Materials and Design 172 (2019) 107755

Contents lists available at ScienceDirect

Materials and Design

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

The GAP methodology: A new way to design composite structures

Florian Neveu, Bruno Castanié ⁎, Philippe Olivier
Institut Clément Ader (ICA), Université de Toulouse, CNRS UMR 5312 - INSA - ISAE Supaero - IMT Mines d'Albi - UPS, Toulouse, France

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• A new methodology is proposed for de-

signing composite structures based on
generation of concepts and specific as-
pects of composite structures
• Key information is provided on the rela-
tionships between design geometry, ar-
chitectures and manufacturing
processes for composite materials
• Two examples illustrate the methodol-
ogy revealing the large number of solu-
tions that composite structures can offer

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

Article history: With the growing demand for composite parts, more and more designs or re-designs are required to use such
Received 15 February 2019 materials. This paper presents a new methodology for designing composite structures by considering three es-
Received in revised form 21 March 2019 sential variables: the geometry, the architecture and the manufacturing process. The methodology proves to be
Accepted 22 March 2019
useful during the pre-design phase of the project, when important decisions have to be taken regarding the
Available online 28 March 2019
choice of the manufacturing route and the investments needed. The methodology starts from the design (the ge-
Keywords:
ometry of which relies on the designer's creativity) and then gives clear insight into the possibilities offered by
Design methodology the concepts from a composite point of view. These include the types of architecture and technologies that are
Composite materials compatible as well as the manufacturing processes. The GAP methodology (Geometry, Architecture, Process) is
Manufacturing processes intended to provide an efficient way to start a composite project by giving a broad overview of the possibilities
Conceptual design and by reducing the risks of poor decisions at the very beginning of the project. In this article, innovative compos-
ite structure design issues will be presented before the methodology itself is explained. Two generic case studies
are then presented to illustrate how the GAP methodology works.
© 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction significant for specific applications: in high-end products for example,

where the mechanical properties of the structure are of utmost impor-
During recent decades, the use of composite materials has been con- tance, the mass to stiffness ratio can be a strong argument for choosing
stantly increasing in various fields of applications from the recreational a composite structure over a metallic one. Some figures have been given
to the aerospace sector. The advantages of composite materials can be to demonstrate the positive impacts of weight savings in the automo-
tive industry, regarding decreases in fuel consumption and CO2 emis-
⁎ Corresponding author. sions [1]. The aim of lightweight construction is to preserve or even
E-mail address: [email protected] (B. Castanié). expand a product functionality while reducing its overall weight.

https://doi.org/10.1016/j.matdes.2019.107755
0264-1275/© 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2 F. Neveu et al. / Materials and Design 172 (2019) 107755

Existing approaches for reducing mass include the use of less dense ma- Table 1
terials (metal foams, composite materials) or a decrease in the material Savings with the “Starflex” conception.

volume by reducing wall thickness in key structural components. This Alouette III (1957) Ecureuil (1974) Savings (%)
has led to a huge development of composite parts in recent vehicles Number of parts 377 70 81
and the trend is still rising. Despite the increase of both manufacturing Weight (kg) 105 (kg) 55 48
and maintenance costs, the benefits obtained on the mass are still Number of bearings 30 0 100
largely profitable. For example in the two most recent programmes of Number of seals 45 0 100
Grease points 22 0 100
Boeing and Airbus, for the 787 and the A350 respectively, composites
Cost 100 45 55
make up about 50% of the mass and about 90% of the aerodynamic
wet surface. Another major reason for the use of composite materials
is that they make it possible to create simpler multifunctional structures obtained after a manufacturing phase, which depends on industrial con-
thus drastically reducing the number of parts. A historic example is the straints such as the scale of production, the certification requirements,
new rotor hub for helicopters, designed by Aérospatiale (now known as the availability of materials, the size of the part, the cost, and many
Airbus Helicopters) in the 1970s, see Fig. 1 [2]. The benefits of this new other constraints. Nevertheless, a limited number of authors have pro-
design are shown in Table 1. It was possible to make weight savings and posed general methods. Monroy Aceves et al. [23] have developed a
reduce part manufacturing costs while also greatly reducing mainte- methodology to help designers select a shortlist or optimum design of
nance costs. For instance, the introduction of composite blades in heli- composite structure from a large number of alternatives and have ap-
copters permitted a cost reduction of 13% [2]. The cost of possession is plied it to a wind turbine blade [24]. The link between composite struc-
often lower for composite structures thanks to their non-sensitivity to ture design and manufacturing methods remains to be established from
corrosion and fatigue if they are correctly designed. More recently, a general point of view. Hambali et al. [25] have proposed a selection
Carello et al. developed an innovative CFRP transverse leaf spring [3] method for ranking five types of processes for manufacturing an auto-
with a weight reduction of 75% and a number of parts that was reduced motive bumper beam based on six main selection factors and twelve
from four to one (Fig. 2). Elixir Aircraft is currently designing an aircraft subfactors. Bader [26] has assessed the manufacturing costs of a simple
for general aviation implementing carbon fibres and a “one shot” wing. component with different composite materials and by different a priori
The company claims to have reduced the number of parts from 17,000 manufacturing routes in the context of aeronautics. A simple compari-
(for an equivalent metallic built aircraft) to 600 with their full carbon son methodology is introduced on the basis of cost-performance effi-
composite solution [4]. ciency. Eventually, engineers wishing to totally or partially redesign
Thus, composite part manufacturers or industries aiming to move to their product are faced with a very complex set of problems involving
composite solutions are facing challenges in composite design. The large a large number of possibilities and challenging objectives such as a re-
choice of material compositions (i.e. matrix and fibres combinations), duction of the number of parts (and, if possible, converging to a “one
architectures and manufacturing processes makes the design process shot” solution [4]) with lower manufacturing costs. Fig. 3 summarizes
complex and difficult, as the designers face a hyper-choice of materials the dilemma introduced here:
and technologies that can be overwhelming. Most of the time, designing Good knowledge and discussion with experts in the field of compos-
composite is understood as, and limited to, the choice of stacking and ite materials and structures will certainly help the engineer to take the
sizing using the TSAI method or derivative, with or without an optimi- right decisions at the right time in the choice of materials and architec-
zation scheme [5–11] because the geometry is already constrained. tures. However, according to the experience of the authors in many
For example, in the latest generation of aircraft, like the B787 or the industrially-driven collaborative projects, the solutions may be strongly
A350, the fuselage geometry is almost the same as in the previous gen- influenced by the past experiences of the experts or the pseudo-experts
eration of metallic aircraft with a thin skin, stiffeners and circumferen- according to the proverb “If all you have is a hammer, everything looks
tial frame [12]. Thus this step in the history of composite in aircraft is like a nail”. This proverb can of course encourage the authors of the pa-
known as “black metal”. In car bodies, the evolution is almost the pers themselves to keep an open mind. So the GAP composite design
same since new composite parts are proposed to directly replace metal- methodology (acronym of Geometry, Architecture, Process) aims to be
lic parts or subcomponents [13]. Of course, introducing such complex a starting point in a composite design process and, in this methodology,
materials remains a strong challenge and this step is necessary. In this we would like to emphasize the importance of creating concepts in suf-
context, some authors have proposed very interesting and useful ficient number and variety to tackle the issue of hyper-choice raised
methods for selecting materials [14–20], in general cases or more partic- above. To introduce the method, this paper first presents its key feature:
ularly, for sandwich structures [21,22]. These approaches will help the creativity in a design process. This is followed by a general overview of
designers but it is also well known that “In composite, materials do composite structure geometries, architectures and processes with key
not pre-exist the structure”, highlighting that a composite part is only information for the development of the methodology. The GAP

