1 Introduction and Aims

There are several methods, with different approaches and itemization, for the assessment and evaluation of physical workload and stress in manual tasks. But there are restrictions: none of the existing methods have yet been validated for short cycle activities of 30 s or less.

The outcome of two surveys (see state of research for further details) reveals that the results of the different methods of assessment vary. Hence it is unclear which of the existing methods should be chosen in practice to get conclusive results regarding the intensity of the workload and the implications of this intensity for the analyzed workplace.

But to be able to validate those existing methods for repetitive assembly tasks the level of stress and strain of the employees working regularly on such workplaces must have been quantified. To do so, as part of this study and as a first step, 249 associates have been interviewed with a questionnaire and 66 different workplaces have been analyzed. The results are presented in this paper. As a next step, the statistical analysis of these results will show potentials in the methods as well as suggestions as to which method is the best one to use for the assessment of repetitive assembly tasks.

2 State of Research

A significant increase in productivity and a reduction in costs was achieved in recent years through great progress in the field of automation of production process. Yet automation is often not cost effective because of short innovation cycles and high investment demand. The advantage of manual production in manufacturing lies in higher flexibility and skills of employees. However, in order to plan long term with the staff, the assembly tasks must be executable long into an employees’ career without causing long-term health conditions.

In order to detect and prevent improper biomechanical stress in time, an explicit ergonomic evaluation of the workplaces is needed, preferably in the planning phase. There are a variety of methods available, for instance OCRA checklist (CL), ISO 11228-3, DIN EN 1005-5, RULA or EAWS, most of which have been theoretically and generally compared by Takala et al. (2010) and Steinberg et al. (2011). The assessment methods show a high convergent validity in these theoretical precomparisons. In particular, it was found earlier (Steinberg et al. 1998) that the different methods show similar results in average conditions. The results differ increasingly only in an extreme expression of individual characteristics.

Within the manual assembly tasks analyzed in the study, the repetitive work provides an extreme expression of individual characteristics when the activity rate is elevated. Silverstein et al. (1986) propose that manual activity be described as highly repetitive when a continuously repeating cycle takes less than 30 s. This limit also appears as an approximate assessment value in the evaluation of machine related repetitive work in the standard DIN EN 1005-5 (2007). Studies in which the results of the methods are compared with each other based on real, repetitive (with cycle times of 30 s or less) workplaces, show large differences in the assessments (Bruder and Kaiser 2012; Heidl 2013). These differ so much, that the same activity is classified in a green zone in a first method, in a yellow in a second and in a red zone in a third method.

Although effects from warming up (Debitz 2005) and training (Joiko et al. 2010) can be observed, experts say that frequent similar movements in the hand/arm area can lead to a variety of ailments. Hartmann et al. (2013) mention styloiditis, carpal tunnel syndrome and tendonitis as examples of diseases that result from repetitive actions. It makes sense in theory, but is there a difference in practice? Employers must be keen to ensure that the workload can be assessed realistically during all manual assembly tasks.

Wittig et al. (2013), in their study of the effects of repetitive jobs, provide statistics regarding the occurrence of complaints and pain sensations, as they collected data of the complaints in the back, neck, shoulders, arms, hands and fingers of volunteers of different professions. Kotzab (2014) and Steinberg et al. (2011) also investigated the complaint perception of production employees and included the back, neck, shoulders, elbows, forearms and hands. The exact values are given in the discussion.

3 Method

To create a database for stress and strain in repetitive assembly work, 66 workplaces in five different plants of a German automotive supplier were recorded and evaluated. Therefore, all default values, such as exercise frequency, weights, distances, breaks etc. were taken.

In addition, a total of 249 subjects, who regularly work in the selected workplaces, were asked about their subjective perceived exertion and their complaints. For this, a questionnaire was composed, which included the following matters:

  • General information on the person, such as age, gender, handedness, working hours, work experience and additional sport load on the upper extremities

  • A BORG-scale (Borg 2004) for the self-assessment of stress and strain of the daily workload and the specific load of the selected tasks

  • Selected features of the questionnaire for the subjective assessment of workplace exposure of Slesina (1987, 2009) (see Figs. 1 and 3)

    Fig. 1
    figure 1

    Header of the questionnaire for the subjective assessment of workplace exposure

    Full size image

  • An expanded form of the Nordic questionnaire (following Kuorinka (1987), see Fig. 2) including a detailed form of the upper extremities

    Fig. 2
    figure 2

    Excerpt of the Nordic questionnaire, expanded by more detailed upper extremities

    Full size image

The response to the questionnaire was voluntary and anonymous. After a brief introduction the forms were filled in independently by the subjects. The investigator was nearby, in case of questions.

The mental aspects were left aside in the subjective assessment of workplace exposure. The questionnaire, as well as the study, was limited to the assessment of physical stress and strain, therefore the questionnaire of Slesina was reduced to a total of 29 features. Interviewing remained the same, both the occurrence as well as the subjective stress/strain was requested.

Once the analysis is completed, the selected workplaces will be evaluated with the assessment methods and the results of those will be compared. In recent studies (Heidl 2013) the methods EAWS (Ergonomic Assessment Worksheet), OCRA checklist (Occupational Risk Assessment), LMM MA (Key Indicator Method for assessing physical workload during Manual Handling Operations), HARM (Hand Arm Risk Assessment) and SI (Strain Index) have shown the best results with repetitive assembly tasks and therefore were selected to analyze the workplaces in this study.

