The relationship between work productivity and acute responses at different levels of production standard times

International Journal of Industrial Ergonomics 56 (2016) 59e68

Contents lists available at ScienceDirect

International Journal of Industrial Ergonomics journal homepage: www.elsevier.com/locate/ergon

The relationship between work productivity and acute responses at different levels of production standard times Nurhayati M.N.a, b, *, Siti Zawiah M.D.a, Mahidzal D.a a b

Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603, Lembah Pantai, Kuala Lumpur, Malaysia Aerospace Department, Universiti Kuala Lumpur, Malaysian Institute of Aviation Technology, Jenderam Hulu, 43800, Dengkil, Selangor, Malaysia

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 December 2014 Received in revised form 10 August 2016 Accepted 24 September 2016

Work productivity is typically associated with production standard times. Harder production standards generally result in higher work productivity. However, the tasks become more repetitive in harder production standard time and workers may be exposed to higher rates of acute responses which will lead to higher risks of contracting work-related musculoskeletal disorders (WMSDs). Hence, this paper seeks to investigate the relationship between work productivity and acute responses at different levels of production standard times. Twenty industrial workers performed repetitive tasks at three different levels of production standard time (PS), corresponding to “normal (PSN)”, “hard (PSH)” and “very hard (PSVH)”. The work productivity and muscle activity were recorded along these experimental tasks. The work productivity target was not attainable for hard and very hard production standard times. This can be attributed to the manifestations of acute responses (muscle activity, muscle fatigue, and perceived muscle fatigue), which increases as the production standard time becomes harder. There is a strong correlation between muscle activity, perceived muscle fatigue and work productivity at different levels of production standard time. The relationship among these variables is found to be significantly linear (R ¼ 0.784, p < 0.01). The findings of this study are indeed beneficial to assess the existing work productivity of workers and serves as a reference for future work productivity planning in order to minimize the risk of contracting WMSDs. © 2016 Elsevier B.V. All rights reserved.

Keywords: Work productivity Acute response Muscle activity Work-related musculoskeletal disorders

1. Introduction Work productivity is a determinant factor in the manufacturing industry, it is measured periodically in order to monitor workers' performance, which in turn, reflects the performance of the organization. Work productivity is crucial in the assembly line since the process directly involves workers, manual handling and repetitive tasks. In addition, most of the tasks in the manufacturing industry are repetitive and previous studies have shown that repetitive tasks are associated with WMSDs (Escorpizo and Moore, 2007; Nordander et al., 2009; You and Kwon, 2005). Therefore, work productivity planning should be done in advance in scenarios involving high repetition of task in order to sustain a good work

* Corresponding author. Aerospace Department, Universiti Kuala Lumpur, Malaysian Institute of Aviation Technology, Jenderam Hulu, 43800 Dengkil, Selangor, Malaysia. E-mail addresses: [email protected] (M.N. Nurhayati), sitizawiahmd@ um.edu.my (M.D. Siti Zawiah), [email protected] (D. Mahidzal). http://dx.doi.org/10.1016/j.ergon.2016.09.009 0169-8141/© 2016 Elsevier B.V. All rights reserved.

productivity level whilst minimizing the risks of developing WMSDs. The current trend in industrial tasks is moving towards more time-intensive production with standardized, short cycle time (Neumann et al., 2002) and limited completion time (Wartenberg et al., 2004) since the aim of the manufacturing industry is to attain a high work productivity levels. Process standard time, such as work pace or duty cycle time for a particular task is determined by a process engineer based on task time analysis. Furthermore, the workplace is not chosen by the worker and it is obligatory for a worker to follow a predetermined task time (Sundelin and Hagberg, 1992). Hence, the worker's capacity and productivity state are often overestimated during the planning stage. The work productivity target is set based on a specific production standard time and therefore, work productivity can be evaluated directly through the number of products produced per day or per hour. The value of productivity loss can be calculated directly if the workers do not perform well according to predetermined standards. Such work productivity assessment and monitoring method is applied specifically for assembly workers (Escorpizo,

60

M.N. Nurhayati et al. / International Journal of Industrial Ergonomics 56 (2016) 59e68

2008). However, the work productivity assessment method is insufficient when it comes to identifying work productivity due to the functional incapacity of workers (Allesina et al., 2010). In addition, the worker's capacity is typically underrated due to the constant drive for organizations to achieve high work productivity (Escorpizo, 2008). Thus, workers may be exposed to a higher risk of contracting WMSDs while carrying out tasks during high work productivity targets or hard production standard times. It is therefore imperative that work productivity assessment is evaluated based on multiple perspectives, and several ergonomics assessment should be contemplated during the initial stage of work productivity planning. The combination of ergonomics and assembly engineering is vital to ensure that the work productivity targets assigned to workers are within their capacity without causing adverse health effects, in particular, WMSDs. The performance or personal capacity of each individual worker is critical since it affects the overall productivity of workers. Previous studies have shown that workers who work beyond their capacity levels result in work productivity loss (Finneran & O'Sullivan, 2010). Work productivity loss is not only characterized by time losses, but also the functional incapacity of workers. In such cases, the workers are present for work, but they are functionally limited due to WMSD-related pain or discomfort (Finneran & O'Sullivan, 2010). Work productivity is one of the measures which have been commonly investigated in various studies related to musculoskeletal disorders. However, studies that focus on work productivity and musculoskeletal disorders (MSDs) among industrial workers are rather sparse, whereas the number of MSD cases increases each year (SOCSO, 2011). Musculoskeletal disorders refer to conditions which involve muscles, nerves, tendons and other soft tissues (NIOSH, 1997). In general, WMSDs develop over time (Putz-Anderson, 1988) and therefore the peak and cumulative musculoskeletal discomfort experienced by workers may be used to predict future musculoskeletal pain (Hamberg-van Reenen et al., 2008). Previous studies have shown that exposure to short-term responses (acute responses) indicates the development of WMSD risks in the short term (Westgaard and Winkel, 1996). Most of the studies on work productivity and acute responses are conducted independently (Alavinia and Burdoff, 2009; Lerner et al., 2003; Resnick and Zanotti, 1997; Shikdar and Das, 2003; Van den Heuvel et al., 2010) and there is a lack of studies which focuses on the relationship between work productivity and acute responses as well as the relationship between work productivity and WMSD risks (Conway and Svenson, 2001; Finneran & O'Sullivan, 2010). Knowledge on the relationships will indeed be beneficial as it forms a groundwork to develop a model to assess work productivity at different levels of production standard time and serves as a reference for industries to design tasks which will optimize and sustain work productivity while minimizing the risk of contracting WMSDs among workers. 2. Methodology 2.1. Acute response variables The acute responses involved depend on the tasks executed. Performing industrial repetitive tasks involve muscle activity and thus muscle fatigue occurs when the muscle fails to maintain the required force or the desired work output level. The accumulation of muscle fatigue causes functional disability, which later develops into WMSDs (Ma et al., 2009) as a long term effect. Therefore, in relation to repetitive tasks, the acute response variables being investigated are muscle activity, muscle fatigue and perceived

