Applied Ergonomics xxx (2013) 1e7

Contents lists available at SciVerse ScienceDirect

Applied Ergonomics
journal homepage: www.elsevier.com/locate/apergo

Muscle fatigue based evaluation of bicycle design

V. Balasubramanian*, M. Jagannath, K. Adalarasu
Rehabilitation Bioengineering Group, Department of Engineering Design, IIT Madras, Chennai 600036, India

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

Article history: Bicycling posture leads to considerable discomfort and a variety of chronic injuries. This necessitates a
Received 8 July 2011 proper bicycle design to avoid injuries and thereby enhance rider comfort. The objective of this study was
Accepted 17 April 2013 to investigate the muscle activity during cycling on three different bicycle designs, i.e., rigid frame (RF),
suspension (SU) and sports (SP) using surface electromyography (sEMG). Twelve male volunteers
Keywords: participated in this study. sEMG signals were acquired bilaterally from extensor carpi radialis (ECR),
Bicycle design
trapezius medial (TM), latissimus dorsi medial (LDM) and erector spinae (ES), during 30 min of cycling on
Surface electromyography (sEMG)
each bicycle and after cycling. Time domain (RMS) and frequency domain (MPF) parameters were
Posture
RGB pain scale
extracted from acquired sEMG signals. From the sEMG study, it was found that the fatigue in right LDM
Muscle fatigue and ES were signicantly (p < 0.05) higher in SP bicycle. This was corroborated by a psychophysical
assessment based on RBG pain scale. The study also showed that there was a signicantly lesser fatigue
with the SU bicycle than the RF and SP bicycles.
2013 Elsevier Ltd and The Ergonomics Society. All rights reserved.

1. Introduction
Bicycle is a popular and economical mode of human-powered
transportation. This also enables it to be used as effective equip-
ment towards tness and rehabilitation (Balasubramanian and
Jayaraman, 2009). However, bicycling demands one to bend for-
ward while pedaling; this prolonged forward exion posture may
increase the risk of chronic injuries such as musculoskeletal dis-
orders (MSD), compression neuropathies, and so on. Hence, proper
bicycle design is necessary to reduce MSD and enhance comfort for
riders (Laios and Giannatsis, 2010; Mellion, 1994; De Vey Mestdagh,
1998).
Cyclists adopt a round-back or at-back posture to reach the
handlebars by exing their pelvis and spine (Burke, 1996; Usabiaga
et al., 1997). Cyclists who maintain a prolonged aerodynamic
posture experience pronounced stress on their shoulder, neck and
low back pain (Mellion, 1994; Burnett et al., 2004). For example, the
working schedule for a bicycle messenger is normally around 40 h
per week. This prolonged bicycling demands a very high physical
capacity with high energy expenditure which often exceeds the
recommended level (mean oxygen uptake (VO2) during an 8 h
working day should not exceed 30% of the maximum capacity of
oxygen uptake) and subsequently develops fatigue (Bernmark et al.,
2006). An individuals control over the bicycle depends on his
* Corresponding author. Tel.: 91 44 2257 4117, fax: 91 44 2257 4732.
E-mail addresses: chanakya@iitm.ac.in, chanakya@live.com (V. Balasubramanian).
capability to balance the upper body and the strength of his core
muscles. When less energy is spent on balancing the upper body, an
individual can spend more energy on powering the bicycle (using
the lower body). This is the main reason why cycling training
centers are considered important; to improve the efciency and
health of the upper body, especially because of the strength they
provide to the lower back muscles which is one of the primary
posture section of a cyclists body (Li and Caldwell, 1998).
Muscle fatigue is an important feature affecting cycling perfor-
mance. It has been reported that muscle fatigue would alter the
cycling motion and muscle activation pattern (So et al., 2005).
Houtz and Fisher (1959) investigated muscles activity patterns on a
stationary bicycle using surface electromyography (sEMG). During
a fatiguing contraction, the sEMG amplitude has been established
to increase in submaximal contractions (De Luca, 1984; Krogh-Lund
and Jorgensen, 1992; Van Dieen et al., 1993; Ferguson et al., 2012).
At the same time, there is an increase in the lower frequency
components of the sEMG spectrum during sustained contractions.
These frequency changes have been used as an indicator of muscle
fatigue (De Luca, 1984, 1993). During prolonged bicycling, muscle
fatigue results from metabolic changes in the recruited muscle -
bers, causing a decrease in the force-generating capacity of the
skeletal muscle (Gibson et al., 2001).
Endurance cycling requires prolonged exed posture, which
appears to be one of the main reasons for lower limb and lower
body problems (Callaghan, 2005; Hausswirth et al., 2010). Since
cycling relies almost entirely on the lower body for propulsion
across the terrain in a seated position, numerous cycling studies