Fig. 1. Rotor of the Alouette III (a) and “Starflex” rotor of the Ecureuil (b).
 F. Neveu et al. / Materials and Design 172 (2019) 107755 3

Fig. 2. Initial design (a) and final composite design (b) of an automotive damping system from Carello et al. [3].

methodology itself is then presented and, in the last part, its use is illus- No judgement should be made on the ideas and it is crucial to let the
trated through two case studies. members propose unrealistic ideas which could be a source of inspira-
tion. One alternative to the brainstorming method is the brainwriting,
2. Key features of the GAP methodology also called the 6-3-5 method [30], which is based on the same principle
as brainstorming but can ensure equal participation for all the group
2.1. Creativity in the design process members. Working with analogies can also be a good source of inspira-
tion. It implies finding solutions based on those already known for sim-
It can be said that creativity is the primary quality needed by a de- ilar functions. These analogies are particularly used in bio-inspired
signer in order to imagine innovative solutions. Howard defines the cre- solutions, which are premised on the observation of nature [31]. In con-
ative process as “a cognitive process culminating in the generation of an trast, it is also possible to think by contradiction and then generate solu-
idea” and points out the differences with the design process defined as tions to answer a given problem. A method called Evaporating Cloud
“a labour intensive process culminating in the proposal of a product or [32] takes advantage of these contradictions; its name comes from the
process” [27,28]. While these definitions show a real distinction be- fact that the contradictions will disappear at the end of the methodol-
tween the two processes, we can be certain that the creative process ogy. The TRIZ method (theory of inventive problem solving) [33] is
is clearly embedded within the design process. Many design procedures also well-known for its strong ability to solve technical problems by
are governed by simulations and calculations nowadays. Topology opti- generating innovative solutions based on existing solutions. Creativity
mization [29], for instance, is commonly used and integrated in the de- is fundamental for the methodology presented in this paper as it is the
sign process. It certainly gives a good optimization for a given volume of starting point of the GAP methodology. The first step is to create a
matter, but the resulting shape needs to be reworked for manufactur- large number of pre-concepts with information on only geometric
ability. There are many methods intended to stimulate creativity and boundaries, named the functional surfaces. It is important to scan
help generate ideas and solutions to a given problem. The best known every possibility when it comes to the design of a new product, simply
and most commonly used technique to date is brainstorming, which because the probabilities of finding a good solution are higher if you
brings a group of persons together to work on a subject and think of a pick up ideas from a large number of designs. As stated before, the use
large number of different ideas, possibly creating links between them. of various methods to stimulate creativity can thus prove useful here,

Fig. 3. Dilemma arising in the process design for a composite part.

4 F. Neveu et al. / Materials and Design 172 (2019) 107755

Fig. 4. Shape classification according to CES EduPack [35].

the important thing being to think with total liberty and no restrictions. categories in this classification; prismatic shapes, thin-walled shapes
The importance of concepts should never be underestimated because and bulk shapes, as shown in Fig. 4 [35], where a geometric example
they are the starting point of the final product. Having too few concepts is given for each shape. As a matter of fact, we will consider here that
strongly increases the risk of wasting time on a first bad idea and can re- all these shapes could possibly be manufactured in composite materials.
sult in a poor final solution as stated by professor Ullman [34]: Prismatic parts can be referred to as beams, thin-walled parts as plate or
shell structures and bulk parts could be seen as solids. For example, a
composite fuselage is made of a thin skin (shell), circular frame
(beams) and stringers (beams). These basic shapes have been
manufactured since the beginning of the composite area. It is also possi-
ble to make solid shaped composite parts with very thick laminate like
the Starflex shown in Fig. 1, which is made with about 400 plies of glass
fibres or with recent ultra-thick laminates described in [36,37]. It is also
possible to obtain solid, shaped composite parts by using very thick
sandwich cores, forged carbon or 3D composites [38], or by stitching
2.2. Geometry of composite parts composite subparts together. Thus, today, shape should not be consid-
ered a limitation in composite design and nearly every part shape can
The shapes of metal parts have already been classified in mechanical be achieved. To sum up, in the creative design process of a new compos-
industry in order to be used with different manufacturing routes, espe- ite structure, the geometry can be a good starting point, by thinking
cially in metal and plastic forming industries. There are three main about a beam-, plate- or solid-dominated shape.

Fig. 5. Composite construction technologies or architecture for evaluation of a fairing.

(Reproduced from [39])
 F. Neveu et al. / Materials and Design 172 (2019) 107755 5

Fig. 6. Architectures for an aircraft wing (a) standard sandwich box, (b) Geodesic structure, (c) Multi-web or corrugated integral sandwich structure (reproduced from [40]).

2.3. Architectures of composite parts • Unidirectional or NCF based laminates [41–46]

• Thin or ultra-thin ply based laminates [47–52]
To illustrate the concept of architecture of a composite structure, the • Textile based laminates [53,54]:
reader is invited to consider the work of Krivanek and Yount from NASA - Woven [55–61]
on the issue of “Composite payload fairing structural architecture as- - Braids [61–64]
sessment and selection” [39]. With the shape of the fairing fixed, there - Knitted [65,66]
are many possibilities for making the structure in composite materials,
as illustrated in Fig. 5.
• Glare [67–70]
The authors classified the possible solutions in three relevant catego-
• Fibre steering [71,72]
ries: stiffened, sandwich and hybrid. Another interesting case study is
• 2.5D solutions: laminate with stitching or Z-Pinning [42,73–75]
that of Urik and Malis [40] on an innovative wing for light aircraft. The
solutions compared in this example are a traditional design with sand-
wich skins (Fig. 6a), a geodesic solution (Fig. 6b) and a corrugated sand-
wich structure (Fig. 6c). The second most popular and historic solution is sandwich struc-
This kind of classification can be extended to various other architec- tures which are always a light and soft core with two strong faces. Sand-
tures and as in the preliminary and non-exhaustive review below. The wich structures can themselves be classified as symmetric and
first and most popular architecture is, of course, the laminate but it en- asymmetric, which have different uses in practice. Symmetric sand-
compasses various sub-categories of technological solutions: wiches are for bending or pressurized structures and asymmetric

Table 2
Processes vs prismatic shape.

Prismatic shapes

Solid Hollow

Circular Non-circular Circular Non-circular

Plain Stepped Plain Stepped Plain Stepped Plain Stepped

Manual processes
Hand lay-up x x – – – – – –
Spray-up x x x x x x x x

Compression-based processes
Thermoplastic stamping x x x x x x x x
Compression moulding x x x x x x x x

Injection processes
Reinforced thermoplastics x x • • • • – –
Resin injection moulding x x • • • • – –
Resin transfer moulding x x • • • [124] • [124] – –

Continuous processes
Pultrusion • x • x • x • x
Filament winding x x x x • • • •
Pullwinding • x • x • x • x

Automated processes
Automated tape laying x x x x x x x x
Automated fibre placement x x x x x x x x

•: Normal practice, –: Less-common practice, x: Not applicable.

6 F. Neveu et al. / Materials and Design 172 (2019) 107755

Table 3
Processes vs thin-walled shape.

