Downloaded from SAE International by University of Wisconsin - Madison , Monday, September 10, 2018
SAE TECHNICAL
PAPER SERIES
Seat Comfort
Janilla Lee and Paul Ferraiuolo
Ford Motor Co.
Reprinted from:
Seat System Comfort and Safety
**
and =For
The Engineering Society
Advancing Mobility
sea AB and Space,
International Congress and Exposition
Detroit, Michigan
March 1-5,1993
400 Commonwea!th Drive,Warrendale, PA 15096-0001U.S.A. Tel: (412)776-4841 Fax:(412)776-5760
Downloaded from SAE International by University of Wisconsin - Madison , Monday, September 10, 2018
The appearanceof the ISSNcode at the bottomof this page indicatesSAE's consent
that copies of the paper may be made for personal or internal use of specific clients.
Thisconsent is givenon the condition, however, that the copier pay a $5.00per article
copy fee through the Copyright Clearance Center, Inc. Operations Center, 27
CongressSt., Salem, MA 01970 for copying beyondthat permitted by Sections 107
or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of
copying such as copying for general distribution, for advertising or promotional
purposes, for creating new collective works, or for resale.
SAE routinely stocks printed papers for a period of three years following date
of publication. Direct your orders to SAE Customer Sales and Satisfaction Depart-
ment.
Quantity reprint rates can be obtained from the Customer Sales and Satisfaction
Department.
To request permissionto reprint a technical paper or permissionto use copyrighted
SAE publications in other works, contact the SAE Publications Group.
GLOBAL MOBIU7Y DATABASE
All SAEpapers, standards, and selected
books are abstractedand indexed in the
SAE GlobalMobility Database.
No part of this publicationmay by reproduced in any form, in an electronic retrieval
system or othetwise, without the prior written permission of the publisher.
ISSN 0148-7191
Copyright 1993 Society of Automotive Engineers, Inc.
Positions and opinions advanced in this paper are those of the author(s) and not
necessarily those of SAE. The author is solely responsible for the content of the
paper. A process is available by which discussions will be printed with the paper if
it is published in SAE transactions. For permission to publish this paper in full or in
part, contact the SAE Publications Group.
Persons wishing to submit papers to be considered for presentation or publication
through SAE should send the manuscript or a 300 word abstract of a proposed
manuscript to: Secretary, Engineering Activity Board, SAE.
Printed in USA 901 ZO~BIPG
Downloaded from SAE International by University of Wisconsin - Madison , Monday, September 10, 2018
Seat Comfort
Janilla Lee and Paul Ferraiuolo
Ford Motor Co.
ABSTRACT office or household furniture, automotive seats are
usually confined and always under dynamic conditions
As customer expectations rise, automotive seat during their usage. Not only does an automotive seat
comfort is becoming an increasingly important design need to provide comfortable support, but it has to allow
goal. This paper explores the application of two for comfortable operation of the vehicle. Therefore,
objective measurables to the design of automotive seats. measurables for automotive seats will need more than
These measurables are EMG (electromyograph) data just static evaluations and the comfort requirements will
and seat pressure distribution data. An attempt is made be very different than those for furniture.
to correlate these measurables with subjective comfort. If positive correlations5 can be established between
An experiment is designed to collect the required subjective and objective measurables of seat comfort the
analytical and subjective data. This data is then task of designing a comfortable seat will be greatly
analyzed statistically to discover any correlations that simplified. If negative correlations between subjective
may exist. The resulting correlations found in the and objective measurable are found the engineer will be
statistical analysis are not large enough to be the basis required to optimize the seat to specific applications. For
for seat design, but indicates that with further work seat example long term drive comfort might only be achieved
comfort may be measured objectively. The results of this at the expense of static "showroom' comfort. It is then
research will lead the direction for further work. up to the designer to determine the balance which is
best for his application.
