Whole-body vibration exposure experienced by motorcycle. riders An evaluation according to ISO and ISO standards

Transcription

1 Whole-body vibration exposure experienced by motorcycle riders An evaluation according to ISO and ISO standards Hsieh-Ching Chen 1, Wei-Chyuan Chen 1, Yung-Ping Liu 1, Chih-Yong Chen 2, Yi-Tsong Pan 2 1 Department of Industrial Engineering and Management, Chaoyang University of Technology, Taiwan 2 Institute of Occupational Safety and Health, Council of Labor Affairs, Executive Yuan, Taiwan Address correspondence to: Hsieh-Ching Chen Department of Industrial Engineering and Management Chaoyang University of Technology No. 168 Jifong E. Rd., Wufong Taichung County 41349; Taiwan Tel: ext 4255 Cell phone: Fax: hcchen@cyut.edu.tw First Ed.: December 12, 2007 R1: March 14, 2009

2 Abstract Riders of twelve motorcycles, comprising 6 full-scale motorbikes and 6 motor-scooters, and 5 sedan vehicles, performed test runs on a 20.6 km paved road composed of 5 km, 5 km, and 10.6 km of rural, provincial and urban routes, respectively. Each test-run of motorcycle was separately performed under speed limits of 55 km/h and 40 km/h. Tri-axial accelerations of whole-body vibration (WBV) were obtained by using a seat pad and a portable data logger, and the driver s view was videotaped with a portable media recorder. Root mean square (RMS) acceleration, 8-h estimated vibration dose value (VDV (8) ) and 8-h estimated daily dose of static compression dose (S ed ) were determined from the collected data in accordance with ISO and ISO standards. Experimental results indicate that the WBV values of the sedan vehicle drivers have low RMS, VDV (8) and S ed values (RMS m/s 2 ; VDV (8) m/s 1.75 ; S ed MPa). However, over 90% of the motorcycle riders had VDV (8) (mean 23.5 m/s 1.75 ) exceeding the upper boundary of health guidance caution zone (17 m/s 1.75 ) recommended by ISO , or had S ed (mean 1.17MPa) exceeding the value associated with a high probability of adverse health effects (0.8MPa) recommended by ISO Over 50% of the motorcycle riders reached these boundary values for VDV and S e in less than 2 hours. The WBV exposure levels of the full-scale motorbikes riders and motor-scooter riders were not significantly different. However, the RMS and VDV (8) values of motorcycle riders indicate significant roadway effect (p<0.001), while their S ed values indicate significant speed limit effect (p<0.05). This study concludes that the WBV exposure levels of common motorcycle riders are distinctively higher than those of sedans, even on a regular paved road. The impact on health of WBV exposure in motorcycle riders should be carefully addressed with reference to ISO and ISO

3 Relevance to industry This study compares the predicted health risks of motorcycle riders according to ISO and ISO standards. Experimental data suggest that the vibration dose value of ISO and daily dose of equivalent static compression stress of ISO have roughly equivalent boundaries for probable health effects. Keywords: Whole-body vibration, vibration exposure, ISO2631-1, ISO2631-5, transportation 1. Introduction The motorcycle is the primary vehicle for short-distance transportation in Taiwan. The Department of Statistics, Ministry of Transportation and Communications in Taiwan, reported more than 13.8 million registered motorcycles by the end of October 2007, and estimated that over 11 million of them were in use (Taiwan MOTC, 2007). Some other Asian countries also have many motorcycles, with estimates of over 33 million motorcycles in mainland China, 18 million in Vietnam, 13 million in Japan, 5.8 million in Malaysia, 2.7 million in Korea and 1.8 million in the Philippines (IRF, 2006). Although most riders use motorcycles only for short-distance transportation, some people, such as postmen, deliverymen and some urban couriers use a motorcycle as the major vehicle for work. Consequently, many workers are likely to experience high levels of WBV caused by motorcycle riding. Most motorcycles in Taiwan can be categorized as scooters (no clutch, sit-on riding) and motorbikes (with clutch, straddle 2

4 riding). These motorcycles generally have engines of 125cc or smaller, are ridden on the shoulder or outside reserved lane of the road, and are convenient for accessing driving lanes. Our experience and observations indicate that a motorcycle can be ridden at an equivalent or slightly higher speed than a car in a metropolitan or urban area with heavy traffic. Riding a motorcycle raises concerns not only about traffic safety, but also about health relating to WBV exposure. Previous studies have identified increased low back pain (Palmer et al., 1999), finger and shoulder symptoms (Mirbod et al., 1997), and a high rate of erectile dysfunction (Ochiai et al., 2006) among motorcyclists. Experienced motorcycle riders assert that continual riding of a standard motorcycle for extended periods can result in stress and medical complaints. A motorcycle rider may encounter multiple shocks introduced from uneven road surfaces due to irregularities such as potholes, humps and manhole covers. Previous studies have reported that a driver s vibration exposure level depends on the road or traffic conditions, vehicle characteristics such as speed, type, weight, seat, maintenance and engine size, and driver characteristics such as age, characteristics, experience, sitting posture and body weight (Peitte and Malchaire, 1992; Ozkaya et al.,1994; Malchaire et al., 1996; Donati, 1998; Chen et al., 2003; Mansfield and Griffin, 2002). Numerous investigations have reported possible discomfort, musculoskeletal problems, muscular fatigue, reduced stability and altered vestibular function caused by WBV exposure (Seidel, 1993; Wasserman et al., 1997; Bongers et al., 1988; Griffin, 1998). Most previous evaluations of WBV exposure measured and analyzed the acceleration data in terms of ISO (1997). However, in ISO (1997) 3

