Brief report

Modeling the body/chair interaction an integrative

experimentalnumerical approach
T. Brosh
a
, M. Arcan
b,
*
a
School of Dental Medicine, Tel Aviv University, Israel
b
Faculty of Eng., Tel Aviv University, P.O. Box 39040, Tel Aviv 69978, Israel
Received 26 April 1999; accepted 14 September 1999
Abstract
Objective. To develop a systematic methodology towards a realistic model, of the body/chair interaction as a tool to analyze
sitting posture and the cushioning system inuence on the pelvis/lower back stress development.
Background. The model concept comprises an integrative structure including both the pelvis girdle and the lower spine, as well as
the surrounding soft tissues with their specic characteristic behavior during sitting.
Methods. In vivo measurements were performed, in order to obtain a set of contact moduli dening the behavior of the soft
tissues under tension during the sitting down process. Additionally in vivo indentation of a metal ball into the same soft tissues
during sitting were performed in order to obtain the characteristic moduli. A nite element model was also developed for the specic
analyses.
Results. Validation of the model was achieved by comparing its results with in vivo measurements of contact stresses developed
between the body and a sti target seat area. Loading the model using two alternative cushioning materials lead to dierent sets of
stresses within the model; as the stiness of the seat decreased, the peak contact stresses, as well as the internal body stresses
substantially decreased.
Relevance
Body/chair interaction model methodology may be a practical means to analyze and design structures and materials in order to
achieve more comfort during work or leisure, as well as for geriatric or impaired subjects. 2000 Elsevier Science Ltd. All rights
reserved.
Keywords: Body/chair interaction; Sitting down mechanics; Soft tissue mechanics; Contact stress; Contact moduli
1. Introduction
This work presents a systematic methodology towards
developing a realistic nite element model of the body/
chair interaction. The model comprises the pelvis girdle
and the lower spine, as well as the surrounding soft tis-
sues. Soft tissue properties had to be measured and as-
sessed in situ. For the sitting contact, these tissues are
rst stretched and only afterwards perpendicularly
loaded. A set of experiments was aimed for these tissues
behavior determination, during sitting. The strain energy
function W = (Ga2)(I
1
3), where I
1
, is the rst in-
variant of the extension ratio matrix, can characterize
such properties [13]. Thus, the shear modulus of the
relevant tissues should be dened. Validation of the
model is obtained, for one case as an example, by com-
paring the results with in vivo measurements of contact
stresses developed between the body and the seat area.
2. Methods
Two approaches were used for analyzing shear
modulus of the soft tissues during sitting: (i) a sitting
down experiment which provided a set of contact
moduli that were then transformed into a set of shear
moduli and (ii) an indentation test which yielded an
additional set of shear moduli.
(i) The Contact Pressure Display (CPD) method [4,5]
was used for experimental measurements of contact
stresses between a seated subject and the horizontal sti
area the laboratory chair. These data served also for
Clinical Biomechanics 15 (2000) 217219
www.elsevier.com/locate/clinbiomech
*
Corresponding author.
E-mail address: arcan@eng.tau.ac.il (M. Arcan).
0268-0033/00/$ - see front matter 2000 Elsevier Science Ltd. All rights reserved.
PII: S 0 2 6 8 - 0 0 3 3 ( 9 9 ) 0 0 0 7 3 - X
calculating the contact moduli. A loaded strip, including
the two peak zones under the ischial tuberosities (Fig. 1)
characterizes the buttocks/chair contact stress distribu-
tion. The subject, dressed with tted elastic pants, with
marks along his side, had to change his posture from
standing to sitting on the laboratory chair. A reference
lateral frame on the chair was scaled using two mea-
suring strips with milimetric marks and the CPD plate
was also marked with reference points. These reference
marks were used during video analysis, for locating
body position related to the chairs frame and the con-
tact area of the buttocks related to the seat area during
the sitting down process. Two video cameras (National
Panasonic, F10, PAL), equipped with a digital clock,
each connected to a videotape (SONY 5830, U-Matic
3a4
//
), recorded simultaneously the process during two
experimental setups. One camera recorded the kine-
matics of the subject's body (rst setup-side view, second
setup-back view). The other camera recorded the con-
tact patterns observed through the CPD plate in both
setups. Image processing and analysis were performed at
0.1 s intervals. By digitizing each of the marks on the
subjects side, using a laboratory program analysis, the
body positions through each time interval were ac-
quired, with a resolution of 1 mm, related to the refer-
ence frame. Vertical displacements of the Ilium during
the process and the contours of pelvis movement, re-
sulting from back view recordings, were obtained. The
image processing of the CPD [4] scanned and analyzed
the contact pattern for the same times as for the kine-
matics. The characteristic loaded strip (Fig. 2) was ob-
served during the whole sitting down process. A contact
stressstrain curve was constructed. The contact stress
distribution diagrams obtained for the loaded strip
could be considered the result of a 2-D pelvis structure
coming into contact with the chair seat area. The model
contour during sitting was plotted using the data ob-
tained from the back view of the nal stage of the sit-
ting down process experiment. X-ray radiographs
obtained during basic sitting posture of the subject
allowed location of the bony parts within the above
contour line while sitting. The thickness of the com-
pressed soft tissues was increased with 14 mm due to the
amount of vertical deformation (kinematics analysis),
from rst contact to full sitting posture achieved. This
procedure allowed evaluating the initial thickness of soft
tissues under the ischial tuberosities after stretching,
before contact occurred. Strains were calculated using
the ratio of vertical displacements to the initial thickness
of these tissues. The resulting strains were related to the
average contact stresses developed during the sitting
down process experiment at the above mentioned peak
zone; this way the nal contact stressstrain diagram
was obtained. The curve could be characterized as bi-
linear, with two slopes dened as a set of Contact
moduli and an intersecting point S using the Macauley
notation:
r = C
1
e C
1
e e
+
) kC
1
e e
+
)Y (1)
where C
1
= 31X9 kPa, kC
1
= C
2
= 87X3 kPa with the
coordinates of S: e
+
= 0X17, r
+
= 5X5 kPa.
