- Email: [email protected]
Physiological and psychophysical responses in handling maximum acceptable weights under different footwear–floor friction conditions
ARTICLE IN PRESS
Applied Ergonomics 38 (2007) 259–265 www.elsevier.com/locate/apergo
Physiological and psychophysical responses in handling maximum acceptable weights under different footwear–floor friction conditions Kai Way Lia,, Rui-feng Yub, Xiao L. Hanb a
Liberty Mutual Research Institute for Safety, Centre for Safety Research, 71 Frankland Road, Hopkinton, MA 01748, USA b Department of Industrial Engineering, Tsinghua University, Beijing, China Received 27 January 2006; accepted 12 June 2006
Abstract A study on combined manual materials-handling tasks performed on floors under three friction levels was conducted. Eight male subjects participated in the study. The maximum acceptable weight of handling, including lifting, carrying for 3 m, lowering, and walking 3 m back at twice per minute was determined. The subject then performed the same tasks for 10 min. Heart rate, Vo2, energy efficiency, perceived sense of slip, and rating of perceived exertion for whole body strain were measured. The results showed that the effects of friction level on the maximum acceptable weights of handling, perceived sense of slip, Vo2, and energy efficiency were statistically significant (pp0.0006). As the friction level increased from low to high, the maximum acceptable weights of handling increased from 8.15 to 9.34 kg. The energy efficiency on the low friction condition (12.58 kg/L/min) was significantly lower than those of the medium (15.73 kg/L/min) and high (15.38 kg/L/min) friction conditions. The perceived sense of slip was the highest (5.44) on the low-friction condition, followed by the medium-friction condition (3.58), and last the high-friction condition (1.84). The implication of this study was that friction level should be regarded as one of the major environmental factors in designing MMH tasks as it affected both physiological and psychophysical responses of the subjects. Low-friction footwear–floor interface should be avoided as it resulted in not only high scores of perceived sense of slip but also in low-energy efficiency utilized in the body. r 2006 Elsevier Ltd. All rights reserved. Keywords: Manual material handling; Slipping and falling; Footwear–floor slipperiness; Perceived sense of slip
1. Introduction Manual materials handling (MMH) tasks are very common in workplaces. One of the most widely accepted approaches in designing MMH tasks is to design the job so as not to exceed the capabilities of the materials handlers (Ciriello and Snook, 1999; Ciriello et al., 1999). Considerable effort has been expended in determining the factors affecting the capacities on human subjects in performing the MMH tasks. Floor slipperiness, or slip-resistance, is one of the major environmental factors affecting walking and materialshandling behaviors. Floor slipperiness may be quantified using the coefficient of friction (COF). Both static COF (SCOF) and dynamic COF (DCOF) have been used to Corresponding author. Tel.: +1 508 497 0212; fax: +1 508 435 8136.
E-mail address: [email protected] (K.W. Li). 0003-6870/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.apergo.2006.06.006
measure floor slipperiness. In the USA, a SCOF of 0.5 has been recommended as the slip-resistant standard for unloaded, normal walking conditions (Miller, 1983). Higher SCOF values may be required for safe walking when handling loads. In Europe, DCOF is widely used to indicate floor slipperiness. Gro¨nqvist (1995) suggested that a floor was ‘‘very slip-resistant’’ if the DCOF was 0.3 or more. A floor with a DCOF between 0.2 and 0.29 was ‘‘slip resistant’’. A floor was classified as ‘‘unsure’’ if its DCOF was between 0.15 and 0.19. A floor was ‘‘slippery’’ and ‘‘very slippery’’ if the DCOF was lower than 0.15 and 0.05, respectively. These classifications were established to quantify the risk associated with slipping and falling. In addition to objective measures of floor slipperiness, subjective assessments of floor slipperiness have also been discussed. Myung et al. (1993) compared subjective ranking of floor slipperiness with SCOF and found that the two measures were consistent. They concluded that
ARTICLE IN PRESS 260
K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
human subjects could discriminate floor slipperiness reliably. Swensen et al. (1992) conducted a study to collect subjective rating of surface slipperiness of steel beams with different coatings from ironworkers and college students. They found that correlation between subjective rating and measured COF were high for both ironworkers (r ¼ 0.75) and college students (r ¼ 0.90). Similar results were reported by Chiou et al. (2000), Chang et al. (2004), and Li et al. (2004). The MMH tasks may be performed on slippery or nonslippery floors. Ciriello et al. (2001) investigated maximum acceptable forces of cart pushing on high and low COF floors at a frequency of once per minute. They found that the maximum acceptable weights of the pushcart tasks on the low COF floor were significantly lower than those on the high COF floor. Haslam et al. (2002) performed an infrequent pushing and pulling experiment on floor surfaces with good and reduced slip resistance. Their results indicated that the maximum acceptable loads were independent of the floor surfaces used despite differences in ground reaction forces, and ratings of perceived slipperiness between the two floor conditions were significantly different. Psychophysical methodology has been used extensively to establish the maximum acceptable weights of lift (MAWL) (Snook, 1978; Garg and Badger, 1986; Ayoub and Mital, 1989; Snook and Ciriello, 1991). The psychophysical data on MAWL has been obtained in laboratory studies (Garg and Saxena, 1980; Snook and Ciriello, 1991; Mital, 1992; Lee et al., 1995; Wu and Chen, 2003). These data, however, were based on studies conducted on nonslippery floors. Floor slipperiness has been ignored in many studies concerning lifting and load carrying. For example, one of the assumptions for the NIOSH 1991 equation is that the COF between the shoe sole and the floor is 0.4 or more (Waters et al., 1993). When the COF value is below 0.4, usage of such an equation may be inappropriate. Another example was Chung and Wang’s (2001) study where human subjects were tested in simulated stair climbing with lifting and load carrying, in semi-conductor-manufacturing operations, to determine the maximum acceptable weight. They used wooden stairs in the experiment, neglecting the fact that all the stairs in those workplaces were steel. The frictions on steel are, in general, lower than those on the wooden surfaces. A few studies concerning carrying tasks were conducted on slippery floors. Li (1991) had male subjects carry loads up to 40% of their body weight and walk on a greased steel floor. He found that load carriage significantly (po0.05) affected some gait parameters such as heel velocity, foot stance time, slip distance, and the time from heel strike to slip-start. Significantly (po0.05) longer slip distances were found for heavier load carrying conditions. The study implied that load carriage increased the risk of slipping/ falling on slippery floor surfaces. Slips and falls were more likely to occur when heavier loads were carried.
