations for self-supporting fused deposition manufacture: a High-value SLM aerospace components: from design to manufac-
novel approach to space filling tessellated geometries. Adv Mater ture. Adv Mater Res 633:135–147 Trans Tech Publications,
Res 633:148–168 Trans Tech Publications, Switzerland Switzerland
16. Chahine G, Smith P, Kovacevic R (2010) Application of topology 30. Cooper DE, Stanford M, Kibble KA, Gibbons GJ (2012) Additive
optimization in modern additive manufacturing. In: Solid freeform manufacturing for product improvement at Red Bull Technology.
fabrication symposium, Austin, Texas Mater Des 41:226–230
17. Doubrovski Z, Verlinden JC, Geraedts JM (2011) Optimal design 31. Wohlers T (2012) Additive manufacturing and 3D printing—state
for additive manufacturing: opportunities and challenges. In: of the industry. In: Wohlers Report 2012. Annual worldwide prog-
ASME 2011 international design engineering technical conferences ress report
and computers and information in engineering conference. 32. Gardana N, Schneider A Topological optimization of internal pat-
American Society of Mechanical Engineers, p.635–46 terns and support in additive manufacturing, Journal of
18. Sigmund O (2000) Topology optimization: a tool for the tailoring of Manufacturing Systems, article in press
structures and materials. Philos Trans R Soc A Math Phys Eng Sci 33. HyperWorks User Manual
358(1765):211–227 34. The Data Warehouse Toolkit, second edition (Wiley), Hoepli,
19. Jakiela MJ, Chapman C, Duda J, Adewuya A, Saitou K (2000) Kimball Ralph, Ross Margy 2003.
Continuum structural topology design with genetic algorithms. 35. Kanellos A (2007) Topological self-organisation: using a particle-
Comput Methods Appl Mech Eng 186(2):339–356 spring system simulation to generate structural space-filling lattices
20. Pedersen P, Pedersen NL (2006) Reflections on truss and continu- University College London
um topology optimal designs. In: IUTAM symposium on topolog- 36. Hanna S, Haroun Mahdavi S Modularity and flexibility at the small
ical design optimization of structures, machines and materials. scale: evolving continuous material variation with
Netherlands: Springer, p. 67–76. stereolithography ACADIA: New Materials and New Industry.