muscle fatigue. 2.2. Subjects A total of 20 subjects were recruited for the series of experimental tasks, comprising 10 male and 10 female industrial workers. The subjects were between the ages of 22 and 45. All the subjects had no previous history of musculoskeletal injuries. The subjects gave their written consent prior to the start of the study. The study was approved by the local Ethics Committee. The descriptive data of the subjects are given in Table 1. 2.3. Tasks and muscles involved The tasks involved repetitive assembly actions, similar to the actual industrial assembly task. The subjects were given two types of component, plastic clips and plastic foam rings. These components were placed into a polybox and plastic container, respectively. The subjects were instructed to assemble the ring foam onto the plastic clip using a simple jig, which pushes the foam onto the clip, as shown in Fig. 1. The task is categorized as light assembly task with high repetition. The muscles involved were identified based on the task executed. The involved muscles are identified from consultation with an anatomist. The muscles which are the focus of this study are the forearm muscles, as listed follows: i Right and Left Flexor Carpi Radialis (FCRR, FCRL) ii Right and Left Extensor Carpi Radialis (ECRR, ECRL) According to Milerad and Ericson (1994), the extensor muscles are one of the most active muscle groups and these muscles have been the focus of several studies related to repetitive tasks (Bosch et al., 2011; Mananas et al., 2005; Nag et al., 2009; Rietveld et al., 2007). The flexor muscles are one of the forearm muscle groups which have been investigated among subjects while carrying out assembly tasks (Finneran & O'Sullivan, 2013; Gooyers and Stevenson, 2012; Nag et al., 2009). The muscle activity was recorded using surface electromyography (EMG). EMG is an experimental technique concerned with recording and analysis of myoelectric signals (Konrad, 2005). EMG detects myoelectric signals generated by muscle cells when these cells contract and at rest, and the signals are recorded using the instrument's reading software. 2.4. Production standard times The levels of production standard time (PS) used in the experimental tasks were listed as follows:1 Normal production standard time (PSN) 2 Hard production standard time (PSH) 3 Very hard production standard time (PSVH) Normal production standard time was based on the 100% normal standard time. A hard production standard (PSH) time was

Table 1 Descriptive data of the subjects.

Mean Standard deviation

Age (Year)

Weight (kg)

Stature (cm)

Male

Female

Male

Female

Male

Female

31.90 8.98

30.10 6.98

71.4 9.15

54.9 5.79

168.7 5.68

152.6 8.78

M.N. Nurhayati et al. / International Journal of Industrial Ergonomics 56 (2016) 59e68

61

Fig. 1. Photograph of the experimental tasks.

equivalent to 126% of the normal standard time (Bosch et al., 2011). A very hard production standard time (PSVH) was 140% of the normal standard time (Shikdar and Das, 2003). 2.5. Experimental task procedure The subjects were first briefed on the experimental task process flow and equipment used prior to the series of experimental tasks. Each subject was given an information sheet, which outlines their involvement as well as potential risks of the study. Written informed consent was obtained from each subject to ensure that they fully agree to participate in the study. The subjects were requested to fill in the questionnaires on descriptive data and subjective measurements prior to the series of experimental tasks. The subjects' skin were cleaned thoroughly and prepared before placing the electrodes. The surface electrodes were attached to the belly of the forearm muscles bilaterally (Lehman et al., 2001), as shown in Fig. 2. The subjects were then instructed to adopt a comfortable sitting posture and the sitting height was adjusted individually in order to obtain a knee angle of 90 . The working height was standardized by placing the table surface 5 cm below the position of the wrist when the elbow was flexed at 90 (Bosch et al., 2011). The subjects were instructed to perform maximum voluntary contraction (MVC) task once the signals from all sensors were stable. The subjects performed the MVC task three times, whereby the duration of each task was approximately 10 s with 30 s of rest in between contractions. The 30 s of rest serves as recovery time after each task. The MVC task is illustrated in Fig. 3, and it can be seen that the task was performed when the subjects were in sitting

position. A stable forearm support was arranged and manual resistance was used. The MVC measurement procedure used in this study was based on Konrad's guidelines (Konrad, 2005). The MVC refers to the highest electromyography (EMG) amplitude obtained from the three recordings and is expressed as the percentage of maximum voluntary contraction (%MVC). The MVC was used to normalize the surface EMG signals recorded during the series of experimental tasks. The subjects were required to perform the experimental tasks after familiarizing themselves with the tasks for 30 min. The subjects performed the tasks according to assigned production standard times over a 1-h period for each production standard time. The muscle signals were recorded using surface electromyography (EMG). The work productivity of the subjects was recorded every 30 min. The subjects were required to rate their perceived muscle fatigue upon completion of the experimental tasks. 2.6. Electromyography data analysis The raw EMG data obtained from muscle activity measurements were processed using MyoResearch XP software in order to filter low and high frequencies (20 and 400 Hz, respectively) from the signals (Bosch et al., 2009). The electrocardiogram (ECG) spikes present in the signals due to EMG artifacts were filtered out without affecting the true EMG amplitudes and power spectra. Here, the RMS amplitude was analysed in this study. The RMS value corresponds to the square root of the average power of the raw EMG signal over a given period of time (De Luca, 2003). RMS offers a more precise index of primary physiological actions (Basmaijan and De Luca, 1985). The RMS amplitude of the EMG signal

Fig. 2. Forearm muscles.