5 Section 6.2.2, the applicability of a basic evaluation is limited for vibrations with crest factors 9. Additionally, ISO (1997) suggests that an alternative evaluation should be considered under conditions of transient shocks, particularly those that occur rarely. Troup (1988) reported that transmitted road-shock is a source of back problems. Wikström et al. (1994) reviewed studies on the effect of long-term exposure to WBV, and concluded that many repeated shocks of sufficient level and duration might lead to back problems. In considering the health problem that shocks might cause, ISO (2004) was proposed to assess the adverse effects on health from vibration containing multiple shocks by considering 2 biomechanical models of the lumbar response. The authors are not aware of any previous studies of WBV exposure of motorcycle riders while performing riding tasks for an extended period of time, namely 1 hour or more. Most scientific investigations have addressed the WBV exposure from vehicles such as trucks, taxi and industrial vehicles (Malchaire et al., 1996; Maeda and Morioka, 1998; Chen et al., 2003; Cann et al., 2004; Rehn, 2005; Hoy et al., 2005), or specialist vehicles, such as terrain vehicles (Goglia and Grbac, 2005; Johanning et al., 2006; Scarlett et al., 2007; Eger et al., 2006, 2008). This study measures the WBV exposure of motorcycle riders while driving on rural, provincial and urban roads with standard paved surfaces. Additionally, the WBV exposures from driving of regular sedan vehicles on the same route are evaluated to serve as a control. This study compares measured WBV exposure with the upper boundary of health guidance caution zone (HGCZ) recommended by ISO (1997, Fig. B.1) and with the limit value associated with a high probability adverse health effects recommended by ISO (2004, Figs. A.1 and A.2). 4

6 2. Method 2.1 Participants and vehicles Twelve senior university students, with a mean age of 22 years (SD=1.5), were recruited to participate in motorcycle riding tests. Five experienced sedan drivers, with a mean age of 29 years (SD=2.4), were also recruited for driving tests. All participants were male, with an average height and mass of cm (SD=4.3 cm) and 67.8kg (SD=8.6) respectively. Each motorcycle driver had at least 2 years experience of motorcycle riding, and was familiar with the test route adopted in this study. Each participant self-reported riding his motorcycle in an upright sitting position without leaning forward or backward by more than 10. Participants were divided into groups of 6 scooter riders and 6 motorbike riders. All motorcycles had the same engine size of 125cc. The sedan vehicles had engine sizes in the range cc. Table 1 presents the characteristics of the participants and the tested vehicles in detail. 2.2 Equipments A tri-axial ICP seat pad accelerometer (model 356B40, Larson Davis Inc., USA) was employed to measure vibrations transmitted to the seated human body as a whole through the supporting surface of the buttock. The accelerometer had a frequency sensitivity range of Hz, and was pre-calibrated for excitation of 1g RMS/ Hz with a hand-held calibrator (model 394C06, PCB Piezotronics Inc., USA). Seat pad outputs were connected to a 3-channel amplifier (model 480B21, PCB Piezotronics Inc., USA) with a signal conditioning gain of 10. The outputs of the amplified signals were recorded on a portable data logger. This data logger was a 5

7 modified version of that used by Liu (2006); it can acquire three analog signals each at a rate of 5000 samples per second, and store collected data on a 2GB compact flash (CF) memory card (Fig. 1a). The logged data were downloaded onto a personal computer using a card reader for further data processing. The view of a motorcycle rider was recorded by a portable media recorder (CNF-200, Carry Media Electronic Ltd., Fig. 1b) and a mini camera. The recorded video was synchronized with the logged acceleration data using a remote transmitter, which sent radio frequency signals at the beginning and end of the riding task. 2.3 Field testing procedure The test runs were performed on a 20.6 km paved route, comprising 5 km rural roadway, 5 km provincial roadway and 10.6 km urban roadway. The route started from the university, and followed the main road to the Taichung City central railway station, and then back to the university the same way. Each test was undertaken on weekdays between 9:30 a.m. and 11:30 a.m., or between 1:30 p.m. and 4:30 p.m., to ensure consistent traffic conditions, and to avoid traffic jams. The standard testing procedure of the riding/driving test and detailed instructions was explained to all participants before the test. Each motorcycle rider performed two riding tasks, one with a speed limit of 55 km/h, and the other with a speed limit of 40 km/h, on a different days. All motorcycle riders rode their own motorcycles, and the speed limits were randomly assigned to their first tasks. For each riding task, a seat pad accelerometer was placed on the seat beneath the 6

8 subject s buttocks in accordance with the ISO standard (1997, Fig. 1a); the positive x and z directions were anterior and upward, respectively. Thus, acceleration in the positive z direction generates a compressive loading on the subject s spine. Each motorcycle rider wore a helmet with a mini-camera taped on its frontal side, and a shoulder strap with a case carrying a portable media recorder (Fig. 2). Each rider also wore a packsack to carry the signal amplifier, the data logger and a rechargeable battery set. The overall weight of the packsack was 1.5 kg. Each participant was asked to remain on the seat, without rising during the riding task. To avoid harmful shocks, the riders were expected to go round, or slow down over, potholes, manhole covers, humps or uneven surfaces that they encountered. The riding and road conditions were videoed during the tasks. Each task took about minutes. The sedan drivers performed the test on the same route, only under the speed limit of 55 km/h, with the mini camera taped on the windshield, and other equipment secured on the passenger seat. Additionally, a remote synchronous transmitter sent a registering signal at each complete stop before a stop light. 2.4 Data processing procedure Analysis software Viewlog software programmed with LabVIEW 7.0 (National Instruments, USA) was applied to download the logged data from the CF card, and to synchronize the data with the taped video (Fig. 3a). Viewlog consists of calibration, vibration analysis, script interpretation and batch processing modules to facilitate analysis and processing of data in bulk. (Chen, 2006) The vibration analysis module, developed under a research contract of Taiwan IOSH (2007), evaluates the exposure level of WBV in terms of both ISO (1997) and ISO (2004) 7