0003-6870/$ e see front matter 2013 Elsevier Ltd and The Ergonomics Society. All rights reserved.
http://dx.doi.org/10.1016/j.apergo.2013.04.013

Please cite this article in press as: Balasubramanian, V., et al., Muscle fatigue based evaluation of bicycle design, Applied Ergonomics (2013),
http://dx.doi.org/10.1016/j.apergo.2013.04.013
2 V. Balasubramanian et al. / Applied Ergonomics xxx (2013) 1e7

have examined mostly the lower extremity muscles (Brown et al., each participant riding all three bicycles as shown in Fig. 1. The
1996; Gregor et al., 1982). This has not been the case for the ef- bicycle dimensions of RF and SU bicycles were the same. They were
fects of upper body orientation on neuromuscular control during having a wheel diameter of 0.60 m, seat height of 0.85 m, handlebar
cycling (Chapman et al., 2008). Balasubramanian and Jayaraman height of 0.95 m and seat-handlebar distance of 0.60 m. SP bicycle
(2009) investigated sEMG based muscle activity analysis on upper however had a wheel diameter of 0.70 m, seat height of 0.95 m,
body muscle-groups for aerobic cyclists. The results suggested that
there is higher muscle fatigue in low back pain group when
compared to their cohorts. Researchers have reported that the low
back pain occurs due to cyclists posture on the bicycle (De Vey
Mestdagh, 1998). Hence, an appropriate bicycle design is very
important to avoid health problems and discomforts associated
with cycling (Baker, 2000).
The effective progress of studies on cycling over several decades
has led to various bicycle designs like sport bicycles, road bicycles
and mountain bicycles (Wilson, 2004). Bicycles are designed to
meet specic needs that are usually associated with cycling ef-
ciency, safety and comfort. Each bicycle designer designs different
frame sizes and dimensions for seat and handlebar, reecting the
qualities of their brands and satisfying the demands of their users.
Some cyclists who suffer from back-pain modify their bicycle
design with extra-high handlebars so that they can sit upright, with
their backs straight. High handlebars decrease the load on the
lower cervical spine and reduce the risk of prolonged neck exten-
sion (Kolehmainen et al., 1989).
Majority of the bicycle design studies focus on race cycling and
the relationship between posture and prevention of injuries caused
during long periods of cycling (Berry et al., 2000). There have been
fewer investigations on the relationship between posture and the
cyclists comfort factor and bicycle design (Laios and Giannatsis,
2010; Christiaans and Bremner, 1998). Hence, there is a need for a
comparison of the physiological responses of cyclists riding
different bicycle categories. We considered three categories of bi-
cycle design, namely:

(1) Rigid frame bicycle (RF): It has a frame with a rigid fork, xed
rear and no suspension.
(2) Suspension bicycle (SU): It uses linkages to attach the rear tri-
angle to the rear shock for a progressive spring rate.
(3) Sports bicycle (SP): It has a (racing) drop handlebar, narrow
tyres and no suspension.

The main objective of the study was to evaluate bicycle design

using sEMG, by examining whether differences in muscle activity
existed in cyclists whilst performing a continuous ride on RF, SU
and SP bicycle designs.

2. Methods and materials

2.1. Participants

Twelve male volunteers participated in this study after they

were briefed extensively on the scope and objectives of this study.
Five participants were more familiar with racing handlebars, and
remaining seven with standard handlebar congurations. Partici-
pants had the following physical characteristics (mean  SD): body
mass 63.5  5.4 kg, body height 1.74  0.04 m,
age 23.2  3.3 yrs. These volunteers were from the Rehabilitation
Bioengineering Group laboratory at IIT Madras, India. They signed
the informed consent that conformed to the ethical guidelines of IIT
Madras.

