Journal of Materials Processing Technology 210 (2010) 363369

Contents lists available at ScienceDirect

Journal of Materials Processing Technology

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

Experimental investigations and numerical analysis for improving knowledge of

incremental sheet forming process for sheet metal parts
S. Dejardin a, , S. Thibaud b , J.C. Gelin a , G. Michel a
a
b

ENSMM National Engineering Institute in Mechanics and Microtechnologies, 26 rue de lEpitaphe, 25000 Besancon, France
FEMTO-ST Institute Applied Mechanics Department, 24 rue de lEpitaphe, 25000 Besancon, France

a r t i c l e

i n f o

Article history:
Received 2 July 2008
Received in revised form
17 September 2009
Accepted 26 September 2009

Keywords:
Incremental sheet forming
FE model
Tool trajectory
Springback
Deformation mechanics

a b s t r a c t
The paper is related to the analysis of shape distortions and springback effects arising in single point
incremental sheet forming in order to study the use of a FE model based on shell elements to perform
simulation of the process. A comparison between numerical and experimental results is made to assess
the suitability of the model. The measurements of geometrical prole of a truncated cone and springback
of cut rings show that the FE model allows to predict accurate results for a set of well dened process
parameters. The deformation mechanism of ISF is taken into consideration to determine the limits of the
model studied.
2009 Elsevier B.V. All rights reserved.

1. Introduction
The requirements associated with customized production and
cost reduction are still growing and new technologies seem to offer
a sustainable approach to answer to such requests. This is a reason
for the increasing interest in exible forming processes like incremental sheet forming (ISF). Indeed, the ISF concept consisting in a
progressive and localized deformation of a sheet metal part is exible because no specialized tooling is required. The movements over
the surface of the sheet of a simple forming tool cause to a highly
localized plastic deformation. Regarding the concept of such a technology, a wide range of 3D shapes can be formed with a correct
denition of the forming tool path controlled by a CNC machine.
In addition to this, ISF also leads to a higher material formability
in comparison with conventional forming processes like stamping.
These advantages bring to a growing interest in both academic and
industrial research centers as it is demonstrated in literature. Automotive panels manufactured by Amino et al. (2002) using ISF or
customized medical products produced by Ambrogio et al. (2005a)
are relevant examples. A global review of works and achievements
made in incremental sheet forming has been proposed by Jeswiet
et al. (2005).

Corresponding author. Tel.: +33 6 27 81 70 29; fax: +33 3 81 40 29 02.

E-mail address: steeve.dejardin@gmail.com (S. Dejardin).
0924-0136/$ see front matter 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmatprotec.2009.09.025

Despite numerous researches in ISF during the last decade, the

deformation mechanism is not fully determined. A deep understanding of the deformation mechanism is important to allow an
accurate numerical modelling of the process to predict the sheet
behaviour and to make possible the process control. In this sense,
recent results obtained by Jackson and Allwood (2009) clearly show
that there is a relevant amount of shear in ISF. So it becomes clear
now that the deformation mechanism of ISF is not almost pure
stretching as it has been described in Filice et al. (2002) or Hirt
et al. (2003) for example. Therefore, these results have raised the
question of the validity of numerical model based on shell elements
to simulate the process. Despite the previous considerations, the
reported study is focused on the possibility to use an adapted FE
model of the process based on shell elements to predict accurate
results associated with a well dened set of process parameters. A
short review of previous works in ISF is given in Section 2 in order
to clearly dene the proposed study.
2. Review
Many types of incremental sheet forming processes have been
investigated, including the use of a exible support or a counter
pressure, the development of multipoint incremental forming as
it is described in Micari et al. (2007), but the most widely used
tool is a hemispherical punch. The main congurations of ISF are
single point incremental sheet (SPIF) and two-point incremental
sheet forming (TPIF). In SPIF, a sheet is clamped around its edges

