Design for additive manufacturing: trends, opportunities, considerations, and constraints

TLDR: In the case of aircraft components, AM technology enables low-volume manufacturing, easy integration of design changes and, at least as importantly, piece part reductions to greatly simplify product assembly.

Abstract: The past few decades have seen substantial growth in Additive Manufacturing (AM) technologies. However, this growth has mainly been process-driven. The evolution of engineering design to take advantage of the possibilities afforded by AM and to manage the constraints associated with the technology has lagged behind. This paper presents the major opportunities, constraints, and economic considerations for Design for Additive Manufacturing. It explores issues related to design and redesign for direct and indirect AM production. It also highlights key industrial applications, outlines future challenges, and identifies promising directions for research and the exploitation of AM's full potential in industry.

Figures

Fig. 16. Brackets before and after topology optimization: Airbus A320 nacelle hinge brackets as-designed for cast steel and optimized for titanium (left) and Airbus A380 brackets as designed and optimized for stainless steel (right) [105]. The optimized brackets were produced by direct metal laser sintering (DMLS).

Fig. 19. Customized laser sintered foot orthoses from Materialise’s AFootprint project (left) and customized selective laser sintered wrist splint produced by Fraunhofer IPA. Images via [251].

Fig. 17. Types of customization. Redrawn from [59].

Fig. 14. Thermal conditioning ring with milled cooling channels enclosed by a welded cover (left) and with additively manufactured conformal cooling channels (right). Courtesy of ASML.

Fig. 18. Titanium implants for the skull (left, [103]) and pelvis (right, [107]) produced using and EOSINT M 280.

Fig. 15. Temperature plots from finite element models of the milled conditioning ring (left) and the additively manufactured conditioning ring (right). Shown with the same temperature scale. Courtesy of ASML.