Myung and Smith (1997) had subjects carry loads up to 40% of their body weight while walking on dry and oily floors. Both load carrying and floor slipperiness significantly (pp0.0001) affected heel landing velocity and stride length. The subjects reduced their stride length when they walked on oily floors. Their heel-landing velocities, on the other hand, were higher on oily floors as compared to dry floors. When carrying load, the subjects tended to reduce their stride lengths so as to transfer their centers of mass faster. This resulted in higher landing velocity of the heel. It is known that the subjects changed their walking patterns when they approached and walked on a slippery floor (Li, 1991; Cham and Redfern, 2002). These changes might affect their capabilities in handling weights. Neither Li (1991) nor Myung and Smith (1997) investigated human capabilities in lifting and load carrying. How the friction at the foot–floor interface affects the burden on the materials handler’s body has not been reported. The aims of the study were to determine the maximal acceptable weight of handling (MAWH) under different footwear–floor slipperiness conditions; and to investigate the physiological and psychophysical responses for subjects who handle materials under different footwear–floor slipperiness conditions. 2. Method 2.1. Subjects Eight males, recruited from the campus at Tsinghua University in Beijing, China, participated in this study as human subjects. They were free from any musculoskeletal disorders. Their mean (7standard deviation) age, height, body weight, and resting heart rate, were 21.6 years (71.2), 172.2 cm (75.4), 65.0 kg (76.8), and 69.6 beats/min (78.1), respectively. 2.2. Footwear and floor environments A walkway of 2 m long, covered with vinyl, was prepared for the experiment. Before each trial, the floor was washed, dried, and waxed manually. The subjects were required to wear the same type of lab shoes with one of the three sole materials: cloth, plastic, or rubber. The footwear and the waxed vinyl floor combinations provided low, medium, and high-friction levels at the footwear–floor interface. A FALEXs 6 Friction and Wear Test Machine was used to measure both the static and dynamic friction coefficients of the three footwear–floor combinations. This machine applied a circular footwear sample, with a diameter of approximately 5 cm, on the floor with a normal force of 490 N with a rotary speed of 3 rpm. The SCOF readings for the cloth, plastic, and rubber soles on the floor were 0.19, 0.43, and 0.89, respectively. The DCOF readings for the three sole material–floor combinations were 0.16, 0.39, and 0.81.
ARTICLE IN PRESS K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
2.3. MMH tasks A plastic container (44 33 25 cm) with handles commonly used in local industry was used. Iron blocks of 2.3, 1.4, and 0.6 kg and bottles of water, each weighting 0.1 kg, were prepared as the materials to be handled. The subject was required to lift the container with a certain amount of weight, from a table 30 cm in height, to his elbow height and carry it for 3 m, then lower the container onto a table of the same height. The subject then walked back 3 m, empty handed, to the original starting position. Due to the limitation of the metabolic cart, the subjects walked 1.5 m round trip to and from the table to reach a 3 m walking distance. In other words, one task included a lifting, carrying for 3 m, lowering, and walking empty handed for 3 m. The frequency of this task was 2 per minute. The walking speeds of the subjects were not controlled. However, they were required to complete two tasks each minute and were instructed to maintain consistent working and walking pace during the experiment. A safety harness was used to prevent the subject from falling. A cushion was attached to the bottom of the
261
container to protect the subject’s feet in the event that he dropped the container during the trial. Elbow and kneepads were also used as a safety precaution (see Fig. 1). The temperature and relative humidity when performing the MMH tasks were approximately 25 1C and 30%, respectively. 2.4. Procedure All subjects signed an informed consent for their participation in the study. Each subject attended three sessions on three different days. In each session, the subject was wearing shoes with one of the sole materials. The subject was informed that a slip was likely when walking on the walkway with the shoes, but the safety harness would prevent a fall. The subjects were instructed to refrain from heavy physical activity before attending the experiment. Before the experiment, the resting heart rate of the subject was measured and the researcher explained the purposes and procedure of the study to the subject. Each session started with the determination of the MAWH. The MAWH was the terminology used to indicate the maximum acceptable weight that a subject could handle
Fig. 1. The MMH tasks: (a) lifting; (b) holding; (c) carrying; and (d) lowering.
ARTICLE IN PRESS 262
K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
under the lifting, carrying, lowering, and walking back scenario described in the previous section. There was a 5-min warm-up period prior to the beginning of each session. During this period, the subject performed light physical exercise including movements of adductions and abductions of the extremities, flexions, extensions, and rotations of the trunk and neck. After the warm-up period, the subject started to handle an initial weight following the described protocol in the previous section. The initial weight was in the range of 8–10 kg. This range was determined in a pilot study in which a research personnel conducted the tasks under the experimental conditions repeatedly. The minimum and maximum weights obtained were used as the lower and upper limits of the initial weight, respectively. The purpose of determining such a range was to make the MAWH adjustment process more efficient. In each MAWH determination trial, an initial weight was randomly assigned using a weight either near the lower limit or near the upper limit. Approximately, half of the trials started at a weight near the lower limit while the other half started at a weight near the upper limit. The subject was instructed to handle the weight as if working on an incentive basis for an 8-h work-shift, working as hard as he could without straining himself, or becoming unusually tired, weakened, overheated or out of breath (Snook, 1985). He was encouraged to adjust the weight by adding or removing iron or bottles during the trials to approximate his MAWH. The entire weight adjustment took about 20 min. At the end of the period, the weight determined was recorded as his MAWH. The subject took a 5-min break upon completion of the MAWH determination. He then performed the same MMH task at his MAWH for 10 min after the break (see Fig. 1). During this period, heart rate was measured using a POLARs vantage XL HR monitor and Vo2 was measured using a PHYSIO-DYNEs MAX II metabolic cart. The means of his heart rate and Vo2 during the last 5 min of the test period were used for statistical analysis. Upon completion of this period, the subject immediately reported his rating of perceived exertion (RPE) (Borg, 1985) of whole body strain ranging from 6 ‘‘no exertion at all’’ to 20 ‘‘maximal exertion’’ and his perceived sense of slip (PSOS) (Chiou et al., 2000) while handling the loads (see Table 1). The PSOS was determined by adding the four ratings in Table 1. A high PSOS score implies a high subjective perception of slip and loss of balance during the MMH tasks. 2.5. Experiment design and data analysis The experiment was a single factor (friction level) experiment. The order of the three friction levels was randomly arranged within each subject. Analysis of variance (ANOVA) was performed for the heart rate, Vo2, MAWH, RPE, and PSOS. Tukey’s honestly significant difference (HSD) test was performed if a variable was
Table 1 Perceived sense of slip (PSOS) rating scale (Chiou et al., 2000) (1) How much did you feel yourself slip (i.e. loss of foot traction)? A little Some A lot 0 0.5 1 1.5 2 (2) Did you have any difficulty in maintaining balance (how much did you or your muscles compensate for your movement)? A little Some A lot 0 0.5 1 1.5 2 (3) Did you feel at any time that you would slip? A little Some 0 0.5 1 1.5
A lot 2
(4) What would you say was the overall difficulty of this task? A little Some A lot 0 0.5 1 1.5 2
statistically significant at po0.05. Pearson’s correlation coefficients between the dependent variables with significant results were calculated. Statistical analysis was performed using the SPSSs 10.0 software. 3. Results Table 2 shows the ANOVA results of the dependent variables. Tukey’s HSD tests were performed for MAWH, PSOS, and Vo2. The results are shown in Tables 3–5. The MAWH, which ranged from 7.8–10.7 kg, was significantly (p ¼ 0.0006) affected by friction level. The MAWH at highfriction level was significantly higher than that of the lowfriction level. The PSOS was significantly (po0.0001) affected by friction level. The PSOS at low-friction level was significantly higher than those of the medium and high-friction ones. The PSOS at medium-friction level was also significantly higher than that of the high-friction level. The mean heart rates for the low-, medium-, and highfriction conditions were 96.3, 97.1, and 97.4 beats/min, respectively. This corresponded to heart rate increases of 26.8%, 27.4%, and 26.1% as compared to the resting heart rates. The differences among these heart rates were not statistically significant. The friction level affected Vo2 significantly at po0.0001. The Vo2 at the medium friction level was significantly lower than those of the low- and high-friction levels. The Vo2 at the high-friction level was significantly lower than that of the low-friction level. To quantify the efficiency of metabolic energy consumed for the materials-handling tasks, the term ‘‘energy efficiency’’ was defined as the amount of weight handled per unit oxygen consumed per minute. It was calculated by dividing the MAWH by the Vo2. The ANOVA results indicated that friction level affected energy efficiency significantly (po0.0001). Table 6 showed Tukey’s HSD test results for energy efficiency. Pearson’s correlation coefficients between PSOS and energy efficiency and between PSOS and MAWH were 0.720 (po0.0001) and 0.611 (p ¼ 0.0015), respectively. Pearson’s correlation coefficients between Borg’s RPE
ARTICLE IN PRESS K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
263
Table 2 ANOVA results of the dependent variables
MAWH
Vo2
HR
RPE
PSOS
Source of variation
Sum of squares
Df
Mean square
F
p-value
Between Between Error Between Between Error Between Between Error Between Between Error Between Between Error
5.60 4.16 2.95 32,100.09 96,049.45 4604.68 5.75 2844.94 643.82 3.25 14.50 14.75 51.85 13.98 16.34
2 7 14 2 7 14 2 7 14 2 7 14 2 7 14
2.80 0.59 0.21 16,050.05 13,721.35 328.90 2.87 406.42 45.99 1.63 2.07 1.05 25.93 1.99 1.17
13.30 2.82
0.0006 0.0467
48.80 41.72
o0.0001 o0.0001
0.06 8.84
0.9396 0.0003
1.54 1.97
0.2481 0.1334
22.21 1.71
o0.0001 0.1856
friction subject friction subject friction subject friction subject friction subject
Table 3 Tukey’s HSD test results for MAWH
Table 6 Tukey’s HSD test results for energy efficiency
Friction level
MAWH (kg)
Tukey’s groupa
Friction level
Energy efficiency (kg/L/min)
Tukey’s groupa
Low Medium High
8.15 8.74 9.34
A AB B
Low Medium High
12.58 15.73 15.38
A B B
a
Number in each cells were the means for eight subjects; Different letters in Tukey’s group showing they were significantly different at po0.05. Table 4 Tukey’s HSD test results for PSOS a
Friction level
PSOS
Tukey’s group
Low Medium High
5.44 3.58 1.84
A B C
a Number in each cells were the means for eight subjects; Different letters in Tukey’s group showing they were significantly different at po0.05.