2.2. Experimental design and protocols

Fig. 1. Three bicycle design concepts. (a) Rigid Frame (RF): It has a frame with a rigid
fork, xed rear and no suspension, (b) Suspension (SU): It uses linkages to attach the
All participants were fully aware of the experiments before rear triangle to the rear shock for a progressive spring rate, (c) Sports (SP): It has a
participating in it. A within-participant design was utilized, with (racing) drop handlebar, narrow tires and no suspension.

Please cite this article in press as: Balasubramanian, V., et al., Muscle fatigue based evaluation of bicycle design, Applied Ergonomics (2013),
http://dx.doi.org/10.1016/j.apergo.2013.04.013
 V. Balasubramanian et al. / Applied Ergonomics xxx (2013) 1e7
handlebar height of 0.82 m and seat-handlebar distance was
0.85 m.
The order in which the participants rode the bicycles was ran-
domized. They were instructed to sit and maintain a comfortable
posture and a constant speed throughout the testing session. Test
track was chosen within the Institute premises, which has a near
at terrain at an altitude of about 10e20 m above mean sea level.
This black topped terrain had non-periodic speed bumps and a low
trafc density. Time bound for testing duration (30 min total)
allowed the participants to cover a distance of 8 km. The study was
carried out for each participant on consecutive days. Each partici-
pant completed maximal voluntary contraction (MVC) test before
and after riding each time. Changes in muscle activity during on-
road cycling were quantied each time a participant rode a bicycle.
2.3. sEMG data acquisition and analysis
The hands, arms, shoulders and back form a muscular sling,
which travels back and forth in supporting the upper body and
pelvis during the pedaling motion (Schmidt, 1994;
Balasubramanian and Jayaraman, 2009). Hence, sEMG activity from
those muscle groups (extensor carpi radialis (ECR), trapezius
medial (TM), latissimus dorsi medial (LDM) and erector spinae (ES))
was bilaterally recorded as shown in Fig. 2 sEMG signals were
recorded using Myomonitor Wireless EMG Systems (Delsys Inc.,
Chicago) with single differential electrode conguration. The
reference electrode was placed on wrist bone (electrically neutral
tissue). The skin was shaved, abraded and cleaned with ethanol
before placing the electrodes to get better contact between the skin
and the electrodes and also to reduce skin impedance.
Prior to and after the experiment, participants were asked to
exert MVC for each muscle group chosen. For measurement of
voluntary activation of muscle groups during maximal contraction
test, participants were asked to lift the weights from the ground as
fast as possible (Fig. 3). All MVCs were collected for 5 s and three
trials were performed. The highest generated contraction from
each muscle group during the three trials was averaged to be the
Fig. 2. Physical locations of the surface electrodes on a typical participant. Muscles
that were examined included extensor carpi radialis (ECR); trapezius medial (TM);
latissimus dorsi medial (LDM); erector spinae (ES).
Please cite this article in press as: Balasubramanian, V., et al., Muscle fatigue based evaluation of bicycle design, Applied Ergonomics (2013),
http://dx.doi.org/10.1016/j.apergo.2013.04.013
3
MVC. Then, they were instructed to start the on-road cycling. sEMG
signals were recorded to determine the activity levels of the muscle
groups during riding.
All raw sEMG signals were digitized with a sampling rate of
1000 Hz and ltered using band pass lter (20e450 Hz) and band
stop lter (47e51 Hz). An ensemble mean of sEMG amplitude was
calculated for each muscle group. To reduce within-participant
variability, the ltered signal was amplitude normalized to the
calculated mean value. The normalized signal was segmented into
10 s each for interval of 1 min and then stored for subsequent
analysis. The sEMG data were extracted and processed in MATLAB
(Mathworks, Inc., USA).
2.3.1. Time domain analysis
sEMG behaviour during fatigue is associated with increase in the
root mean square (RMS) of the time signal (Krogh-Lund and
Jorgensen, 1993; Balasubramanian and Jayaraman, 2009). Hence
RMS of sEMG signal is one of the most reliable parameters in the
time domain analysis (Basmajian and De Luca, 1985). Slope of RMS
with respect to time is an indicator of neuromuscular efciency;
thus weaker muscles are described by positive slopes which indi-
cate that the muscles are able to generate less force to maintain the
level of contraction to accomplish the desired task (Gaudreault
et al., 2005). In this study, an RMS change during MVC was
computed as the difference in RMS value before and after riding. It
was expressed in terms of percentage deviation as depicted in
equation (1):
RMSafter  RMSbefore
% Deviation in RMS 100 (1)
RMSafter
The changes in RMS during on-road cycling were calculated and
analysed using linear regression and the corresponding slopes were
determined.
2.3.2. Frequency domain analysis
Method for evaluating the frequency domain power spectrum of
sEMG signal to muscle fatigue has been well-studied (De Luca,
1984). In this study, power spectrum was determined using a
spectral hamming window with a length of 128 samples for each
segment. Mean power frequency (MPF) was estimated from the
power spectrum. The MPF changes during MVC test were
computed using equation (2). The slope of MPF versus time (fatigue
rate), was also calculated and analysed by linear regression. The
negative slope of the regression line would indicate fatigue rate
(Gaudreault et al., 2005). Hence RMS and MPF of the sEMG signal
were used as the fatigue indicator parameters in this study.
MPFafter  MPFbefore
% Deviation in MPF 100 (2)
MPFafter
2.4. Statistical analysis
SamplePower 2.0 software (SPSS, Inc. USA) was used to
determine the number of participants in this study. The analysis
used the mean and SD values of the extracted features with an
alpha level of 0.05 and power at 84.6% was achieved. The per-
centage deviations extracted from MVC test and the slopes of RMS
and MPF of sEMG signals extracted from real-time monitoring
during on-road cycling were statistically analysed. The extracted
features should be normally distributed set for parametric test.
Normality test on the features-to-be-evaluated yielded a negative
result. Hence, a non-parametric test (Friedman test) was performed
on the features extracted from MVC test and on-road cycling, to
evaluate the level of signicance of muscle activity, riding among
4 V. Balasubramanian et al. / Applied Ergonomics xxx (2013) 1e7