Full text

Design for Additive Manufacturing: Trends, Opportunities, Considerations and Constraints
Mary Kathryn Thompson
a
, Giovanni Moroni
b
(2), Tom Vaneker
c
(2), Georges Fadel
d
, R. Ian Campbell
e
, Ian Gibson
f
,
Alain Bernard
g
(1), Joachim Schulz
h
(3), Patricia Graf
h
, Bhrigu Ahuja
i
, Filomeno Martina
j
a
Department of Mechanical Engineering, Technical University of Denmark, Kgs. Lyngby 2800, Denmark
b
Department of Mechanical Engineering, Politecnico di Milano, Italy
c
Faculty of Engineering Technology, University of Twente, Netherlands
d
Department of Mechanical Engineering
,
Clemson University, Clemson, SC 29634, USA
e
Loughborough Design School, Loughborough University, UK
f
School of Engineering, Deakin University, Australia
g
Université Bretagne Loire, Ecole Centrale de Nantes, IRCCyN UMR CNRS 6597, France
h
Aesculap AG, Germany
i
LPT, Institute of Photonic Technologies, Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany
j
Welding Engineering and Laser Processing Centre, Cranfield University, Cranfield, MK43 0AL, UK
The past few decades have seen substantial growth in Additive Manufacturing (AM) technologies. However, this growth has mainly been process-driven. The
evolution of engineering design to take advantage of the possibilities afforded by AM and to manage the constraints associated with the technology has lagged
behind. This paper presents the major opportunities, constraints, and economic considerations for Design for Additive Manufacturing. It explores issues related to
design and redesign for direct and indirect AM production. It also highlights key industrial applications, outlines future challenges, and identifies promising
directions for research and the exploitation of AM’s full potential in industry.
Design, Manufacturing, Additive Manufacturing
1. Introduction
The evolution of Additive Manufacturing (AM) over the past three
decades has been nothing less than extraordinary. AM has
experienced double-digit growth for 18 of the past 27 years, taking it
from a promising set of uncommercialized technologies in the early
1980s to a market that was worth over $4 billion in 2014. The AM
market is expected to grow to more than $21 billion by 2020
[354][355]. This growth has been made possible by improvements in
AM materials and technologies and is being driven by the market
factors that necessitate its use such as shorter product development
cycles, increasing demand for customized and personalized products,
increased focus and regulations on sustainability, reduced
manufacturing cost and lead times, and the introduction of new
business models [13][354][355].
During the past thirty years, the use of AM technology has also
undergone a transformation. Early AM applications focused on
models and prototypes [178][179]. As the technology matured, AM
played a major role in producing rapid and soft tooling (e.g. vacuum
and silicone casting molds) [187]. Today it is also used for the
production of end use parts and products. It is estimated that the
market for AM end use parts was worth $1.748 billion in 2014 - up
66% from the previous year. Strong double-digit growth in this area is
expected to continue for the next several years [355]. Leveraging the
geometric and material freedoms of AM for end use parts creates a
world of opportunity. However, not all parts are possible or cost
effective to produce using AM. This necessitates a better
understanding of when, why, and how to (re)design for the
opportunities and constraints associated with these technologies.
The CIRP community has previously reported on advances in AM
processes [178][179][187][181][152], their role in rapid product
development [42], and how they have been used in the biomedical
[36] and turbomachinery [176] industries. This paper explores the
opportunities, constraints, and economic considerations related to
Design for Additive Manufacturing (DfAM). It begins with a brief
overview of Additive Manufacturing, Design for Manufacturing, and
the need for DfAM. It presents the main design opportunities,
considerations and constraints related to AM technologies, including
production time and cost. It presents DfAM success stories from a
number of industries. Finally, it identifies promising directions for
research and development that will enable Design for Additive
Manufacturing to reach its full potential in industry.
2. Additive Manufacturing
Additive Manufacturing processes produce physical objects from
digital information piece-by-piece, line-by-line, surface-by-surface, or
layer-by-layer [178][130]. This simultaneously defines the object’s
geometry and determines its material properties. AM processes place,
bond, and/or transform volumetric primitives or elements (voxels) of
raw material to build the final part. Each voxel’s shape and size and
the strength of the bonds between the voxels are determined by the
raw material(s), the manufacturing equipment (e.g. the build platform
precision, nozzle geometry, light or laser beam wavelength, etc.), and
the process parameters (e.g. the nozzle temperature, light or beam
intensity, traverse speed, etc.). The overall part geometry is
determined by tool paths, projection patterns (digital masks), or a
combination of the two. This allows AM technologies to fabricate
parts without the need for intermediate shaping tools [155].
AM processes are characterized by increasing workpiece mass.
They represent one of three major classes of manufacturing
technologies, along with subtractive processes where the workpiece
mass is reduced and formative processes where the workpiece mass is
conserved [125][26]. Additive Manufacturing processes are also
distinct from chemical and thermal processes such as etching, plating,
oxidation, and heat treatment, which act on all exposed (reactive)
surfaces and traditional processes to create composite materials.
Contents lists available at SciVerse ScienceDirect
CIRP Annals Manufacturing Technology
Journal homepage: www.elsevier.com/locate/cirp