62

M.N. Nurhayati et al. / International Journal of Industrial Ergonomics 56 (2016) 59e68

Fig. 3. Illustration of the maximum voluntary contraction task.

provides more information compared to other available methods (Potes, 2008) as it provides consistent, valid and accurate measurements of noise and problematic signals (Gerleman and Cook, 1992; Konrad, 2005). The RMS value was normalized to the highest MVC value obtained from the MVC test and was expressed as the percentage of maximum voluntary contraction (%MVC) (Bao et al., 2001). The normalized RMS (%MVC) was averaged for every 5 min. The mean value of the normalized RMS represents the muscle activity in this study, while the rate of muscle fatigue is represented by the linear regression slope of the normalized RMS.

2.7. Statistical analysis Statistical analysis was carried out to analyse data derived from the experimental task. The data is summarized in Table 2. The data were tested for normality prior to analysis using Shapiro-Wilk test. Descriptive analysis was applied to all variables. Repeated measures ANOVA was applied to determine the differences in acute responses among different levels of production standard time. The relationship between work productivity and acute responses was investigated through Pearson correlation and regression analysis.

3. Results 3.1. Work productivity The work productivity data were recorded in terms of quantity per hour and the percentage of normal standard time achieved. The average work productivity for different levels of production standard time is summarized in Table 3. It can be seen that the work productivity increases as the production standard time becomes harder. The very hard production standard time (PSVH: 140% of the normal standard time) results in the highest output, followed by hard production standard time (PSH: 126% of the normal standard time) and normal production

Table 3 Means of work productivity for different production standard times. Production standard (PS)

Work productivity (mean)

PSN PSH PSVH

851 890 928

Table 2 Experimental data. Subject

Work productivity

Muscle Activity (% MVC)

PSN

PSH

PSVH

PSN

PSH

PSVH

PSN

PSH

PSVH

PSN

PSH

PSVH

PSN

PSH

PSVH

781 650 654 852 787 910 826 959 905 815 673 885 789 893 972 873 1011 928 883 974

830 674 711 902 802 914 829 973 907 862 703 938 830 894 1020 1021 1090 954 919 1033

980 686 757 972 829 1004 851 1024 940 893 729 942 841 900 1030 1039 1160 1010 930 1053

10.516 8.168 8.747 7.437 11.066 8.612 7.882 7.918 9.159 7.481 14.193 6.261 6.058 8.856 8.146 7.666 10.508 9.646 9.814 8.361

10.769 10.346 9.505 8.188 11.254 8.890 7.841 8.138 7.011 7.069 15.925 6.664 6.598 9.993 10.407 10.199 12.659 10.476 11.013 8.988

10.691 9.587 10.164 8.243 11.546 8.763 8.729 8.337 6.851 7.621 16.499 7.553 7.054 10.957 10.141 9.176 14.350 11.342 11.560 9.352

8.763 11.669 8.307 11.539 11.087 11.217 8.773 10.056 12.675 12.768 11.856 12.499 12.465 13.214 10.591 8.376 11.860 11.186 11.479 9.836

11.387 13.182 9.653 12.189 11.679 11.396 9.964 9.794 12.179 13.725 12.854 12.740 12.843 13.342 11.928 12.119 12.413 11.643 12.313 10.476

10.852 13.700 10.329 12.262 11.324 11.950 9.387 12.498 11.103 13.577 16.969 13.705 13.090 13.582 11.396 13.274 13.054 12.646 13.613 11.197

15.041 11.734 5.793 8.720 11.506 8.432 10.392 10.458 11.072 12.269 15.943 9.611 8.930 8.382 9.182 9.849 8.632 10.678 12.120 8.813

16.049 14.119 6.604 8.837 12.572 8.463 11.190 11.380 9.228 12.835 16.463 9.125 10.527 9.345 9.789 11.922 8.855 11.089 12.836 9.475

19.812 14.911 7.137 9.511 14.152 9.186 10.930 11.732 9.858 12.528 17.874 11.189 10.322 8.801 10.576 14.891 9.352 11.432 13.578 10.314

10.203 18.516 13.868 6.550 14.663 9.133 6.471 9.962 11.073 10.412 17.527 12.250 9.925 10.499 16.623 6.616 10.968 10.520 12.702 9.715

11.602 19.120 15.105 6.851 12.519 9.413 7.179 9.227 10.834 11.205 18.325 12.959 9.516 10.153 17.850 10.121 11.770 11.168 14.243 10.339

12.325 20.973 16.564 7.765 12.380 10.071 7.582 10.341 11.308 12.473 19.471 13.723 10.435 11.737 16.130 10.522 13.522 11.489 14.926 10.310

FCRR

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

FCRL

ECRR

ECRL

M.N. Nurhayati et al. / International Journal of Industrial Ergonomics 56 (2016) 59e68

63

Table 4 Work productivity in the percentage of normal standard time. Production standard (PS)

Work productivity target

Work productivity achieved

PSN PSH PSVH

100% 129% 140%

118.0 123.0 129.0

standard time (PSN: 100% of the normal standard time). The work productivity achieved in the percentage of normal standard time were then determined and the values are tabulated in Table 4. The work productivity performance for different levels of production standard time is plotted in Fig. 4. It can be seen that the work productivity achieved at the normal standard time is higher than the target value. However, the work productivity achieved for hard and very hard production standard times are found to be less than the targeted value. The work productivity achieved is 123% for hard production standard time whereas the value is 129% for very hard production standard time. Therefore, the maximum work productivity achieved from this study is 129% of the normal production standard time. 3.2. Muscle activity The muscle activity observed in this study is represented by the normalized root mean square (RMS) from the electromyography signals. The mean value and standard deviation of the normalized RMS for all muscles are summarized in Table 5. It can be observed that the muscle activity increases as the production standard time becomes harder. The very hard production standard time (PSVH) shows the highest RMS value for all muscles. The highest RMS value is obtained for Extensor Carpi Radialis-Left (ECRL), followed by Flexor Carpi Radialis-Left (FCRL), Extensor Carpi Radialis-Right (ECRR) and Flexor Carpi RadialisRight (FCRR). Repeated measures ANOVA was conducted to investigate the effect of production standard time on muscle activity and the results reveal the production standard time has a significant effect on the mean RMS (muscle activity) for all muscles. The results are summarized in Table 6. Table 7 presented the percentage of increment in muscle activity at different levels of production standard times. The increment of muscle activity in hard production standard times was in the range of 4.9%e8.7%. The provision of an assigned very hard production standard time showed that the muscle activity was