Table 5 Tukey’s HSD test results for Vo2 Friction level
Vo2 (L/min)
Tukey’s groupa
Low Medium High
0.651 0.562 0.613
A B C
a Number in each cells were the means for eight subjects; Different letters in Tukey’s group showing they were significantly different at po0.05.
and energy efficiency and between MAWH and energy efficiency were 0.574 (p ¼ 0.0033) and 0.424 (p ¼ 0.0388), respectively. 4. Discussion The psychophysical approach measures the perceived stress in MMH tasks. It is based on the assumption that
a
Number in each cells were the means for eight subjects; Different letters in Tukey’s group showing they were significantly different at po0.05.
both biomechanical and physiological stresses are present when people are handling materials, and these stresses may be combined under the measure of perceived stress (Ayoub and Mital, 1989). The maximum acceptable weights are the maximal loads that will not result in over-exertion or excessive fatigue for the workers. Ciriello et al. (2001) found that the maximum acceptable weights in pushing selected by subjects on low friction floors were most likely limited by the inadequate friction rather than the physical limitations. Miller (1983) mentioned that the required COF to prevent slipping is highly task-dependent. He indicated that a SCOF higher than 0.5 was required for load handling tasks. This implied that both the medium and low friction conditions provided inadequate slip-resistance for the subject’s materials-handling tasks. One of the major hypotheses of the present study was that the friction level affects the subject’s capability in the combined MMH tasks. This hypothesis was supported by the results of the study. In addition to the physical burden, floors with lower, inadequate friction levels apparently placed limitations on the subjects in the selection of their MAWH. It was believed that the subjects could differentiate friction level of the footwear–floor interface at the beginning or after a few MMH trials. The perception of floor slipperiness might affect their decision in determining their MAWH. The results of the Borg–RPE rating were not significant indicating that the subjects perceived no difference in their
ARTICLE IN PRESS 264
K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
whole body strain in the MMH tasks under the three friction conditions. The RPE results, ranging from 11 (or ‘‘light’’) to 16 (between ‘‘hard’’ and ‘‘very hard’’), indicated that different subjects would work at different RPE levels. This was consistent with the findings in the literature (Garg and Saxena, 1982; Garg and Banaag, 1988). The mode of RPE was 13, which corresponded to ‘‘somewhat hard’’. Two-thirds of the trials had a RPE less than or equal to 13. As the MAWH was determined by the subject, this implies that most of the subjects selected a weight that they felt was ‘‘somewhat hard’’ or lighter when they were performing the MMH tasks for an 8-h shift. This was consistent with the findings in the literature (Wang et al., 2000; Wu and Chen, 2003). The Vo2 was significantly (po0.0001) affected by friction level. The Vo2 at the medium friction condition was significantly lower than those of the low- and high-friction conditions. This implied the subjects were experiencing higher physiological strain when they handled loads on either low- or high-friction floors as compared to that of the medium one. It should be noted that the effects of friction level on Vo2 was confounded with workload as the MAWH were different among the friction levels. However, the trends between the MAWH (see Table 3) and the Vo2 (see Table 5) under the friction levels were not consistent. In Table 3, the MAWH increased as the friction level increased from low to high. But the Vo2 in Table 5 showed that the medium friction level resulted in a lowest Vo2 level among the three friction levels. It was, therefore, concluded that friction level did affect the physiological strains under the experimental conditions of the study. Chiou et al. (2000) indicated that the subjects experienced a high postural instability when the PSOS were high. The extra physiological burden on low friction floor might be attributed to the different walking and materials handling strategies adopted in order to avoid a fall. The literature (Li, 1991; Myung and Smith, 1997; Cham and Redfern, 2001) has shown that the subjects adapted to the slippery floor by increasing the knee flexion and hip extension moments and the co-contraction of the leg muscles so as to maintain balance. An additional muscular effort and physiological cost was required for such a gait strategy. This was supported by our results of energy efficiency where the low-friction condition had significantly lower value (12.58 kg/L/min) as compared to those of the medium (15.73 kg/L/min) and high-friction (15.38 kg/L/ min) conditions. The mean heart rates for the three friction levels (97.4 beats/min or lower) were lower than those reported by Garg and Banaag (1988) (119–126 beats/min) indicating that the subjects selected a workload that was lower than those in the literature. Chiou et al. (2000) suggested that the PSOS be used as a predictor of both slip occurrence and postural instability. In the present study, the subjects reported their PSOS after each trial. The PSOS scores ranged from 0.5–7.5 with means of 5.44, 3.59, and 1.84 for the low, medium, and
high-friction levels, respectively. They were significantly (po0.0001) different among one another. This was consistent with the findings of Chiou et al. (2000). Chiou et al. (2000) indicated that the percentage of slip occurrence increased from 0% to 100% when a PSOS increased from 0 to 4.5. A slip always occurred when the PSOS exceeded 4.5. In the present study, seven out of eight subjects reported a PSOS higher than 4.5 on the low-friction condition. Two out of eight subjects reported a PSOS of 4.5 or higher on medium friction condition. All the subjects reported a PSOS of 3.5 or lower on high friction condition. The PSOS results correlated well with the friction levels. According to Chiou et al. (2000), the subjects were not only more likely to slip but also experienced more postural instability at high PSOS levels. According to Gro¨nqvist (1995), the slip-resistance of the low-friction condition (with a DCOF of 0.16) in the present study was ‘‘unsure’’. The slip-resistances of the medium and high-friction conditions (both had a DCOF higher than 0.3) were both ‘‘very slip-resistant’’. In our study, there were two subjects reported a PSOS of 4.5 on medium friction condition and one subject reported a 5 on high friction condition. This implied that Gro¨nqvist’s (1995) categories were not consistent with our PSOS results. The discrepancy between the two might be attributed to the fact that Gro¨nqvist’s (1995) categories considered only the risk of slipping while the PSOS considered the effects of postural instability in performing the task as an addition. Gro¨nqvist’s (1995) categories, however, supported our results in the physiological burden where the energy efficiency was significantly lower (12.58 kg/L/min) on the ‘‘unsure’’ condition as compared to those of the ‘‘very slipresistant’’ conditions (15.73 and 15.38 kg/L/min). In other words, the amount of weight handled per unit oxygen consumed on the ‘‘unsure’’ slip-resistance condition was significantly lower than those on the ‘‘very slip-resistant’’ conditions. There were limitations to this study. First of all, the strategies for a walking pattern when carrying a load that the subjects might adopt could be very complicated, especially on low-friction floors. The subjects were required to maintain a consistent working and walking pace during the experiment. However, their walking speeds and gait patterns were not measured or controlled. It was possible that they reduced their stride length and increased cadence (or vice versa) at the same time to maintain the same walking speed in different trials. This type of gait adaptation could affect the experimental results. Secondly, whether the subject slipped or not was not recorded and percentage of slip occurrence was not measured. Thirdly, only one lifting height (55 cm from floor), one carrying distance (3 m), and one frequency (2 per min) were tested. Data on more experimental conditions are required in the future to provide more information in understanding the interactions between friction levels and the MAWH and the physiological responses.