9 (Fig. 3b). The module was specifically designed to perform batch computing and export the results to a user-defined MS Excel template. Batch processing Following the calibration process, acceleration data were divided into periods of 30 seconds, and was batch processed using the vibration analysis module. For each ith data period, the root mean square ( a, ) and vibration dose values (VDV i ) of accelerations in the x and y axes, and in the z-axis, were individually computed from the W d and W k frequency weighting functions of ISO (1997), respectively. The computation of acceleration dose (D k,i ) for the x- and y-axis were based on a biomechanical model with a single degree-of-freedom and the ISO procedure (2004, 5.2.2, 5.3, 5.4). The computation of D k.i for the z-axis started with down-sampling acceleration to 160 Hz, which was then used as input for the recurrent neural network model described in ISO (2004, 5.2.3, 5.3, 5.4). The model calculation results are response accelerations. The D k.i is calculated by summing the response acceleration peaks. According to ISO (2004, 5.3), a peak is the maximum absolute value of response acceleration between consecutive zero crossings. Although peaks in the positive and negative directions are considered for response accelerations in the x and y directions, only positive peaks (compression of the spine) count for response acceleration in the z direction. The vibration exposures from each 30-sec period were exported to a pre-defined MS Excel template to generate the WBV exposure report. w i Artefact removal An experienced experimenter manually inspected the Excel report to identify every record period containing atypically large VDV i, D k,i and a w, i values. The experimenter then inspected the raw acceleration data and synchronized 8

10 video of each period with abnormal data, sometimes helped by the test participant, to determine whether the abnormal data represented an artefact, or a natural feature of the task. An abnormal record confirmed to be owing to an artefact, caused by ingress/egress or the participant instinctively rising from the seat, was removed by replacing all VDV i, D k,i and a w, i values of the record with zero values, instead of by deleting the record. This procedure ensured that the overall period of the measurement remained the same. Exposure report - The Excel report employed an embedded macro program to compute the total RMS, VDV and D k of each axis by combining data of all 30-sec exposure periods using Eqs. (1), (2), and (3) given below, 1/ 2 2 a w, i Ti RMS = (1) T i 4 1/ 4 [ VDV ] VDV (2) = i 6 1/ 6 [ ] D (3) k = D k, i where a w, i, VDV i and D k,i denote the frequency-weighted root mean square acceleration, VDV and acceleration dose of the ith exposure period, respectively, and T i denotes the exposure duration (30 sec) of the ith period. The axis with the highest (most severe) total RMS acceleration, obtained by running embedded math functions in Excel, was compared with the upper boundary of HGCZ ( RMS ) defined in ISO (1997, Figure B.1), which is 0.8 m/s 2. The multiplying factor k=1.4, as suggested by ISO (1997, 7.2.3), in weighting transversal (x and y) acceleration was not applied in this study. To compare vibration levels of motorcycles with those of other vehicles, we assume a daily exposure time of 8 hours. The estimated 8-h 9

11 VDV (VDV (8) ) was thus derived from the axis with the greatest total VDV using Eq. (4), and compared with the upper boundary of HGCZ (VDV ) defined in ISO (1997, B.3.1), which is 17 m/s VDV T = (4) 8 1/ 4 ( 8) VDV ( ) T m In Eq. (4), T 8 denotes the 8-h time period, and VDV and T m denote the total VDV and the measurement duration, respectively. The equivalent static compressive stress (S e ) (ISO , 2004, A.2) during the measurement period was then calculated from the total D k of each axis with the following equation: e / 6 [ 0.015D ) + (0.035D ) (0.032D ) ] S = + (5) ( x y z The value of S e was then scaled by Eq. (6) to obtain estimated daily equivalent static compression dose (S ed ), and compared with the limit value of 0.8 MPa ( S ed ), which is defined by ISO (2004, A2 and Figure A.1) as indicating high probability of an adverse health effect. 1/ 6 T 8 S = ed Se (6) Tm In Equation (6), T 8 represents an 8-h time period, and T m represents the measurement duration. The allowable duration (T a ) for RMS, VDV, and S e to reach the exposure value (EV) of RMS, VDV, and S ed (7):, respectively, were calculated using Eq. T a EV n = ( ) T (7) VL where VL denotes the vibration level or dose of RMS, VDV and S e, with n = 2, 4, and 6, 10

12 respectively, and T = T 8 for n = 2 and T = T m for n = 4 and 6. A risk factor (R) was computed using Eqs. (8) and (9) (ISO , 2004, A.3) sequentially considering increased age as the exposure time increases: 6 1/ 6 n 1/ 6 Sed N R = (8) j= 1 Suj c S uj = ( b + j) (9) where N is the number of exposure days per year, j is the year counter, n is the number of years of exposure, c is a constant representing static stress due to the gravitational force, S uj is the ultimate strength of the lumber spine for a person aged (b+j) years, with b as the age at which exposure started. The characteristics of a typical motorcycle rider (b=20 years; N=240 days; c=0.25 MPa) and 4 S ed scaled from each trial S e by Eq. (6) with estimated daily exposure periods of 1, 2, 4, and 8 were applied in computing R. For each S ed, the age (b+ j) in years at which R reaches 1.2 ( R ) was utilized for statistical analysis. Notably, R, which is defined by ISO , indicates a high probability of an adverse health effect. 2.5 Data analysis The measurement duration for each task was counted from the time the participant drove out of the parking lot to the time that he returned to the point of departure. Sub-periods of riding or driving on the rural, provincial and urban roadway for all subjects and tasks in each measurement period were determined from the synchronized video and logged data provided by an experimenter. The data of an exposure report was further divided into WBV exposures on each roadway, based on the duration of each sub-period. The number of complete stops was determined by 11

13 counting the number of registered wireless indents, and double-checked from the video data. The statistical analysis was performed with SPSS 10 for Windows. Group differences in WBV exposure of sedan and motorcycle were assessed by the independent t-test. The differences among all measurements obtained in the motorcycle ride tests were compared by repeated-measures ANOVA, using the vehicle type as the between-subjects factor, and the roadway and speed limit as the within-subjects factors. A difference at a level of p<0.05 was regarded as significant. 3. Results No significant differences in measurement period and average speed were observed among the sedan (mean 55.0±1.1 min) and the two types of motorcycles at the speed limit of 55 km/h. The average measurement periods of the motorcycles were 51.8 and 61.8 minutes for the tests at speed limits of 55 km/h and 40 km/h, respectively. The average riding speed of the motorcycles was 23.9 km/h and 20.0 km/h at speed limits of 55 km/h and 40 km/h, respectively (Table 2). Roadway and speed limit significantly affected the duration and average riding speed of the motorcycles (p<0.001). The driving speed was highest on the rural roadway, with average speeds of 31.6km/h and 25.0 km/h for the speed limits of 55 km/h and 40 km/h, respectively, and was the slowest on the urban roadway, with an average of 21.8 km/h and 19.0 km/h, respectively (Table 2). This difference was partly due to the different numbers of complete stops required on urban and rural roadways. 3.1 WBV in sedans and motorcycles 12

