lable at ScienceDirect

International Journal of Industrial Ergonomics 41 (2011) 625e633

Contents lists avai

International Journal of Industrial Ergonomics

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

Objective and subjective responses of seated subjects while reading Hindinewspaper under multi axis whole-body vibration

M.K. Bhiwapurkar*, V.H. Saran, S.P. HarshaMechanical and Industrial Engineering Department, Indian Institute of Technology, Roorkee, India

a r t i c l e i n f o

Article history:Received 27 June 2010Received in revised form18 January 2011Accepted 17 June 2011Available online 5 August 2011

Keywords:Multi axis vibrationReadingHindi newspaperRandom vibrationWhole-body vibration

* Corresponding author. Tel.: þ91 1332 285682; faxE-mail addresses: mbhiwapurkar@gmail.com (M.

iitr.ernet.in (V.H. Saran), surajfme@iitr.ernet.in (S.P. H

0169-8141/$ e see front matter � 2011 Elsevier B.V.doi:10.1016/j.ergon.2011.06.004

a b s t r a c t

Train passengers often read newspapers while traveling. Vibration is one of the key factors that mayoccasionally inhibit this activity. An experimental study was, therefore, conducted to investigate theextent of interference perceived in reading task by seated subjects in two postures under randomvibration. 30 healthy male subjects were exposed to vibration magnitudes of 0.4, 0.8 and 1.2 m/s2 inmono, dual and multi axis in the low frequency range 1e20 Hz. The task required subjects to read a givenparagraph of Hindi national newspaper, in two seated postures (lap posture with backrest support andtable posture with leaning over the table). The reading performance was evaluated by both degradationin performance in terms of time required to complete the task and subjective rating using Borg CR10scale. Both the methods of reading performance evaluation exhibit progressive increase with an increasein vibration magnitude for both the subject postures in all the direction of vibration and are found to behigher in lateral and vertical direction among mono axes. The effects of multi axis vibration on perceiveddifficulty have been found to be similar to dual axes vibration and greater than mono axes vibration;however degradation in reading performance in multi axis vibration was also found to be similar to thatfor lateral direction. A comparison of the effect of postures by both evaluation methods revealed that thereading performance was adversely affected for table posture in all direction of vibration, however for lapposture, only the X-axis vibration effect was more severe.Relevance to industry: Available ride comfort standards for vehicles do not include the effects of vibrationson passenger activities. Assessment of activity discomfort would be useful for vehicle design optimiza-tion to facilitate activity comfort.

� 2011 Elsevier B.V. All rights reserved.

1. Introduction

In a moving vehicle, any handheld material such as a newspaperis subject to vibration. The human visual system may also beaffected by vibrations. Such a situation may arise when a personreads a newspaper in a moving train. Reading small characters onthe newspaper in dynamic conditions is not as effortless as in staticconditions.

The reading of handheld material, such as a newspaper, ina vibrating environment may require quite an effort owing toa combination of head and hand motions. The displacement of theeye is often greatest in the range 2e5 Hz and so it is around thesefrequencies that reading difficulty is likely to be greatest (Griffin,2003). When reading a newspaper on a train, the motion of the

: þ91 1332 285665.K. Bhiwapurkar), saranfme@arsha).

All rights reserved.

arms may result in the motion of the paper being different inmagnitude and phase from the motions of both the seat and thehead of the observer. The dominant axis of motion of the news-paper may be different from the dominant axis of motion of theperson (Griffin and Hayward, 1994).

A literature review shows that a considerable amount of vibra-tion studies were found in military applications (Wollstrom, 2000).However, only a few studies have shown that vibrations aredisturbing a significant number of passengers on performingactivities like reading, writing, sketching and drinking (Corbridgeand Griffin, 1991; Griffin and Hayward, 1994; Sundström andKhan, 2008; Bhiwapurkar et al., 2010a). Whole-body vibrationhas been shown to affect both reading speed and reading accuracyin many studies (Lewis and Griffin, 1978; Moseley and Griffin,1986;Griffin, 2003). Lewis and Griffin (1980) found a degradation ofreading performance during exposure to fore-and-aft (x-axis)vibration at frequencies between 5.6 and 11 Hz, while reading ofleading articles from a national newspaper. However, the effect was

Fig. 1. Subject postures used in the study.