2.1 History of Additive Manufacturing
The foundations of Additive Manufacturing go back almost 150
years, with proposals to build freeform topographical maps and
photosculptures from two-dimensional (2D) layers [40][256][48].
Research efforts in the 1960s and 70s provided proof of concept and
patents for the first modern AM processes including
photopolymerization in the late 1960s [356], powder fusion in 1972
[72], and sheet lamination in 1979 [243]. This work was enabled by
the invention of the computer in the late 1940s, the development of
photopolymer resins by DuPont in the 1950s, and commercial
availability of lasers in the 1960s. It followed advances in computer
aided design (CAD) and manufacturing (CAM), including the
development of numerical control machine tools in the early 1950s,
computer graphics and CAD tools in the early 1960s, CAD/CAM
systems in the late 1960s, and the availability of low cost computer
monitors starting in early 1970s [71][356][258]. However, the
technology was in its infancy with no commercial market and little
support for research and development activities.
The 1980s and early 1990s saw an increase in patents and academic
publications; the development of new technologies such as MIT’s 3D
printing process in 1989 [130], laser beam melting (LBM) processes
in the early 1990s [287], and the successful commercialization of
process technologies including stereolithography (SL) in 1988; fused
deposition modelling (FDM), solid ground curing, and laminated
object manufacturing in 1991 [356]; and laser sintering in 1992 [287].
These advances were made possible, in part, by improvements in
geometric modelling capabilities [71] and the development of
programmable logic controllers [130] during the 1960s and 1970s, the
development of ink jet printing technology in the late 1970s [130],
and by the decreased cost and improved capabilities and availability
of computers and CAD/CAM systems in the 1980s [256]. However,
the high cost, limited material choices, and low dimensional accuracy
of these machines limited their industrial application to rapid
prototyping and model making.
The 1990s and 2000s were a period of growth for AM. New
processes such as electron beam melting (EBM) [22] were
commercialized, existing technologies were improved, and attention
began to shift to developing AM related software. AM-specific file
formats such as STL (StereoLithography), CLI (Common Layer
Interface), LEAF (Layer Exchange Ascii Format), and LMI (Layer
Manufacturing Interface) [256] were introduced. AM-specific
software programs, such as Clemson’s CIDES (1990) and
Materialise’s Magics (1992) were developed. New generations of
commercial systems offered new and improved features. Quality
improved to the point that Additive Manufacturing technologies could
be used to produce patterns, tooling, and final parts. The terms ‘Rapid
Tooling’, ‘Rapid Casting’, and ‘Rapid Manufacturing’ were created
to highlight the ability to use Additive Manufacturing technologies
for production. Cheap, powerful computers helped to make new
generations of AM machines smaller and more affordable [131].
Advances in solid modelling software made it easy and inexpensive
for students and professionals to design and model 3D objects.
Finally, the Internet made knowledge sharing easy and supported the
development of open-source hardware and software. This led to the
development of the first hobby AM machines from the RepRap
project in 2005.
The late 2000s saw the commoditization of the AM processes that
were commercialized in the 1980s and were a period of growth for
the younger metal-based AM processes. The expiration of key patents
for a number of older AM processes opened the market to
competition. This, combined with a growing AM hobby community,
spurred innovation, leading to a major expansion of market supply
and demand. Today, AM products and services support a wide range
of activities including manufacturing, energy, transportation, art,
architecture, education, hobbies, space exploration, and the military.
Wide scale adoption of AM for the direct manufacture of final parts
has occurred in the medical, dental, and aerospace industries.
Meanwhile, commercial hobby printers and entry-level professional
machines have made AM technology available to the masses.
If the current trends continue, we will soon enter a new stage of
evolution where Additive Manufacturing becomes a design paradigm
in addition to a means of production.
2.2 Digital workflow for Additive Manufacturing
Additive Manufacturing processes have a digital dataflow that
generates the instructions for the AM machine followed by a physical
workflow that transforms the raw materials into final parts (Fig. 1).
The process usually begins with a product idea, a 2D image such as a
photograph, a set of 2D images like those derived from Computed
Tomography (CT) scans, or a physical 3D object like a prototype or a
part for reverse engineering. These are transformed into digital
models (e.g. volume models or facet models) using solid modelling,
metrology, or image reconstruction software. Next, the data is
checked for errors, the errors are corrected, and support structures are
added if needed. This is often done with AM-specific software such
as Magics from Materialise NV. Finally, the model is sliced or
otherwise discretized to create instructions for the machine. This is
often done using machine-specific software.
New software formats have been developed and standardized to
support AM data preparation and digital workflow. For example, the
AMF format, which has native support for color, materials, lattices,
and constellations, has been standardized and is intended to replace
the STL format. Other formats such as STEP, STEP-NC, and 3MF
have integrated AM concepts to compete with AM-specific formats.
Kim et al. [174] recently proposed a systems approach for data flow
structuring and decomposition in several steps, clarifying the need for
data generation and transformation along the AM digital chain.
Fig. 1. Digital and physical workflow from product idea to actual component.
Redrawn from [337].
product idea
2D image
of object
physical model
3D digitalization
point cloud
Polygonisation/
triangulation
reverse
engineering
surface model
3D CAD
modeling
volume model
polygonisation/
triangulation
facet model
slicing process
sliced contour
data
additive fabrication
process
component
result
process
ASCII format
DXF
format
STEP, IGES,
VDA-FS
format
STL, VRMLformat
data preparation data acquisition