Fig. 3. Maximum voluntary contraction test for each muscle group chosen were recorded to determine the maximum activity levels of the muscle group.

three bicycle designs. Post hoc analysis was performed using the SP in comparison with SU. RF bicycle had larger fatigue in right TM
Tukey multiple comparisons test and the signicance level was set muscle group as compared to SU and SP (Fig. 7). It is interesting to
to p < 0.05 for all analysis. SPSS 15.0 for Windows (SPSS Inc., note that there was no signicant difference between SU and SP
Chicago, IL) was used for the statistical calculations. bicycles for the same muscle group.

2.5. Psychophysical analysis 3.2. On-road testing

Self administered questionnaire study was conducted among all
the participants to quantify the overall comfort levels at neck,
shoulders, arms, hands, back and legs using RBG pain scale. The
study was administered at the end of each experiment so that the
participant could grade the perceived comfort or pain they expe-
rienced immediately after 30 min of cycling. RBG pain scale criteria
and their corresponding grades are depicted in Table 1.
3. Results
3.1. MVC study
Friedman test was performed for on-road test data. The slopes of
both MPF and RMS versus time revealed a signicant difference
(p < 0.05) between SP and SU for right ES. Besides, it was also
evident that the scale of the fatiguing rate (negative slope for MPF
and positive slope for RMS, versus time) was higher for the above
mentioned muscle group in SP when compared to SU (Figs. 8 and
9). Thus, there was a signicant positive rate of change of RMS with
a concomitant negative rate of change of MPF in SP compared to SU
bicycle design. In all other muscle groups considered in this study,
no such appreciable difference was identied.

In the MVC study, statistical tests were able to signicantly

differentiate (p < 0.05) the sEMG activity of right LDM and right ES
muscle groups in SP bicycle when compared to RF and SU bicycles.
There was a decrease in MPF value of right LDM muscle group in SP
as compared to RF and SU (Fig. 4). Both time and frequency domain
analysis indicated that the right ES was signicantly fatigued
(p < 0.05) in SP over RF and SU bicycles (Figs. 5 and 6). This study
illustrates that signicant fatigue of left ECR muscle group in RF and

Table 1
RBG pain scale criteria adapted from Balasubramanian et al. (2005) that was used for
evaluating perceived discomfort and pain.

Grade Pain scale criteria

0 No pain
1 Very minor, barely felt
2 Minor pain not interfering with the work routine
3 Moderate pain
4 Continuous pain affecting routine cycling Fig. 4. Mean and standard deviation error of % deviation in MPF of right latissimus
5 Unbearable/severe pain dorsi medial muscle group calculated from before and after 30 min of bicycling (MVC
test).