M.K. Bhiwapurkar et al. / International Journal of Industrial Ergonomics 41 (2011) 625e633626

only present when a seat with a backrest was used. In the samestudy, they reported that lateral (y- axis) vibration produced onlya slight degradation in reading performance at about 5.6 Hz.

There are a number of studies (Corbridge and Griffin, 1986;Griffin and Hayward, 1994) on the effects of horizontal (X-, Y-axis)vibrations but none of these studies have been performed in trainsor uses vibration conditions similar to trains. There are unfortu-nately, hardly any studies available on the influence of the vibrationdirection and seated posture on the ability to read. In a recent fieldstudy (Bhiwapurkar et al., 2010b) on various Indian railwaypassenger trains, which included both questionnaire survey andvibration measurements, it was revealed that the maximum diffi-culty was observed in writing activity (72%), subsequently inworking with laptop computer (58%) and reading (56%). It wasobserved from the study that the majority of the passengerspreferred to read the newspaper in the national language Hindi.Further, it reported that the vibration levels measured from floor ofpassenger compartment were in the range of 0.2e0.67 m/s2 rms inlongitudinal (X-axis) direction; 0.23e0.83 m/s2 rms in lateral (Y-axis) direction and 0.38e1.2 m/s2 rms in vertical (Z-axis) direction.Comparing quantitatively with X-axis vibration, the vibration levelswere found to be about 30% higher in the Y- axis and approximately80% higher in Z-axis. Therefore, these vibratory conditions werechosen for the present study.

The objective of the study is to investigate the extent of interfer-ence perceived in reading newspaper by seated subjects in twoposturesundermono, dual andmulti axisGaussian randomvibrationenvironment. It was hypothesized that vibration magnitudes,vibration directions and subject postures would affect reading diffi-culty and this would also be reflected in the reading performance.

2. Methodology

2.1. Subjects

A total of 30 healthy male subjects with age in years (22.9� 4.6),weight in kg (68.9 � 12.0) and height in cms (173.8 � 5.8), allengineering undergraduate/postgraduate students of the institute,participated in the experiment. The subjects participated volun-tarily under informed written consent and were given a smallremuneration. Ethical approval was obtained from IIT RoorkeeHuman Ethical Committee. All the subjects had normal eyesight(normal visual acuity 6/6 vision). A screening questionnaire wascollected from subjects on their personal background: level ofeducation; frequency of traveling in trains; fitness, ‘Hindi’ languagefluency, reading habits and musculoskeletal disorders (Kuorinkaet al., 1987) to assure the suitability of the subjects for experi-mental task.

2.2. Subject postures

In the laboratory study subjects performed reading task in twopostures, representative of passenger like posture as defined below,and shown in Fig. 1.

(i) Lap posture: the subject is seated with backrest support withthe newspaper placed on lap and

(ii) Table posture: the seated subject leans forward with thenewspaper placed on the table

2.3. Vibration environment

The study was conducted on the vibration simulator developedas a mockup of a railway vehicle, in Vehicle Dynamics Laboratory,

IIT Roorkee, India. It consists of a platform onwhich a table and tworigid chairs have been securely fixed, Fig. 2. The backrest of thechair was rigid, flat, and vertical. Neither the seat, nor the backrest,nor the table had any resonances within the frequency rangestudied (up to 20 Hz) in any of the three axes. Three electro-dynamic vibration exciters were used to provide vibration stimulito the platform in three axes; longitudinal (X-axis), lateral (Y-axis)and vertical (Z-axis). For safety reasons the internal positioningaccelerometers of the shakers were continuously used for motionfeedback. The onboard vibrations of the platform were measuredon line for continuous monitoring of the vibration signal by usinga tri-axial accelerometer (KISTLER 8393B10), the signal transmittedto the Labview Signal Express software via a data acquisition card(NI 6218).

The simulator provided a controlled train atmosphere witha working illumination well above 250 lux using both direct andindirect light sources for constant and well-distributed illumina-tion at both seats and tables. The test subjects were seated on thechairs rigidly mounted on the platform of vibration simulator andexcitedwith vibration stimuli of frequencies up to 20 Hz. This rangeis considered critical since the structural dynamics of a passengerrailcar usually gives rise to several resonance peaks in thefrequency range of 0.5e20 Hz (Andersson et al., 2005) and itcoincides with the most vulnerable range for reading activity also(Sundström and Khan, 2008).