Fig. 2. Additive Manufacturing process families and materials [155].
2.3 Additive Manufacturing processes and physical workflow
The physical workflow begins with one of the seven currently
recognized groups of AM technologies: binder jetting, directed
energy deposition, material extrusion, material jetting, powder bed
fusion, sheet lamination, and vat polymerization (Fig. 2) [26][155].
AM processes can be used for the direct production of models,
prototypes, end use parts, and assemblies, as well as fixtures, patterns,
and tooling for indirect production [155][337][66][71]. AM can be
integrated to create hybrid processes [163][166][168][182][317] or
combined with other processes to form longer multi-stage process
chains [149][327][337]. For example, parts can be printed to near net
shape and then post-machined (Fig. 3), molds can be produced by
alternating printing and machining operations (Fig. 4), features can be
printed on top of formed components [14], and components can be
embedded within printed parts (Fig. 5 and Fig. 6).
Each process family has distinct operating principles, production
characteristics and compatible material types. These traits affect the
cost, quality, and sometimes the color and scale of the parts that can
be produced, and therefore can substantially impact design decisions.
The consideration of process specific characteristics during the design
process is even more important when AM is combined with other
direct manufacturing processes (e.g. machining) and indirect
manufacturing processes (e.g. molding or casting) [43].
2.4 Current AM standards
Working groups for the development of AM-related standards have
been organized by the International Organization for Standardization
(ISO/TC 261) and the American Society for Testing and Materials
(ASTM F42). To date, they have produced standards related to
terminology, individual processes, chains of processes (hardware and
software), test procedures, quality parameters, customer-supplier
agreements, and fundamental elements. Recent additions address data
processing [156] and consider the relevance of and specify variations
to existing standards [27][28] (Fig. 7). In 2013, ISO and ASTM
defined a common goal to produce one set of global standards
including general standards that are applicable to most AM materials,
processes, and applications; category standards that define the
requirements for a material or a process category; and specialized
standards for specific requirements to a material, process or
application [158]. AM standardization efforts are also taking place in
Germany (VDI FA 105 and DIN NA 145-04-01AA), Spain
(AEN/CTN 116), France (AFNOR UNM 920), Sweden (SIS/TK
563), the US (SAE AMS-AM) and the UK (BSI AMT/8). The
Association of German Engineers published VDI 3404 and VDI 3405
as part of this work.
AM standards provide a common understanding of the field and a
shared lexicon from which to work. This is important for developing
and using AM-related design tools and methodologies. It is also a pre-
requisite for developing design related AM standards. For example,
ISO/ASTM DIS 20195 “Guide for Design for Additive
Manufacturing” [157] is currently under development.
Fig. 3. Outboard landing gear rib (24 kg) produced in Ti–6Al–4V by Wire +
Arc Additive Manufacturing (WAAM): CAD model (left, courtesy of the
Welding Engineering and Laser Processing Centre at Cranfield University)
and printed part before machining (right, [352]).
Fig. 4. Injection molding tooling produced by 3-axis Hybrid Layered
Manufacturing (Gas Metal Arc Welding plus CNC machining): CAD model
(left), near net shape molds (center), and finished molds (right) [317].
Fig. 5. Conformal cooling channels in an injection molding die. The cooling
tubes were inserted into the substrate mold (left), the tubes were ‘buried’ and
the die was completed using a laser-aided metal-based AM process (center),
and the final tool was post-machined (right). Adapted from [59].
Fig. 6. Timer circuit with embedded electronic components produced using a
hybrid stereolithography / direct print (SL/DP) machine [193].
3. Design for Additive Manufacturing
The term ‘Design for Additive Manufacturing’ has been used
extensively in the literature [10][19][31][70][77][74][91][122][142]
[150][262][284][335][336], however there have been only a few
attempts to define it [271][272][130]. This section provides an
overview of classical Design for Manufacturing and Assembly
(DfMA), examines the suitability of that definition and framework for
AM applications, and outlines the need for the development of
Design for Additive Manufacturing expertise and education.
3.1 Design for Manufacturing and Assembly
DfMA is the practice of designing and optimizing a product
together with its production system to reduce development time and
cost, and increase performance, quality, and profitability. This is done
Vat photo-
polymer-
ization
Material
jetting
Binder
jetting
Powder
bed
fusion
Material
extrusion
Directed
energy
deposition
Sheet
lamination
Thermoset
Polymers
Epooxies and
acrylates
X X
Thermo-
plastic
polymers
Polyamide,
ABS, PPSF
X X X X X
Wood paper X
Metals
Steel,
Titanium
alloys, Cobalt
chromium
X X X X
Industrial
ceramic
materials
Alumina,
Zirconia,
Silicone
nitride
X X X X
Structural
ceramic
materials
Cement,
Foundry sand
X X X
Process categories
Materials
Example
materials
Note: Combinations of the above material classes, e.g. a composite, are possible