ARTICLE IN PRESS K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
5. Conclusion This study investigated the effects of the friction level at footwear–floor interface on the MAWH, heart rate, Vo2, energy efficiency, PSOS, and RPE. The results indicated that friction level was a significant factor affecting the MAWH, Vo2, energy efficiency, and PSOS. When working on low-friction floor surfaces, the subjects selected less weight as compared to that on the high friction floors. The Vo2 values in handling the weight at the MAWH level for ten minutes at the low- and high-friction conditions were significantly higher than that on the medium-friction condition. The energy efficiency at the low-friction condition was significantly lower than those on the medium- and high-friction conditions. The PSOS in the MMH tasks on the low-friction condition was the highest, next being the medium-friction condition, and last the high-friction condition. The results of this study showed that floorfriction level should be regarded as one of the major environmental factors in designing similar combined MMH tasks as it affected both the physiological and perceptual responses of the subjects. Acknowledgments The authors thank Ray McGorry of Liberty Mutual Research Institute for Safety for his thoughtful comments of the earlier drafts of the manuscript. References Ayoub, M.M., Mital, A., 1989. Manual Materials Handling. Taylor & Francis, London. Borg, G., 1985. An Introduction to Borg’s RPE Scale. Movement Publications, Ithaca, NY. Cham, R., Redfern, M.S., 2001. Lower extremity corrective reactions to slip events. J. Biomech. 34, 1439–1445. Cham, R., Redfern, M.S., 2002. Heel contact dynamics during slip events on level and inclined surfaces. Saf. Sci. 40, 559–576. Chang, W.R., Li, K.W., Huang, Y.H., Filiaggi, A., Courtney, T.K., 2004. Assessing floor slipperiness in fast-food restaurants in Taiwan using objective and subjective measures. Appl. Ergon. 35, 401–408. Chiou, S.-Y., Bhattacharya, A., Succop, P.A., 2000. Evaluation of Workers’ perceived sense of slip and effect of prior knowledge of slipperiness during task performance on slippery surfaces. Am. Ind. Hyg. Assoc. J. 61, 492–500. Chung, H.C., Wang, M.J., 2001. The effects of container design and stair climbing on maximal acceptable lift weight, wrist posture, psychophysical, and physiological responses in wafer-handling tasks. Appl. Ergon. 32, 593–598. Ciriello, V.M., Snook, S.H., 1999. Survey of manual handling tasks. Int. J. Ind. Ergon. 23, 149–156.
265
Ciriello, V.M., McGorry, R.W., Martin, S.E., 2001. Maximum acceptable horizontal and vertical forces of dynamic pushing on high and low coefficient of friction floors. Int. J. Ind. Ergon. 27, 1–8. Ciriello, V.M., Snook, S.H., Hashemi, L., Cotnam, J., 1999. Distributions of manual material handling task parameters. Int. J. Ind. Ergon. 24, 379–388. Garg, A., Saxena, U., 1980. Container characteristics and maximum acceptable weight of lift. Hum. Factors 22, 487–495. Garg, A., Saxena, U., 1982. Maximum frequency acceptable to female workers for one-handed lifts in the horizontal plane. Ergonomics 25 (9), 839–853. Garg, A., Badger, D., 1986. Maximum acceptable weights and maximum voluntary isometric strength for asymmetric lifting. Ergonomics 29 (7), 879–892. Garg, A., Banaag, J., 1988. Maximum acceptable weights, heart rates and RPEs for one hour’s repetitive asymmetric lifting. Ergonomics 31 (1), 77–96. Gro¨nqvist, R., 1995. Mechanisms of friction and assessment of slip resistance of new and used footwear soles on contaminated floors. Ergonomics 38, 224–241. Haslam, R.A., Boocock, M., Lemon, P., Thorpe, S., 2002. Maximum acceptable loads for pushing and pulling on floor surfaces with good and reduced resistance to slipping. Saf. Sci. 40, 625–637. Lee, Y.H., Wu, S.P., Hsu, S.H., 1995. The psychophysical lifting capacities of Chinese subjects. Ergonomics 38, 671–683. Li, K.W., 1991. A biomechanical study of slipping and falling with load carriage, unpublished doctoral dissertation, Department of industrial engineering. Texas Tech University, Lubbock, TX. Li, K.W., Chang, W.R., Leamon, T.B., Chen, C.J., 2004. Floor slipperiness measurement: friction coefficient, roughness of floors, and subjective perception under spillage conditions. Saf. Sci. 42, 547–565. Miller, J.M., 1983. ‘‘Slippery’’ work surface: toward a performance definition and quantitative coefficient of friction criteria. J. Saf. Res. 14, 145–158. Mital, A., 1992. Psychophysical capacity of industrial workers for lifting symmetrical and asymmetrical loads symmetrically and asymmetrically for 8 h work shifts. Ergonomics 35, 745–754. Myung, R., Smith, J.L., 1997. The effect of load carrying and floor contaminants on slip and fall parameters. Ergonomics 40 (2), 235–246. Myung, R., Smith, J.L., Leamon, T.B., 1993. Subjective assessment of floor slipperiness. Int. J. Ind. Ergon. 11, 313–319. Snook, S.H., 1978. The design of manual handling tasks. Ergonomics 21, 963–985. Snook, S.H., 1985. Psychophysical considerations in permissible loads. Ergonomics 28, 327–330. Snook, S.H., Ciriello, V.M., 1991. The design of manual tasks: revised tables of maximum acceptable weights and forces. Ergonomics 34, 28–34. Swensen, E.E., Purswell, J.L., Schlegel, R.E., Stanevich, R.L., 1992. Coefficient of friction and subjective assessment of slippery work surface. Hum. Factors 34 (1), 67–77. Wang, M.J., Chung, H.C., Chen, H.C., 2000. The effect of handle angle on MAWL, wrist, posture, RPE, and heart rate. Hum. Factors 42 (4), 553–565. Waters, T.R., Putz-Anderson, V., Garg, A., Fine, L., 1993. Revised NIOSH equation for the design and evaluation of manual lifting tasks. Ergonomics 36 (7), 749–776. Wu, S.P., Chen, J.P., 2003. Effects of the adjustment period on psychophysically determined maximum acceptable weight of lift and the physiological cost. Int. J. Ind. Ergon. 31, 287–294.