364

S. Dejardin et al. / Journal of Materials Processing Technology 210 (2010) 363369

and deformed by a simple forming tool (generally of hemispherical

shape) which presses on one side of the sheet and moves around
from a proper forming tool path, whereas in TPIF, the sheet is
formed against a die or a second mobile tool. In both cases, the
most common forming strategies used are contours or spirals combined with an increasing depth, following the prole of the product.
Associated with these different versions of ISF, the numerous process parameters play an important role in material formability
and deformation mechanism. Jackson and Allwood (2009) have
recently reviewed the previous research made into the deformation
mechanics and have demonstrated, by a rigorous measurement of
a cross-section of a 3 mm thick copper plate, that the deformation mechanisms of both SPIF and TPIF are increasing stretching
and shear in the plane perpendicular to tool path direction and
shear in the tool path direction. This conclusion has been already
experimentally demonstrated by Jackson et al. (2008) through the
detection of the presence of through-thickness shear in the direction of punch movement by measuring the relative displacement
of both surfaces of a sandwich panel deformed by SPIF. Bambach et
al. (2003) have also remarked through numerical analysis, that the
occurrence of shear through the thickness and have observed that
the level of shear depends both of the forming tool diameter and
the vertical forming step increment.
Such results have to be taken into account to develop accurate
models of the process. Numerical modelling of ISF based on the
idealized deformation mechanism seems to be questionable. Until
now, it has been accepted that the deformation mechanism resulting from ISF is largely one of plane and in plane stretching has been
noticed for a straight sided pyramid formed by TPIF by Bambach et
al. (2003), for a truncated cone formed by SPIF from various materials by Fratini et al. (2004), and for various shapes formed by SPIF
by Jeswiet and Young (2005).
Considering recent results demonstrated by Jackson and
Allwood (2009), shell elements seem not to be the best choice to
predict shear mechanism deformation on the thickness of the sheet
metal part. Bambach et al. (2003) suggested using a 3D FE model
with four continuum elements through the thickness. This model
predicted that shear in the plane perpendicular to the tool direction dominates in TPIF. In the sheet metal plane parallel to the tool
direction, it was predicted that shear strains occur, increasing when
decreasing the head forming tool diameter, which is in contradiction with the plane strain assumption. However, the model used
also predicted that stretching is signicant in the plane perpendicular to the tool direction in SPIF but also that all strain components
are negligible in the direction parallel to tool path. Jackson and
Allwood (2009) mentioned that it is not necessarily a contradiction to the experimental measurements of Allwood et al. (2007)
because the tool path used by Bambach et al. (2003) alternated in
direction. This forming strategy leads to cancel any shear on successive laps as a result of friction, whereas the tool always moved
in the same direction in the strategy used by Allwood et al. (2007).
In addition, the results described by Jackson and Allwood (2009)
have been obtained through an experimental campaign performed
on a 3 mm thick copper plate which represents the thickness limit
for thin metal part.
2.1. Context of present work
This paper is focused on the necessity to dene a reliable
relationship between ISF parameters, including both material
parameters (thickness, ow stress, hardening, friction) and process parameters (forming tool path or forming tool diameter), to
develop suitable models for the process. Despite the previous considerations, the reported study analyses the possibility to use a
simple FE model of the process based on shell elements to predict accurate results in terms of geometrical accuracy (nal shape,

Fig. 1. (a) Experimental setup for SPIF experiments (three-axis milling machine
toolSPIF tooling system). (b) Tooling system used for SPIF experiments.

springback), in the case of thin metal sheet. The forming strategy is

also investigated to underline its inuence on the resulting formed
part. The rst aim is to measure geometrical proles of truncated
cones resulting from SPIF and to compare the experimental results
with the numerical ones. The study is then completed by a springback analysis of circumferential rings by focusing on the inuence
of the forming tool path on the resulting parts.
The investigations that are related in this paper result from
the previous work carried out at FEMTO-ST Institute and from the
Dejardin (2008) PhD thesis.
3. Experimental investigations
3.1. Experimental platform
Experiments were carried out using a three-axis CNC milling
machine as a platform to develop the ISF process. An experimental
tooling system, where the sheet metal part is supported around its
contour, was xed on the horizontal movable table of the milling
machine. Simple clamps allow to maintain the sheet metal blank
in position during the movement of the tooling system (Fig. 1).
The forming tool consists in a barrel with 10 mm diameter
and hemispherical end shape. A subroutine has been developed
to describe tool path from CAM procedure for simple geometries. In order to compare experimental and numerical results, this

S. Dejardin et al. / Journal of Materials Processing Technology 210 (2010) 363369

365

Fig. 2. Schematic description of the tooling pathstrategy 1.

subroutine generates the same tool path for both CNC milling
machine and numerical simulation in term of discretisation of
forming strategy.