by “simultaneously considering design goals and manufacturing
constraints” [168] such as “user and market needs, materials,
processes, assembly and disassembly methods,” maintenance
requirements, etc. [228]. DfMA can be viewed from three levels of
abstraction. At the first level, DfMA offers concrete tools, techniques,
and guidelines to adapt a design to a given set of downstream
constraints. These are usually process-specific (e.g. Design for
Injection Molding) [46][260], feature-specific (e.g. how part size,
weight, and symmetry affect insertion/assembly time) [46], or
activity-specific (e.g. how to calculate the theoretical minimum
assembly time) [45]. At the next level of abstraction, DfMA aims to
understand and quantify the effect of the design process on
manufacturing (and vice versa). This is needed to improve the
performance of the manufacturing system, the execution qualities of
the product (cost, functionality, customer satisfaction, etc.), the
evolution (through-life) qualities of the product (safety, reliability,
service and repair costs, etc.), and the long-term potential of the
associated business case (e.g. the ability to respond to unexpected
surges in product demand) [20]. In this context, DfMA is a subset of
Design for X [183]. At the highest level, DfMA explores the
relationship between design and manufacturing and its impact on the
designer, the design process, and design practice. In this context, it
addresses topics such as material and process selection, concurrent
engineering [231][291], and how to improve CAD to support DfMA
[46].
General AM Standards (general concepts, common requirements, generally applicable)
Terminology
Processes / Materials
Test Methods
Design / Data Format
ASTM F 2792
ISO 17296-2
ISO 17296-3
ISO 17296-4
ISO / ASTM 52921
ASTM F 2971
ISO / ASTM 52915
ASTM F 3122
ISO / ASTM DIS
20195 DRAFT
Raw Materials
Process / Equipment
Finished Parts
Materials Category-
Specific
Process Category / Materials
Specific
Standard Protocols for
Round Robin Testing
Metal powders, polymer
powders, polymer resins,
ceramics, etc.
Powder Bed Fusion, Material
Extrusion, Directed Energy
Deposition, etc.
Mechanical Test Methods,
Parts Specification, etc.
ASTM F 3049
ASTM F 3091 / F3091M
Materials-Specific
Standards
Process/Materials-Specific
Standards
Application-Specific
Standards
Material-Specific Size
Specification, Material-
Specific Chemical
Composition, Material-
Specific Viscosity
Specification, etc.
Process-Specific Performance
Test Methods, Process-
Specific Performance Test
Artifacts, System Component
Test Methods, etc.
Aerospace, Medical,
Automotive, etc.
ASTM F 2924
ASTM F 3001
ASTM F 3055
ASTM F 3056
Fig. 7. ASTM and ISO standards for AM. Updated and modified from [158].
3.2 The need for Design for Additive Manufacturing
The definition of DfMA above is valid for all processes and process
chains that involve AM. However, in practice the design knowledge,
tools, rules, processes and methodologies at all three levels of
abstraction will be substantially different for DfAM than traditional
DfMA. For example, AM can create different types of features and
impose different types of constraints than other manufacturing
processes. Therefore, they require different process-specific design
rules and tools [10][70][74][77][130][139][142][150][261][262][335]
[336]. At the same time, the freedoms of AM reduce the need for, and
therefore the importance of, designing for activities such as assembly
[149]. AM processes have different batch sizes, production times, and
cost drivers than traditional processes [29][148][275][276][366] and
require different approaches to metrology and quality control