2.4. Vibration stimuli

In the study, for each subject postures, a continuous Gaussianrandom signal over the frequency range 1e20 Hz was generatedusing random vibration controller. Power spectral density curve(g2/Hz) of the signal generated by the exciter over the frequencyspectrum of interest is shown in Fig. 3. The three levels of RSSacceleration amplitudes viz. 0.4, 0.8 and 1.2 m/s2 were excited inmono, dual andmulti axis. Inmono axis, the three RMS accelerationamplitudes viz. 0.4, 0.8 and 1.2 m/s2 were given independently. Asdiscussed above, the vibration levels in Indian railway passengertrains were found to be about 30% higher in the Y-axis andapproximately 80% higher in Z-axis in comparison with X-axisvibration level. Therefore, in dual axis, any two mono axes wereexcited simultaneously fulfilling the above relation, so as toproduce equivalent RSS magnitudes of 0.4, 0.8 and 1.2 m/s2. Thisprocedure was similarly followed for multi axis to yield the threegiven RSS magnitudes, Table 1. The RSS is the vibration total valuewhich is obtained from the square root of the sum of the squares ofthemeasured RMS values in the X-, Y-, and Z-axis (Mansfield, 2005).

2.5. Reading task and performance

The subjects were asked to read a given text paragraph ofapproximately 150 words from a Hindi newspaper at a comfortablespeed with a 1 min break between consecutive sessions. For thestudy, various articles of general interest from ‘Amar Ujala’,a leading Hindi national newspaper were selected. A separatearticle was assigned for each vibration condition and posture toprevent learning effect. The reading performance was evaluated by

Fig. 2. Schematic presentation of the equipment used for vibration measurements.

M.K. Bhiwapurkar et al. / International Journal of Industrial Ergonomics 41 (2011) 625e633 627

measuring the time required to complete the reading paragraphwith a digital stop watch.

2.6. Test procedure

Each subject began the experiment by filling out a generalquestionnaire about his personal information. Following a briefintroduction about the experiment, the subject was asked toperform the experimental task. Each subject was required toparticipate in the experiment only about an hour per day to avoidthe influence of fatigue and was given three consecutive days tofinish the task.

Since the test was conducted on two subjects at a time, the taskwas required to be performed silently. Each subject was exposed toa total of 42 conditions from a combination of three levels ofvibration magnitudes, two subject postures and seven levels ofvibration directions (mono, dual and multi axis) with a 1 min breakbetween consecutive sessions. A static conditionwith no vibrationswas also used in both postures. The vibration conditions werepresented in random with Latin Square Design to minimize ordereffects. The test subjects were instructed to occupy themselves

with the prescribed task during the vibration exposure and ratetheir perceived difficulty of reading using Borg CR 10 scale.

3. Data analysis

A factorial analysis of variance (ANOVA) was performed toevaluate subject’s response for which results at the p < 0.05 levelare referred as significant. The statistical package for social sciences(SPSS Inc., Chicago, USA, version 16) was used for all statisticalanalysis.

ANOVA was performed with the repeated measure within-subjects design for the independent variables: vibration magni-tudes, subject postures and directions of vibration. The repeated-measures design is well suited since the judgments from eachparticipant were collected repetitively for all the test conditions. Allthe collected responses were manually coded and analyzed withthe statistical software SPSS. Two other statistical measures wereconsidered for interpreting the ANOVA, i.e. the estimate of effectsize (partial eta squared) and the observed power. Partial etasquared was chosen since it is not dependent on the number offactors for explaining the main effects or interactions. The

Fig. 4. Effect of mono and multi axis vibration and vibration magnitudes on level ofdifficulty.

Fig. 3. Power spectral density (g2/Hz) curve.

M.K. Bhiwapurkar et al. / International Journal of Industrial Ergonomics 41 (2011) 625e633628

advantage of this measure lies in its ability to avoid masking effectsfrom the most powerful variables in the analysis. Observed powerwas calculated to increase the certainty of correct detection of anobserved effect. In the range from 0 to 1, an observed power of 0.95would mean a 5% chance of detecting an effect that is not true. Apost hoc test was needed after completing an ANOVA in order todetermine which groups differ from each other.

4. Results

The reading difficulty evaluated by subjective evaluation wasrepresented as mean values of the level of difficulty as shown fromFigs. 4e9. Similarly, the reading performance was evaluated by

Table 1Summary of various vibration stimuli used in the experiment.