Thin-walled shapes

Curved Flat

Axisymmetric Non-axisymmetric

Shallow Deep Re-entrant Shallow Deep Re-entrant No cutouts Cutouts

Manual processes
Hand lay-up • • x • [125] • [125] x • [125,126] •
Spray-up • • x • • x • [127,128] •

Compression-based processes
Thermoplastic stamping • [129] – x • [129] – [130] x • [129] –
Compression moulding • – x • – x • –

Injection processes
Reinforced thermoplastics • – x • – x • •
Resin injection moulding • – x • – x • •
Resin transfer moulding • – x • – x • •

Continuous processes
Pultrusion x x x x x x x x
Filament winding x x x x x x x x
Pullwinding x x x x x x x x

Automated processes
Automated tape laying • • x • [131] – x • [131] •
Automated fibre placement • [132] • x • [133] – x • [133] •

•: Normal practice, –: Less-common practice, x: Not applicable.

sandwiches may replace stiffened structures for very lightweight, non- • X-Cor [100]
pressurized parts of helicopters or drones [76,77]. As for laminates, • Hierarchical [99,101]
there are, in fact, many solutions for the core and a non-exhaustive list • Nap Core [102]
is proposed here with some examples for reference: • And probably more…

• Foams, balsa, cork [78–84] In aeronautics or, more generally, for lightweight structures, stiff-
• Honeycomb and other shapes [76–81,85,86] ened structures have been widely used since the 1930s and the inven-
• Plywood [87,88] tion of the “monocoque” fuselage. The main categories are
• Entangled [89,90] [12,39,103–105]:
• Kagome, tetrahedral, pyramidal or other lattice cores [91–95]
• Corrugated [40,96,97] • L or T stringers
• Folded [98,99] • Omega stringers

Table 4
Processes vs bulk shape.

Bulk Shapes

Parallel features Transverse features

Solid Hollow Solid Hollow

Simple Complex Simple Complex Simple Complex Simple Complex

Manual processes
Hand lay-up x x – – – – – –
Spray-up x x x x x x x x

Compression-based processes
Thermoplastic stamping x x x x x x x x
Compression moulding x x x x x x x x

Injection processes
Reinforced thermoplastics • – x x • – x x
Resin injection moulding • – x x • – x x
Resin transfer moulding • [134] – x x • – x x

Continuous processes
Pultrusion • [135] x • [135] x • x • x
Filament winding x x x x • • x x
Pullwinding • x • x x x x x

Automated processes
Automated tape laying x x x x x x x x
Automated fibre placement x x x x x x x x

•: Normal practice, –: Less-common practice, x: Not applicable.

 F. Neveu et al. / Materials and Design 172 (2019) 107755 7

Table 5
Processes vs architectures.

Architectures

Thin laminate Thick laminate Sandwich Tube Geodesic Overmoulded 3D

Manual processes
Hand lay-up • [125,126] – • [136] x x x x
Spray-up • [127,128] – x x x x x

Compression-based processes
Thermoplastic stamping • [129] x – [137] x x • [138] x
Compression moulding x x – [136] x x x x

Injection processes
Reinforced thermoplastics • – – [137] • [135] x • [138] – [135]
Resin injection moulding • • [132] • [136,137] x x x •
Resin transfer moulding • • [134] • [137] • [135] x x • [135]

Continuous processes
Pultrusion • • – [139] • [135] x x • [135]
Filament winding x x x • [135] • [140] x x
Pullwinding x x x • [135] x x x

Automated processes
Automated tape laying • [131] – x x x x x
Automated fibre placement • [132,133] – x x x x x

•: Normal practice, –: Less-common practice, x: Not applicable.

Table 6
Processes vs reinforcements.

Reinforcements

Mat Short fibres UD Fabric Pre-preg Non-crimp fabric 3D braiding 3D weaving

Manual processes
Hand lay-up • [126] • [126] – • [125] • [125] x x x
Spray-up • [127] • [127] x x x x x x

Compression-based processes
Thermoplastic stamping x x • [129] • [129,130] x x x x
Compression moulding • • • • • x x x

Injection processes
Reinforced thermoplastics • • • • • x • [135] –
Resin injection moulding x x • • • • [128] – –
Resin transfer moulding x – [124] • [134] • [134] – [141] • [134] • [135] –

Continuous processes
Pultrusion • • – x • x • [135] x
Filament winding x x – x • x x x
Pullwinding x x – x • x x x

Automated processes
Automated tape laying • • – x – [131] x x x
Automated fibre placement • • x x – [132] x x x

•: Normal practice, –: Less-common practice, x: Not applicable.

• Isogrid or orthogrid Russian industry in this domain have been decisive for this renewal
• PRSEUS [106–109]. A short, non-exhaustive list of technologies is given here:

A renewal of geodesic structures, which were applied for the first • Isogrid
time on the Wellington aeroplane during World War II [12] can be ob- • Lattice structure
served nowadays. In particular, the successful developments by • Isotruss

Fig. 7. GAP methodology structure. Fig. 8. GAP methodology.

8 F. Neveu et al. / Materials and Design 172 (2019) 107755

Fig. 9. Integration of the GAP methodology in the design flow for a generic project [14].

Other kinds of architectures exist, in particular to make thick parts: Thus, the term “architecture” in this paper will refer to these
macro-scale definitions and tentative classifications. For a given
• Forged carbon [110,111]
geometry, many architectures are possible but the inverse is rarely
• 4D thermostructural composites [112]
true. The choice of the architecture is also rather important as it de-
• Textile preform [38,53,54,113–115]
termines the mechanical behaviour of the part (i.e. a sandwich
- Interlock [115]
structure takes advantage of a high bending stiffness compared to
- Braids: tubular, 3D weaving [99,113,114,116,117]
thin laminates), and it also drives the possible manufacturing
- Knitted fabrics [118]
routes.

Fig. 10. Beam-based solutions for an economy seat.

 F. Neveu et al. / Materials and Design 172 (2019) 107755 9

Fig. 11. Shell and plate-based solutions for an economy seat.

2.4. Manufacturing processes thermoset based composites, the following main families can be found
[112,120,121]:
Here, too, the possibilities are numerous. However, the manufactur-
ing processes are closely bound to the materials, the architecture, the • Hand lay-up and autoclave
way fibres are placed and the use of open or closed moulds. According • ATL or AFP and autoclave
to Astrom [119], manufacturing processes are mainly classified as • Filament winding
thermoset-matrix techniques or thermoplastic-matrix techniques. For • Out-of-autoclave processes:

Fig. 12. Shell and plate-based solutions for an economy seat.

10 F. Neveu et al. / Materials and Design 172 (2019) 107755

Fig. 13. Mixed solutions beam/shell/solid for an economy seat.

- Resin Transfer Moulding different approach to classify these manufacturing processes. It depends
- Liquid Resin Infusion on the main characteristic of the process: manual, compression, injec-
- Resin Film Infusion tion, continuous, and automated processes will be added to this list. Ad-
ditive manufacturing is also under development, mainly for short-fibres
composites [123] but it will not be considered here since it is still at a
fairly early stage.
In order to highlight the influence of the manufacturing processes on
For thermoplastics, compression moulding on laminate preforms the design process of a composite part, we propose tables to help during
can be applied after a heating phase. Recently, ATL or filament winding the analysis. The first three tables below (Tables 2, 3, 4) give the combi-
has become possible with thermoplastic composites. To end, two other nations between manufacturing processes and part geometry. The
processes have to be taken into account: pultrusion and overmoulding. shape categories are taken from the shape classification presented in
They are able to produce “net shaped” parts which are, of course, of Fig. 4 and the processes are those most commonly used for manufactur-
great interest for industrial purposes. Other authors [120,122] use a ing composite parts. Three symbols are used in the table to indicate

Fig. 14. Functional surfaces for the support.