14 Analytical results indicate that acceleration in the z-axis generated the most severe total RMS and VDV levels. Therefore, RMS and VDV (8) was determined from RMS z and VDV z, respectively, and then compared with the corresponding upper boundary of HGCZ defined by ISO (Table 3). The statistical results demonstrate no significant difference between the two motorcycle types in terms of any driving metric (duration, average speed, number of complete stops) or level of RMS, VDV (8) and S ed. The statistical analysis was thus undertaken by treating all motorcycles as one group. The WBV exposure values for RMS, VDV (8), and S ed in the sedan drivers were much lower than those of the motorcycle riders (p<0.001, t-test). For the sedan drivers, the RMS ranged from 0.27 m/s 2 to 0.32 m/s 2 ; VDV (8) ranged from 6.3 m/s 1.75 to 8.3 m/s 1.75, and S ed ranged from 0.21 MPa to 0.26 MPa. These values were all lower than the corresponding values of RMS, VDV, and S ed proposed in the ISO standards. In contrast, all motorcycles on the 20.6 km ride had the S e and VDV values greater than 0.5 m/s 2 and 8.5 m/s 1.75, respectively. Furthermore, 11 out of the 24 values for S e exceeded S ed, and 2 out of the 24 VDV exceeded VDV (Fig. 4). If estimated for 8 hours of exposure, over 90% of all motorcycle rides in this study would produce VDV (8) (mean 23.1 m/s 1.75 ) values exceeding VDV in ISO , and S e(8) (mean 1.15 MPa) over S ed in ISO (Fig. 4). However, the RMS parameter underestimated the WBV exposure resulting from motorcycle rides, since 38% of all tasks fell within the caution range of m/s 2, and only 62% of all tasks exceeded 0.8 m/s 2. The motorcycle riders had more variable vibration exposure than the sedan 13

15 drivers at the same driving speed. Notably, a significant positive correlation was noted between S e and VDV for all motorcycle measurements with the regression line passing through the conjunction of the ISO HGCZ ( m/s 1.75 ) and ISO caution range ( MPa) (Fig. 4). Analytical results indicate that the crest factors in the most sever axis were greater than 9, and the values of VDV/(a w T 1/4 ) were greater than 1.75, in all motorcycle rides. This evidence implies that the WBV exposure of the motorcycle ride involved occasional shocks, meaning that alternative parameters to RMS should also be considered. Visual inspection of the acceleration signals also confirmed the presence of multiple shocks. 3.2 Effect of roadway condition and driving speed Significant roadway effect (p<0.001) was found in RMS and VDV (8), and significant speed limit effect (p<0.05) was found in and S ed, respectively. (Table 4) The greatest mean RMS (1.08 m/s 2 ), VDV (8) (26.55 m/s 1.75 ) and S ed (1.24 MPa) was obtained on the rural roadway with the speed limit at 55km/h. The average motorcycle speed in this case was 31.6 km/h (Table 2). A slower average riding speed or lower speed limit lowered the vibration to which the riders were exposed. Lowering the speed limit from 55 km/h to 40km/h caused the mean S ed, VDV (8) and RMS to fall on average by 21%, 7%, and 3%, respectively (Table 4). However, riding speed was not the only factor influencing the WBV exposure of the motorcycle riders. The WBV exposure on the urban roadway was greater than that on the provincial roadway, even at the slowest riding speed. 14

16 3.3 Comparison of the ISO standards The value of S e, according to the total vibration dose received by motorcycle riders, can be employed to obtain the shortest exposure period in which S ed (0.8 MPa) can be reached. The average exposure period of a motorcycle drive to reach value of S ed and VDV were estimated from S e and VDV as 2.26 hours (SD=4.1) and 3.3 hours (SD=2.3), respectively, at the speed limit of 55km/h, and 3.2 hours (SD=3.0) and 3.9 hours (SD=2.5), respectively, at the speed limit of 40 km/h (Fig. 5). In contrast, the RMS overestimated the riding time needed to reach RMS as 9.3 (SD=3.3) and 11.0 (SD=3.7) hours at speed limits of 55 km/h and 40 km/h, respectively. Although estimates produced by S e provided the most rigorous estimation among all vibration parameters on the time taken to reach S ed, its estimates were also more variable than those of either VDV or RMS. Nevertheless, the time taken to reach S ed or VDV was shorter than 2 hours in 50% of the tasks, according to estimated values obtained from S e and VDV (Fig. 5). Thus, S e and VDV produced closer estimates than RMS of the time taken by a rider to reach the corresponding S ed and VDV. According to the sequential computation of the R factor, the age at which a rider reaches the value of R reduced significantly as speed limit and daily exposure increased (Fig. 6). For example, 75% (9/12) and 42% (5/12) of experimental trials at 40 km/h predicted that R reaches R until a rider reaches an age of 65 years of older for an assumed daily exposure of 1 and 2 hours, respectively. The corresponding percentages reduced to 33% and 25% for trials at 55 km/h. Moreover, 17% (2/12) and 33% (3/12) of riders traveling at 55 km/h speed limit had R reach R before the rider turned 45 with 1 and 2 hours of daily exposure, respectively. 15