Fig. 4. Work productivity performance at different levels of production standard time.

higher than assigned normal production standard time and hard production standard time with an increment in the range of 11.3%e 14.7% and 3.4%e7.8% respectively, depending on the type of muscles. On average, the muscle activity increased 6.9% in hard production standard times and 12.9% in very hard production standard times. 3.3. Muscle fatigue The muscle fatigue rate was determined from the linear regression slopes of the normalized EMG RMS versus time and the values are summarized in Table 8. It can be seen that the muscle fatigue rate is higher at harder production standard times. Table 9 presented the percentage of increment in muscle fatigue rate at different levels of production standard times. The increment of muscle fatigue rate in hard production standard times was in the range of 2.2%e77.9%. The provision of an assigned very hard production standard time showed that the muscle fatigue rate was higher than assigned normal production standard time and hard production standard time with an increment in the range of 11.1%e 94.9% and 2.3%e11.9% respectively, depending on the type of muscles. On average, muscle fatigue rate increased 31.3% in hard production standard times and 41.8% in very hard production standard times. 3.4. Perceived muscle fatigue Perceived muscle fatigue ratings were measured before and after the experimental tasks using the Borg CR-10 scale. It is found that all subjects rated ‘no fatigue’ prior to the series of experimental tasks. It is observed that the perceived muscle fatigue increases when the subjects performed their tasks at harder production standard times, and the perceived muscle fatigue varies depending on the type of muscles. In this study, the subjects rated their perceived muscle fatigue as ‘weak to moderate’, ‘moderate’ and ‘moderate to strong’ when they performed the experimental tasks at normal, hard and very hard production standard time, respectively. The mean and standard deviation for the perceived muscle fatigue are summarized in Table 10. It can be observed that the highest perceived muscle fatigue is observed for the Extensor Carpi Radialis-Left (ECRL) muscle when the subjects performed the experimental tasks in very hard production standard time. The results are consistent with the EMG results, whereby the highest muscle fatigue occurs at the Extensor Carpi Radialis-Left (ECRL) muscle. Comparison of the perceived muscle fatigue between production standard times was analysed using repeated measures ANOVA and the results are presented in Table 11. The results show that there is a significant increase in perceived muscle fatigue for all muscles when the production standard time becomes harder. Table 12 presented the percentage of increment in perceived muscle fatigue at different levels of production standard times. The increment of perceived muscle fatigue in hard production standard times was in the range of 22.6%e29.2%. The provision of an assigned very hard production standard time showed that the

64

M.N. Nurhayati et al. / International Journal of Industrial Ergonomics 56 (2016) 59e68

Table 5 Mean and standard deviation of normalized RMS. Production standard (PS)

Mean RMS (standard deviation)

PSN PSH PSVH

FCRR

FCRL

ECRR

ECRL

8.824 (1.827) 9.596 (2.257) 9.925 (2.402)

11.010 (2.809) 11.890 (2.496) 12.475 (2.921)

10.377 (2.344) 11.035 (2.552) 11.904 (3.154)

11.409 (3.416) 11.974 (3.431) 12.702 (3.486)

Table 6 Summary of muscle activity results obtained from repeated measures ANOVA. Muscle

Wilk's Lambda

F-Value

p-Value

Partial eta squared

FCRR FCRL ECRR ECRL

0.478 0.410 0.420 0.222

6.179 8.142 7.819 19.830

0.005 0.001 0.002 0.000

0.522 0.590 0.580 0.778

Table 7 Production standard time and % of increase in the muscle activity. Comparison between standard

% Increase in muscle activity (RMS) FCRR

FCRL

ECRR

ECRL

PSN vs PSH PSN vs PSVH PSH vs PSVH

8.7 12.5 3.4

8.0 13.3 4.9

6.3 14.7 7.8

4.9 11.3 6.0

Table 12 Production standard time and % of increase in the perceived muscle fatigue. Comparison between standard

% Increase in perceived muscle fatigue FCRR

FCRL

ECRR

ECRL

PSN vs PSH PSN vs PSVH PSH vs PSVH

24.4 60.0 28.6

29.2 58.3 22.6

22.6 54.7 26.2

26.7 63.3 28.9

perceived muscle fatigue was higher than assigned normal production standard time and hard production standard time with an increment in the range of 54.7%e63.3% and 22.6%e28.9% respectively, depending on the type of muscles. On average, the perceived muscle fatigue increased 25.7% in hard production standard times and 59.1% in very hard production standard times. 3.5. Correlation between acute responses and work productivity

Table 8 Muscle fatigue rate. Production standard (PS)

Muscle fatigue rate

PSN PSH PSVH

FCRR

FCRL

ECRR

ECRL

0.045 0.046 0.050

0.061 0.086 0.088

0.088 0.092 0.103

0.059 0.105 0.115

Correlation analysis was performed the results are summarized in Table 13. The results show that there is a significant correlation between muscle activity and work productivity as well as between perceived muscle fatigue and work productivity for the Extensor Carpi Radialis-Left (ECRL) muscles. The maximum muscle activity and perceived muscle fatigue were also observed for this muscle. 3.6. Relationship among muscle activity, perceived muscle fatigue and work productivity

Table 9 Production standard time and % of increase in the muscle fatigue. Comparison between standard

% Increase in muscle fatigue rate FCRR

FCRL

ECRR

ECRL

PSN vs PSH PSN vs PSVH PSH vs PSVH

2.2 11.1 8.7

40.9 44.2 2.32

4.5 17.0 11.9

77.9 94.9 9.5

Table 10 Mean and standard deviation values for perceived muscle fatigue. Production standard (PS)