INTRODUCTION AND BACKGROUND This paper will discuss two objective measures of
seat comfort. As the first part of a two part study this
The importance of comfortable seat design is paper only addresses static comfort. This initial study of
increasing as customer expectations rise. In the static comfort will establish the background information
automotive seating industry there has been a recent needed to attack the problem of seat comfort under
struggle to define seat comfort in an objective manner. dynamic conditions, which will be the second part of the
This is a very difficult task, as seat comfort is a highly study.
subjective matter. Due to the lack of analytical
measurables, the seating industry has relied on jury DESIGN OF EXPERIMENT
evaluations as the main measure of seat comfort. This
trial and error approach is very time consuming, Designing for seat comfort is a difficult task. In order
expensive, and prone to measurement error. to begin to understand how various automotive seat
Wih the advancement of technology, several design parameters affect seat comfort more information
objective measurables of seat comfort have evolved.'p2f3 on this relationship is needed. A Taguchi approach was
None of these objective measurables has emerged as a used to design an experiment to investigate the
singular predictor of seat comfort, but in combination contribution and significance of various seat design
may lead to a better understanding of how to design a parameters. This consists of a two-level factorial design.
comfortable seat. As the technologies progress it is This type of experiment is extremely useful for this
almost certain that in the future there will evolve certain purpose. Factorial designs not only estimate the main
standard objective measures of seat comfort. effects, but also interactions between variables. This
The objective measurables of seat comfort should be allows determination of how a combination of different
focused on occupant's orthopedic requirements4 A seat design parameters may affect seat comfort where a
good vehicle seat should be perceived as comfortable by single parameter has little effect.
the customer under varying conditions. Different from
Downloaded from SAE International by University of Wisconsin - Madison , Monday, September 10, 2018
DESIGN AND BUILD - To explore the effects of letters for record keeping purposes and to eliminate any
various design parameters on seat comfort several test ordering bias.
seats were constructed using the two-level factorial SURVEY - In order to determine the subjective
approach. The choice of parameters was based on the comfort of each of the test seats, participants were
known contributors to seat comfort and the conventional surveyed. Due to the subjective nature of seat ~0mf0R
design variables that an engineer may alter. The a large sample size of 100 participants was used to
parameters that were varied were as follows: achieve statistical relevance. The survey consisted of an
initial tutorial for the participants and an information
foam thickness gathering phase prior to the subjective rating of the
foam hardness seats.
back contour In this initial tutorial and information gathering phase
back angle6 the participants were given a description of the actions
cushion angle required. Limited anthropometric data was gathered on
spring suspension rates each participant. This data included the participant's
side support age, height, weight, and sex.
As the participants were lead through all of the test
Each of these parameters was set at two levels (above seats they were requested to provide a numerical rating,
and below the currents design for a particular seat on a scale from 0 to 10, of the perceived comfort at
chosen as the standard for this study). All structures of various body regions as they sat in each seat.7 These
the test seats use the standard seat structure. The regions included the following:
amount of deviation of each of the parameters from the
current design was determined by engineering neck
judgement to be within practical limits and generated a upper back
significant change. For a full factorial design this would middle back
incorporate 128 different designs. This was reduced to lower back
a more manageable size of 16 seats for practicality thighs
purposes. Designs were selected which included an buttocks
adequate mix of the 7 design parameters in a proper calves
statistical array. The selection of these parameters is chest
based on previous experience of interactions that are shoulder
believed to have a significant effect on seat comfort. sides
The 16 test seats were fabricated from production overall seat comfort
representativeequipment and processes to simulate real
world conditions. The seats were constructed to be The participants were allowed only to vary the seat track
similar in appearance to eliminate any aesthetic factors positionto provide the proper relationship to the steering
from the analysis. In order to get the proper seating wheel and floor board. The seats were surveyed in a
position, bucks were build to simulate the floor pan random order which differed for each participant. The
geometry and steering wheel position for a current participants were required to sit in each seat for a
vehicle (Figure 1). The seats were assigned random
I
I 1
Figure 1, Seating Buck.
2
Downloaded from SAE International by University of Wisconsin - Madison , Monday, September 10, 2018
minimum of 2 minutes. This was done in an attempt to Electromyography (EMG) was also used as a
simulate a 'showroommcomfort analysis. possible predictor to measure the comfort index of a seat
design. EMG measures the electrical activity of the
muscle action potential. Although the relationship
PRESSURE DISTRIBUTION between EMG activity and mechanical force of muscle
contraction is far from straightforward, it has been a
Pressure distribution information was employed to useful tool in predicting muscle activities. It is assumed
attempt to apply objective data to the subjective that too much EMG activity will lead to muscle fatigue.