Fig. 3. Investigated geometries through SPIF: a truncated cone with inclination

equal 45 .

3.2. SPIF experiments

The material used for the investigations corresponds to a square
aluminum plate (AA 1050).
The size of the sheet metal part corresponds to 300 mm
300 mm 1 mm, and it is clamped around its contour with simple
clamps as shown in Fig. 1b. The top plate has a 200 mm 200 mm
square orice where the blank could be formed. A mineral oil was
used as a lubricant.
The tool moves along a path that determines the contour of the
required shape at a feed rate equal to 500 mm min1 and a tool rotation equal 400 rpm is imposed. The tool path consists of a series of
circular movements around the vertical axis of the milling machine
combined with an in plane transversal incremental displacement.
The tool rst has a vertical movement by moving along the sheet
metal over a distance equal to the desired vertical step (step 1 in
Fig. 2), then the tool moves along the circumferential path (step
2) as shown in Fig. 2. Once the full circular rotation is completed,
the tool moves rst in the horizontal direction (step 3) and penetrates the blank in the vertical direction to proceed with the next
circumferential tool path.
The investigated data are summarized in Table 1.
Reported experiments are mainly related to the forming of relatively simple part geometries that underlines the possibilities and
the difculties associated with the ISF process (see Fig. 3).
Several experiments were carried out and the resulting sheet
metal proles were measured. The experimental prole was
obtained after a set of several series of measurements by means of
laser method taking into account the transverse sections indicated
in Fig. 4.
The results of the prole measurements as detailed in Fig. 4
are related in Fig. 5. As it can be observed, geometrical proles are
almost exactly the same in both XZ plane (prole 1) and YZ plane
(prole 2) meaning that the part is properly formed and maintains
its symmetry during forming. The only difference noticed is located
along section AB (see Fig. 4), that corresponds to the area where
the circumferential movement of the tool is completed and where
it moves in Z direction to perform the next contour (steps 1, 2 and 3

Fig. 4. Denition of proles used for geometrical measurements.

in Fig. 2). The relative variation between both proles reaches 3.7%.
This difference can be related to the deformation line associated
with the tool vertical movement along this section.
4. Modelling and numerical analysis of SPIF
4.1. Incremental sheet forming conditions and material behaviour
The sheet metal is considered as isotropic and the ow has been
accounted by means of a Swift type hardening law expressed as
0 = k(0 + )n where k is the strain hardening coefcient, n the
power law coefcient and is the effective accumulated plastic
strain.

Table 1
ISF process parameters and geometry.
Tool diameter
Tool depth step
Rotation speed
Feed rate
Major base
Height
Wall inclination

10 mm
0.2 mm
400 rpm
500 mm min1
140 mm
50 mm
45

Fig. 5. Comparison of geometrical proles obtained on the conical geometry considering two axes (XZ and YZ).

366

S. Dejardin et al. / Journal of Materials Processing Technology 210 (2010) 363369

Fig. 6. Deformed mesh of the tronconical part. axisymmetric part.

This choice has been validated through a bulging test that has
been used to determine the material coefcients. The parameters
are as follow: k = 111 MPa, n = 0.14 and 0 = 75 MPa.
Concerning the processing conditions, tool is considered as a
rigid body and the corresponding boundary conditions are related
to the dened path. Friction conditions between the forming tool
and the sheet metal part have been accounted by considering sliding friction with a friction coefcient equal to 0.3.
In a previous work, Ambrogio et al. (2005b) pointed out that a
pressure equal to 2% of the material yield strength (1.5 MPa in the
present case) is enough to clamp the sheet metal part. This result
was conrmed by Dejardin et al. (2007) who performed several
tests in order to analyse the inuence of the holding pressure on
the formed parts.
4.2. Finite element simulation of SPIF
As incremental forming is a progressive sheet metal forming
process characterized by large displacements and large localized
strains, an explicit solution scheme was adopted, resulting in the
choice of LS-Dyna as the FEM simulation software. Explicit FE
method is suited to analyse the forming of simple geometries like
cones or pyramids as it is well described by Bambach and Hirt
(2007) who reviewed the work done in the modelling of ISF.
According to the experimental tests and related data reported
in Section 3.2, the investigated shape used to perform simulation
is a truncated cone with 140 mm major base diameter and 50 mm
height. The slope of the truncated cone is xed to 45 .
Due to the 3D tool path movement, a fully three-dimensional
spatial analysis has been realized. As a consequence, shell elements
with 4 nodes and 6 degrees of freedom per node (Belytschko-Tsai
shell elements) and ve integration points through the thickness
were used. These elements are the so-called 3D reduced shell
elements and are well suited to properly considered thickness variations through the deformation process. Furthermore, an adaptive
mesh renement was performed in order to reduce element size
when the distortion level reaches a maximum value. These ingredients allow a proper modelling and simulation of the progressive
deformation of the sheet metal part by increasing the number
of nodes in contact with the tool surface. The blank was initially
meshed with 3600 nite elements.
The deformed mesh corresponding to the investigated shape
is reported in Fig. 6. One can observe that the expected shape is
globally well reproduced.