17 4. Discussion The participants found that the route tested in this study was not significantly different from the roadway conditions that they generally encountered, such that it can represent a typical roadway for most motorcycle riders. The participants rarely reached the speed limit of 55 km/h during their rides, in contrast with the speed limit of 40 km/h. These participants were more likely to ride at their daily comfortable speeds in the 55 km/h speed limit trial. Thus, the experimental result of the 55 km/h speed limit tasks demonstrates WBV exposure might create significant health hazards in Asian countries with a large number of motorcycles. Long-term exposure to WBV is associated with early spinal degeneration (Frymoyer et al. 1984), low back pain and herniated lumbar disc (Bovenzi and Zadini, 1992; Boshuizen et al., 1992). Most motorcycle riders do not expose themselves to WBV for extended periods. However, the large population and gradually aging society means that many motorcycle riders require special attention for potential health problems resulting from WBV exposure. The ISO was first published in 2004 to deal with WBV with multiple shocks. This study produced similar results to several previous investigations that also applied ISO standard in assessing exposures to WBV. From a study of 158 measurements, Marjanen (2005) showed that the RMS does not properly address transient shocks, and that the S ed value is more rigorous than RMS when evaluating health effects. Khorshid et al. (2007) conducted tests in which participants drove over various shapes of speed control hump at speeds of km/h, and indicated that higher speeds generally produced larger values of VDV and S e. Alem (2005) demonstrated that the S ed value was positively correlated with the VDV (8), and was 16

18 more sensitive to shock-containing vibrations than either the VDV or the RMS. However, some studies have also been undertaken and yielded opposite results. Eger et al. (2008) explored WBV exposure in load-haul-dump (LHD) mining vehicles, and concluded that the ISO criteria always predicted lower health risks than ISO (Fig. 7). Nevertheless, Alem (2005), who observed a low VDV (8) in army vehicles, suggested that HGCZ of VDV should be reduced to m/s 1.75 to have equivalent risk assessment as S ed does in detecting high shock content in a vibration signature (Fig. 7). Despite the fact that these studies obtained conflicting experimental results, this study demonstrated that S e produced prediction of health risks slightly more severe than VDV for a 50- to 70-min period of assessment. However, when exposure time is extended to, say, 8 hours, evaluation of health risks based on ISO and can be highly consistent for motorcycle riders, as shown by the original 20.6-km regression line inclining toward an increased VDV and passing through the conjunction of the two caution zones (Fig. 4). The observed tilt of the VDV-S e regression line can be explained by the greater time scaling factor for VDV ( 4 T 8 / Tm, Eq. (4)) than that for S ed ( 6 T 8 / Tm, Eq. (6)). A similar trend for VDV-S e regression lines existed for sedans (Fig. 4). The disagreement between the results of Eger et al. (2008) and of this study may be due to the different vehicles evaluated. For a comparable VDV (8) exposure range of m/s 1.75, Eger et al. (2008) found a significantly lower S ed (type A, C; range= MPa) for LHD mining vehicles than that for motorcycles (range= MPa). A rational explanation is that a four-wheel vehicle typically has better suspension than a motorcycle, and the driver sits far from the wheel lines. 17

19 This design dampens and decreases shocks introduced by the wheels, as demonstrated by the experimental results for sedans in this study, in which the VDV-S e regression line leans toward that of LHD vehicles. In contrast, a motorcycle rider always sits on the line of the wheels, and is thus confronted with larger shocks introduced from the wheels than a four-wheel car driver at the same speed. As for army vehicles results reported by Alem (2005), a large S ed and relatively low VDV (8) may result from short daily exposure under high jolt conditions, the use of an incongruous formula (Alem, 2005, Annex A, Eq. A-3), or some other unknown causes. According to the ISO (2004) standard, the assessment of WBV is based on an upright posture. A bent forward or twisted posture is likely to increase the adverse health effects of drivers while exposing to WBV (ISO , 2004, Annex A). Hoy et al. (2005) also noted a higher rate of low back pain among forklift drivers than among non-drivers. Additionally, a driving posture with the trunk twisted or bent forward is associated with increased risk of low back pain. Although riding posture in individual subjects was not restrained in this field study, each subject claimed he drove with an upright sitting posture and rarely twisted or bent his trunk. Therefore, unreliably measured body responses due to inconsistent posture with the ISO model constraints are negligible. However, motorcycle riders typically sit higher than most vehicle drivers and generally do not have backrests. Hinz et al. (2002) conducted an study of 39 males sitting on a suspension seat with and without a backrest exposed to random WBV with a w =0.6 m/s 2 at a relaxed or bent forward posture. They demonstrated that peak transmissibilities between accelerations at the seat base and the compressive force at L5/S1 were highest for those without the backrest during bending. Wang et al. (2004) determined that increasing seat height 18

20 increases peak magnitude response as a relatively larger portion of body mass is supported by the seat. Wang et al. (2006) also demonstrated that sitting on a flat pan without back support causes greater energy absorption of vertical vibrations than when sitting with a back support. These experimental findings indicate that motorcycle riders likely have higher health risks than that of those driving 4-wheel vehicles with comparable WBV exposure. Further studies of the effects of posture and seat support on WBV-related health risks among motorcycle riders are warranted. The results of this study show that applying different evaluation methods may lead to different conclusions about the WBV exposure. This study found that RMS produced much lower results for WBV exposure in motorcycle riding than VDV and S e. This observation indicates that a motorcycle rider encounters many shocks. Vibration peak distributions encountered by motorcycle riders result from a combination of riding speeds and road profiles (Khorshid et al., 2007). In real environments, riding speed is typically related to traffic conditions, road profiles, imposed speed limits, and driver behavior. In this field study, average riding speed was significantly affected by roadway factors, speed limit and the interaction between the two (Table 2). Due to technical difficulties, this study was unable to measure actual speed and road profiles experienced by individual subjects or to further explore the possible effects of individual factors on RMS, VDV (8), and S ed. Nevertheless, the roadway significantly impacted RMS and VDV (8). In contrast, S ed only identified the effect of speed limit. This may suggest that the VDV parameter in ISO is more sensitive to vibrations resulting from road profile or surface characteristics, while the S e parameter in ISO may be more sensitive to effects resulting from the riding speed. For example, S ed was 21% lower at a speed limit of 40 km/h than at 55 km/h, 19