The contact moduli were transformed into shear
moduli using the relation G = (Ea2)a(1 t), with
E = C
i
and m = 0X5 resulting in G
1
= 10X6 kPa and
G
2
= 29X1 kPa.
(ii) An indentation test of a sphere into the soft tissues
during sitting was used. In such an approach, the rela-
tionship [6] between indentation d, shear modulus G,
radius of the sphere R, and total applied load, P is given
by: (Rd)
3a2
= (3a8)(1a2G)RP. A lever system that in-
dents a rigid smooth metal ball into the buttock tissue
was set up under a chair with a small part of the seat
Fig. 2. Development of contact stress distribution on the loaded strip
during the sitting-down process and during dierent stages of model
loading due to symmetry vs. the sagittal center line (C
L
) only half di-
agrams are plotted. Fig. 1. Characteristic contact stress pattern on the sitting area as ob-
served through the CPD technique. The loaded strip and peak zones
are marked.
218 T. Brosh, M. Arcan / Clinical Biomechanics 15 (2000) 217219
area cut out. Loading the lever, at the other end, caused
the ball to indent the soft tissue of the seated subject.
Displacement indentation measurements were perform-
ed using a Linear Variable Dierential Transducer
(LVDT). Substituting the data (loads and indentations)
into the equation revealed again a bi-linear diagram,
representing the behavior of these tissues when stret-
ched, with two slopes dening a similar set of Shear
moduli: G
1
= 11X7 kPa, G
2
= 33X8 kPa.
3. The model
The nite element analysis adina program, allowing
calculations of contact stresses at the body/target inter-
face was used for this model. The 2-D, plane stress,
symmetrical model (Fig. 3) consists of both bony parts
(E = 20 GPaY m = 0X3) of the pelvis (including the lower
part of the spine) and the soft tissues (using the above G
moduli). The model was loaded in 14 increments of
1 mm. Three plane support targets (seats) with 30 mm
thickness were used: sti (E = 10 GPaY m = 0X3), semi-
soft (E = 20 MPaY m = 0X2) and soft (E = 3 MPaY
m = 0X1).
Contact stress distribution diagrams developed be-
tween the model soft tissues and the at sti target were
compared with experimentally obtained data. The
distribution was characterized by a peak zone with high
contact stresses (~40 kPa) when compared to the
average contact stress distribution (less than 10 kPa)
(Fig. 2). The principal stresses were characterized by
high values in the zone of soft tissues under the ischial
tuberosities and almost negligible at the very lateral area
of soft tissues. The results obtained from the model,
using semi-soft and soft targets showed that peak
contact stresses were reduced with 54% and 80%
when the elastic modulus of the target body was
changed from sti to semi-soft and soft, respectively.
Similar changes were found in the maximal spine
compressive stresses.
4. Discussion and conclusions
The data obtained from the two experimental ap-
proaches (i) and (ii) dening the basic mechanical
properties of soft tissues, provided similar G
i
values. In
addition, these tissues are of lower stiness when under
small contact stresses and of much higher (threefold)
stiness when under relatively higher stresses. Since,
bones geometry and position are signicantly inuenc-
ing the contact stresses developed in the soft tissues
during sitting (see also [7]), the advantages of such in
vivo measurement methodologies are obvious and ap-
pear as a way of approaching dierent categories of
body/chair interaction and nd the optimally tted chair
design solutions.
The good agreement between the results of the ex-
perimental and numerical approaches of contact stress
analysis could not have been achieved without the
accurate data concerning the soft tissue mechanical
behavior during sitting. The results also validate the bi-
modulus concept and that of the strain energy function,
which describe the mechanical behavior of the soft
tissues during sitting.
The whole methodology developed along the above
described process allows an integral evaluation of the
body/chair system and implicitly of sitting as a basic
human posture. The model is also relevant in designing
modern cushion structures and can serve as an aid for
selecting dierent material components in order to re-
duce contact stress concentrations and the related low
back stresses.
References
[1] Chow WW, Odell EI. Deformations and stresses in soft body
tissues of a sitting person. J Biomech Eng 1978;100:7987.
[2] Reddy NP, Patel H, Cochran GVB, Brunski JB. Model experi-
ments to study the stress distributions in a seated buttocks. J
Biomech 1982;15(7):493504.
[3] Candadai RS, Reddy NP. Stress distribution in a physical buttock
model: eect of simulated bone geometry. J Biomech
1992;25(12):140311.
[4] Brosh T, Arcan M. Toward early detection of the tendency to stress
fractures. Clin Biomech 1994;9:1116.
[5] Arcan M. Non invasive and sensor techniques in contact mechan-
ics: a revolution in progress. In: Proceedings of the Nineth
International Conference on Experimental Mechanics, vol. 1,
Copenhagen, Denmark, 1990;118 (invited paper).
[6] Johnson KL. Contact Mechanics. Cambridge: Cambridge Univer-
sity Press, 1985;15396.
[7] Sacks AH. Theoretical prediction of a time-at-pressure curve for
avoiding pressure sores. J Rehabilitation Res and Develop
1989;26(3):2734.
Fig. 3. Finite element model during sitting posture: (a) before loading
(b) after loading on sti target. (Note: the vertical and lateral defor-
mations of soft tissues).
T. Brosh, M. Arcan / Clinical Biomechanics 15 (2000) 217219 219