 F. Neveu et al. / Materials and Design 172 (2019) 107755 11

Fig. 15. Beam solutions for a composite support.

whether the combination is used in normal practice, barely used (which included in our thinking because this has already been addressed in
could mean that the process is not perfectly suitable for the shape) or the literature. Moreover, it can also be considered as too constraining
not applicable at all. References are also provided inside the tables; during the creativity process and this choice should ultimately be
they give some examples or applications to which the reader can refer. made according to the industrial context and the mechanical con-
The second part of the analysis concerns the relationship between straints of the product. In the following section, the GAP methodology
the architecture and the manufacturing processes (Table 5). The third is explained and illustrated by two examples.
part is the relationship between the reinforcements and the
manufacturing processes (Table 6). The different types of reinforcement 3. The GAP methodology
that exist in the market cannot be used with every manufacturing pro-
cess. This table gives an idea of the possibilities available for each pro- 3.1. The scope of the methodology
cess. The choice of the reinforcement has to be governed by the
mechanical strength desired, on the one hand, and by the manufactur- The GAP methodology defines a way to create of innovative compos-
ing process used on the other. ite material based solutions. As stated in the previous sections, to face
These tables are neither exhaustive nor definitive in this article and the dilemma (Fig. 3) of the design or re-design of a part or a complete
are likely to be improved in later publications. They are presented mechanical subassembly, a resolution scheme is proposed and includes
here as an illustration of the GAP methodology proposed in the present the three axes: Geometry, Architecture and Process. One of the direct
article. They can be used in two configurations in the methodology: if consequences of this approach is that thinking first about the geometry
there is a manufacturing constraint, they can help with the choice of gives more freedom for creativity to generate as many solutions as pos-
the geometry during the creative phase, and, they can be used as a sible. We thus suggest the following methodology (Fig. 8):
tool for selecting geometries during the screening phase (more details
• Create innovative solutions by thinking about geometry first
about this phase are given in Part 3). Finally, it should be noted that
• These solutions can be imagined according to three possible
other solutions exist. For example, some manufacturing processes can
configurations:
be combined and adapted directly to the manufacturing process of
one part, as is the case for the frames manufactured by Airbus [142]. - No constraints at all: free design
- Architecture constraints: for example, laminates or sandwiches re-
2.5. Summary quired
- Manufacturing objectives: for example, “one shot” parts
In this section, we propose to tackle the complex problem of the de-
sign of a composite part following three main axes of development: the
shape, the manufacturing process and the architecture of the structure.
The GAP methodology can thus be represented simply by the diagram Once these solutions have been created, a more classic design proce-
shown in Fig. 7. Some of the interdependencies between these axes dure must be performed to make a go/no go decision on each design.
were underscored previously but not in an exhaustive way. The com- This can be based, among other things, on Figure of Merit, Key Perfor-
plex issue concerning the choice of materials is intentionally not mance parameters [39,143], material selection procedure [14–24],
12 F. Neveu et al. / Materials and Design 172 (2019) 107755

manufacturing process selection, sizing, or cost evaluation depending or sandwich structures. Multi-material solutions could also be
on the industrial context. Thus, the GAP methodology is part of the con- considered…
ceptual and preliminary design as defined in the design flow by Ashby As for case study number 2, these solutions have to be screened with
[14] shown in Fig. 9. At this stage of the project, the idea is to imagine particular parameters belonging to each industrial context.
as many concepts as possible in the conceptual part in order to obtain
a short list of concepts at the end of the preliminary phase. 3.2.2. Case study 2: composite support
The subject was: “Make a composite solution for a support. Use the
3.2. Case studies GAP methodology and propose solutions based on plate or shell shapes
and others based on beam shapes. Then explain how to manufacture
Two purely academic examples are proposed in this section. The your solution.” The functional surfaces for this case study are shown in
GAP methodology was tested on bachelor's degree students of INSA Fig. 14. In order to give an idea of the size of the composite part to be de-
Toulouse who had a very limited knowledge of composite materials signed, we specify that the dimensions between the corners are 700
(10 h of courses). × 400 mm. A selection of the solutions imagined by the students is pre-
sented in Figs. 15 to 17 and shows the wide variety of solutions pro-
3.2.1. Case study 1: economy seat posed that are a good starting point for such a project. These solutions
The subject was: “Make a composite solution for an economy class were then screened with the constraints of manufacturing cost for a me-
seat of a short range aircraft. It is attached to the floor by four point dium series.
links located at the centre of the structure. Use the GAP methodology Unlike the situation in case study 1, the geometry is constrained by
then propose three solutions based on plate or shell shapes and three the imposed functional surfaces here. This adds difficulties for the de-
others based on beam shapes. For dimensions, use the dimensions of signer but such situations are common in the industry. The rules for de-
your own chair. Then explain how to manufacture your solution.” signing the structure are simple: the geometry must necessarily
A selection of the solutions imagined by the students is shown in coincide with the functional surfaces and the design space falls within
Figs. 10 to 13 and illustrates the wide variety of design solutions pro- the limits set by these surfaces. Otherwise, the design remains free
posed that are a good starting point for such a project. and all kinds of shapes can be imagined to join these surfaces. There is
For the beam-based concepts, the students proposed two kinds of no limit on the number of components, i.e. the structure could be imag-
solutions with various geometries: solutions composed of simple com- ined in one shot manufacturing or by assembling several parts. Among
posite beams joined together (Fig. 10a to f) and solutions using more the beam-based solutions (Fig. 15), some may appear too fragile
complex beams in order to reduce the number of parts (Fig. 10g to (Fig. 15a, e, f and j) and reinforcements might be needed. The solution
m), even going as far as a one shot solution (Fig. 10k). in Fig. 15g appears to be too complex and should be discarded. In con-
For the shell-based and plate-based concepts, there were more pos- trast, plate and shell-based solutions (Fig. 16) or mixed concepts
sibilities (Figs. 11 and 12). For each solution, there were several ways of (Fig. 17) look simpler to manufacture. The generation of the first con-
manufacturing the seat: assemblies, one shot injection, folding of ther- cepts must not impose limits on the manufacturing possibilities, the
moplastic plates, etc. After designing the first concepts, each solution next screening step will determine and eliminate the geometries con-
should be analysed considering the manufacturing processes suitable sidered as too costly to manufacture.
in an industrial context. It can also be noted that, at that step of the An overview of the concepts created in the first instance is given in
methodology, some solutions could be realized with either laminates Figs. 15 to 17. It is interesting to note the diversity of architectures and

Fig. 16. Shell and plate-based solutions for a composite support.

 F. Neveu et al. / Materials and Design 172 (2019) 107755 13

Fig. 17. Mixed beam/shell/solid solutions for a composite support.