Stimulus Vibration magnitude (m/s2, unweighted)

X-axis RMS Y-axis RMS Z-axis RMS RSS ¼ Paxes

1 0.4 e e 0.42 0.8 e e 0.83 1.2 e e 1.24 e 0.4 e 0.45 e 0.8 e 0.86 e 1.2 e 1.27 e e 0.4 0.48 e e 0.8 0.89 e e 1.2 1.210 0.25 0.32 e 0.411 0.5 0.63 e 0.812 0.75 0.94 e 1.213 0.2 e 0.35 0.414 0.4 e 0.7 0.815 0.6 e 1.0 1.216 e 0.24 0.33 0.417 e 0.45 0.65 0.818 e 0.7 1.0 1.219 0.17 0.22 0.3 0.420 0.33 0.43 0.6 0.821 0.5 0.63 0.9 1.2Static e e e e

RMS ¼ root mean square; RSS ¼ root sum of squares.

objective measure of time to complete the reading task for eachvibration condition and the increase in timewas represented as themean percentage decrement in reading performance with respectto static condition as shown from Figs. 10e15.

4.1. Subjective evaluation of reading difficulty

The mean values of level of reading difficulty by subjectiveevaluation in mono, dual and multi axis vibration for table postureare shown in Figs. 4 and 5. Also the effect due to subject posturescan be distinguished from Figs. 6e9.

4.1.1. Influence of vibration magnitudes on subjective evaluationFigs. 4e9 shows the effect of vibration magnitudes on reading

difficulty in all the mono, dual and multi axis for both the subjectpostures. It was observed that the level of difficulty progressively

Fig. 5. Effect of dual axes vibration and vibration magnitudes on level of difficulty.

0

1

2

3

4

5

0 0.4 0.8 1.2

Vibration Magnitude (m/s2

)

Le

ve

l o

f D

iffic

ulty

X_T X_L Y_T Y_L

Fig. 6. Effect of subject postures on level of difficulty in X- and Y-axis.

0

1

2

3

4

5

0 0.4 0.8 1.2

Vibration Magnitude (m/s2

)

Le

ve

l o

f D

iffic

ulty

XZ_T XZ_L YZ_T YZ_L

Fig. 8. Effect of subject postures on level of difficulty in XZ-axis and YZ-axis.

M.K. Bhiwapurkar et al. / International Journal of Industrial Ergonomics 41 (2011) 625e633 629

increases with an increase in vibration magnitude, also the signif-icant difference in level of difficulty was observed between givenrange of vibration magnitudes for both the subject postures and inall the direction of vibration (p < 0.05).

4.1.2. Influence of vibration directions on subjective evaluationIt is necessary to keep the posture constant to show the effect of

vibration directions on perceived difficulty; therefore table posturewas selected based on maximum severity on perceived difficulty.

Among all mono axes, the highest level of difficulty wasobserved for Z-axis vibration and least difficulty with X-axisvibration (Fig. 4). While significant difference in level of difficultywas found between vibration in X- and Y-axis; X- and Z-axis(p< 0.05), however, it remained insignificant between Y- and Z-axisvibration (p > 0.05). Statistically, it implies that effect of Z-axisvibration is comparable to Y-axis vibration on perceived difficulty ofreading.

0

1

2

3

4

5

0 0.4 0.8 1.2

Vibration Magnitude (m/s2

)

Le

ve

l o

f D

iffic

ulty

Z_T Z_L XY_T XY_L

Fig. 7. Effect of subject postures on level of difficulty in Z- and XY-axis.

Similarly, while reading newspaper among all given combina-tions of dual axes vibration (Fig. 5), no significant difference in levelof difficulty was observed (p> 0.05). All given combinations of dualaxes vibration showed greater effect of perceived difficulty ascompared to all mono axes vibration (p < 0.05). It can also seenfrom Figs. 4 and 5 that the higher level of reading difficulty wasobserved in multi axis vibration as compared to mono axes vibra-tion (p < 0.05), but significant difference in level of difficulty wasnot found between multi axis and dual axis vibrations (p > 0.05).

4.1.3. Influence of subject postures on subjective evaluationWhile comparing the effect of subject postures on level of

difficulty in each mono axes, the X-axis vibration yielded higherlevel of difficulty on lap thanwith table posture, Fig. 6 (p < 0.05). Incontrast (Figs. 6,7 and 9), reading on table produced greater diffi-culty as compared to that on lap in Y-, Z- and multi axis vibration(p < 0.05). In dual XY- and XZ-axis vibration (Figs. 7 and 8),

0

1

2

3

4

5

0 0.4 0.8 1.2

Vibration Magnitude (m/s2

)

Le

ve

l o

f D

iffic

ulty

XYZ_T XYZ_L

Fig. 9. Effect of subject postures on level of difficulty in XYZ- axis.