[224][274]. Therefore a new body of knowledge is required to
support DfAM. Finally, the unique characteristics of AM processes
allow for and require different approaches to the design process and
design practice [31][138][130][284][126]. This includes new
approaches to explore large, complex design spaces
[70][271][272][348]; to incorporate material, mesostructures and
multi-scale design considerations [130][271][272]; and to overcome
the “cognitive barriers” imposed by past experience and the
conventional fabrication techniques [284].
The development of DfAM knowledge, tools, rules, processes and
methodologies has been cited as one of the technical principle
challenges of AM [19]. Insufficient understanding and application of
DfAM is said to be limiting the overall penetration of AM in industry
[122], holding back the use of AM for the production of end-use parts
[10][122], preventing designers from fully benefitting from AM
[91][126], and preventing AM from reaching its full potential in
general [31][74]. Once Design for Additive Manufacturing is well
understood, that knowledge must be disseminated to current and
future members of industry. Thus, AM-specific design education
[19][122][150] and design standards [19] are also needed.
4. Design opportunities, benefits, and freedoms of AM
This section provides an overview of design opportunities, benefits
and freedoms associated with Additive Manufacturing. These have
been divided into three levels: the part level with macro scale
complexity, the material level with micro scale complexity, and the
product level with multi-scale complexity. Production and business
level benefits are discussion in section 6.
4.1 Design freedoms at the part level with macro scale complexity
Incorporating the material and geometric freedoms of AM into
macro scale parts can provide a variety of aesthetic, functional,
economic, emotional, and ergonomic benefits.
4.1.1 Material choice
AM technologies can process a large range of materials.
Commercial AM machines can process polymers, metals, and
ceramic materials [155]. Sheet lamination processes are compatible
with paper, wood, cork, foam, and rubber [34]. Investment casting
molds and cores have been printed in sand [343] and large structures
have been printed in clay and concrete [171][173]. Research to print
Lunar and Martian habitats using locally available materials such as
lunar regolith is also underway [172]. Various AM processes have
been used to print edible items such as chocolate, sugar, frosting,
pasta, spreads, cheese, scallop puree, ground beef, egg whites, insect
powders, and an entire pizza. Much of this work is motivated by the
desire to produce novel shapes, flavors and textures; to provide
personalized nutrition; to enhance the quality of life for individuals
who have difficulty swallowing; to increase food supply security; and
to improve dining in outer space [350][315][192]. (Some AM foods
must be cooked, baked, or fried before consumption.) AM has also
been used to print biological and bio-compatible materials such as
cells, proteins, synthetic hydrogels, biological hydrogels, and
bioactive glasses [36]. This work could ultimately enable additive
manufacture of tissues and organs.
4.1.2 Color
Some AM processes can create products in full color (Fig. 8). This
can be done by adding color to the raw materials (e.g. by ink jet
printing on paper or powder), by using different color feedstock for
different parts of the model, or by inducing color change in a single
feedstock (e.g. resin) by in-process activation of pigments
[169][263][318]. Additively manufacturing parts in color can reduce
or eliminate downstream painting and decoration steps during