approaches encompassed by the three families of forms defined above. CRediT authorship contribution statement
The majority of the solutions tend to use shells, which is understandable
as composite structures are mainly oriented towards these forms. Given Florian Neveu: Writing - original draft. Bruno Castanié: . Philippe
the large choice of architecture and processes available for composite Olivier:
materials, having a clear vision of the possibilities of manufacturing at
this early stage is a great asset for pre-concepts. At this point, the GAP
methodology suggests reviewing every pre-concept imagined and Acknowledgements
screening with the help of important criteria relative to the composite
part. Criteria could be anything from the manufacturing cost to the The authors acknowledge Bpifrance Financement for its financial
mass, the structure resistance, the technological difficulties etc. As pre- support in the FUI contract N°F1504022M.
concepts are rather sketchily detailed geometries, one criterion should
be based on the technological difficulties that may arise later in the References
manufacture of the part. In one of the solutions proposed here, there
may be a bonding issue between the omega-shaped beams and the sup- [1] J. Fan, J. Njuguna, An introduction to lightweight composite materials and their use
in transport structures, Lightweight Composite Structures in Transport (J. Njuguna,
port that has not been considered. The designer's experience is of great
Ed.), ch. 1, pp. 3–34, Woodhead Publishing, 2016.
value here to judge whether the concepts have good potential. In cases [2] Brochure Aérospatiale, Matériaux composites: dans la grande mutation des struc-
where the concepts are not mature enough to properly indicate the se- tures. DCT 1135, 01/04/1986.
lection criteria, other design iterations can be considered to facilitate [3] M. Carello, A.G. Airale, A. Ferraris, A. Messana, L. Sisca, Static design and finite ele-
ment analysis of innovative CFRP transverse leaf spring, Appl. Compos. Mater. 24
this screening step. At the end of that phase, a short list of solutions (Dec. 2017) 1493–1508.
should be given. The concepts chosen are obviously those that obtain [4] Elixir Aircraft. http://elixir-aircraft.com/en/. Accessed: 2018-12-13.
the best score and two or three concepts should be enough to continue. [5] S.W. Tsai, J.M. Patterson, Simplified design of composite materials, Mater. Des. 8
(May–June 1987).
Iterative design loops must then be carried out along with numerical [6] L. Wang, A. Kolios, T. Nishino, P.-L. Delafin, T. Bird, Structural optimisation of
simulations to validate the geometry and find the best stacking and vertical-axis wind turbine composite blades based on finite element analysis and
thickness for the part. genetic algorithm, Compos. Struct. 153 (2016) 123–138.
[7] S.R. Henrichse, E. Lindgaard, E. Lund, Free material stiffness design of laminated
composite structures using commercial finite element analysis codes, Struct.
Multidiscip. Optim. 51 (5) (2015) 1097–1111.
4. Conclusion [8] J.L. Galos, M. Sutcliffe, Development of a structural optimisation methodology for
use in the design of a composite semi-trailer chassis, 21st International Conference
on Composite Materials Xi'An, Aug. 2017.
The task of designing composite parts is made challenging by the [9] F. Léné, G. Duvaut, M. Olivier-Mailhé, M. Ben Chaabane, S. Grihon, An advanced
complexity of the materials and the large number of decisions that methodology for optimum design of a composite stiffened cylinder, Compos.
Struct. 91 (2009) 392–397.
have to be taken throughout the project. The GAP methodology pro-
[10] J. Lee, D. Lee, J. Park, I. Choi, J.W. Lim, S. Kim, D.G. Lee, Carbon/epoxy composite foot
vides a solution to help the right decisions to be taken during the structure for biped robots, Compos. Struct. 140 (June 2016) 344–350.
early pre-concept phase by enabling the emergence of numerous solu- [11] P.V. Vijay, P.R. Soti, H.V.S. GangaRao, R.G. Lampo, J.D. Clarkson, Design and evalua-
tions. Considering this step as one of the most important, since bad de- tion of an integrated FRP composite wicket gate, Compos. Struct. 145 (June 2016)
149–161.
cisions at this stage could lead to a mediocre final result, the GAP [12] D. Paul, D. Pratt, History of flight vehicle structures 1903–1990, J. Aircr. 41 (Sep–
methodology provides tools to perform a rapid analysis of the first con- Oct 2004) 969–977.
cepts by integrating the three main sets of problems impacting the de- [13] P. Feraboli, A. Masini, L. Taraborrelli, A. Pivetti, Integrated development of CFRP
structures for a topless high performance vehicle, Compos. Struct. 78 (2007)
sign of composite structures. The methodology relies heavily on 495–506.
imagination and creativity for the design part. It also relies on a specific [14] M. Ashby, K. Johnson, Materials and Design: The Art and Science of Material Selec-
analysis of each concept giving enough information to select the most tion in Product Design, Butterworth-Heinemann, Oxford, 2002.
[15] M. Noryani, S.M. Sapuan, M.T. Mastura, M.T. Zuhri, E.S. Zainudin, Material selection
valuable concepts that will be developed more precisely later on. Early criteria for natural fibre composite in automotive component: a review, IOP Con-
decisions that have to be taken are decisive for any project. Bad deci- ference Series: Materials Science and Engineering 368 (1) (2018).
sions at this point imply a waste of time and money that could become [16] D. Zindani, S.R. Maity, S. Bhowmik, A decision-making approach for material selec-
tion of polymeric composite bumper beam, Composite and Advanced Materials for
critical at some point. The GAP methodology is a solution for reducing
Industrial Applications (2018) 112–128.
the risks of poor decisions when it comes to designing - or re- [17] Y. Swolfs, I. Verpoest, L. Gorbatikh, Recent advances in fibre-hybrid composites:
designing - a mechanical product in composite materials. materials selection, opportunities and applications, Int. Mater. Rev. (2018) 1–35.
14 F. Neveu et al. / Materials and Design 172 (2019) 107755