interpretation of seat comfort. Each of the test seats was However, a certain minimum amount of muscle activity is
equipped with a pressure sensitive mat on the seat needed to ensure blood circulation and to avoid
cushion and back. Each participant was asked to numbness. Therefore, when a person sits at the most
assume a driving posture to aid in the evaluation of the relaxed posture with comfortable support from the seat
test seat for comfort. While in this posture, pressure cushion, all muscles have a limited amount of muscle
readings were taken. This was done in a manner that activity.
was transparent to the participant so that they were not This study was in conjunction with Wayne State
inhibited while completing the subjective evaluation. Universrty (WSU) Bioengineering Center, Department of
The pressure sensitive mats were securely attached Orthopaedic Surgery and Department of Physical
to the seat cushion and back. Care was taken to ensure Therapy. Before the tests, the entire procedure was
that the mats were attached in a consistent location from explained to the subject in detail. Subjects were then
seat to seat (Figure 2). prepared by a single investigator. Seven muscles in a
The pressure sensitive mats consist of an array of unilateral side were cleaned with rubbing alcohol to
pressure transducers arranged in 44 rows by 48 remove body oils. Surface EMG electrodes were then
columns. The pressure is measured at 10mm intervals. fixed to the skin over each muscle. The electrode set
This array of pressure transducers produces 2112 contains a grounding electrode and the EMG is recorded
numbers representing the pressures at each position on as the voltage difference between the two remaining
the mat. This information may be displayed graphically electrodes. The seven muscles studied are listed below:
or analyzed numerically to produce various quantitative
values for each body region. The quantitative values neck (semispinalis cervices)
included in this analysis were: maximum pressures, shoulder (trapezius)
average pressures, average gradients, and maximum lumbar (erector spinae)
gradients. quadriceps (rectus femoris)
lateral hamstrings (biceps femoris)
ELECTROMYOGRAPHY medial hamstrings (semitendinosus)
Figure 2, Pressure Sensitive Mat on Seat Cushion.
3
Downloaded from SAE International by University of Wisconsin - Madison , Monday, September 10, 2018
anterior tibia (tibialis anterior) group 2: chest, neck, shoulder, upper back
group 3: middle back, lower back, buttocks, sides
This selection is based on previous experience and
experimentation. Automotive seating is believed to The degree of comfort at body regions in each group
influence these muscles the most. are strongly correlated. The correlations between the
Once the test participant is instrumented with overall seat comfort and the comfort at the 10 body
electrodes to measure the muscle activity, a maximum regions are all rather strong, while the correlations
voluntary isometric contraction (MVIC) was recorded to between overall comfort and sex, weight, height, and age
be used in later data analysis as the 100% muscle are all very weak.
activity. The participant was then led through the 16 test The method of stepwise regression is used to
seats in the same procedure as used in the pressure express, quantitatively, the overall seat comfort in terms
distribution survey. While the participant was seated in of the degrees of comfort at the 10 body regions and
a driving posture their muscle activities were recorded at other variables. This method results in the following
2000 Hz simuJtaneously with pressure distribution data equation:
recording. After the subjective evaluationwas completed
the subject then moved to the next seat. -
CWeraII= 1.317 0.006 AGE - 0.0159 SEX
The raw EMG data was rectified and then a linear +
-0.004 WEIGHT 0.080 C1 + 0.124 C2
envelope procedure was employed. The resting EMG of -0.086 C, +
0.144 C, 0.187 C, +
0.066 C, +
each subject was averaged to a single number for each i 0.1 42 C, +
0.107 C, 0.207 C, +
seat. Thus, each subject has 16 values for each of the
seven muscles. The data was then averaged over the Where C,, C,, ...,C, are the 10 body regions: chest,
subjects for each muscle. These number were finally thigh, calf, neck, upper back, middle back, lower back,
compared to the subjective evaluation results. buttocks, shoulder, and side respectively. This equation
explains about 80% of the total variation. Considering
RESULTS the highly subjective nature of the data this is a very
good model.