Fig. 7. Comparison between numerical, theoretical and experimental prole measurements.

Fig. 8. Location of rings cut from the formed part.

model has been carried out through the comparison between geometrical prole of the part calculated through FE simulation and
the one experimentally measured. Then the investigated results
are focused on the inuence of the tool path on elastic springback
related to the formed parts.
5.1. Geometrical accuracy
In order to setup and to validate the numerical simulation
model, experimental proles were measured and compared with
numerical results. For better clarity of the paper, the results measured on only one prole are presented.
Fig. 7 relates the comparison between the original prole
dened through a CAM system (theoretical prole), the one
obtained through SPIF experiments (experimental prole) and the
one resulting from FE simulations (numerical prole).
In summary about the comparison between numerical and
experimental results, a proper correlation could be observed. Considering the geometrical error dened by the difference between

5. Results and discussion

This section provides an analysis of the obtained results with
regards to the effect of different tool paths characterizing the incremental sheet forming process. In a rst step, a validation of the FEM

Fig. 9. Model used for the simulation of the elastic springback obtained by cutting
and relaxing rings in the wall of the conical part obtained by ISF.

S. Dejardin et al. / Journal of Materials Processing Technology 210 (2010) 363369

367

Fig. 10. (a) Results from springback simulation of tronconical rings extracted from the truncated conetop view (XY plan). (b) Results from springback simulation of
tronconical rings extracted from the truncated coneright view (YZ plan).

two curves, the maximum error no more exceeds 3%. In this sense,
the SPIF Finite Element simulation is suitable to check the feasibility of the required shape of standard workpiece geometries, with a
set of process parameters carefully dened and a forming tool path
as prescribed in Section 3.2.

5.2. Springback analysis

The following section focuses on springback analysis concerning truncated rings cut in the wall of the cone in order to qualify
geometry variations due to residual stresses. This analysis allows to
underline the necessity to take into account the forming strategy
as an important process parameter when analysing ISF deformation mechanism and consequently to use a suitable modelling of
the process.
Springback analysis is carried from rings cut from the formed
part as shown in Fig. 8.
The trimming operation consists in a digital cutting of sheet
metal rings resulting from the complete incremental sheet forming
simulation. This step do not led to more mechanical constraints into
the rings but allows to obtain a new mesh adapted to the geometry of the each considered rings. Then, the principle consists in
cutting rings along a circumferential line in perpendicular plane
to the vertical Z axis. This step is led by doubling nodes localized
along the circumferential line and creating new elements which
allow to relax the sheet metal ring. The nodes that are localized on
the opposite axis are rigidly constrained to allow the development
of springback and to avoid rigid movement during the simulation
stage (Fig. 9).

Fig. 11. Negative elastic springback resulting from the cutting of rings in experimental part.

A negative elastic springback (closing of the ring) can be noticed

in Fig. 10a. Fig. 10b indicates a bigger gap in the horizontal plan (XY)
for rings D, E and F (see Fig. 8) corresponding to the middle of the
wall.
A similar experimental analysis performed on sheet metal part
obtained from incremental forming conrms numerical results
previously described. In this case, rings were cut by means of a wire
electrical discharge machining equipment (WEDM). Fig. 11 shows
the negative elastic springback observed on the experimental
rings.
Measurements of the separated rings position were performed
by using a 3D optical measurement equipment. Table 2 reports the

Table 2
Numerical and experimental elastic springback comparison of rings (forming tool path number 1).
Ring

X num (mm)

X exp (mm)