Applied Ergonomics 38 (2007) 259–265 www.elsevier.com/locate/apergo
Physiological and psychophysical responses in handling maximum acceptable weights under different footwear–floor friction conditions Kai Way Lia,, Rui-feng Yub, Xiao L. Hanb a
Liberty Mutual Research Institute for Safety, Centre for Safety Research, 71 Frankland Road, Hopkinton, MA 01748, USA b Department of Industrial Engineering, Tsinghua University, Beijing, China Received 27 January 2006; accepted 12 June 2006
Abstract A study on combined manual materials-handling tasks performed on floors under three friction levels was conducted. Eight male subjects participated in the study. The maximum acceptable weight of handling, including lifting, carrying for 3 m, lowering, and walking 3 m back at twice per minute was determined. The subject then performed the same tasks for 10 min. Heart rate, Vo2, energy efficiency, perceived sense of slip, and rating of perceived exertion for whole body strain were measured. The results showed that the effects of friction level on the maximum acceptable weights of handling, perceived sense of slip, Vo2, and energy efficiency were statistically significant (pp0.0006). As the friction level increased from low to high, the maximum acceptable weights of handling increased from 8.15 to 9.34 kg. The energy efficiency on the low friction condition (12.58 kg/L/min) was significantly lower than those of the medium (15.73 kg/L/min) and high (15.38 kg/L/min) friction conditions. The perceived sense of slip was the highest (5.44) on the low-friction condition, followed by the medium-friction condition (3.58), and last the high-friction condition (1.84). The implication of this study was that friction level should be regarded as one of the major environmental factors in designing MMH tasks as it affected both physiological and psychophysical responses of the subjects. Low-friction footwear–floor interface should be avoided as it resulted in not only high scores of perceived sense of slip but also in low-energy efficiency utilized in the body. r 2006 Elsevier Ltd. All rights reserved. Keywords: Manual material handling; Slipping and falling; Footwear–floor slipperiness; Perceived sense of slip
1. Introduction Manual materials handling (MMH) tasks are very common in workplaces. One of the most widely accepted approaches in designing MMH tasks is to design the job so as not to exceed the capabilities of the materials handlers (Ciriello and Snook, 1999; Ciriello et al., 1999). Considerable effort has been expended in determining the factors affecting the capacities on human subjects in performing the MMH tasks. Floor slipperiness, or slip-resistance, is one of the major environmental factors affecting walking and materialshandling behaviors. Floor slipperiness may be quantified using the coefficient of friction (COF). Both static COF (SCOF) and dynamic COF (DCOF) have been used to Corresponding author. Tel.: +1 508 497 0212; fax: +1 508 435 8136.
E-mail address: [email protected] (K.W. Li). 0003-6870/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.apergo.2006.06.006
measure floor slipperiness. In the USA, a SCOF of 0.5 has been recommended as the slip-resistant standard for unloaded, normal walking conditions (Miller, 1983). Higher SCOF values may be required for safe walking when handling loads. In Europe, DCOF is widely used to indicate floor slipperiness. Gro¨nqvist (1995) suggested that a floor was ‘‘very slip-resistant’’ if the DCOF was 0.3 or more. A floor with a DCOF between 0.2 and 0.29 was ‘‘slip resistant’’. A floor was classified as ‘‘unsure’’ if its DCOF was between 0.15 and 0.19. A floor was ‘‘slippery’’ and ‘‘very slippery’’ if the DCOF was lower than 0.15 and 0.05, respectively. These classifications were established to quantify the risk associated with slipping and falling. In addition to objective measures of floor slipperiness, subjective assessments of floor slipperiness have also been discussed. Myung et al. (1993) compared subjective ranking of floor slipperiness with SCOF and found that the two measures were consistent. They concluded that
ARTICLE IN PRESS 260
K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
human subjects could discriminate floor slipperiness reliably. Swensen et al. (1992) conducted a study to collect subjective rating of surface slipperiness of steel beams with different coatings from ironworkers and college students. They found that correlation between subjective rating and measured COF were high for both ironworkers (r ¼ 0.75) and college students (r ¼ 0.90). Similar results were reported by Chiou et al. (2000), Chang et al. (2004), and Li et al. (2004). The MMH tasks may be performed on slippery or nonslippery floors. Ciriello et al. (2001) investigated maximum acceptable forces of cart pushing on high and low COF floors at a frequency of once per minute. They found that the maximum acceptable weights of the pushcart tasks on the low COF floor were significantly lower than those on the high COF floor. Haslam et al. (2002) performed an infrequent pushing and pulling experiment on floor surfaces with good and reduced slip resistance. Their results indicated that the maximum acceptable loads were independent of the floor surfaces used despite differences in ground reaction forces, and ratings of perceived slipperiness between the two floor conditions were significantly different. Psychophysical methodology has been used extensively to establish the maximum acceptable weights of lift (MAWL) (Snook, 1978; Garg and Badger, 1986; Ayoub and Mital, 1989; Snook and Ciriello, 1991). The psychophysical data on MAWL has been obtained in laboratory studies (Garg and Saxena, 1980; Snook and Ciriello, 1991; Mital, 1992; Lee et al., 1995; Wu and Chen, 2003). These data, however, were based on studies conducted on nonslippery floors. Floor slipperiness has been ignored in many studies concerning lifting and load carrying. For example, one of the assumptions for the NIOSH 1991 equation is that the COF between the shoe sole and the floor is 0.4 or more (Waters et al., 1993). When the COF value is below 0.4, usage of such an equation may be inappropriate. Another example was Chung and Wang’s (2001) study where human subjects were tested in simulated stair climbing with lifting and load carrying, in semi-conductor-manufacturing operations, to determine the maximum acceptable weight. They used wooden stairs in the experiment, neglecting the fact that all the stairs in those workplaces were steel. The frictions on steel are, in general, lower than those on the wooden surfaces. A few studies concerning carrying tasks were conducted on slippery floors. Li (1991) had male subjects carry loads up to 40% of their body weight and walk on a greased steel floor. He found that load carriage significantly (po0.05) affected some gait parameters such as heel velocity, foot stance time, slip distance, and the time from heel strike to slip-start. Significantly (po0.05) longer slip distances were found for heavier load carrying conditions. The study implied that load carriage increased the risk of slipping/ falling on slippery floor surfaces. Slips and falls were more likely to occur when heavier loads were carried.
Myung and Smith (1997) had subjects carry loads up to 40% of their body weight while walking on dry and oily floors. Both load carrying and floor slipperiness significantly (pp0.0001) affected heel landing velocity and stride length. The subjects reduced their stride length when they walked on oily floors. Their heel-landing velocities, on the other hand, were higher on oily floors as compared to dry floors. When carrying load, the subjects tended to reduce their stride lengths so as to transfer their centers of mass faster. This resulted in higher landing velocity of the heel. It is known that the subjects changed their walking patterns when they approached and walked on a slippery floor (Li, 1991; Cham and Redfern, 2002). These changes might affect their capabilities in handling weights. Neither Li (1991) nor Myung and Smith (1997) investigated human capabilities in lifting and load carrying. How the friction at the foot–floor interface affects the burden on the materials handler’s body has not been reported. The aims of the study were to determine the maximal acceptable weight of handling (MAWH) under different footwear–floor slipperiness conditions; and to investigate the physiological and psychophysical responses for subjects who handle materials under different footwear–floor slipperiness conditions. 2. Method 2.1. Subjects Eight males, recruited from the campus at Tsinghua University in Beijing, China, participated in this study as human subjects. They were free from any musculoskeletal disorders. Their mean (7standard deviation) age, height, body weight, and resting heart rate, were 21.6 years (71.2), 172.2 cm (75.4), 65.0 kg (76.8), and 69.6 beats/min (78.1), respectively. 2.2. Footwear and floor environments A walkway of 2 m long, covered with vinyl, was prepared for the experiment. Before each trial, the floor was washed, dried, and waxed manually. The subjects were required to wear the same type of lab shoes with one of the three sole materials: cloth, plastic, or rubber. The footwear and the waxed vinyl floor combinations provided low, medium, and high-friction levels at the footwear–floor interface. A FALEXs 6 Friction and Wear Test Machine was used to measure both the static and dynamic friction coefficients of the three footwear–floor combinations. This machine applied a circular footwear sample, with a diameter of approximately 5 cm, on the floor with a normal force of 490 N with a rotary speed of 3 rpm. The SCOF readings for the cloth, plastic, and rubber soles on the floor were 0.19, 0.43, and 0.89, respectively. The DCOF readings for the three sole material–floor combinations were 0.16, 0.39, and 0.81.