COMFORT SURVEY RESULTS - The seat comfort The method of analysis of variance is used to find
survey participants were asked to rate the 16 test seats whether the effects of the design factors and the seats
in 11 different areas. These include 10 body regions and are significant to the overall comfort. At the 5%
overall comfort as well as the anthropometric data. This significance level the following is found:
data was analyzed using three different statistical The total effect of the design factors and the seats
methods: correlation analysis, regression equation fi, is significant.
and analysis of variance. The effect of the factors of foam thickness, foam
The correlation coefficients between any two of these hardness, side support, the interaction of back contour
variables is shown in Table 1. This table shows that and suspension rate, and the interaction of back angle
these variables may be divided into 3 groups due to their and cushion angle are each significant.
high correlation among the individual variables: a There is a significant difference between each of
the 16 test seats.
group 1: thighs, calves
Correlation Coeficients Between Subject Variables
Tablel.
4
Downloaded from SAE International by University of Wisconsin - Madison , Monday, September 10, 2018
The effects of back angle and the interaction parameters. The work presented in this paper is only the
between cushion angle and side support are each nearly foundation for future studies in EMG.
significant at the 5% level.
It is found that lighter participants, shorter ACKNOWLEDGEMENTS
participants, and female participants are more sensitive
to the changes of the levels of the design parameters. Many thanks to Roger Burnett, Majdi Zahra, and
PRESSURE DISTRIBUTION RESULTS - In order to Wayne State University for their contributions.
statistically interpret the pressure distribution data,
quantitative numbers must be generated from the large REFERENCES
array of pressures. To achieve this the pressure
sensitive mat was divided into various regions where 1. Althoff et al, An Improved Method of Stature
numbers corresponding to maximum pressure, average Measurement for Quantitative Determination of
pressure, average gradients, and maximum gradients Spinal Loading, Spine, Vol 17, No 6, 1992.
could be calculated. These regions were determined by
visually inspectingthe pressure distributions for areas of 2. Nagashima, Hideyuki, Technological Development of
interest, taking into account the geometric properties of Automotive Seat Comfort, JSAE, Vol 45, No 12,
the seat. These numbers were then statistically analyzed 1991.
to determine the effects on perceived comfort. This was
performed with limited success. The correlations found 3. Sheridan, Thomas B. et all Physiological and
between the various pressure terms and subjective PsychologicalEvaluations of Driver Fatigue During
comfort were not high enough to be a basis for any Long Term Driving, SAE Paper 910116, 1991.
design decisions. Further work is currently being done
to extract more relevant data from these pressure arrays. 4. Viano, David C. and Andrzejak, Dennis V., Research
Issues on the Biomechanicsof SeatingDiscomfort:
EMG RESULTS - It was found that the seats with the An Overview with Focus on Issues of the Elderly
highest six EMG numbers for the neck muscles are the and Low-Back Pain, SAE Paper 920130, 1992.
best six seats rated by the subjective survey (n=79).
Shoulder and medial hamstrings predictedfour out of the 5. Kamijo, Ken et al. Evaluation of Seating Comfort, SAE
top six seats ranked by the subjective survey while the Paper 820761, 1982.
other muscles predicted three out of the six. For the
worst six seats rated by the subjective survey, the neck 6. Treaster, Delia E. and Marras, William S., Seat
muscle and medial hamstrings identified five while the Pressure: Measurement and Analysis, SAE paper
shoulder muscle identified four, lumbar and quadriceps 890849, 1989.
identified three, with the remaining muscles identified
two. More tests are needed to establish the relationship 7. Arrowsmith, M. J., The Design and Development of the
between EMG and subjective seat comfort. More tests U40 Seating System, Proc lnstn Mech Engrs, Vol
are also needed to study the long term driving situation. 200, No D5, 1986.
CONCLUSIONS
From the subjective evaluation the importance of
each design parameter was determined. The ranking of
these variables allows the most important parameter to
be selected for the concentration of design efforts to
produce a more comfortable seat. This result is still
more qualitative than quantitative and is constrained to
design parameters within the range of the study.
, The analysis of the pressure distribution data does
not show that pressure distributions and seat comfort are
correlated highly enough to be the basis for design.
Further work in breaking the pressure array into
meaningful data may result in stronger correlations.
Although the current correlations are weak they do
indicate the direction for further studies in seat pressure
distributions.
At this stage the EMG data is not conclusive.
Different approaches to data analysis are necessary to
fully understand the correlation between EMG and
comfort. Once this relationship is understood more
experiments will be needed to study specific seat design