Relative variation (%)

Y num (mm)

Y exp (mm)

Relative variation (%)

D
E
F

1.402
0.202
0.112

1.504
0.242
0.135

6.78
16.5
17.0

3.115
3.538
3.140

3.350
3.050
3.574

7.01
16.0
12.1

368

S. Dejardin et al. / Journal of Materials Processing Technology 210 (2010) 363369

Fig. 12. Schematic description of the tooling path associated to strategy 2.

movement of the end of cut rings resulting from both experimental

and numerical measurements for rings D, E and F. An error estimation between numerical and experimental results shows a
maximum relative variation of 17% in X direction and 7% in Y direction. These results allow to conclude that the proposed numerical
modelling used is suitable to simulate SPIF process, with a tool
path in alternating directions, a vertical tool pitch equal to 0.2 mm
associated with a 10 mm forming tool diameter and a 1 mm thick
aluminium sheet.
In order to analyse the potential inuence of tool path on the use
of a numerical modelling based on thin shell elements, the following analysis compare results obtained from two forming strategies.
The rst one correspond to the incremental forming strategy (strategy 1) used to performed simulation, that is the same as described
above (cf. Fig. 2). The second strategy (strategy 2) adopted is similar to the rst one, but in this case the digital tool path keep the
same direction to perform each revolution around the conical part
(Fig. 12).
Fig. 13 indicates numerical elastic springback of rings D, E and F
in the horizontal plane XY measurements obtained from both forming strategies 1 and 2. In each case, negative elastic springback can
be noticed. Nodes that are initially joined deviate from each other
in X direction with greater amplitude in the case of strategy 2 than
in strategy 1. In opposite the deviation in Y direction presents a
bigger closing in the case of strategy 1.
Such a result is not necessary amazing. In the case of strategy
1, tool path alternates in direction. This alternative path change
tends to cancel out any shear during successive revolutions. This
hypothesis is in agreement with assumption made by Jackson and
Allwood (2009) by reviewing numerical results from Bambach et al.
(2003) and experimental ones from Allwood et al. (2007). Indeed,
the former predicts that shear dominates whilst stretching is also
signicant in the plane perpendicular to the tool direction. This
remark is also related to the fact that all strain components are
negligible in the direction parallel to tool path direction. Indeed,
in the model used by Bambach et al. (2003), tool path alternates in direction between two revolutions and in experiments by
Allwood et al. (2007), the tool always moves in the same direction.
These observations led the authors to wonder if the use of FE
model with shell elements is suitable in the case of strategy 2.

Fig. 13. Tool path inuence on elastic springback of rings localized in the middle of
the wall of the formed part.

Indeed, with this tool path, transverse shear should be accounted.

The same study as the one described above was performed. Table 3
compares the movement of the cut rings resulting from both experimental results and numerical measurements in the case of forming
tool path number 2. Error estimation shows a relative variation
in the range between 20% and 73%. These results demonstrate
that FE model used is not suitable to predict the behaviour of the
sheet formed by SPIF in the case of a forming tool moving in the
same direction. It also seems to conrm that the choice of alternating tool path could cancel out the inuence of shear, probably
resulting from friction. In a consequence, shell elements are not
suitable to perform SPIF simulation where shear effects are dominants.
5.3. Future investigations
Three suggestions for future work can be raised from the above
discussion. First of all, the results described in this paper that
are based on a FE model using shell elements provide accurate
results in certain cases, but are unsatisfactory in other cases. So
a future work could be associated with the use of a FE model
using elements able to properly account shear behaviour and to
verify assumptions made in the present study. Besides, observations described in this paper are associated with a set of dened
material and process (SPIF, sheet thickness, tool diameter, and
vertical pitch). The second suggestion is to investigate the inuence of forming tool path on a wide range of process variables
and to carry out an analysis in order to evaluate the sensitivity
associated with shear through sheet metal thickness and in plane
shear. The third suggestion is to develop well focused numerical models to simulate incremental sheet forming process and
to develop tool path design, process control, in order to build
a better understanding of the process and as well to optimize
it.