Mean (standard deviation) FCRR

FCRL

ECRR

ECRL

PSN PSH PSVH

2.25 (1.25) 2.80 (1.00) 3.60 (0.75)

2.40 (1.23) 3.10 (1.02) 3.80 (0.89)

2.65 (1.04) 3.25 (0.91) 4.10 (0.78)

3.00 (1.02) 3.80 (1.00) 4.90 (0.91)

Table 11 Results on perceived muscle fatigue obtained from repeated measures ANOVA. Muscle

Wilk's Lambda

F-value

p-Value

Partial eta squared

FCRR FCRL ECRR ECRL

0.132 0.158 0.093 0.106

37.29 30.26 55.21 57.61

0.000 0.000 0.000 0.000

0.868 0.842 0.907 0.894

The experimental results reveal that the work productivity target is unattainable for hard and very hard production standard times as the acute responses increase in these scenarios. The results of the correlation analysis also indicate that there is a significant correlation between muscle activity and work productivity as well as between perceived muscle fatigue and work productivity. Thus, linear regression analysis was conducted to investigate and verify these relationships. The regression model summary is shown in Table 14. It can be seen from Table 14 that, there is a strong correlation between the muscle activity, perceived muscle fatigue, production standard time and work productivity, with an R-value above 0.5 (R ¼ 0.784). In general, an R-value above 0.5 indicates that there is a strong correlation among variables (Cohen, 1988; Pallant, 2013). The adjusted R Square is always used to verify the accuracy of the model since the R Square value always increase when the regressor variable is added to the regression model and it is, therefore, a bad indicator of model accuracy (Montgomery, 2014). In this case, however, both parameters can be used since the difference between R Square and adjusted R Square is very small. The R Square value is only an indicator of completeness of the regression model. The statistical significance of the result is given by the p-values in the regression ANOVA as well as p-values of the coefficients. It is found that the p-values for both regression ANOVA and coefficients are statistically significant (p < 0.005), which indicate that correlation is reliable and can be used to make predictions. Hence, the following regression equation is proposed to predict work

M.N. Nurhayati et al. / International Journal of Industrial Ergonomics 56 (2016) 59e68 Table 13 Results of the Pearson correlation test.

65

Table 15 Descriptive data of workers.

Factor

Correlation

P

Significance level

Muscle Activity Perceived Muscle Fatigue

0.630 0.510

0.000 0.000

P < 0.001 P < 0.001

Items

Mean S.D

Age (Year)

Stature (cm)

Weight (kg)

Male

Female

Male

Female

Male

Female

30.0 6.782

29.6 6.309

169.0 3.240

157.4 6.985

67.6 7.635

52.2 5.118

productivity performance: Y ¼ 0.361e0.039 (X1) þ 0.087 (X2) þ 0.20 (X3)

(1)

where: Y Work productivity X1 Muscle activity X2 Perceived muscle fatigue X3 Production standard time The regression model developed in this study is a ‘Work Productivity Prediction Model’ which predicts work productivity performance as a function of muscle activity, perceived muscle fatigue and production standard time. The model is then validated with data from 10 industrial workers. The descriptive data of the workers are presented in Table 15. The Standard Error of Estimate (SEE) was used to validate the data predicted by the model and the results are tabulated in Table 16.

Table 16 Standard error of estimates for the Work Productivity Prediction Model. Model

Production standard time

SEE

Work Productivity

PSN PSH PSVH

0.1237 0.1337 0.1266

harder production standard times due to WMSD risks. The results obtained are highly consistent with the findings of previous studies which showed that workers tend to slow down when they are fatigued due to WMSDs (Resnick and Zanotti, 1997). Several studies have also reported that the workers are exposed to a higher risk of contracting WMSDs in highly repetitive tasks (Gooyers and Stevenson, 2012; You and Kwon, 2005). Previous studies have also suggested that the development of WMSD risks is indicated by acute responses (Bosch et al., 2011; Westgaard and Winkel, 1996). Hence, the findings on work productivity can be attributed to the variations of acute responses at different levels of production standard time.

4. Discussion 4.1. Work productivity at different levels of production standard time In general, the frequency of repetitive movements is higher in harder production standard times. As expected, the results show that work productivity increases as the production standard time becomes harder. This indicates that the workers were able to achieve higher work productivity at harder production standard times compared to the normal production standard time. This observation agrees well with the findings of Shikdar and Das (2003) whereby work productivity increases at harder production standard times. Higher work productivity achieved in harder production standard times is considered desirable in terms of productivity in the organization as well as for economic reasons. However, the results revealed otherwise from an ergonomics' point of view. In this study, work productivity target was assigned for each level of production standard time. The results show that work productivity target is attainable for normal production standard time, such is not the case for hard and very hard production standard times. The work productivity target for hard production standard time is 126% of the normal standard time. The results show that the workers only achieved 123% of the normal standard time. Similarly, the work productivity target for very hard production standard time is 140% of the normal standard time and it is found that the workers were only able in achieving 129% of the normal standard time. In general, workers performing tasks with higher repetitions in harder production standard times are exposed to a higher risk of WMSDs. The results indicate that the workers tend to slow down in

4.2. Muscle activity, muscle fatigue and perceived muscle fatigue It is known that repetitive tasks involve muscle activity and the muscles involved are dependent upon the type of task executed. In this study, the muscles involved were the forearm muscles, Flexor Carpi Radialis (left and right) and Extensor Carpi Radialis (left and right). The muscle activity, expressed as the RMS value (%MVC), is observed to increase significantly as the production standard time becomes harder. The results of this study also revealed that the average muscle activity increases by 6.9 and 12.9% for hard and very hard production standard time, respectively, whereby the values were determined relative to the normal production standard time. This indicates that the muscle activity is twice its initial value as the production standard time shifts from hard to very hard. The result is consistent with the findings of previous studies which discovered that an increase in work pace will increase the muscle activity (Escorpizo and Moore, 2007; Gooyers and Stevenson, 2012). In this study, the maximum muscle activity occurs at the Extensor Carpi Radialis-Left muscle with a value of 12.702%MVC at 140% of the normal standard time. The Extensor Carpi Radialis-Left muscles exhibit the highest muscle activity, followed by the Flexor Carpi Radialis-Left, Extensor Carpi Radialis-Right and Flexor Carpi Radialis-Right. The results are also consistent with the findings of previous studies which were focused on repetitive hand tasks, in which the forearm extensor muscle activity increases with respect to time (Finneran & O'Sullivan, 2013; Gooyers and Stevenson, 2012). The highest indication of muscle fatigue rate was also detected