[18] S. Kazemzadeh Azad, T. Akiş, Automated selection of optimal material for pressur- [54] S. Rana, R. Fangueiro, BRAIDED STRUCTURES and COMPOSITES Production, Proper-
ized multilayer composite tubes based on an evolutionary approach, Neural ties, Mechanics and Technical Applications, CRC Press, 2016.
Comput. & Applic. 29 (7) (2018) 405–416. [55] A. Malpot, F. Touchard, S. Bergamo, Influence of moisture on the fatigue behaviour
[19] M. Noryani, S.M. Sapuan, M.T. Mastura, Multi-criteria decision-making tools for of a woven thermoplastic composite used for automotive application, Mater. Des.
material selection of natural fibre composites: a review, Journal of Mechanical En- 98 (May 2016) 12–19.
gineering and Sciences 12 (1) (2018) 3330–3353. [56] B. Vieille, V.M. Casado, C. Bouvet, Influence of matrix toughness and ductility on the
[20] H. Wargnier, F.X. Kromma, M. Danis, Y. Brechet, Proposal for a multi-material de- compression-after-impact behavior of woven-ply thermoplastic and thermoset-
sign procedure, Mater. Des. 56 (2014) 44–49. ting composites: a comparative study, Compos. Struct. 110 (2014) 207–218.
[21] O.T. Thomsen, Sandwich materials for wind turbine blades: present and future, J. [57] A. Rashidi, A.S. Milani, A multi-step biaxial bias extension test for wrinkling/de-
Sandw. Struct. Mater. 11 (2009) 7. wrinkling characterization of woven fabrics: towards optimum forming design
[22] J. Pflug, I. Verpoest, Sandwich materials selection charts, J. Sandw. Struct. Mater. 8 guidelines, Mater. Des. 146 (May 2018) 273–285.
(407) (2006). [58] F. Gamdani, A. Boukhili, R. ad Vadean, Tensile strength of open-hole, pin-loaded
[23] C. Monroy Aceves, A.A. Skordos, M.P.F. Sutcliffe, Design selection methodology for and multi-bolted single-lap joints in woven composite plates, Mater. Des. 88
composite structures, Mater. Des. 29 (2008) 418–426. (Dec. 2015) 702–712.
[24] C. Monroy Aceves, M.P.F. Sutcliffe, M.F. Ashby, A.A. Skordos, C. Rodríguez Román, [59] F. Pascal, A. Rogani, B. Mahmoud, P. Navarro, S. Marguet, J.-F. Ferrero, Impact dam-
Design methodology for composite structures: a small low air-speed wind turbine age prediction in thin woven composite laminates - part II: application to normal
blade case study, Mater. Des. 36 (2012) 296–305. and oblique impacts on sandwich structure, Compos. Struct. 190 (Apr. 2018)
[25] A. Hambali, S.M. Sapuan, N. Ismail, Y. Nukman, Composite manufacturing process 43–51.
selection using analytical hierarchy process, International Journal of Mechanical [60] F. Pascal, O. Dorival, P. Navarro, S. Marguet, J.-F. Ferrero, Impact damage prediction
and Materials Engineering (IJMME) 4 (1) (2009) 49–61. in thin woven composite laminates - part I: modeling strategy and validation,
[26] M.G. Bader, Selection of composite materials and manufacturing routes for cost- Compos. Struct. 190 (Apr. 2018) 32–42.
effective performance, Compos. Part A 33 (2002) 913–934. [61] Z.M. Huang, The mechanical properties of composites reinforced with woven and
[27] S. Howard, T. Culley, E. Dekoninck, Creativity in the engineering design process, In- braided fabrics, Compos. Sci. Technol. 60 (4) (2000) 479–498.
ternational Conference on Engineering Design, Aug. 2007. [62] C. Wang, A. Roy, Z. Chen, V.V. Silberschmidt, Braided textile composites for sports
[28] S. Howard, T. Culley, E. Dekoninck, Describing the creative design process by the protection: energy absorption and delamination in impact modelling, Mater. Des.
integration of engineering design and cognitive psychology literature, Des. Stud. 136 (Dec. 2017) 258–269.
29 (2008) 160–180. [63] O. Zebdi, R. Boukhili, F. Trochu, An inverse approach based on aminate theory to
[29] S. Niemann, et al., The use of topology optimisation in the conceptual design of calculate the mechanical properties of braided composites, J. Reinf. Plast. Compos.
next generation lattice composite aircraft fuselage structures, Aeronaut. J. 117 28 (23) (2009) 2911–2930.
(Nov. 2013). [64] C. Zhao, Y. Huang, Z. Chen, S.K. Ha, Progressive failure prediction of a landing gear
[30] C. Wilson, Brainstorming and Beyond: A User-centered Design Method, Morgan structure of braided composites, Compos. Struct. 161 (2017) 407–418.
Kaufmann, 2013. [65] P.L. Dewalt, R.P. Reichard, Just how good are knitted fabric, J. Reinf. Plast. Compos.
[31] M.S. Aziz, A.Y. El Sheriff, Biomimicry as an approach for bio-inspired structure with 13 (10) (1994) 908–917.
the aid of computation, Alexandria Engineering Journal 55 (2016) 704–717. [66] D.H. Benson, G.H. Hasko, Mechanical properties and damage tolerance of multiax-
[32] J. Lu, The conflict resolution in product experience design based on evaporating ial warp-knit composites, Compos. Sci. Technol. 56 (3) (1996) 367–380.
cloud of the theory of constraints, Lect. Notes Comput. Sci 9186 (2015) 53–62. [67] T. Sinmazçelik, E. Avcu, M.O. Bora, C. Çoban, A review: fibre metal laminates, back-
[33] L. Chechurin, Y. Borgianni, Understanding TRIZ through the review of top cited ground, bonding types and applied test methods, Mater. Des. 32 (Aug. 2011)
publications, Comput. Ind. 82 (2016) 119–134. 3671–3685.
[34] D. Ullman, The Mechanical Design Process Fourth edition, McGraw-Hill, 2010. [68] A. Asundi, A.Y.N. Choi, Fiber metal laminates: an advanced material for future air-
[35] CES EduPack, http://www.grantadesign.com/education/InDepth/html/indepth/ craft, J. Mater. Process. Technol. 63 (1997) 384–394.
shapeinfo/shape_classification.htm, Accessed date: 13 December 2018. [69] S. Christke, A.G. Gibson, K. Grigoriou, A.P. Mouritz, Multi-layer polymer metal lam-
[36] K. Zimmermann, D. Zenkert, M. Siemetzki, Testing and analysis of ultra-thick com- inates for the fire protection of lightweight structures, Mater. Des. 97 (May 2016)
posites, Compos. Part B 41 (June 2010) 326–336. 349–356.