Please cite this article in press as: Balasubramanian, V., et al., Muscle fatigue based evaluation of bicycle design, Applied Ergonomics (2013),
http://dx.doi.org/10.1016/j.apergo.2013.04.013
 V. Balasubramanian et al. / Applied Ergonomics xxx (2013) 1e7
Fig. 5. Mean and standard deviation error of % deviation in MPF of right erector spinae
muscle group calculated from MVC test.
3.3. RBG pain score result
Pain scores from the psychophysical study, that is, RBG pain
scale grading, are tabulated in Table 2. Perceived discomfort on
riding SP bicycle in the mid back and low back muscle groups were
signicantly higher (p < 0.05) from RF and SU bicycles. Additionally,
the overall discomfort/pain score was also signicantly higher
(p < 0.05) in SP bicycle.
4. Discussion
The purpose of the study was to determine the muscle activity
of cyclists, riding on three different bicycle design concepts: RF, SU
and SP bicycles using bilateral sEMG. Experiment duration was
conned to 30 min of road cycling, since the average speed could
not be maintained thereafter (Balasubramanian and Jayaraman,
2009). The performance observed on these bicycle congurations
indicated that they were signicantly different from each other;
this is not to justify that each bicycle is not equally manoeuvrable. It
is evident from the results of this study that each bicycle design
Fig. 6. Mean and standard deviation error of % deviation in RMS of right erector spinae
muscle group calculated from MVC test.
Please cite this article in press as: Balasubramanian, V., et al., Muscle fatigue based evaluation of bicycle design, Applied Ergonomics (2013),
http://dx.doi.org/10.1016/j.apergo.2013.04.013
5
Fig. 7. Mean and standard deviation error of % deviation of MPF of right trapezius
medial muscle group calculated from MVC test.
exhibited a difference in muscle activity of the muscle groups
chosen.
EMG-based studies rely on many parameters that are consid-
ered as a basis for evaluating fatigue: an increase in EMG signal
amplitude or a decrease in the number of zero crossings (McLean
et al., 1998), downward shifts in mean/median power frequency
(Arendt-Nielsen and Mills, 1988; Srinivasan and Balasubramanian,
2007). In the present study, muscle activity was estimated in
both the time and frequency domains by measuring the rate of
decrease of MPF and the rate of increase of RMS. This study used
percentage deviation in MPF and RMS to signicantly differentiate
the muscle activity before and after 30 min of cycling; slope of MPF
and RMS to signicantly evaluate the fatigue rate during on-road
cycling on the three different bicycle designs.
Cycling in the dropped bar (SP bicycle) posture results in a
signicantly higher submaximal VO2 than cycling in the standard
bar posture. The reason is that the arms, shoulder and lower back
muscle groups are being engaged to a greater extent to overcome
aerodynamic force. However, the mechanical work spent with SP
bicycle is higher when riding with the rider crouched (Faria et al.,
1978; Capelli et al., 2008). Kirkendall (1990) reported that fatigue
ultimately results in reduced performance and function; thereby
the present study conrmed the above-mentioned issues that in SP
Fig. 8. Mean and standard deviation of slope of MPF from right erector spinae muscle
group calculated from on-road bicycling.
6 V. Balasubramanian et al. / Applied Ergonomics xxx (2013) 1e7
Fig. 9. Mean and standard deviation of slope of RMS from right erector spinae muscle
group calculated from on-road bicycling.
bicycles, participants have reduced performance in right LDM and
right ES muscle groups after 30 min of cycling when compared to
RF and SU bicycles. The SP bicycle appeared to lead to a higher
muscular fatigue among the three designs, corroborating this result
is another study (Purswell et al., 1974) stating that SP (racing) bi-
cycle offers the least manoeuvrability. Additionally, results from
psychophysical analysis of our study provide corroborative evi-
dence to the above ndings (Table 2).
Graves (2000) reported that only minimal advantages could be
expected using suspension systems on bumpy tracks, but the re-