Table 14 Regression model summary for work productivity. Model

R

R square

Adjusted R square

Std. Error of estimate

Work productivity

0.784

0.614

0.594

0.1135

66

M.N. Nurhayati et al. / International Journal of Industrial Ergonomics 56 (2016) 59e68

in the Extensor Carpi Radialis-Left muscle with 94.9% increment in very hard production standard time. The left muscles are found to be more active compared to right muscles, which is due to the possibility that the subjects are all right handed. Moreover, the left muscles are used to grip and hold the components while carrying out a task and thus they have a greater number of active motor units (Gnitecki et al., 2002). The greater the motor unit recruitment and firing rate, the greater the force generated (Merletti and Parker, 2004). The force generated shows a rapid decline when the muscles are stimulated continuously at a frequency close to the maximal force and leads to muscle fatigue (Allen et al., 2008). It is found in this study that the RMS value increases with respect to time for all muscles, which signifies the development of muscle fatigue. This observation is in agreement with the results of previous studies which highlighted that higher muscle activity (RMS) corresponds to higher discomfort (Kuijt-Evers et al., 2007) and an increase in the muscle fatigue rate. Muscle fatigue is an initiating factor of WMSDs (Takala, 2002) and the rate of muscle fatigue is reflected by the variations of muscle activity with respect to time (Konrad, 2005). In general, it can be deduced that the muscle fatigue rate is higher in harder production standard times due to the shorter task time and higher frequency of movement. The muscle fatigue rate increases with an increase in muscle activity. An increase in muscle activity indicates the development of muscle fatigue. Ma et al. (2009) also found that the accumulation of muscle fatigue results in functional disability and musculoskeletal disorders. The results agree well with the results of previous studies in which an increase in muscle activity results in increased muscle fatigue (Basmaijan and De Luca, 1985) and WMSD risks (Selen et al., 2006; Visser, 2004). Muscle fatigue and WMSD risks can also be assessed based on perception. In the workplace, the perceived exertion and perceived muscle fatigue that a person experiences can be a greater indicator of musculoskeletal risks (Borg, 1998). Perceived muscle fatigue is the psychological acute response investigated in the study, and it is found that the perceived muscle fatigue is higher at harder production standard times. It is observed from the results that the Extensor Carpi RadialisLeft exhibit the highest muscle activity, and thus the perceived muscle fatigue is also the highest for this muscle in hard and very hard production standard times. The percentage of increment in very hard production standard time is the highest with 63.3%. The perceived muscle fatigue increases significantly with an increase in the work productivity target for all muscles. These results conform with the muscle activity and muscle fatigue results obtained from objective measurements (surface electromyography). The average discomfort level for all muscles is close to ‘scale 5’ (strong), which reveals that the workers experienced higher discomfort while working in harder production standard times and they were more at risk of contracting WMSDs. The results for perceived muscle fatigue strengthen the objective measurements using EMG, which indicates the presence of muscle fatigue after the workers performed repetitive tasks. This suggests that higher perceived muscle fatigue is indicative of higher muscle fatigue rate and WMSD risks. The results are in agreement with the observations of Adams et al. (2010), in which the rating for the perceived muscle fatigue is higher as the tasks become more repetitive. In relation to the work productivity results discussed earlier, it is proposed that the reduction in work productivity performance is associated with an increase in muscle activity and muscle fatigue rate which is prevalent in harder production standard times. Past studies have also shown that workers tend to slow down when they feel fatigued due to WMSD symptoms, which reduces work productivity (Resnick and Zanotti, 1997; Xu et al., 2012).

4.3. Relationship between variations of acute responses and work productivity at different levels of production standard time Correlation and regression analysis was performed to further investigate the relationship between work productivity and acute responses at different levels of production standard time. The correlation results form the groundwork in investigating the relationship between all factors. There is significant correlation between muscle activity and work productivity (R ¼ 0.630, p < 0.01) as well as between perceived muscle fatigue and work productivity (R ¼ 0.510, p < 0.01). Linear regression analysis was implemented to examine the relationship among muscle activity, perceived muscle fatigue and work productivity. It is found that there is a strong relationship between these variables (R ¼ 0.784) at different levels of production standard time. A regression model which predicts work productivity has been developed. The Pearson correlation coefficient (R) is most frequently used to determine how well a model fits a set of data. In general, an R-value above 0.5 indicates that there is a strong correlation among variables. Thus, it is evident that multiple linear regression model will give the best fit to predict work productivity (Cohen, 1988; Pallant, 2013). The adjusted R Square value represents the variance of work productivity, and the value is obtained to be 0.594. This indicates that approximately 60% of the total variance in work productivity can be explained by the regression model. The work productivity regression model is directly related to muscle activity and muscle fatigue. The results are supported by the findings of Coorevits et al. (2005) which also showed that linear regression techniques are still the preferred techniques to study muscle fatigue. Solnik et al. (2010) also developed a linear regression model which describes the variations of electromyography signal frequency during submaximal isometric contractions. The results show that acute responses (muscle fatigue and perceived fatigue) have a significant relationship with work productivity. Acute responses are indicators of WMSD risks, and it can be inferred that work productivity is correlated with WMSDs. The findings are consistent with the results of previous studies (Resnick and Zanotti, 1997; Xu et al., 2012). The regression model developed in this study predicts the work productivity of workers as a function of muscle activity, perceived muscle fatigue and production standard time, and it shall be highlighted that this relationship has never been proposed within other previous studies. This model is useful in assessing work productivity for an existing or new product in order to ensure that the productivity of workers is sustainable while minimizing the risk of WMSDs. The model was validated using data from another group of industrial workers. The results show that the SEE values for the model are relatively small and fairly close to the SEE of the model, which validates the model. SEE value is evidence of validation (Portney and Watkins, 2009). According to Kutner et al. (2004), if the SEE of prediction is fairly close to SEE of the regression model, then the SEE for the regression model is not seriously biased and gives an appropriate indication of the predictive ability of the model. It shall be noted that there is no upper limit for the SEE values. The SEE measures the differences between the measured and predicted value and the smallest value for the SEE is 0, which indicates that all of the points fall along the equation line. 5. Conclusion The work productivity target (in terms of the percentage of normal production standard time) is not attainable for hard and very hard production standard times. This is attributed to the manifestations of acute responses (muscle activity, muscle fatigue, and perceived muscle fatigue), which increase as the production