production and reduce chipping and flaking. In rapid prototyping and
model making, color can be used as a communication tool to
highlight features such as tumors in medical models and to map
analytical data onto objects to make the information easier to
understand and discuss [303][332].
Fig. 8. AM objects in full color: frog and toad models printed using paper-
based selective deposition lamination on an Mcor IRIS and colored to appear
as aged copper (top left) [215]; bicycle seat colored to show simulated
pressure distribution from a rider printed on an Objet Connex3 (top right)
[294]; plates showing a 9x9x9 set of color options from a ZCorp ZPrinter 650
before and after brushing (lower left) [92]; and a surgical planning model of a
human liver printed on an Objet Connex3 in clear and colored resins [303].
Fig. 9. Jewelery produced with AM: award winning Tiger Ring from OG-Art -
pattern printed in wax on a Solidscape machine (via [34]) (left), Kinetic Ring
from Vulcan Jewelry (available for purchase) (center, courtesy of Vulcan
Jewelery); custom R2D2 inspired ring from Uptown Diamond and Jewelry -
pattern printed in wax on a 3D Systems ProJet machine [4] (right).
Fig. 10. Home furnishings produced with AM: the Monarch Stool from Future
Factories (left, via [90]), Quin.Mgx Pendant Light from Bathsheba Grossman
printed in polyamide using SLS (available for purchase) (center, courtesy of
Bathsheba Sculpture LLC), and decorative bowl by Carl Bass printed in
stainless steel and bronze on an ExOne metal binder jet printer (available for
download) (right, [114]).
Fig. 11. AM in the fashion industry: dress from Iris van Herpen’s Voltage
haute couture collection produced using laser sintering (left [208]), one-of-a-
kind purse from Kipling produced using laser sintering (center, [210]), and
Mutatio shoes by Francis Bitonti produced using SLS and then gold plated
(available for purchase) (right, courtesy of Francis Bitonti Studio).
4.1.3 Freeform geometry for art and aesthetics
AM’s ability to create unique, intriguing, and appealing geometric
forms has led to its adoption by artists, artisans, and industrial
designers. For example, AM is used in the jewelry industry for direct
production [104][218] and to produce patterns for investment casting
[94][97] (Fig. 9). It is also being used to enrich interior design with
high-end furniture, lighting fixtures, and accessories (Fig. 10) and to
explore new forms for clothing, shoes, purses, and other accessories
in the fashion industry (Fig. 11). In the past, AM applications that
emphasized form were mainly intended for exploration and
exhibition. However, additively manufactured designs are becoming
increasingly available for purchase and use.
4.1.4 Internal freeform geometry for functionality and performance
Additive Manufacturing enables the creation of complex internal
features to increase functionality and improve performance. For
example, AM has been used to create integrated air ducts
[41][101][311][209] and wiring conduits [209] for industrial robots;
3D flexures for integrated actuators and universal grippers [134],
complex internal pathways for acoustic damping devices [285];
optimized fluid channels (Fig. 12), and internal micro vanes for
ocular surgical devices [69]. However, one of the most widely studied
applications is conformal cooling. Conformal cooling channels follow
the external geometry to provide more effective and consistent heat
transfer (Fig. 13). Early research [280][359][129][267] showed that
conformal cooling in injection molding tooling improves process
efficiency and quality. Industrial injection molding case studies have
confirmed these benefits with reports of reduced lead time, more
uniform temperature distributions, reduced cycle times, improved
quality, reduced reject rates, reduced corrosion, longer maintenance
intervals, and overall cost savings [98][108][112].
Fig. 12. Solid model of a water redistribution manifold redesigned for AM:
original design made in PEEK with perpendicular drilled channels (left) and
optimized version printed in titanium (right). The redesign reduced turbulence
induced vibration forces by 90%. Images courtesy of ASML.
Fig. 13. Schematic of conventional cooling channel (left) and conformal
cooling channel (right). Adapted from [17].
Conformal cooling is not limited to tooling. Fig. 14 shows two
versions of a thermal conditioning ring from the semiconductor
industry. The original design has circular cooling channels milled into
the outer circumference of the ring and enclosed by a welded cover
plate. The redesigned version was optimized for performance by
incorporating additively manufactured conformal cooling channels on
the top and side surfaces of the ring. The thermal behaviour of the
two rings is shown in Fig. 15. The redesign improved temperature
uniformity across the top surface of the ring by more than 6x,
reducing the temperature range across the top face from 13.8 milli-
Kelvin (mK) to 2.3mK and the temperature range over the thickness
of the ring from 22mK to 3.7mK.