21 while the corresponding reduction in VDV (8) was only 7%. Therefore, both ISO and ISO should be considered when assessing the WBV exposure. In other words, since ISO and ISO both produce dose values with great variabilities, either method might underestimate the health risks if applied on its own to measure WBV exposure (Alem, 2005; Marjanen, 2005; Eger et al. 2008). Analytical results show that very large values of VDV or S e can be introduced simply through artifacts. The greatest artifact resulting from ingress of the participants generated D z =47 m/s 2 and VDV=12 m/s The greatest D z caused by a rider voluntarily rising from the seat during a motorcycle ride was over 30 m/s 2. This observation was larger than that reported by Newell and Mansfield (2004), and indicates that all artifacts need to be identified and removed, since a single artifact may distort a WBV assessment. With the artifact removal procedure, the decision to segment data segmentation into periods of 30 sec represented a trade-off between efficient data processing and minimizing data loss caused by artifact removal. The data processing architecture employed in this study enables software to bulk-process data obtained over an extended period, so that a more rational and consistent result of WBV exposure can be obtained. The results of the artifact removal procedure were regarded as conservative because of the zero introducing method adopted. This study found that WBV exposure levels of sedan drivers and motorcycle riders were significantly different. The RMS acceleration levels of the sedan driving (ranging from 0.27 m/s 2 to 0.32 m/s 2 ) were within the range of urban taxi driving ( m/s 2 ) in Taiwan (Chen et al., 2003). Whereas all sedan drivers required exposure times of more than 8 hours to reach RMS, VDV, and S ed, motorcycle 20

22 riding required considerably shorter periods of WBV exposures. Over 50% of the motorcycle riding tasks in this study had exposure period limits of 2 hours or less, according to the VDV levels recommended in ISO , and particularly the S e guidance in ISO Nonetheless, most riders disregard the potential risks of WBV exposure in motorcycle riding. According to computational results for the R factor, 13% of motorcycle riders, even when riding only 1 hour daily, can have an R value exceeding R before the age of 45 when speed is not constrained. Riding a motorcycle at high speed can cause vigorous shocks or jolts to the riders; however, riding at a lower speed will increase the duration of exposure, which may raise subjective discomfort. A trade-off exists between the speed and exposure duration for motorcycle riders. Unfortunately, many people are unwilling to lower speed, and thus increase travel time, in order to reduce the possibility of adverse health effects. Effectively reducing the health risks of motorcycle riders associated with WBV may still require good maintenance of road surfaces, and improved design of vehicle cushion systems. Acknowledgements The authors would like to thank the Institute of Occupational Safety and Health for financially supporting this research under Contract No. IOSH96-H318. Ted Knoy is appreciated for his editorial assistance. References Alem N., Application of the new ISO to health hazard assessment of repeated shocks in U.S. army vehicles. Industrial Health 43, Bongers, P.M., Boshuizen, H.C., Hulshof, T.J., Koemeester, A.P., Back disorders in crane operators exposed to whole-body vibration. International 21

23 Archives on Occupational Environmental Health 60, Boshuizen, H.C., Bongers, P.M., Hulshof, C.T., Self-reported back pain in fork-lift truck and freight-container tractor drivers exposed to whole-body vibration. Spine 17, Bovenzi, M., Zadini, A., Self-reported low back symptoms in urban bus drivers exposed to whole-body vibration. Spine 17, Cann, A.P., Salmoni, A.W., Eger, T.R., Predictors of whole-body vibration exposure experienced by highway transport truck operators. Ergonomics 47, Chen, J.C., Chang, W.R., Shih, T.S., Chen, C.J., Chang, W.P., Dennerlein, J.T., Ryan, L.M., Christiani, D.C., Predictors of whole-body vibration levels among urban taxi drivers. Ergonomics 46, Chen H.C., Chen C.Y., Lee C.L., Wu H.C., Lou S.Z., Data logging and analysis tools for worksite measurement of physical workload. In: Proceedings, 16th World Congress of the IEA, July 10-14, Maastricht, Netherlands. Donati, P.,1998. A procedure for developing a vibration test method for specific categories of industrial trucks. Journal of Sound and Vibration 215, Eger, T., Salmonib, A., Cann, A., Jack, R., Whole-body vibration exposure experienced by mining equipment operators. Occupational Ergonomics 6, Eger, T., Stevenson, J., Boileau, P.-É., Salmoni, A., VibRG., 2008 Predictions of health risks associated with the operation of load-haul-dump mining vehicles: Part 1 Analysis of whole-body vibration exposure using ISO and ISO standards. International Journal of Industrial Ergonomics 38, Frymoyer, J.W., Newberg, A., Pope, M.H., Wilder, D.G., Clements, J., et al., Spine radiographs in patients with low-back pain. An epidemiology study in men. Journal of Bone & Joint Surgery- American volume 66, Goglia, V., Grbac, I., Whole-body vibration transmitted to the framesaw operator. Applied Ergonomics 36, Griffin, M.J., General hazards: vibration, Encyclopedia of Occupational Health and Safety (International Labour Organization Geneva), Hinz, B., Seidel, H., Menzel, G., Blüthner, R., Effects related to random whole-body vibration and posture on a suspended seat with and without backrest. 22

24 Journal of Sound and Vibration 253, Hoy, J., Mubarak, N., Nelson, S., Sweerts, de Landas M., Magnusson, M., Okunribido, O., Pope, M., Whole body vibration and posture as risk factors for low back pain among forklift truck drivers. Journal of Sound and Vibration 284, International Organization for Standardization, ISO Mechanical vibration and shock evaluation of human exposure to whole-body vibration. Part 1: General requirements. Geneva: ISO. International Organization for Standardization, ISO Mechanical vibration and shock evaluation of human exposure to whole-body vibration. Part 5: Method for evaluation of vibration containing multiple shocks. Geneva: ISO. IRF, World road statistics 2006 data 1999 to International Road Federation (IRF), Geneva, Switzerland. Johanning, E., Landsbergis, P., Fischer, S., Christ, E., Göres, B., Luhrman, R., Whole-body vibration and ergonomic study of US railroad locomotives. Journal of Sound and Vibration 298, Khorshid, E, Alkalby, F, Kamal, H., Measurement of whole-body vibration exposure from speed control humps. J Sound and Vibration 304, Liu, Y.P., Chen, H.C., Chen, C.Y., Multi-transducer data logger for worksite measurement of physical workload. J Medical and Biological Engineering 26, Maeda, S. and Morioka, M., Measurement of whole-body vibration exposure from garbage trucks. Journal of Sound and Vibration 215, Malchaire, J., Piette, A., Mullier, I., Vibration exposure on fork-lift trucks. Ann Occup Hyg. 40, Mansfield, N. J. and Griffin, M. J., Effects of posture and vibration magnitude on apparent mass and pelvis rotation during exposure to whole-body vertical vibration. Journal of Sound and Vibration 253, Marjanen, Y., Using ISO as an additional whole body vibration evaluation method with ISO to include also transient shocks to the analysis. In: Proceeding, 12 th International Congress on Sound and Vibration, July, Lisbon, Portugal. Mirbod, S. M., Yoshida, H., Jamali, M., Masamura, K., Inaba, R., Iwata, H., Assessment of hand-arm vibration exposure among traffic police motorcyclists. 23