0

5

10

15

20

25

30

35

40

45

0 0.4 0.8 1.2

Vibration Magnitude (m/s2

)

% P

erfo

rm

an

ce

D

ec

re

me

nt

X Y Z XYZ

Fig. 10. Effect of mono and multi axis vibration and vibration magnitudes on decre-ment in reading performance.

Fig. 12. Effect of subject postures on decrement in reading performance in X- andY-axis.

M.K. Bhiwapurkar et al. / International Journal of Industrial Ergonomics 41 (2011) 625e633630

insignificant difference in level of difficulty due to subject postureswas observed (p > 0.05), however vibration in YZ-axis (Fig. 8)produced greater difficulty while reading on table as compared tothat on lap (p < 0.05).

4.2. Reading performance as percentage decrement

The mean values of percentage decrement in reading perfor-mance in mono, dual and multi axis vibration for table posture areshown in Figs. 10 and 11. Also the effect attributable to subjectpostures can be distinguished from Figs. 12e15.

4.2.1. Influence of vibration magnitudes on decrement in readingperformance

Figs. 10e15 show the effect of vibration magnitudes onpercentage decrement in reading performance for all mono, dual

Fig. 11. Effect of dual axis vibration and vibration magnitudes on decrement in readingperformance.

and multi axis vibration. It was observed that the percentagedecrement in reading performance increases with an increase invibration magnitude, also statistically significant difference inreading performance was observed between given range of vibra-tionmagnitude for both the subject postures and in all the directionof vibration (p < 0.05).

4.2.2. Influence of vibration directions on decrement in readingperformance

It is necessary to keep the posture constant to show the effect ofvibration directions on percentage decrement in reading perfor-mance; therefore table posture was selected based on maximumseverity on percentage decrement in reading performance.

Among all mono axes vibrations, a higher percentage decrementin reading performance was observed with Y-axis and least withX-axis vibration (Fig. 10). While significant difference in readingperformance was observed between vibration in X- and Y-axis;

Fig. 13. Effect of subject postures on decrement in reading performance in Z- andXZ-axis.

Fig. 14. Effect of subject postures on decrement in reading performance in YZ- and XY-axis.

M.K. Bhiwapurkar et al. / International Journal of Industrial Ergonomics 41 (2011) 625e633 631

X-and Z-axis (p < 0.05), however, no significant difference wasobserved between Y- and Z-axis vibration (p > 0.05). Statistically,this implied that effect of Z-axis vibration is similar to Y-axisvibration on reading performance. Similarly, among all dual axesvibration, the percentage decrement in reading performance wasfound to be higher with YZ-axis vibration and least with XZ- andXY-axis (Fig. 11), but no significant difference in reading perfor-mance was observed between given combination of them(p > 0.05).

The effect of vibration in dual axes has been compared with itsassociated mono axes in order to study their combined effect ondecrement in reading performance, Figs. 10 and 11. As compared toXY-axis vibration, the percentage decrement in reading perfor-mance was found to be lower in X-axis vibration (p < 0.05), butinsignificant with Y-axis vibration (p > 0.05). Similarly, thepercentage decrement in reading performance was found to behigher in YZ-axis vibration as compared to associated Z-axisvibration (p < 0.05), but insignificant with Y-axis vibration

Fig. 15. Effect of subject postures on decrement in reading performance in XYZ- axis.

(p > 0.05). Moreover, the percentage decrement in readingperformance was found to be higher in XZ-axis vibration ascompared to both associated X- and Z-axis vibration (p < 0.05).

It can be seen from Figs. 10 and 11 that the percentage decre-ment in reading performance for multi axis (XYZ) vibration wasfound to be similar with Y-axis vibration (p> 0.05) and higher thanX- and Z-axis vibration (p < 0.05). Also similar effect of multi axisvibration and dual axis vibration on percentage decrement inreading performance was observed (p > 0.05).

4.2.3. Influence of subject postures on decrement in readingperformance

While comparing the postural differences (Fig. 12), the vibrationin X-axis indicates the higher percentage decrement in readingperformance for reading the newspaper on lap than with table(p < 0.05). In contrast, the percentage decrement in readingperformance was found to be higher while reading the newspaperon table than with lap in all other direction of vibration (p < 0.05).