ARTICLE IN PRESS K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
2.3. MMH tasks A plastic container (44 33 25 cm) with handles commonly used in local industry was used. Iron blocks of 2.3, 1.4, and 0.6 kg and bottles of water, each weighting 0.1 kg, were prepared as the materials to be handled. The subject was required to lift the container with a certain amount of weight, from a table 30 cm in height, to his elbow height and carry it for 3 m, then lower the container onto a table of the same height. The subject then walked back 3 m, empty handed, to the original starting position. Due to the limitation of the metabolic cart, the subjects walked 1.5 m round trip to and from the table to reach a 3 m walking distance. In other words, one task included a lifting, carrying for 3 m, lowering, and walking empty handed for 3 m. The frequency of this task was 2 per minute. The walking speeds of the subjects were not controlled. However, they were required to complete two tasks each minute and were instructed to maintain consistent working and walking pace during the experiment. A safety harness was used to prevent the subject from falling. A cushion was attached to the bottom of the
261
container to protect the subject’s feet in the event that he dropped the container during the trial. Elbow and kneepads were also used as a safety precaution (see Fig. 1). The temperature and relative humidity when performing the MMH tasks were approximately 25 1C and 30%, respectively. 2.4. Procedure All subjects signed an informed consent for their participation in the study. Each subject attended three sessions on three different days. In each session, the subject was wearing shoes with one of the sole materials. The subject was informed that a slip was likely when walking on the walkway with the shoes, but the safety harness would prevent a fall. The subjects were instructed to refrain from heavy physical activity before attending the experiment. Before the experiment, the resting heart rate of the subject was measured and the researcher explained the purposes and procedure of the study to the subject. Each session started with the determination of the MAWH. The MAWH was the terminology used to indicate the maximum acceptable weight that a subject could handle
Fig. 1. The MMH tasks: (a) lifting; (b) holding; (c) carrying; and (d) lowering.
ARTICLE IN PRESS 262
K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
under the lifting, carrying, lowering, and walking back scenario described in the previous section. There was a 5-min warm-up period prior to the beginning of each session. During this period, the subject performed light physical exercise including movements of adductions and abductions of the extremities, flexions, extensions, and rotations of the trunk and neck. After the warm-up period, the subject started to handle an initial weight following the described protocol in the previous section. The initial weight was in the range of 8–10 kg. This range was determined in a pilot study in which a research personnel conducted the tasks under the experimental conditions repeatedly. The minimum and maximum weights obtained were used as the lower and upper limits of the initial weight, respectively. The purpose of determining such a range was to make the MAWH adjustment process more efficient. In each MAWH determination trial, an initial weight was randomly assigned using a weight either near the lower limit or near the upper limit. Approximately, half of the trials started at a weight near the lower limit while the other half started at a weight near the upper limit. The subject was instructed to handle the weight as if working on an incentive basis for an 8-h work-shift, working as hard as he could without straining himself, or becoming unusually tired, weakened, overheated or out of breath (Snook, 1985). He was encouraged to adjust the weight by adding or removing iron or bottles during the trials to approximate his MAWH. The entire weight adjustment took about 20 min. At the end of the period, the weight determined was recorded as his MAWH. The subject took a 5-min break upon completion of the MAWH determination. He then performed the same MMH task at his MAWH for 10 min after the break (see Fig. 1). During this period, heart rate was measured using a POLARs vantage XL HR monitor and Vo2 was measured using a PHYSIO-DYNEs MAX II metabolic cart. The means of his heart rate and Vo2 during the last 5 min of the test period were used for statistical analysis. Upon completion of this period, the subject immediately reported his rating of perceived exertion (RPE) (Borg, 1985) of whole body strain ranging from 6 ‘‘no exertion at all’’ to 20 ‘‘maximal exertion’’ and his perceived sense of slip (PSOS) (Chiou et al., 2000) while handling the loads (see Table 1). The PSOS was determined by adding the four ratings in Table 1. A high PSOS score implies a high subjective perception of slip and loss of balance during the MMH tasks. 2.5. Experiment design and data analysis The experiment was a single factor (friction level) experiment. The order of the three friction levels was randomly arranged within each subject. Analysis of variance (ANOVA) was performed for the heart rate, Vo2, MAWH, RPE, and PSOS. Tukey’s honestly significant difference (HSD) test was performed if a variable was
Table 1 Perceived sense of slip (PSOS) rating scale (Chiou et al., 2000) (1) How much did you feel yourself slip (i.e. loss of foot traction)? A little Some A lot 0 0.5 1 1.5 2 (2) Did you have any difficulty in maintaining balance (how much did you or your muscles compensate for your movement)? A little Some A lot 0 0.5 1 1.5 2 (3) Did you feel at any time that you would slip? A little Some 0 0.5 1 1.5
A lot 2
(4) What would you say was the overall difficulty of this task? A little Some A lot 0 0.5 1 1.5 2
statistically significant at po0.05. Pearson’s correlation coefficients between the dependent variables with significant results were calculated. Statistical analysis was performed using the SPSSs 10.0 software. 3. Results Table 2 shows the ANOVA results of the dependent variables. Tukey’s HSD tests were performed for MAWH, PSOS, and Vo2. The results are shown in Tables 3–5. The MAWH, which ranged from 7.8–10.7 kg, was significantly (p ¼ 0.0006) affected by friction level. The MAWH at highfriction level was significantly higher than that of the lowfriction level. The PSOS was significantly (po0.0001) affected by friction level. The PSOS at low-friction level was significantly higher than those of the medium and high-friction ones. The PSOS at medium-friction level was also significantly higher than that of the high-friction level. The mean heart rates for the low-, medium-, and highfriction conditions were 96.3, 97.1, and 97.4 beats/min, respectively. This corresponded to heart rate increases of 26.8%, 27.4%, and 26.1% as compared to the resting heart rates. The differences among these heart rates were not statistically significant. The friction level affected Vo2 significantly at po0.0001. The Vo2 at the medium friction level was significantly lower than those of the low- and high-friction levels. The Vo2 at the high-friction level was significantly lower than that of the low-friction level. To quantify the efficiency of metabolic energy consumed for the materials-handling tasks, the term ‘‘energy efficiency’’ was defined as the amount of weight handled per unit oxygen consumed per minute. It was calculated by dividing the MAWH by the Vo2. The ANOVA results indicated that friction level affected energy efficiency significantly (po0.0001). Table 6 showed Tukey’s HSD test results for energy efficiency. Pearson’s correlation coefficients between PSOS and energy efficiency and between PSOS and MAWH were 0.720 (po0.0001) and 0.611 (p ¼ 0.0015), respectively. Pearson’s correlation coefficients between Borg’s RPE
ARTICLE IN PRESS K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
263
Table 2 ANOVA results of the dependent variables
MAWH
Vo2
HR
RPE
PSOS
Source of variation
Sum of squares
Df
Mean square
F
p-value
Between Between Error Between Between Error Between Between Error Between Between Error Between Between Error
5.60 4.16 2.95 32,100.09 96,049.45 4604.68 5.75 2844.94 643.82 3.25 14.50 14.75 51.85 13.98 16.34
2 7 14 2 7 14 2 7 14 2 7 14 2 7 14
2.80 0.59 0.21 16,050.05 13,721.35 328.90 2.87 406.42 45.99 1.63 2.07 1.05 25.93 1.99 1.17
13.30 2.82
0.0006 0.0467
48.80 41.72
o0.0001 o0.0001
0.06 8.84
0.9396 0.0003
1.54 1.97
0.2481 0.1334
22.21 1.71
o0.0001 0.1856
friction subject friction subject friction subject friction subject friction subject
Table 3 Tukey’s HSD test results for MAWH
Table 6 Tukey’s HSD test results for energy efficiency
Friction level
MAWH (kg)
Tukey’s groupa
Friction level
Energy efficiency (kg/L/min)
Tukey’s groupa
Low Medium High
8.15 8.74 9.34
A AB B
Low Medium High
12.58 15.73 15.38
A B B
a
Number in each cells were the means for eight subjects; Different letters in Tukey’s group showing they were significantly different at po0.05. Table 4 Tukey’s HSD test results for PSOS a
Friction level
PSOS
Tukey’s group
Low Medium High
5.44 3.58 1.84
A B C
a Number in each cells were the means for eight subjects; Different letters in Tukey’s group showing they were significantly different at po0.05.