M.N. Nurhayati et al. / International Journal of Industrial Ergonomics 56 (2016) 59e68

standard time becomes harder. The muscle activity increases twice from its original percentage value (normal production standard time) as the production standard time shifts from hard (6.9%) to very hard (12.9%). The increase in muscle activity increases the muscle fatigue rate, which in turn decreases work productivity performance. There is a strong relationship between muscle activity, perceived muscle fatigue and work productivity at different levels of production standard time. The relationship among these variables is found to be significantly linear (R ¼ 0.784, p < 0.01). Hence, a regression model has been developed which predicts work productivity as a function of muscle activity, perceived muscle fatigue and production standard time. The model has been validated using SEE and the model can be used as a reference for work productivity planning, which will be beneficial to organizations to minimize the risk of WMSDs among workers who constantly perform repetitive tasks in the manufacturing industry. Relevance to industry This study highlights the relationship between work productivity and acute responses at different levels of production standard time. The knowledge on the relationship is important to design tasks according to the capability of the workers to sustain their work productivity and to minimize the risk of developing WMSDs. Acknowledgement The authors gratefully acknowledge the financial support provided by the Ministry of Higher Education, Malaysia, under the High Impact Research Grant, UM.C/HIR/MOHE/ENG/35. References Adams, K., DeBeliso, M., Sevene-Adams, P., Berning, J., Miller, T., Tollerud, D., 2010. Physiological and psychophysical comparison between a lifting task with identical weight but different coupling factors. J. Strength Cond. Res. 24 (2), 307e312. Alavinia, S.M., Burdoff, A., 2009. Productivity loss in the workforce: associations with health, work demands, and individual characteristics. Am. J. Ind. Med. 52 (1), 49e56. Allen, D.G., Lamb, G.D., Westerblad, H., 2008. Skeletal muscle fatigue: cellular mechanisms. Physiol. Rev. 88 (1), 287e332. Allesina, S., Azzi, A., Battini, D., Regattieri, A., 2010. Performance measurement in supply chains: new network analysis and entropic indexes. Int. J. Prod. Res. 48 (8), 2297e2321. Bao, S., Silverstein, B., Cohen, M., 2001. An electromyography study in three high risk poultry processing jobs. Int. J. Ind. Ergon. 27 (6), 375e385. Basmaijan, J., De Luca, C.J., 1985. Muscles Alive;: Their Functions Revealed by Electromyography, fifth ed. Williams and Wilkins, Baltimore. Borg, G., 1998. Borg's Perceived Exertion and Pain Scales. Human Kinetics Publisher. €n, J.H., 2009. ElectromyoBosch, T., de Looze, M.P., Kingma, I., Visser, B., van Diee graphical manifestations of muscle fatigue during different levels of simulated light manual assembly work. J. Electromyogr. Kinesiol. 19 (4), e246ee256. €n, J.H., 2011. The effect Bosch, T., Mathiassen, S.E., Visser, B., de Looze, M.P., van Diee of work pace on workload, motor variability and fatigue during simulated light assembly work. Ergonomics 54 (2), 154e168. Cohen, J., 1988. Statistical Power Analysis for the Behavioral Sciences, second ed. Lawrence Erlbaum Associates. Conway, H., Svenson, J., 2001. Musculoskeletal disorders and productivity. J. Labor Res. 22 (1), 29e54. Coorevits, P.L.M., Danneels, L.A., Ramon, H., Van Audekercke, R., Cambier, D.C., Vanderstraeten, G.G., 2005. Statistical modeling of fatigue-related electromyographic median frequency characteristics of back and hip muscles during a standardized isometric back extension test. J. Electromyogr. Kinesiol. 15, 444e451. De Luca, G., 2003. Fundamental Concepts in EMG Signal Acquisition. Delsys Inc. Escorpizo, R., 2008. Understanding work productivity and its application to workrelated musculoskeletal disorders. Int. J. Ind. Ergon. 38, 291e297. Escorpizo, R.S., Moore, A.E., 2007. Quantifying precision and speed effects on muscle loading and rest in an occupational hand transfer task. Int. J. Ind. Ergon. 37, 13e20. Finneran, A., O'Sullivan, L., 2010. Force, posture and repetition induced discomfort as a mediator in self-paced cycle time. Int. J. Ind. Ergon. 40, 257e266.