25 International Archives of Occupational and Environmental Health 70, Newell, G.S. and Mansfield, N.J., Exploratory study of whole-body vibration artefacts experienced in a wheel loader, mini-excavator, car and office worker s chair. 39th United Kingdom Group Meeting on Human Responses to Vibration, September, held at Ludlow, Shropshire, England. Ochiai, A., Naya, Y., Soh, J., Ishida, Y., Ushijima, S., Mizutani, Y., Kawauchi, A., Miki, T., Do motorcyclists have erectile dysfunction? A preliminary study. International Journal of Impotence Research 18, Ozkaya, N., Willems, B. and Goldsheyder, D., Whole-body vibration exposure: a comprehensive field study. American Industrial Hygiene Association Journal 55, Palmer, K.T., Coggon, D.N., Bendall, H.E., Pannett, B., Griffin, M.J., Haward, B., Whole-body vibration: occupational exposures and their health effects in Great Britain. Health and Safety Executive Contract Research Report 233/1999, HSE Books, ISBN: Peitte, A. and Malchaire, J., Technical characteristics of overhead cranes influencing the vibration exposure of operators. Applied Ergonomics 22, Rehn B., Lundström R., Nilsson L., Liljelind I., Jörvholm B., 2005, Variation in exposure to whole-body vibration for operators of forwarder vehicles - aspects on measurement strategies and prevention. International Journal of Industrial Ergonomics. 35: Scarlett, A.J., Price, J.S., Stayner, R.M., Whole-body vibration: Evaluation of emission and exposure levels arising from agricultural tractors. Journal of Terramechanics 44, Seidel H., Selected health risks caused by long-term whole-body vibration. American Journal of Industrial Medicine 23, Taiwan MOTC, Number of registered motor vehicles in Taiwan-fuchien area. Taiwan Ministry of Transportation and Communications (MOTC), Taiwan IOSH, Development of software and field testing tools for assessing whole-body vibration with shocks. Taiwan Institute of Occupational Safety and Health (IOSH), Council of Labor Affairs, Taiwan. Contract Report No. 24

26 IOSH96-H318. (in Chinese) Troup, J., Clinical effects of shock and vibration on the spine. Clinical Biomechanics 3, Wang, W., Rakheja, S., Boileau, P.É., Effects of setting postures on biodynamic response of seated occupants under vertical vibration. International Journal of Industrial Ergonomies Wang, W., Rakheja, S., Boileau, P.E., The role of seat geometry and posture on the mechanical energy absorption characteristics of seated occupants under vertical vibration. International Journal of Industrial Ergonomies 36, Wasserman, D.E., Wilder, D.G., Pope, M.H., Magnusson, M., Aleksiev, A.R., Wasserman, J.F., Whole-body vibration exposure and occupational work-hardening. Journal of Occupational and Environmental Medicine 39, Wikström, B., Kjellberg, A., Landström, U., Health effect of long-term occupational exposure to whole-body vibration: a review. International Journal of Industrial Ergonomics 14, Appendix: Nomenclature a w, i frequency-weighted RMS acceleration of the ith exposure period, m/s 2 T i exposure duration of the ith exposure period, 30 sec RMS root mean square value of acceleration, m/s 2 RMS upper RMS boundary value of ISO (1997) health guidance caution zone, 0.8 m/s 2 VDV vibration dose value, m/s 1.75 VDV i VDV of the ith exposure period, m/s 1.75 VDV (8) predicted vibration dose value for 8-hour exposure, m/s 1.75 VDV upper boundary value of ISO (1997) health guidance caution zone, 15 m/s 1.75 D k,i acceleration dose of the ith exposure period in the k direction, m/s 2 25

27 D k acceleration dose in the k direction, m/s 2 T 8 T m S e S ed 8-h time period, 8 hours overall measurement period, hour equivalent static compression stress, MPa daily dose of equivalent static compression stress, MPa S ed value of S ed indicating high probability of an adverse health effect, 0.8 MPa T a allowable exposure duration for an RMS to reach RMS, an VDV to reach VDV, or an S e to reach S ed, hour EV exposure value of RMS, VDV, or S e VL value of RMS, VDV, or S ed R a non-dimensional factor defined in ISO (2004, Annex A) for use in the assessment of the adverse health effects related to the human response acceleration dose R R value indicating high probability of an adverse health effect, 1.2 N j n c S uj b the number of exposure days per year the year counter the number of years of exposure constant representing the static stress due to gravitational force the ultimate strength of the lumber spine for a person of age (b+j) years the age at which the exposure starts 26