Table 5 Tukey’s HSD test results for Vo2 Friction level
Vo2 (L/min)
Tukey’s groupa
Low Medium High
0.651 0.562 0.613
A B C
a Number in each cells were the means for eight subjects; Different letters in Tukey’s group showing they were significantly different at po0.05.
and energy efficiency and between MAWH and energy efficiency were 0.574 (p ¼ 0.0033) and 0.424 (p ¼ 0.0388), respectively. 4. Discussion The psychophysical approach measures the perceived stress in MMH tasks. It is based on the assumption that
a
Number in each cells were the means for eight subjects; Different letters in Tukey’s group showing they were significantly different at po0.05.
both biomechanical and physiological stresses are present when people are handling materials, and these stresses may be combined under the measure of perceived stress (Ayoub and Mital, 1989). The maximum acceptable weights are the maximal loads that will not result in over-exertion or excessive fatigue for the workers. Ciriello et al. (2001) found that the maximum acceptable weights in pushing selected by subjects on low friction floors were most likely limited by the inadequate friction rather than the physical limitations. Miller (1983) mentioned that the required COF to prevent slipping is highly task-dependent. He indicated that a SCOF higher than 0.5 was required for load handling tasks. This implied that both the medium and low friction conditions provided inadequate slip-resistance for the subject’s materials-handling tasks. One of the major hypotheses of the present study was that the friction level affects the subject’s capability in the combined MMH tasks. This hypothesis was supported by the results of the study. In addition to the physical burden, floors with lower, inadequate friction levels apparently placed limitations on the subjects in the selection of their MAWH. It was believed that the subjects could differentiate friction level of the footwear–floor interface at the beginning or after a few MMH trials. The perception of floor slipperiness might affect their decision in determining their MAWH. The results of the Borg–RPE rating were not significant indicating that the subjects perceived no difference in their
ARTICLE IN PRESS 264
K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
whole body strain in the MMH tasks under the three friction conditions. The RPE results, ranging from 11 (or ‘‘light’’) to 16 (between ‘‘hard’’ and ‘‘very hard’’), indicated that different subjects would work at different RPE levels. This was consistent with the findings in the literature (Garg and Saxena, 1982; Garg and Banaag, 1988). The mode of RPE was 13, which corresponded to ‘‘somewhat hard’’. Two-thirds of the trials had a RPE less than or equal to 13. As the MAWH was determined by the subject, this implies that most of the subjects selected a weight that they felt was ‘‘somewhat hard’’ or lighter when they were performing the MMH tasks for an 8-h shift. This was consistent with the findings in the literature (Wang et al., 2000; Wu and Chen, 2003). The Vo2 was significantly (po0.0001) affected by friction level. The Vo2 at the medium friction condition was significantly lower than those of the low- and high-friction conditions. This implied the subjects were experiencing higher physiological strain when they handled loads on either low- or high-friction floors as compared to that of the medium one. It should be noted that the effects of friction level on Vo2 was confounded with workload as the MAWH were different among the friction levels. However, the trends between the MAWH (see Table 3) and the Vo2 (see Table 5) under the friction levels were not consistent. In Table 3, the MAWH increased as the friction level increased from low to high. But the Vo2 in Table 5 showed that the medium friction level resulted in a lowest Vo2 level among the three friction levels. It was, therefore, concluded that friction level did affect the physiological strains under the experimental conditions of the study. Chiou et al. (2000) indicated that the subjects experienced a high postural instability when the PSOS were high. The extra physiological burden on low friction floor might be attributed to the different walking and materials handling strategies adopted in order to avoid a fall. The literature (Li, 1991; Myung and Smith, 1997; Cham and Redfern, 2001) has shown that the subjects adapted to the slippery floor by increasing the knee flexion and hip extension moments and the co-contraction of the leg muscles so as to maintain balance. An additional muscular effort and physiological cost was required for such a gait strategy. This was supported by our results of energy efficiency where the low-friction condition had significantly lower value (12.58 kg/L/min) as compared to those of the medium (15.73 kg/L/min) and high-friction (15.38 kg/L/ min) conditions. The mean heart rates for the three friction levels (97.4 beats/min or lower) were lower than those reported by Garg and Banaag (1988) (119–126 beats/min) indicating that the subjects selected a workload that was lower than those in the literature. Chiou et al. (2000) suggested that the PSOS be used as a predictor of both slip occurrence and postural instability. In the present study, the subjects reported their PSOS after each trial. The PSOS scores ranged from 0.5–7.5 with means of 5.44, 3.59, and 1.84 for the low, medium, and
high-friction levels, respectively. They were significantly (po0.0001) different among one another. This was consistent with the findings of Chiou et al. (2000). Chiou et al. (2000) indicated that the percentage of slip occurrence increased from 0% to 100% when a PSOS increased from 0 to 4.5. A slip always occurred when the PSOS exceeded 4.5. In the present study, seven out of eight subjects reported a PSOS higher than 4.5 on the low-friction condition. Two out of eight subjects reported a PSOS of 4.5 or higher on medium friction condition. All the subjects reported a PSOS of 3.5 or lower on high friction condition. The PSOS results correlated well with the friction levels. According to Chiou et al. (2000), the subjects were not only more likely to slip but also experienced more postural instability at high PSOS levels. According to Gro¨nqvist (1995), the slip-resistance of the low-friction condition (with a DCOF of 0.16) in the present study was ‘‘unsure’’. The slip-resistances of the medium and high-friction conditions (both had a DCOF higher than 0.3) were both ‘‘very slip-resistant’’. In our study, there were two subjects reported a PSOS of 4.5 on medium friction condition and one subject reported a 5 on high friction condition. This implied that Gro¨nqvist’s (1995) categories were not consistent with our PSOS results. The discrepancy between the two might be attributed to the fact that Gro¨nqvist’s (1995) categories considered only the risk of slipping while the PSOS considered the effects of postural instability in performing the task as an addition. Gro¨nqvist’s (1995) categories, however, supported our results in the physiological burden where the energy efficiency was significantly lower (12.58 kg/L/min) on the ‘‘unsure’’ condition as compared to those of the ‘‘very slipresistant’’ conditions (15.73 and 15.38 kg/L/min). In other words, the amount of weight handled per unit oxygen consumed on the ‘‘unsure’’ slip-resistance condition was significantly lower than those on the ‘‘very slip-resistant’’ conditions. There were limitations to this study. First of all, the strategies for a walking pattern when carrying a load that the subjects might adopt could be very complicated, especially on low-friction floors. The subjects were required to maintain a consistent working and walking pace during the experiment. However, their walking speeds and gait patterns were not measured or controlled. It was possible that they reduced their stride length and increased cadence (or vice versa) at the same time to maintain the same walking speed in different trials. This type of gait adaptation could affect the experimental results. Secondly, whether the subject slipped or not was not recorded and percentage of slip occurrence was not measured. Thirdly, only one lifting height (55 cm from floor), one carrying distance (3 m), and one frequency (2 per min) were tested. Data on more experimental conditions are required in the future to provide more information in understanding the interactions between friction levels and the MAWH and the physiological responses.