Table 3
Numerical and experimental elastic springback comparison of rings (forming tool path number 2).
Ring

X num (mm)

X exp (mm)

Relative variation (%)

Y num (mm)

Y exp (mm)

Relative variation (%)

D
E
F

0.914
1.137
1.122

0.640
0.655
0.723

42.8
65.5
72.3

1.730
1.638
1.185

4.350
2.050
3.574

60.2
20.1
66.8

S. Dejardin et al. / Journal of Materials Processing Technology 210 (2010) 363369

6. Conclusion
The paper investigates and demonstrates the capabilities of
single point incremental sheet forming to provide sheet metal components with accuracy. It also demonstrates the accuracy and the
reliability of 3D nite elements simulation of the process, carefully accounting the tool path strategy. The paper also focuses
on the springback resulting from SPIF. It has been demonstrated
through experiments and nite elements analyses that springback,
characterized through the cut rings method, can be accurately
predicted from numerical simulations based on shell elements
associated with a suitable forming tool path. In the case of springback associated with in plane shear, it is necessary to perform
simulation based on 3D elements accounting shear. In summary,
the paper demonstrates the possibility offered by single point
incremental forming as well as nite elements possibilities and
requirements.
References
Allwood, J.M., Shouler, D.R., Tekkaya, A.E., 2007. The increased forming limits of
incremental sheet forming processes. In: SheMet07 International Conference
on Sheet Metal, Palermo, Italy, pp. 621628.
Ambrogio, G., Napoli, L.De., Filice, L., Gagliardi, F., Muzzupappa, M., 2005a. Application of incremental forming process for high customized medical product. J.
Mater. Process. Technol. 162163, 156162.
Ambrogio, G., Filice, L., Micari, F., Gaglardi, F., 2005b. Sheet incremental forming: a
new process conguration allowing a sheet material controlled ow under the
blank-holder. In: Proceedings of the 8th International Conference on Technology
of Plasticity, Verona, Italy, pp. 351352.

369

Amino, H., LU, Y., Ozawa, S., Fukuda, K., Maki, T., 2002. Dieless NC forming of automotive service panels. In: Proceedings of the Conference on Advanced Techniques
of Plasticity, pp. 10151020.
Bambach, M., Hirt, G., Junk, S., 2003. Modelling and experimental evaluation of the
incremental CNC sheet metal forming process. In: Proceedings 7th International
Conference on Computational Plasticity, Barcelona, Spain, pp. 115.
Bambach, M., Hirt, G., 2007. Error analysis in explicit nite element analysis of incremental sheet forming. In: Proceedings of International Numiform 07 Conference,
Porto, Portugal, pp. 859864.
Dejardin, S., Thibaud, S., Gelin, J.-C., 2007. Experimental and numerical investigations
in single point incremental. In: Proceedings of the 10th International ESAFORM
Conference. American Institute of Physics, 907, pp. 933938.
Dejardin, S., 2008. Experimental investigations, modelling, simulation and control
of incremental sheet forming process of thin metal part. PhD Thesis, FEMTOST Institute, Department of Applied Mechanic, University of Franche-Comte,
France.
Filice, L., Fratini, L., Micari, F., 2002. Analysis of material formability in incremental
sheet forming. Ann. CIRP, 199202.
Fratini, L., Ambrogio, G., Lorenzo, R.D., Filice, L., Micari, F., 2004. Inuence of mechanical properties of the sheet material of formability in single point incremental
forming. Ann. CIRP 53 (1), 207210.
Hirt, G., Junk, S., Bambach, M., Chouvalova, I., 2003. Process limits and material behavior in incremental sheet forming with CNC-tools. Mater. Sci. Forum
426432, 38253830.
Jackson, K.P., Allwood, J.M., Landert, M., 2008. Incremental forming of sandwich
panels. J. Mater. Process. Technol. 204, 290303.
Jackson, K.P., Allwood, J., 2009. The mechanics of incremental sheet forming. J. Mater.
Process. Technol. 209, 11581174.
Jeswiet, J., Micari, F., Hirt, G., Bramley, A., Duou, J., Allwood, J., 2005. Asymmetric
single point incremental forming of sheet metal. Ann. CIRP 54 (1), 623650.
Jeswiet, J., Young, D., 2005. Forming limit diagrams of single-point incremental
forming of aluminium sheet metal. Proc. IMechE: J. Eng. Manufact. B 219,
16.
Micari, F., Ambrogio, G., Filice, L., 2007. Shape and dimensional accuracy in single point incremental forming: state of the art and future trends. Int. J. Mater.
Process. Technol..