4.3. Results from data analysis

In the within-subject test, general effects on the reading diffi-culty and percentage decrement in reading performance werefound for all the independent variables, i.e. vibration magnitudes,directions of vibration and subject postures (Tables 2 and 3). TheTable 2 shows all the independent and interacted variables up tothree levels with its significant value (p < 0.05), indicating that allthe main parameters are significantly responsible for the judgmentof perceived difficulty. In general, the observed power attained highvalues for all the independent and interacted variables. The resultshows that the vibration magnitude is the variable that contributesthe most to the perceived difficulty and that posture comes secondfollowed by direction of vibration. It was also observed from Table 2that the overall effects of all independent variables are higher thaninteracted variables.

From the within-subject test for percentage decrement inreading performance, similar outcome was observed for all theindependent and interacted variables. The interacted variables(D � P) and (V � P) also show comparable contribution withdirection of vibration, Table 3.

5. Discussions

Various international and national standards provide guidanceon the measurement, evaluation and assessment of whole-bodyvibration in respect to perceived discomfort but the standardsdiffer in evaluation and assessment of vibration (Henrietta, 2004).These standards are usually used as tools by the train operators andmanufacturers to ensure low vibration levels for ride comfort.Limits and procedures for the evaluation of discomfort caused bywhole-body vibration are given in standards such as ISO 2631-4,ENV 12299 (ISO 2631-4, 1998 and ENV 12299, 1999). However,

Table 2Within-subjects effect of test parameters for reading difficulty.

Source Type IIIsum ofsquares

df Meansquare

F Sig.(p) Partial etasquared

Observedpower

Direction (D) 131.9 6 21.98 24.84 <0.001 0.76 1.0Vibration (V) 695.42 3 231.81 538.86 <0.001 0.98 1.0Posture (P) 13.83 1 13.83 71.21 <0.001 0.89 1.0D � V 61.61 18 3.42 10.17 <0.001 0.56 1.0D � P 6.79 6 1.13 5.36 <0.001 0.40 0.99V � P 2.88 3 0.96 7.11 ¼0.001 0.47 0.96D � V � P 8.72 18 0.48 2.96 <0.001 0.27 0.99

Table 3Within-subjects effect of test parameters for decrement in reading performance.

Source Type IIIsum ofsquares

df Meansquare

F Sig.(p) Partial etasquared

Observedpower

Direction (D) 2643.28 6 440.55 13.61 <0.001 0.69 1.0Vibration (V) 59696.09 3 19898.69 627.66 <0.001 0.99 1.0Posture (P) 1452.84 1 1452.84 41.62 ¼0.001 0.87 0.99D � V 1342.30 18 74.57 3.46 <0.001 0.37 1.0D � P 1260.13 6 210.02 12.14 <0.001 0.67 1.0V � P 495.92 3 165.31 12.44 <0.001 0.67 0.99D � V � P 596.35 18 33.13 2.62 ¼0.001 0.30 0.99

M.K. Bhiwapurkar et al. / International Journal of Industrial Ergonomics 41 (2011) 625e633632

these standards have very little use in determining the extent ofdifficulty in performing sedentary activities like reading or writingdue to vibrations.

5.1. Effect of vibration magnitudes

On the whole, among all independent and interacted variables,it was observed from data analysis of both the subjective evaluationand reading performance that the highest contribution comes fromthe independent variable i.e. vibration magnitudes, for which theobserved power attains high values. For all mono, dual and multiaxis vibration, the perceived difficulty of reading was found toincrease with an increase in vibration magnitude for both thesubject postures by subjective evaluation, which matches with theresults of Mansfield and Maeda (2005), where subjective ratings ofintensity increased with vibration magnitude for both single axisand dual axis vibration conditions. Similar outcomes was alsoreflected in reading performance where percentage decrement inreading performance increases progressively with an increase invibration magnitude for both the subject postures and in all thedirection of vibration.

5.2. Effect of subject postures

The second higher contribution comes from both the perceiveddifficulty and reading performance was subject postures. Ina vibrating environment, the posture becomes even more impor-tant in suppressing and compensating the motions to limit theireffect on the performance of the work. The posture has, thus, a vitalrole in transmitting vibrations to the different body segments, aswell as to the reading material.