[37] M. Siemetzki, K. Zimmermann, P. Middendorf, Ultra-thick laminates chances and limita- [70] K. Giasin, S. Ayvar-Soberanis, A. Hodzic, The effects of minimum quantity lubrica-
tions, 17th International Conference on Composite Materials, Edinburgh, UK, 2009. tion and cryogenic liquid nitrogen cooling on drilled hole quality in GLARE fibre
[38] A.P. Mouritz, M.K. Bannister, P.J. Falzon, K.H. Leong, Review of applications for ad- metal laminate, Mater. Des. 89 (Jan. 2016) 996–1006.
vanced three-dimensional fibre textile composites, Compos. Part A 30 (1999) [71] C. Lopes, P. Camanho, Z. Gürdal, B. Tatting, Progressive failure analysis of tow-
1445–1461. placed, variable-stiffness composite panels, Int. J. Solids Struct. 44 (Dec. 2007)
[39] T.M. Krivanek, B.C. Yount, Composite payload fairing structural architecture assess- 8493–8516.
ment and selection, SAMPE Conference, Baltimore, MD, USA, May 2012. [72] R. Li, D. Kelly, A. Crosky, Strength improvement by fibre steering around a pin
[40] T. Urik, M. Mališ, Innovative composite structures for small aircraft, 26th Interna- loaded hole, Compos. Struct. 57 (July 2002) 377–383.
tional Congress of the Aeronautical Sciences, Anchorage, Alaska, USA, Sept. 2008. [73] J. Toral, B. Castanié, J.-J. Barrau, N. Swiergel, Multi-level analysis of low-cost Z-
[41] C. Soutis, Fibre reinforced composites in aircraft construction, Prog. Aerosp. Sci. 41 pinned composite joints; part 2: joint behaviour, Compos. A: Appl. Sci. Manuf. 42
(2005) 143–151. (2011) 2082–2092.
[42] A. Njionhou, F. Berthet, B. Castanié, C. Bouvet, Relationships between LRI process [74] J. Toral, B. Castanié, J.-J. Barrau, N. Swiergel, Multi-level analysis of low-cost Z-
parameters and impact and post-impact behaviour of stitched and unstitched pinned composite joints; part 2: single Z-pin behaviour, Compos. A: Appl. Sci.
NCF laminates, Appl. Compos. Mater. 19 (2012) 885–899. Manuf. 42 (2011) 2070–2081.
[43] L. Gigliotti, S.T. Pinho, Translaminar fracture toughness of NCF composites with [75] I. Gnaba, X. Legrand, P. Wang, D. Soulat, Through-the-thickness reinforcement for
multiaxial blankets, Mater. Des. 94 (March 2016) 410–416. composite structures: a review, J. Ind. Text. (2018)https://doi.org/10.1177/
[44] A. Shipsha, M. Burman, J. Ekh, Failure of cross-ply NCF composites under off-axis 1528083718772299.
compressive loads - an experimental study and a new strength prediction model [76] B. Castanié, J.-J. Barrau, J.-P. Jaouen, S. Rivallant, Combined shear/compression
including fibre bundle waviness, Compos. Part B 153 (Nov. 2018) 49–56. structural testing of asymmetric sandwich structures, Exp. Mech. 44 (2004)
[45] A. Njionhou, F. Berthet, B. Castanié, Effects of process parameters on the mechani- 461–472.
cal properties and morphology of stitched and non-stitched carbon/epoxy liquid [77] B. Castanié, J.-J. Barrau, J.-P. Jaouen, Theoretical and experimental analysis of asym-
resin-infused NCF laminate, out of autoclave and out of oven, Int. J. Adv. Manuf. metric sandwich structures, Compos. Struct. 55 (2002) 295–306.
Technol. 65 (2013) 1289–1302. [78] L.L. Gibson, M.F. Ashby, Cellular Solids, Structure and Properties, Cambridge Uni-
[46] D. Wilhelmsson, R. Gutkin, F. Edgren, L.E. Asp, An experimental study of fibre wav- versity Press, 1999.
iness and its effects on compressive properties of unidirectional NCF composites, [79] D. Zenkers, An Introduction to Sandwich Construction, E-MAS Publishing, 1995.
Compos. A: Appl. Sci. Manuf. 107 (Feb. 2018) 665–674. [80] J.R. Vinson, The Behavior of Sandwich Structures of Isotropic and Composite Mate-
[47] R. Amacher, J. Cugnoni, J. Botsis, L. Sorensen, C. Smith, W. Dransfeld, Thin ply com- rials, CRC Press, 1999.
posites: experimental characterization and modeling of size-effects, Compos. Sci. [81] S. Abrate, B. Castanié, Y.D.S. Rajapakse, Dynamic Failure of Composite Sandwich
Technol. 101 (2014) 121–132. Structures, Springer, 2012.
[48] A. Arteiro, G. Catalanotti, J. Xavier, P. Linde, P.P. Camanho, A strategy to improve the [82] I.M. Daniel, E.E. Gdoutos, Y.D.S. Rajapakse, Major Accomplishments in Composite
structural performance of non-crimp fabric thin-ply laminates, Compos. Struct. 188 Materials and Sandwich Structures: An Anthology of ONR Sponsored Research,
(2018) 438–449. Springer, 2009.
[49] J.D. Fuller, M.R. Wisnom, Pseudo-ductility and damage suppression in thin ply [83] H. Wang, K.S. Ramakrishnan, K. Shankar, Experimental study of the medium veloc-
CFRP angle-ply laminates, Compos. A: Appl. Sci. Manuf. 69 (2015) 64–71. ity impact response of sandwich panels with different cores, Mater. Des. 99 (June
[50] A. Arteiro, G. Catalanotti, J. Xavier, P.P. Camanho, Notched response of non-crimp 2016) 68–82.
fabric thin-ply laminates, Compos. Sci. Technol. 79 (2013) 97–114. [84] Y. Zhang, Z. Zong, Q. Liu, J. Ma, Y. Wu, Q. Li, Static and dynamic crushing responses
[51] G. Bullegas, S.T. Pinho, S. Pimenta, Engineering the translaminar fracture behaviour of CFRP sandwich panels filled with different reinforced materials, Mater. Des. 117
of thin-ply composites, Compos. Sci. Technol. 131 (2016) 110–122. (March 2017) 396–408.
[52] S. Sihn, R.Y. Kim, K. Kawabe, S.W. Tsai, Experimental studies of thin-ply laminated [85] D. Zhang, G. Lu, D. Ruan, Q. Fei, W. Duan, Quasi-static combined compression-shear
composites, Compos. Sci. Technol. 67 (6) (May 2007) 996–1008. crushing of honeycombs: an experimental study, Mater. Des. 167 (April 2019).
[53] R. Kamiya, B.A. Cheeseman, P. Popper, T.-W. Chou, Some recent advances in the [86] Y. Aminanda, B. Castanié, J.-J. Barrau, P. Thevenet, Experimental analysis and
fabrication and design of three-dimensional textile preforms: a review, Compos. modeling of the crushing of honeycomb cores, Appl. Compos. Mater. 12 (3–4)
Sci. Technol. 60 (2000) 33–47. (May 2005) 213–227.
 F. Neveu et al. / Materials and Design 172 (2019) 107755 15