sults were not conclusive. Hence, present study considered the
physiological responses (sEMG activity) of cyclists riding on three
bicycle designs. The main factors involving human response to vi-
bration of bicycles are the vibration coming from the bicycle itself,
the posture of the hands and arms during the gripping of the
handlebars and the surrounding environment. Vibration is trans-
mitted from the bicycle from the handlebars to the hand-arms and
the body; from the saddle to buttocks, the waist, and the body and
from the foot rest, the leg and the body. Karnopp and Margolis
(1984) stated that for human response to vibration, the suspen-
sion should isolate the body from high frequency road proles. At
lower frequencies, the body and wheel should closely follow the
vertical prole from the road to improve balancing. When bicycle
wheels with spring suspension encounters a bump, the spring
compresses, hence the force on the frame and cyclist is much
smaller. Therefore, the potential benets of a SU bicycle on a rough
terrain are: reduced energy cost by the rider and better perfor-
mance and comfort.
In SU bicycle, the left ECR, right LDM and right ES muscle groups
have shown no signicant fatigue and hence, considered comfort-
able to ride where the greater stability on the upper extremity and
the back muscles were more likely to lead to better performance. In
this study, right TM muscle group was signicantly fatigued in RF
Table 2
Tabulation of mean and standard error of discomfort scores for various muscle
groups in accordance to the RBG pain scale for RF, SU and SP bicycle designs. Asterisk
indicates statistically signicant difference (p < 0.05).
Muscle-group Rigid frame (RF) Suspension (SU)
Forearms 0.00  0.00 0.00  0.00
Shoulders 1.42  0.52 0.58  0.34
Mid back* 1.08  0.32 0.25  0.19
Low back* 1.42  0.42 0.08  0.08
Overall pain* 0.98  0.07 0.23  0.05
Please cite this article in press as: Balasubramanian, V., et al., Muscle fatigue based evaluation of bicycle design, Applied Ergonomics (2013),
http://dx.doi.org/10.1016/j.apergo.2013.04.013
when compared to SU bicycle. This result suggested that prolonged
stretching of the hand for gripping the handlebar requires the
participation of upper extremity and shoulder muscle groups.
Since, in RF bicycle, the wheels are attached directly to the frame;
when one of the wheels hits a bump, both that wheel and the frame
must accelerate upward together. This exerts an enormous upward
force on cyclist who experiences a sudden upward acceleration. To
reduce the upward acceleration that the cyclist experiences, the
direct connection between the bicycle wheels and the frame can be
replaced by a spring suspension, i.e., SU bicycle design. As a result,
the cycle frame and the cyclist do not bounce signicantly when the
wheel rides on irregular road surfaces.
The results of this study suggest that in SP bicycle, there is a
signicant fatigue in right LDM and right ES muscle groups, which
ultimately shows that the cyclist experiences an asymmetry
loading during prolonged cycling, caused by the dominance of the
natural side, at work. All the participants in the study being right
handed, the results eventually expressed a signicantly higher fa-
tigue in the right side when compared to the left.
5. Limitations of current study
In this study we have not performed a detailed postural and
energy expenditure analysis of a rider while riding on different
bicycle designs. Further, owing to cultural and other considerations,
we had only male participants in this study. Future studies could
take into consideration female participants and perform a detailed
postural and energy expenditure analysis. In addition, the time at
which fatigue is initiated can also be analysed. Future research on
the effect of bicycle designs on bicycling performance over time can
also be made. Familiarity of the participants with a particular type
of bicycle should also be taken into account in the future studies.
6. Concluding remarks
The ndings of this study indicate that there is a substantially
higher muscular fatigue in upper extremity and low back muscle