67

Finneran, A., O'Sullivan, L., 2013. Effects of grip type and wrist posture on forearm EMG activity, endurance time and movement accuracy. Int. J. Ind. Ergon. 43, 91e99. Gerleman, D.G., Cook, T.M., 1992. Instrumentation. In: Soderberg, G.L. (Ed.), Selected Topics in Surface Electromyography for Use in the Occupational Setting: Expert Perspectives. U.S. Department of Health and Human Services-National Institute for Occupational Safety and Health. Gnitecki, J.E., Kler, G.P.S., Moussavi, Z., 2002. EMG signs of fatigue in anterior deltoid and posterior deltoid muscles: questioning the role of RMS during fatigue. In: Canadian Medical and Biological Society (CMBES). Gooyers, C.E., Stevenson, J.M., 2012. The impact of an increase in work rate on task demands for a simulated industrial hand tool assembly task. Int. J. Ind. Ergon. 42, 80e89. Hamberg-van Reenen, H.H., van der Beek, A.J., Blatter, B.M., van der Grinten, M.P., van Mechelen, W., Bongers, P.M., 2008. Does musculoskeletal discomfort at work predict future musculoskeletal pain? Ergonomics 51 (5), 637e648. Konrad, P., 2005. The ABC of EMG. A Practical Introduction to Kinesiological Electromyography. Noraxon Inc. USA. Noraxon Inc., USA. Kuijt-Evers, L.F.M., Bosch, T., Huysmans, M.A., de Looze, M.P., Vink, P., 2007. Association between objective and subjective measurements of comfort and discomfort in hand tools. Appl. Ergon. 38 (5), 643e654. Kutner, M.H., Nachtsheim, C.J., Neter, J., Li, W., 2004. Applied Linear Statistical Models, fifth ed. McGraw-Hill Higher Education. Lehman, K.R., Psihogiost, J.P., Meulenbroek, R.G.J., 2001. Effects of sitting versus standing and scanner type on cashiers. Ergonomics 44 (19), 719e738. Lerner, D., Amick, B.C., Lee, J.C., Rooney, T., Rogers, W.H., Chang, H., Berndt, E.R., 2003. Relationship of employee-reported work limitations to work productivity. Med. Care 41 (5), 649e659. Ma, L., Chablat, D., Zhang, W., 2009. Dynamic muscle fatigue evaluation in virtual working environment. Ergonomics 39 (1), 211e220. Mananas, M.A., Rojas, M., Mandrile, F., Chaler, J., 2005. Evaluation of muscle activity and fatigue in extensor forearm muscles during isometric contractions. In: 27th Annual International Conference of Engineering in Medicine and Biology Society. Shanghai, China. Merletti, R., Parker, P.J., 2004. Electromyography: Physiology, Engineering, and Noninvasive Applications. Wiley-IEEE Press. Milerad, E., Ericson, M., 1994. Effects of precision and force demands, grip diameter, and arm support during manual work: an electromyographic study. Ergonomics 37 (2), 255e264. Montgomery, D.C., 2014. Design and Analysis of Experiments, eighth ed. John Wiley & Sons. Nag, P.K., Pal, S., Nag, A., Vyas, H., 2009. Influence of arm and wrist support on forearm and back muscle activity in computer keyboard operation. Appl. Ergon. 40 (2), 286e291. Neumann, W.P., Kihlberg, S., Medbo, P., Mathiassen, S.E., Winkel, J., 2002. A case study evaluating the ergonomic and productivity impacts of partial automation strategies in the electronics industry. Int. J. Prod. Res. 40 (16), 4059e4075. NIOSH, 1997. Musculoskeletal Disorders and Workplace Factors: a Critical Review of Epidemiologic Evidence for WRMDs. NIOSH Publication No. 97-141, Cincinnati. Nordander, C., Ohlsson, K., Akesson, I., Arvidsson, I., Balogh, I., Hansson, G.A., Skerfving, S., 2009. Risk of musculoskeletal disorders among females and males in repetitive/constrained work. Ergonomics 52 (10), 1226e1239. Pallant, J., 2013. SPSS Survival Manual: a Step by Step Guide to Data Analysis Using IBM SPSS, fifth ed. Open University Press. Portney, L.G., Watkins, M.P., 2009. Foundations of Clinical Research, third ed. Pearson/Prentice Hall. Potes, C.M., 2008. Assessment of Human Muscle Fatigue from Surface EMG Signals Recorded during Isometric Voluntary Contractions by Using a Cosine Modulated Filter Bank. Department of Electrical and Computer Engineering, University of Texas at El Paso (Master thesis). Putz-Anderson, V., 1988. Cumulative Trauma Disorders: a Manual for Musculoskeletal Diseases of the Upper Limbs. Taylor & Francis. Resnick, M.L., Zanotti, A., 1997. Using ergonomics to target productivity improvement. Comput. Ind. Eng. 33 (1e2), 185e188. Rietveld, S., Van Beest, I., Kamphuis, J.H., 2007. Stress-induced muscle effort as a cause of repetitive strain injury? Ergonomics 50 (12), 2049e2058. Selen, L.P.J., van Dieen, J.H., Beek, P.J., 2006. Impedance modulation and feedback corrections in tracking targets of variable size and frequency. J. Neurophysiol. 96 (5), 2750e2759. Shikdar, A., Das, B., 2003. A strategy for improving worker satisfaction and job attitudes in a repetitive industrial task: application of production standards and performance feedback. Ergonomics 46 (5), 466e481. SOCSO, 2011. Annual Report 2011. Social Security Organisation (SOCSO). Solnik, S., Devita, P., Grzegorczyk, K., Koziatek, A., Bober, T., 2010. EMG frequency during isometric, submaximal activity: a statistical model for biceps brachii. Acta Bioeng. Biomech. 12 (3), 21e28. Sundelin, G., Hagberg, M., 1992. Electromyographic signs of shoulder muscle fatigue in repetitive arm work paced by the methods-time measurement system. Scand. J. Work, Environ. Health 18 (4), 262e268. Takala, E., 2002. Static muscular load, an increasing hazard in modern information technology. Scand. J. Work Environ. Health 28 (4), 211e213. Van den Heuvel, S.G., Geuskens, G.A., Hooftman, W.E., Koppes, L.L.J., van den Bossche, S.N.J., 2010. Productivity loss at work; Health-related and work-related factors. J. Occup. Rehabil. 20 (3), 331e339. Visser, B., 2004. Effects of precision demands and mental pressure on muscle

68

M.N. Nurhayati et al. / International Journal of Industrial Ergonomics 56 (2016) 59e68

activation and hand forces in computer mouse tasks. Ergonomics 47, 202e217. Wartenberg, C., Dukic, T., Falck, A.C., Hallbeck, S., 2004. The effect of assembly tolerance on performance of a tape application task: a pilot study. Int. J. Ind. Ergon. 33, 363e379. Westgaard, R.H., Winkel, J., 1996. Guidelines for occupational musculoskeletal load as a basis for intervention: a critical review. Appl. Ergon. 27 (2), 79e88.

Xu, Z., Ko, J., Cochran, D.J., Jung, M., 2012. Design of assembly lines with the concurrent consideration of productivity and upper extremity musculoskeletal disorders using linear models. Comput. Ind. Eng. 62 (2), 431e441. You, H., Kwon, O., 2005. A survey of repetitiveness assessment methodologies for hand e intensive tasks. Int. J. Ind. Ergon. 35, 353e360.