[87] J.P. Susainathan, F. Eyma, E. De Luycker, A. Cantarel, B. Castanié, Experimental in- [116] T. Huang, Y. Wang, G. Wang, Review of the mechanical properties of a 3D woven
vestigation of impact behavior of wood-based sandwich structures, Composite composite and its applications, Polymer - Plastics Technology and Engineering 57
Part A: Applied Science and Manufacturing 109 (2018) 10–19. (8) (2018) 740–756.
[88] J.P. Susainathan, F. Eyma, E. De Luycker, A. Cantarel, B. Castanié, Manufacturing and [117] R. Geerinck, I. De Baere, G. De Clercq, L. Daelemans, K. De Clerck, One-shot produc-
quasi-static bending behavior of wood-based sandwich structures, Compos. Struct. tion of large-scale 3 woven fabrics with integrated prismatic shaped cavities and
182 (2017) 487–504. their applications, Mater. Des. 165 (March 2019).
[89] A. Shahdin, L. Mezeix, J. Morlier, Y. Gourinat, C. Bouvet, Fabrication and mechanical [118] K. Leong, S. Ramakrishna, Z.M. Huang, G.A. Bibo, Potential of knitting for engineer-
testing of a new sandwich structure with carbon fiber network core, J. Sandw. ing composites - a review, Compos. A: Appl. Sci. Manuf. 31 (3) (2000) 197–220.
Struct. Mater. 12 (2010) 569–589. [119] B.T. Astrom, Manufacturing of Polymer Composite, Chapman and Hall, 2002.
[90] A. Shahdin, L. Mezeix, C. Bouvet, J. Morlier, Y. Gourinat, Fabrication and mechanical [120] F.C. Campbell, Manufacturing Processes for Advanced Composites, Elsevier Ad-
testing of glass fiber entangled sandwich beams: a comparison with honeycomb vanced Technology, 2004.
and foam sandwich beams, Compos. Struct. 90 (2009) 404–412. [121] U.P. Breuer, Commercial Aircraft Composite Technology, Springer, 2016.
[91] I. Ullah, J. Elambasseril, M. Brandt, S. Feih, Performance of bio-inspired Kagome [122] D. Gay, Composite Materials: Design and Applications Third edition, CRC Press,
truss core structures under compression and shear loading, Compos. Struct. 118 2015.
(Dec. 2014) 294–302. [123] X. Wang, M. Jiang, Z. Zhou, J. Gou, D. Hui, 3D printing of polymer matrix compos-
[92] H. Wadley, K. Dharmasena, Y. Cheb, P. Dudt, D. Knight, R. Charette, K. Kiddy, Com- ites: a review and prospective, Compos. Part B 110 (2017) 442–458.
pressive response of multilayered pyramidal lattices during underwater shock [124] J.D. Suh, D.G. Lee, Manufacture of composite screw rotors for air compressors by
loading, International Journal of Impact Engineering 35 (2008) 1102–1114. RTM process, J. Mater. Process. Technol. 113 (June 2001) 196–201.
[93] H.L. Fan, F.H. Meng, W. Yang, Sandwich panels with Kagome lattice cores rein- [125] M. Elkington, D. Bloom, C. Ward, A. Chatzimichali, K. Potter, Hand layup: under-
forced by carbon fibers, Compos. Struct. 81 (2007) 533–539. standing the manual process, Advanced Manufacturing: Polymer and Composites
[94] R. Gautam, S. Idapalapati, Performance of strut-reinforced Kagome truss core struc- Science 1 (Nov. 2015) 138–151.
ture under compression fabricated by selective laser melting, Mater. Des. 164 (Feb. [126] T. Kikuchi, H. Hamada, A. Nakai, A. Ohtani, A. Goto, Y. Takai, A. Endo, C. Narita, T.
2019). Koshino, A. Fudauchi, Relationships between degree of skill, dimension stability
[95] H.Z. Jishi, R. Umer, W.J. Cantwell, The fabrication and mechanical properties of and mechanical properties of composite structure in hand lay-up method, The
novel composite lattice structures, Mater. Des. 91 (Feb. 2016) 286–293. 19th International Conference on Composite Materials, 2013.
[96] P.-B. Su, B. Han, M. Yang, Z.-H. Wei, Z.-Y. Zhao, Q.-C. Zhang, Q. Zhang, K.-K. Qin, T.J. [127] T. Kikuchi, Y. Tani, Y. Takai, A. Goto, H. Hamada, Mechanical properties of jute com-
Lu, Axial compressive collapse of ultralight corrugated sandwich cylindrical shells, posite by spray up fabrication method, Energy Procedia 56 (2014) 289–297.
Mater. Des. 160 (Dec. 2018) 325–337. [128] J. Verrey, V. Michaud, J.-A.E. Manson, Dynamic capillary effects in liquid composite
[97] B. Han, K. Qin, B. Yu, B. Wang, Q.C. Zhang, T.J. Lu, Honeycomb-corrugation hybrid as moulding with non-crimp fabrics, Compos. A: Appl. Sci. Manuf. 37 (Jan. 2006)
a novel sandwich core for significantly enhanced compressive performance, Mater. 92–102.
Des. 93 (Dec. 2016) 271–282. [129] B. Vieille, W. Albouy, L. Chevalier, L. Taleb, About the influence of stamping on
[98] Heimbs and S. Foldcore, Sandwich structures and their impact performance: an thermoplastic-based composites for aeronautical applications, Compos. Part B 45
overview, Dynamic Failure of Composite and Sandwiches Structures (S. Abrate, (1) (Feb. 2013) 821–834.
B. Castanie, and Y. Rajapakse, eds.), pp. 491–544, Springer, 2013. [130] B.-A. Behrens, A. Raatz, S. Hübner, C. Bonk, F. Bohne, M. Bruns, C. Micke-Camuz, Au-
[99] M.N. Velea, C. Schneider, S. Lache, Second order hierarchical sandwich structure tomated stamp forming of continuous fiber reinforced thermoplastics for complex
made of self-reinforced polymers by means of a continuous folding process, shell geometries, Procedia CIRP 66 (2017) 113–118.
Mater. Des. 102 (July 2016) 313–320. [131] C. Grant, Automated processes for composite aircraft structure, Industrial Robot:
[100] T.K. O'Brien, I.L. Paris, Exploratory investigation of failure mechanisms in transition The International Journal of Robotics Research and Application 33 (2) (2006)
regions between solid laminates and X-cor truss sandwich, Compos. Struct. 57 117–121.
(2002) 189–204. [132] M. Rakhshbahar, M. Sinapius, A novel approach: combination of automated fiber
[101] L.-J. Feng, Z.-T. Yang, G.-C. Yu, X.-J. Chen, L.-Z. Wu, Compressive and shear proper- placement (AFP) and additive layer manufacturing (ALM), Journal of Composite
ties of carbon fiber composite square honeycombs with optimized high-modulus Sciences 2 (July 2018).
hierarchical phases, Compos. Struct. 201 (Oct. 2018) 845–856. [133] P. Maung, B.G. Prusty, J.M. White, M. David, A.W. Philips, N.A. St John, Structural
[102] G.X. Ha, M.W. Zehn, Experimental study and finite element simulation of novel performance of a shape-adaptive composite hydrofoil using automated fibre place-
napcore sandwich composite, Composite Part B: Engineering 158 (Feb. 2019) ment, Eng. Struct. 183 (March 2019) 351–365.
117–130. [134] E. Poodts, G. Minak, L. Mazzocchetti, L. Giorgini, Fabrication, process simulation and
[103] J. Bertolini, B. Castanié, J.-J. Barrau, J.-P. Navarro, Multi-level experimental and nu- testing of a thick CFRP component using the RTM process, Compos. Part B 56 (Jan.
merical analysis of composite stiffener debonding part 2: element and panel level, 2014) 673–680.
Compos. Struct. 90 (2009) 392–403. [135] Y. Kyosev, Advances in Braiding Technology - Specialized Techniques and Applica-
[104] J. Bertolini, B. Castanié, J.-J. Barrau, J.-P. Navarro, Multi-level experimental and nu- tions, Woodhead Publishing, 2016.
merical analysis of composite stiffener debonding part 1: non-specific specimen [136] A. Krzyzak, M. Mazur, M. Gajewski, K. Drozd, A. Komorek, and P. Przybylek, “Sand-
level, Compos. Struct. 90 (2009) 381–391. wich structured composites for aeronautics: methods of manufacturing affecting
[105] A. Bergan, J. Bakuckas Jr., J. Awerbuch, T.-M. Tan, Assessment of damage contain- some mechanical properties,” International Journal of Aerospace Engineering, vol.
ment features of a full-scale PRSEUS fuselage panel, Compos. Struct. 113 (2014) 2016, May 2016.
174–185. [137] J. Grünewald, P. Parlevliet, V. Altstädt, Manufacturing of thermoplastic composite
[106] V.V. Vasiliev, A.F. Razin, Anisogrid composite lattice structures for spacecraft and sandwich structures: a review of literature, J. Thermoplast. Compos. Mater. 30
aircraft applications, Compos. Struct. 76 (Oct. 2006) 182–189. (4) (Apr. 2017) 434–464.
[107] V.V. Vasiliev, V.A. Barynin, A.F. Razin, Anisogrid composite lattice structures: devel- [138] A.C. Long, Design and Manufacture of Textile Composites, Woodhead Publishing,
opment and aerospace applications, Compos. Struct. 94 (2012) 1117–1127. 2005.
[108] E.V. Morozov, A.V. Lopatin, V.A. Nesterov, Buckling analysis and design of anisogrid [139] S.C. Joshi, The pultrusion process in polymer matrix composites, Manufacturing
composite lattice conical shells, Compos. Struct. 93 (Nov. 2011) 3150–3162. Techniques for Polymer Matrix Composites (PMCs) (S. G. Advani and K.-T. Hsiao,
[109] M. Buragohain, R. Velmurugan, Study of filament wound grid-stiffened composite eds.), ch. 12, pp. 381–408, Woodhead Publishing, 2012.
cylindrical structures, Compos. Struct. 93 (2011) 1031–1038. [140] L. Sorrentino, M. Marchetti, C. Bellini, A. Delfini, F. Del Sette, Manufacture of high
[110] Hexcel, Composites, https://www.hexcel.com/Products/Tooling/HexTool-Tooling- performance isogrid structure by robotic filament winding, Compos. Struct. 164
Material, Accessed date: 13 December 2018. (March 2017) 43–50.
[111] Lamborghini, https://www.lamborghini.com/fr-en/brand/innovation-excellence/ [141] W.D. Li, G. Liu, X.L. Hu, X.F. An, X.Y. Zhong, Y. Li, X.S. Yi, The processing character-
forged-composites, Accessed date: 13 December 2018. istics and mechanical properties of semi-prepreg RTM composites, Adv. Mater. Res.
[112] X. Aubard, C. Cluzel, L. Guitard, P. Ladevèze, Damage modeling at two scales for 4D 721 (2013) 153–158.
carbon/carbon composites, Comput. Struct. 78 (Nov. 2000) 83–91. [142] Composites World, https://www.compositesworld.com/articles/airbus-a350-up-
[113] K. Bilisik, Three-dimensional braiding for composites: a review, Text. Res. J. 83 date-braf-fpp, Accessed date: 13 December 2018.
(2013) 1414–1436. [143] J.C. Mankins, Technology readiness and risk assessments: a new approach, Acta
[114] K. Bilisik, Multiaxis three-dimensional weaving for composites: a review, Text. Res. Astronautica 65 (Nov.–Dec. 2009) 1208–1215.
J. 82 (2012) 725–743.
[115] E. De Luycker, F. Morestin, P. Boisse, D. Marsal, Simulation of 3D interlock compos-
ite preforming, Compos. Struct. 88 (2009) 615–623.