groups, which could reect in poor balancing of bicycle while
manoeuvring SP bicycle compared to RF and the SU bicycles. The
results also suggest that the SU bicycle shows evidence of how well
the suspension prevents muscular fatigue and vibration-induced
low-back pain. These signicant differences among the three bi-
cycle designs have been demonstrated by means of MVC test before
and after cycling and real-time monitoring of muscle activity dur-
ing on-road cycling. This study helps to understand the variation of
muscle activity caused by the mechanical factors of bicycle design.
These inferences could be considered for ergonomics design of
bicycle.
Acknowledgements
The authors would like to acknowledge all the members of
Rehabilitation Bioengineering Group (RBG) at IIT Madras. We also
wish to thank the participants for their valuable contributions and
support.
References
Arendt-Nielsen, L., Mills, K.R., 1988. Muscle bre conduction velocity, mean power
Sports (SP) frequency, mean EMG voltage and force during sub maximal fatiguing con-
tractions of human quadriceps. European Journal of Applied Physiology and
1.08  0.38 Occupational Physiology 58, 20e25.
1.83  0.57 Baker, A., 2000. Medical problems in road cycling. In: Gregor, R.J., Conconi, F. (Eds.),
2.92  0.38 Road Cycling. Blackwell Science Ltd., London, pp. 68e118.
2.58  0.38 Balasubramanian, V., Jayaraman, S., 2009. Surface EMG based muscle activity
2.10  0.03 analysis for aerobic cyclist. Journal of Bodywork and Movement Therapies 13
(1), 34e42.
 V. Balasubramanian et al. / Applied Ergonomics xxx (2013) 1e7
Balasubramanian, V., Swami Prasad, G., Varadhan, S.K.M., 2005. Quantitative eval-
uation of low back pain in bar benders. NICMAR Journal of Construction
Management 20 (4), 31e36.
Basmajian, J.V., De Luca, C.J., 1985. Muscles Alive: Their Functions Revealed by
Electromyography, fth ed. William and Wilkins, Baltimore.
Bernmark, E., Wiktorin, C., Svartengren, M., Lewne, M., berg, S., 2006. Bicycle
messengers: energy expenditure and exposure to air pollution. Ergonomics 49
(14), 1486e1495.
Berry, M.J., Koves, T.R., Benedetto, J.J., 2000. The inuence of speed, grade and mass
during simulated off road bicycling. Applied Ergonomics 31 (5), 531e536.
Brown, D.A., Kautz, S.A., Dairaghi, C.A., 1996. Muscle activity patterns altered during
pedaling at different body orientations. Journal of Biomechanics 29, 1349e1356.
Burke, E., 1996. High Tech Cycling. Human kinetics publishers, Champaign, Illinois.
Burnett, A.F., Cornelius, M.W., Dankaerts, W., OSullivan, P.B., 2004. Spinal kine-
matics and trunk muscle activity in cyclists: a comparison between healthy
controls and non-specic chronic low back pain subjects e a pilot investigation.
Manual Therapy 9, 211e219.
Callaghan, M.J., 2005. Lower body problems and injury in cycling. Journal of
Bodywork & Movement Therapies 9 (3), 226e236.
Capelli, Carlo, Paolo Ardig, Luca, Schena, Federico, Zamparo, Paola, 2008. Energy
cost and mechanical efciency of riding a human-powered recumbent bicycle.
Ergonomics 51 (10), 1565e1575.
Chapman, A.R., Vicenzino, B., Blanch, P., Knox, J.J., Dowlanc, S., Hodges, P.W., 2008.
The inuence of body position on leg kinematics and muscle recruitment
during cycling. Journal of Science and Medicine in Sport 11, 519e526.
Christiaans, H.H.C.M., Bremner, A., 1998. Comfort on bicycles and the validity of a
commercial bicycle tting system. Applied Ergonomics 29 (3), 201e211.
De Luca, C.J., 1984. Myoelectrical manifestations of localized muscular fatigue in
humans. Critical Review in Biomedical Engineering 11, 251e279.
De Luca, C.J., 1993. Use of surface EMG signal for performance evaluation of back
muscles. Muscle Nerve 16, 210e216.
De Vey Mestdagh, 1998. In search of an optimum cycling posture. Applied Ergo-
nomics 29, 325e334.
Faria, I., Dix, C., Frazer, C., 1978. Effect of body position during cycling on heart rate,
pulmonary ventilation, oxygen uptake and work output. Journal of Sports
Medicine 18 (1), 49e56.
Ferguson, S.A., Marras, W.S., Allread, W.G., Knapik, G.G., Splittstoesser, R.E., 2012.
Musculoskeletal disorder risk during automotive assembly: current vs. seated.