ARTICLE IN PRESS K.W. Li et al. / Applied Ergonomics 38 (2007) 259–265
5. Conclusion This study investigated the effects of the friction level at footwear–floor interface on the MAWH, heart rate, Vo2, energy efficiency, PSOS, and RPE. The results indicated that friction level was a significant factor affecting the MAWH, Vo2, energy efficiency, and PSOS. When working on low-friction floor surfaces, the subjects selected less weight as compared to that on the high friction floors. The Vo2 values in handling the weight at the MAWH level for ten minutes at the low- and high-friction conditions were significantly higher than that on the medium-friction condition. The energy efficiency at the low-friction condition was significantly lower than those on the medium- and high-friction conditions. The PSOS in the MMH tasks on the low-friction condition was the highest, next being the medium-friction condition, and last the high-friction condition. The results of this study showed that floorfriction level should be regarded as one of the major environmental factors in designing similar combined MMH tasks as it affected both the physiological and perceptual responses of the subjects. Acknowledgments The authors thank Ray McGorry of Liberty Mutual Research Institute for Safety for his thoughtful comments of the earlier drafts of the manuscript. References Ayoub, M.M., Mital, A., 1989. Manual Materials Handling. Taylor & Francis, London. Borg, G., 1985. An Introduction to Borg’s RPE Scale. Movement Publications, Ithaca, NY. Cham, R., Redfern, M.S., 2001. Lower extremity corrective reactions to slip events. J. Biomech. 34, 1439–1445. Cham, R., Redfern, M.S., 2002. Heel contact dynamics during slip events on level and inclined surfaces. Saf. Sci. 40, 559–576. Chang, W.R., Li, K.W., Huang, Y.H., Filiaggi, A., Courtney, T.K., 2004. Assessing floor slipperiness in fast-food restaurants in Taiwan using objective and subjective measures. Appl. Ergon. 35, 401–408. Chiou, S.-Y., Bhattacharya, A., Succop, P.A., 2000. Evaluation of Workers’ perceived sense of slip and effect of prior knowledge of slipperiness during task performance on slippery surfaces. Am. Ind. Hyg. Assoc. J. 61, 492–500. Chung, H.C., Wang, M.J., 2001. The effects of container design and stair climbing on maximal acceptable lift weight, wrist posture, psychophysical, and physiological responses in wafer-handling tasks. Appl. Ergon. 32, 593–598. Ciriello, V.M., Snook, S.H., 1999. Survey of manual handling tasks. Int. J. Ind. Ergon. 23, 149–156.
265
Ciriello, V.M., McGorry, R.W., Martin, S.E., 2001. Maximum acceptable horizontal and vertical forces of dynamic pushing on high and low coefficient of friction floors. Int. J. Ind. Ergon. 27, 1–8. Ciriello, V.M., Snook, S.H., Hashemi, L., Cotnam, J., 1999. Distributions of manual material handling task parameters. Int. J. Ind. Ergon. 24, 379–388. Garg, A., Saxena, U., 1980. Container characteristics and maximum acceptable weight of lift. Hum. Factors 22, 487–495. Garg, A., Saxena, U., 1982. Maximum frequency acceptable to female workers for one-handed lifts in the horizontal plane. Ergonomics 25 (9), 839–853. Garg, A., Badger, D., 1986. Maximum acceptable weights and maximum voluntary isometric strength for asymmetric lifting. Ergonomics 29 (7), 879–892. Garg, A., Banaag, J., 1988. Maximum acceptable weights, heart rates and RPEs for one hour’s repetitive asymmetric lifting. Ergonomics 31 (1), 77–96. Gro¨nqvist, R., 1995. Mechanisms of friction and assessment of slip resistance of new and used footwear soles on contaminated floors. Ergonomics 38, 224–241. Haslam, R.A., Boocock, M., Lemon, P., Thorpe, S., 2002. Maximum acceptable loads for pushing and pulling on floor surfaces with good and reduced resistance to slipping. Saf. Sci. 40, 625–637. Lee, Y.H., Wu, S.P., Hsu, S.H., 1995. The psychophysical lifting capacities of Chinese subjects. Ergonomics 38, 671–683. Li, K.W., 1991. A biomechanical study of slipping and falling with load carriage, unpublished doctoral dissertation, Department of industrial engineering. Texas Tech University, Lubbock, TX. Li, K.W., Chang, W.R., Leamon, T.B., Chen, C.J., 2004. Floor slipperiness measurement: friction coefficient, roughness of floors, and subjective perception under spillage conditions. Saf. Sci. 42, 547–565. Miller, J.M., 1983. ‘‘Slippery’’ work surface: toward a performance definition and quantitative coefficient of friction criteria. J. Saf. Res. 14, 145–158. Mital, A., 1992. Psychophysical capacity of industrial workers for lifting symmetrical and asymmetrical loads symmetrically and asymmetrically for 8 h work shifts. Ergonomics 35, 745–754. Myung, R., Smith, J.L., 1997. The effect of load carrying and floor contaminants on slip and fall parameters. Ergonomics 40 (2), 235–246. Myung, R., Smith, J.L., Leamon, T.B., 1993. Subjective assessment of floor slipperiness. Int. J. Ind. Ergon. 11, 313–319. Snook, S.H., 1978. The design of manual handling tasks. Ergonomics 21, 963–985. Snook, S.H., 1985. Psychophysical considerations in permissible loads. Ergonomics 28, 327–330. Snook, S.H., Ciriello, V.M., 1991. The design of manual tasks: revised tables of maximum acceptable weights and forces. Ergonomics 34, 28–34. Swensen, E.E., Purswell, J.L., Schlegel, R.E., Stanevich, R.L., 1992. Coefficient of friction and subjective assessment of slippery work surface. Hum. Factors 34 (1), 67–77. Wang, M.J., Chung, H.C., Chen, H.C., 2000. The effect of handle angle on MAWL, wrist, posture, RPE, and heart rate. Hum. Factors 42 (4), 553–565. Waters, T.R., Putz-Anderson, V., Garg, A., Fine, L., 1993. Revised NIOSH equation for the design and evaluation of manual lifting tasks. Ergonomics 36 (7), 749–776. Wu, S.P., Chen, J.P., 2003. Effects of the adjustment period on psychophysically determined maximum acceptable weight of lift and the physiological cost. Int. J. Ind. Ergon. 31, 287–294.