In X-axis vibration, both the perceived difficulty and decrementin reading performance was found to be higher while reading onlap than with table posture. While reading the newspaper on lap,the upper body is supported by the back of the seat, and the legs aresupported by the floor, which will affect the vibration transmittedto the head over a wide range of frequencies (Paddan and Griffin,1988) and perhaps vision. The comments of several subjects wereconsistent with backrest motion being an important factor incausing decrement of reading performance during exposure toX-axis vibration.

In Y-axis vibration, with the newspaper on table, the upper bodywill lack support from the seat backrest and the newspaper willattain an oscillation that is almost equal to that of the table. Theunsupported upper body may have a greater possibility to moveout of phase with the newspaper. For the two postures, the differ-ence in perceived difficulty and decrement of reading performancewas significant in Y-axis vibration. The result revealed that both theperceived difficulty and reading performance was affected for tableposture as compared to lap posture in Y-axis vibration.

Similarly in Z-axis vibration, both the perceived difficulty andreading performance was affected more while reading newspaper

on the table posture. Since the lap posture is comparatively relaxposture under vertical excitation which results in a softening of thebiomechanical system and reduces the resonance frequency of thebody. As the muscles relax, the body stiffness reduces and thedamping increases (Kitazaki and Griffin, 1997). However, thevibrations on the table are considerable compared to lap due todamping in the lap, the perceived difficulty and decrement inreading performance could be expected in vertical excitation.

While in both XY- and XZ-axis vibration, the subject perceivedequal difficulty in both posture, however reading performance wasgreatly hampered while reading on table as compared to lap. It wasalso observed that both the perceived difficulty and decrement inreading performance was affected more while reading newspaperon the table in dual YZ-axis vibration. Similar effect attributable tosubject postures was found with multi axis vibration for both theperceived difficulty and reading performance. Since as discussedpreviously for individual Y- and Z-axis vibration, the effect ofposture was more pronounced on the table, therefore it is expectedthat the same effect can occurs in combined axis.

5.3. Effect of vibration directions

The directions of vibration stands third position in overallcontribution effect of independent variables by both subjectiveevaluation and reading performance (Tables 2 and 3). The inter-acted variable (D � P) also found comparable contribution toreading performance.

While comparing all mono axes vibration, both the perceiveddifficulty and decrement in reading performance was higher inboth Y- and Z-axis vibration. The adverse effects of Y-axis vibrationmay have arisen from increased upper body movement and that ofZ-axis due to themore pitchmovement of head thereby resulting ininterference with vision (Paddan and Griffin, 1988).

The effect of vibration in dual axes was compared with itsassociated mono axes, in order to study their combined effect onreading performance by both subjective and objective methods.The decrement in reading performance shows that the effect ofvibration was similar for all dual axes and Y-axis and greater thanX-and Z-axis vibration. This could be attributable to higherperformance degradation in Y-axis vibration. From the subjectiveevaluation, the results revealed that the subject perceived greaterdifficulty in all dual axes vibration than all mono axes vibration.

Lewis and Griffin (1978) reported that the effects of multi axisvibration have been found to be similar to the effect of single axisvibration corresponding to the RSS of the magnitudes in each axis.It was observed from results of subjective evaluation that the effectof multi axis vibration has been similar to the effect of dual axisvibration (p> 0.05) and higher than mono axis vibration (p< 0.05).The decrement in reading performance in multi axis was alsosimilar with all dual axes and one mono axis (i.e. Y-axis) vibration.

The results from the present study have indicated that low levelsof vibration in mono axes may not adversely affect task perfor-mance as compared to dual and multi axis vibration. This couldsuggest that reading performances are more sensitive to dual andmulti axis than mono axes vibration.

6. Conclusion

This is an important issue since the passengers use most of theirtraveling time for performing sedentary activities. In this study ofreading newspaper in passenger’s compartment like environment,both the perceived difficulty and reading performance shows thatthe extent of interference in reading newspaper increases withvibration magnitudes in all the directions of vibration for bothsubject postures. The subject experience difficulty and decrement

M.K. Bhiwapurkar et al. / International Journal of Industrial Ergonomics 41 (2011) 625e633 633

in reading performance in both lateral and vertical directionequally. The effects of multi axis vibration on perceived difficultyhave been found to be similar to dual axes vibration and greaterthan mono axes vibration; however degradation in readingperformance in multi axis vibration was also found to be similar tothat for lateral direction. The study revealed pronounced effect ofsubject postures in each direction of vibration. The subjectperceived higher difficulty and decrement in reading performancewhile working on lap due to transmitted vibration to the head andinterruption in vision. While in all other directions of vibration, thedifficulty perceived and decrement in reading performance wasfound to be higher on table than with lap posture, except dual XY-and XZ-axis vibration in subjective evaluation, in which equaldifficulty was perceived in both the subject postures.