Applied Ergonomics 43 (4), 671e678.
Gaudreault, N., Arsenault, A.B., Lariviere, C., DeSerres, S.J., Rivard, C.H., 2005.
Assessment of the paraspinal muscles of subjects presenting an idiopathic
scoliosis: an EMG pilot study. BMC Musculoskeletal Disorders 6, 14.
Gibson, A.S., Schabort, E.J., Noakes, T.D., 2001. Reduced neuromuscular activity and
force generation during prolonged cycling. American Journal of Physiology e
Regulatory Integrative Comparative Physiology 281, 187e196.
Graves, P., 2000. Olympic preview. Mountain Biker International 61, 62e64.
Gregor, R.J., Green, D., Garhammer, J.J., 1982. An electromyographic analysis of
selected muscle activity in elite competitive cyclists. In: Biomechanics VII.
University Park, Baltimore, MD, pp. 537e541.
Please cite this article in press as: Balasubramanian, V., et al., Muscle fatigue based evaluation of bicycle design, Applied Ergonomics (2013),
http://dx.doi.org/10.1016/j.apergo.2013.04.013
7
Hausswirth, C., Argentin, S., Bieuzen, F., Le Meur, Y., Couturier, A., Brisswalter, J.,
2010. Endurance and strength training effects on physiological and muscular
parameters during prolonged cycling. Journal of Electromyography and Kine-
siology 20, 330e339.
Houtz, S.J., Fisher, F.J., 1959. An analysis of muscle action and joint excursion on
stationary bicycle. Journal of Bone Joint Surgery 41A, 123e131.
Karnopp, D., Margolis, D., 1984. Adaptive suspension concepts for road vehicles.
Vehicle System Dynamic 13, 145e160.
Kirkendall, D.T., 1990. Mechanisms of peripheral fatigue. Medicine & Science in
Sports & Exercise 22, 444e449.
Kolehmainen, I., Harms-Ringdahl, K., Lanshammart, H., 1989. Cervical spine posi-
tions and load moments during bicycling with different handlebar positions.
Clinical Biomechanics 4 (2), 105e110.
Krogh-Lund, C., Jorgensen, K., 1992. Modication of myo-electric power spectrum in
fatigue from 15% maximal voluntary contraction of human elbow exor mus-
cles, to limit of endurance: reection of conduction velocity variation and: or
centrally mediated mechanisms? European Journal of Applied Physiology 64,
359e370.
Krogh-Lund, C., Jorgensen, K., 1993. Myo-electric fatigue manifestations revisited:
power spectrum, conduction velocity, and amplitude of human elbow exor
muscles during isolated and repetitive endurance contractions at 30% maximal
voluntary contraction. European Journal of Applied Physiology and Occupa-
tional Physiology 66, 161e173.
Laios, L., Giannatsis, J., 2010. Ergonomics evaluation and redesign of children bi-
cycles based on anthropometric data. Applied Ergonomics 41 (3), 428e435.
Li, L., Caldwell, G.E., 1998. Muscle coordination in cycling: effect of surface incline
and posture. Journal of Applied Physiology 85, 927e934.
McLean, L., Tingley, M., Scott, R.N., Rickards, J., 1998. Mean frequency cycling during
sustained postural contractions of the cervical and lumbar extensor muscles. In:
First International Society of Electrophysiology and Kinesiology (ISEK)
Congress, p. 210.
Mellion, M.B., 1994. Neck and back pain in bicycling. Clinics in Sports Medicine 13
(1), 137e164.
Purswell, J.L., Terrell, R., Greene, J., 1974. A study of bicycle design as it affects safety
of operation. In: Proceedings, 18th Annual Meeting of the Human Factors So-
ciety, Huntsville, Alabama, October.
Schmidt, A., 1994. Handbook of Competitive Cycling: Training, Keep Fit, Tactics.
Meyer & Meyer, Oxford, pp. 13e16.
So, R.C.H., Ng, J.K.F., Ng, G.Y.F., 2005. Muscle recruitment pattern in cycling: a re-
view. Physical Therapy in Sport 6, 89e96.
Srinivasan, J., Balasubramanian, V., 2007. Low back pain and muscle fatigue due to
road cycling-a sEMG Study. Journal of Body Work & and Movement Therapy 11
(3), 260e266.
Usabiaga, J., Crespo, R., Iza, I., Aramendi, J., Terrados, N., Poza, J.J., 1997. Adaptation of
the lumbar spine to different positions in bicycle racing. Spine 22, 1965e1969.
Van Dieen, J., Toussaint, H., Thissen, C., Van De Van, A., 1993. Spectral analysis of
erector spinae EMG during intermittent isometric fatiguing exercise. Ergo-
nomics 36, 407e414.
Wilson, D.G., 2004. Bicycling Science, third ed. MIT Press, Cambridge.