28 Figure captions Figure 1. (a) Data loggers and (b) portable media recorders utilized for data collection Figure 2. Experimental setup of a participant Figure 3. (a) Data processing environment in Viewlog; (b) Screen shot of vibration analysis module Figure 4. VDV-S e scatter plot of all tasks, with the caution zones (grey bands) of ISO and ISO (hollow symbol: 20.6-km measured; solid symbol: 8-h predicted) Figure 5. Estimated periods of motorcycle riding needed to reach the individual exposure limits of RMS = 0.8 m/s 2, VDV = 17 m/s 1.75, and S ed = 0.8 MPa Figure 6. Prediction of motorcycle rider age when R=1.2 under speed limits of 55 and 40 km/h and estimated daily exposures of 1, 2, 4, and 8 hours. Figure 7. The VDV-S ed scatter plot of army vehicles (Alem, 2005), LHD mining vehicles (Eger et al., 2008), motorcycles (this study), and sedans (this study) with the caution zones (grey bands) of ISO and ISO Table captions Table 1. Characteristics of the participants and vehicles Table 2. Task condition of different roadways and speed limits Table 3. Mean (s.d.) of RMS, vibration dose value (VDV) and acceleration dose (D k ) of the 20.6 km road tests Table 4. Mean (s.d.) of RMS, estimated 8-h vibration dose value (VDV (8) ) and estimated 8-h static compression dose (S ed ) on different roadways and at speed limits 27

29 Table 1. Characteristics of the participants and vehicles Participant Vehicle Group Age Height Mass Experience Engine Wheel Age Manufacturer (yr) (cm) (kg) (yr) size(cc) size(in) (yr) (n) Scooter (n=6) 24.2 (1.6) (5.0) 67.2 (8.5) 5.7 (1.6) (3.7) Yamaha(3) Sanyang(3) Motorbike (front) 2.4 Sanyang(6) (n=6) (0.8) (3.8) (9.6) (1.8) 17(rear) (0.6) Sedan ~ Toyota(1) (n=5) (6.2) (4.6) (11.2) (6.8) (2.7) Nissan(2) Ford(1) Misubishi(1) 28

30 Table 2. Task condition of different roadways and speed limits Roadway A (rural/ 5km) B (province/ 5km) C (urban/ 10.6km) A+B+C (20.6km) Condition Speed limit (km/h) Duration* (min) 9.75 (1.64) (2.52) (0.81) (2.21) (3.61) (3.42) (5.01) (5.59) Average motorcycle speed* (km/h) (5.15) (4.49) (1.54) (2.51) (2.72) (2.04) (2.35) (1.70) Fully stops 3.1± ± ± ± ± ± ± ±2.7 * p<0.001, significant roadway difference by repeated measures ANOVA p<0.001, significant speed limit difference by repeated measures ANOVA p<0.05, significant roadway*speed limit effect by repeated measures ANOVA 29

31 Table 3. Mean (s.d.) of RMS, vibration dose value (VDV) and acceleration dose (D k ) of the 20.6 km road tests Exposure parameter Axis X Y Z* Speed limit (km/h) RMS (m/s 2 ) scooter (n=6) motorbike (n=6) 0.44 (0.04) 0.39 (0.04) 0.38 (0.03) 0.33 (0.03) 0.17 (0.04) 0.14 (0.08) 0.16 (0.03) 0.12 (0.02) 0.80 (0.12) 0.93 (0.14) 0.80 (0.15) 0.90 (0.15) Sedan (n=5) 0.15 (0.03) 0.11 (0.01) 0.30 (0.02) VDV (m/s 1.75 ) scooter (n=6) motorbike (n=6) 6.30 (0.60) 5.15 (0.39) 5.60 (0.61) 4.52 (0.57) 2.60 (1.02) 3.43 (2.55) 4.25 (2.57) 1.94 (0.59) (2.89) (2.83) (2.61) (1.82) sedan (n=5) 1.93 (0.27) 1.55 (0.11) 4.17 (0.42) scooter (n=6) (6.87) (10.86) 7.26 (4.78) (6.72) (10.11) (3.88) D k (m/s 2 ) motorbike (n=6) (13.00) (4.95) 8.80 (6.17) 6.37 (2.34) (6.68) (3.64) sedan (n=5) (0.31) (0.38) * Axis with the most severe acceleration level/ dose 4.99 (0.47) 30

32 Table 4. Mean (s.d.) of RMS, estimated 8-h vibration dose value (VDV (8) ) and estimated 8-h static compression dose (S ed ) on different roadways and at speed limits Roadway A (rural/ 5km) B (province/ 5km) C (urban/ 10.6km) A+B+C (20.6km) Exposure parameter Speed limit (km/h) RMS* (m/s 2 ) scooter (n=6) motorbike (n=6) 0.97 (0.15) 1.18 (0.20) 0.96 (0.18) 1.05 (0.22) 0.68 (0.11) 0.84 (0.14) 0.68 (0.12) 0.82 (0.22) 0.79 (0.13) 0.90 (0.13) 0.79 (0.16) 0.88 (0.13) 0.80 (0.12) 0.94 (0.14) 0.80 (0.15) 0.90 (0.15) Sedan (n=5) 0.40 (0.03) 0.25 (0.03) 0.27 (0.02) 0.30 (0.02) VDV (8) * (m/s 1.75 ) scooter (n=6) motorbike (n=6) (6.30) (5.96) (4.75) (3.63) (5.24) (4.99) (3.24) (4.52) (4.35) (4.54) (4.53) (2.84) (4.96) (4.89) (4.17) (3.18) sedan (n=5) 8.85 (0.97) 6.28 (0.93) 6.63 (0.55) 7.18 (0.71) S ed (MPa) scooter (n=6) motorbike (n=6) 1.32 (0.70) 1.16 (0.16) 0.91 (0.19) 0.97 (0.26) 1.12 (0.44) 1.15 (0.24) 0.86 (0.14) 0.80 (0.19) 1.26 (0.45) 1.29 (0.28) 1.11 (0.11) 1.00 (0.20) 1.32 (0.46) 1.27 (0.22) 1.06 (0.09) 1.00 (0.17) sedan (n=5) (0.02) (0.03) (0.01) * p<0.001, significant roadway difference by repeated measures ANOVA (motorcycle) p<0.05, significant speed limit difference by repeated measures ANOVA 0.24 (0.02) (motorcycle) p<0.001, significant difference between sedan (n=5) and motorcycles (n=12) by t-test 31

33 Figure 1. (a) Data loggers and (b) portable media recorders utilized for data collection 32

34 Figure 2. Experimental setup of a participant 33

35 (a) (b) Figure 3. (a) Data processing environment in Viewlog; (b) Screen shot of vibration analysis module 34