Acknowledgments

The financial assistance received from Department of Science &Technology, New Delhi (India) for the research work is dulyacknowledged. We also wish to thank all the participants forparticipating in the experiment and openly discussing theiropinions.

References

Andersson, E., Berg, M., Stichel, S., 2005. Rail Vehicle Dynamics. KTH e Div. ofRailway Technology.

Bhiwapurkar, M.K., Saran, V.H., Harsha, S.P., Goel, V.K., Berg, M., 2010a. Effect ofmagnitudes and directions (mono-axis and multi-axis) of whole body vibrationexposures and subjects postures on the sketching performance. Proceedings ofthe Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit225, 71e83.

Bhiwapurkar, M.K., Singh, P.P., Yadav, J., Saran, V.H., Harsha, S.P., 2010b. Influence ofvibration on passenger comforte A survey on Indian train. International

Conference on Advances in Industrial Engineering Applications (ICAIEA 2010).Anna University Chennai, India.

Corbridge, C., Griffin, M.J., 1991. Effects of vertical vibration on passenger activities:writing and drinking. Ergonomics 34, 1313e1332.

Corbridge, C., Griffin, M.J., 1986. Vibration and comfort: vertical and lateral motionin the range 0.5 to 5.0 Hz. Ergonomics 29, 249e272.

ENV 12299, 1999. Railway Applications e Ride Comfort for Passengers eMeasurements and Evaluation.

Griffin, M.J., Hayward, R.A., 1994. Effects of horizontal whole-body vibration onreading. Applied Ergonomics 25, 165e169.

Griffin, M.J., 2003. Handbook of Human Vibration, second ed. Academic PressLimited, London.

Henrietta V.C.H., 2004. A comparison of standardized methods of evaluating railvehicle vibration with respect to passenger discomfort. 39th United KingdomConference on Human Response to Vibration. 395e408.

ISO 2631-4, 1998. Mechanical Vibrations and Shock e Evaluation of Human Expo-sure to Whole Body Vibrations e Part 4: Guidelines for the Evaluation of theEffects of Vibration and Rotational Motion on Passenger and Crew Comfort ofFixed Guide Way Transport Systems Revised draft, Geneva.

Kitazaki, S., Griffin, M.J., 1997. Resonance behaviour of the seated human body andeffects of posture. Journal of Biomechanics 31, 143e149.

Kuorinka, I., Jonsson, B., Kilbom, A., Vinterberg, H., Biering-Sørensen, F.,Andersson, G., Jørgensen, K., 1987. Standardised nordic questionnaires for theanalysis of musculoskeletal symptoms. Applied Ergonomics 18, 233e237.

Lewis, C.H., Griffin, M.J., 1978. A review of the effects of vibration on visual acuityand continuous manual control. II: continuous manual control. Journal of Soundand Vibration 56, 415e457.

Lewis, C.H., Griffin, M.J., 1980. Predicting the effects of vertical vibration frequency,combinations of frequencies and viewing distance on the reading of numericdisplays. Journal of Sound and Vibration 70, 355e377.

Mansfield, N.J., 2005. Human Response to Vibration. CRC Press, London.Mansfield, N.J., Maeda, S., 2005. Comparison of subjective ratings of whole-body

vibration for single and multi-axis vibration. In: Proc.. of 40th UnitedKingdom Conference on Human Response to Vibration Liverpool, England.

Moseley, M.J., Griffin, M.J., 1986. Effects of display vibration and whole bodyvibration on visual performance. Ergonomics 29, 977e983.

Paddan, G.S., Griffin, M.J., 1988. The transmission of translational seat vibration tothe head e II. Horizontal seat vibration. Journal of Biomechanics 21, 199e206.

Sundström, J., Khan, S., 2008. Influence of stationary lateral vibrations on trainpassengers’ ability to read and write. Applied Ergonomics 39, 710e718.

Wollstrom, M., 2000. Effects of vibrations on passenger activities: reading andwriting e a literature study, TRITA eFKT Report 2000:64.