Frequently asked questions

Q1. What are the contributions mentioned in the paper "Design for additive manufacturing: trends, opportunities, considerations and constraints" ?

In this paper, the main design opportunities, considerations and constraints related to AM technologies, including production time and cost, are discussed. 

Q2. What have the authors stated for future works in "Design for additive manufacturing: trends, opportunities, considerations and constraints" ?

This section explores some of the future challenges and coming trends that will shape DfAM and the technology it will enable. 

Q3. What are the three categories of build time models?

Existing build time models can be grouped into 3 categories: models dedicated to one process using a limit set of parameters; generic build time models that use many parameters to estimate build times; and parametric models that use neural networks to predict production times based on historic data. 

Q4. Why is AM being used to create custom objects?

Because AM simultaneously creates an object’s material and geometry, it can be used to create custom alloys and composite materials. 

Q5. What are the two common methods to overcome some of the bulk geometric limitations of legacy CAD systems?

Two common methods to overcome some of the bulk geometric limitations of legacy CAD systems are haptic modelling and reverse engineering. 

Q6. What is the way to maximize the business benefits of AM?

maximizing the business benefits of AM requires a through-life approach that considers production, use, maintenance, repair, and disposal. 

Q7. What are the challenges of creating freeform external geometries?

The organic, freeform external geometries that can be created by AM require more complex measurement techniques and greater data processing capabilities. 

Q8. How many different FDM samples were able to estimate the build time?

After 72 training cases, they were able to estimate the build time of six different FDM samples with errors ranging from 6.07 to 20.3%. 

Q9. What are the main features of the design rules for laser sintering?

In the academic literature, Adam and Zimmer [10] presented a catalogue of design rules for laser sintering, laser melting, and FDM that address geometric constraints such as sharp edges, element transitions, unsupported features, and feature spacing. 

Q10. What is the common strategy for maximizing the strength of a part?

Designers typically compensate for these mechanical effects by orienting the part to maximize its strength during the build, by adding support structures to the part, or by designing the part to be self-supporting throughout the printing process. 

Q11. How many pieces of plastic was economical for the smaller part?

It was estimated that LS was economical up to 14,000 pieces for the smaller part (Fig. 50) and up to 700 pieces for the larger part. 

Q12. How much did they estimate the cost of a single DMLS assembly?

They estimated the cost of a single DMLS assembly to be 526.31 EUR (material cost 5%, pre-processing cost 1%, build costs 90%, post processing cost 4%) vs. 21.29 + 21,000/N EUR for HDPC. 

Q13. What was the cost of a part produced by LS and injection molding?

It indicated that the cost per AM part was driven by the production speed and the break-even point between LS and injection molding was driven by part size. 

Q14. What are the main challenges when designing for heterogeneity?

there remain many challenges when designing for heterogeneity taking into account the shape and material distribution in order to meet the functionality, requirements or constraints of the artifact. 

Q15. How many input variables were used to obtain a good estimate of build time?

This approach requires very few input variables to obtain a good estimate of build time (generally conservative and within 12% of the actual build time), however only one type of machine was used and the settings were kept constant. 

Q16. How much do cost models assume for AM products?

Most cost models assume higher production volumes for a single design and therefore underestimate the labor costs of AM products. 

Q17. What are the options for converting solid models of truss structures?

Past solutions have involved complete solid models of truss structures using geometric modeling kernels such as ACIS [347], algorithms [24][88][140], and unit cell libraries [16][70][246] (Fig. 37). 

Q18. What was the main reason for the rapid prototyping and model making of AM?

the high cost, limited material choices, and low dimensional accuracy of these machines limited their industrial application to rapid prototyping and model making.