4 Results

The sample of surveyed employees consists of 249 production staff from five different plants of a German automotive supplier. The proportion of female subjects is 58.6 %. Furthermore, 60.6 % of the respondents are younger than 50. On average, the interviewed employees have worked in their current roles on the assembly lines for at least 5.54 years (SD: 5.64) and have already been working for 13.16 years (SD: 10.28). 31.7 % of the subjects, in their spare time, perform sport or exercise such as weight training, martial arts, handball, etc. that involved the upper body. In addition to the physically strenuous work, 18.9 % did some upper body exercise or sport in the past.

A total of 66 different work tasks with cycle times between 5 and 64 s were examined. There are six workstations with cycle times of more than 30 s because several work pieces (up to ten) are handled within one cycle. The time per piece therefore is much shorter. The overall average of cycle time amounts to 19.68 s (SD 11.78).

The occurrence and the perceived strain of the 29 items selected for the questionnaire for the subjective assessment of workplace exposure are shown in Fig. 3. The items are sorted by the perceived strain which is shown in red. The posture that causes the most perceived strain is marked as “unfavorable body posture” (73.5 %) and “forced body posture” (59.3 %), and these are the items with the highest scores. Work organization also seems to have an impact on the perceived strain as “shift work” and “weekend working” are rated at 58.2 % and 57.4 % respectively. Next is “standing” (55.7 %), followed by “monotonous work” and “cycle-dependent work” which all are rated over 50 % as causing “strain” by the surveyed employees.

Fig. 3
figure 3

Occurrence and perceived strain of the 29 selected items of the questionnaire for the subjective assessment of workplace exposure

Full size image

The frequency of the occurrence of strain symptoms can also be seen in Fig. 3. Where employees are concerned the most prevalent is “shift work” (91.4 %) and “cycle-dependent work” (85.4 %). Other frequent items are “dependent on the unit speed”, “accurate viewing of details”, “standing” and “weekend working” with 77.7 %, 76.4 %, 74.8 % and 74.7 % respectively. “Hand skill” (72.9 %), “time coping” (70.4 %) and “concentration” (70.2 %) were also rated as occurring over 70 % of the working time.

As explained in Sect. 3, the employees have also been questioned about the occurrence of complaints and pain in the upper extremities as well as neck and back. Only 18.07 % of the subjects stated that they were free of pain. The majority had complaints in the neck, the back and/or the shoulders. A summary is shown in Fig. 4: 63.05 % of the subjects complain about the neck, another 44.98 % indicate pain in the upper as well as the lower back. Over 40 % of the employees feel pain in the left (42.17 %) and right (43.78 %) shoulder. The upper extremities are mentioned much less. The complaints and pains range from 9.64 % (left upper arm) to 19.68 % (right elbow).

Fig. 4
figure 4

The occurrence of complaints and pain in the different body regions, n = 249

Full size image

The subjective rating of stress and strain via the CR10 scale of BORG provides mean values of low 1.33 to high 8. In fact, 5 workplaces are rated between “nothing at all” and “weak (light)” (0–2), 22 workplaces were considered “weak” to “moderate” (>2–3), 16 workplaces were classified between “moderate” and “somewhat strong” (>3–4), 15 places were categorized as “somewhat strong” to “strong (heavy)” (>4–5), and 8 places were rated higher than “strong” (>5–10). The overall average is calculated to be 3.64 (SD 1.50).

Finally, the selected assembly tasks were analyzed using five different assessment methods: EAWS, OCRA CL, LMM MA, SI and HARM. EAWS was used in two different versions (1.3.2c and 1.3.3) in order to see which implications are caused by the changes in praxis. Each assembly task was analyzed by two different assessors to reduce the influence of personal perception. The results can be seen in Table 1.

Table 1 Ratings of the five different assessment methods for the 66 different assembly tasks
Full size table

As you can see from the results above the assessors came to similar results in each area (green, yellow and red). It is only in the Strain index that the assessors’ results are different. The small differences in results for the other methods can be explained by rounding and subjective interpretation (e.g. number of actions, grip conditions).

Furthermore, the new version of EAWS seems to be stricter than version 1.3.2c as less assessment results tend to be in the green zone and more tend to the red. Nevertheless, EAWS rates most assembly tasks, as well as LMM MA and HARM, in between green and yellow. It is only OCRA CL and SI which are stricter. With OCRA, most work tasks are considered as red.

5 Discussion and Outlook

As already assumed in Sect. 2 (state of research), this study confirms the diverse results of the different assessment methods. It is, therefore, not obvious which method is best to evaluate the stress and strain of repetitive assembly tasks. Nor is it clear which one to choose in praxis.

Moreover, the assessment methods for “repetitive work” focus on overload through repetition, movement frequencies and/or on grip conditions. Evidenced by the study, this shows that this is an incomplete perspective.

As this study (with 249 subjects) was able to show, the highest perceived strain was not in the arms, but rather in the entire body. Particularly the neck, the back and the shoulders are stressed, a load that also then extends to the arms, hands and fingers.

Equally unexpected were the results of the 29 selected items of the questionnaire for the subjective assessment of workplace exposure. The perceived items causing the most strain were by far the different forms of body posture. “Standing” and “lack of exercise” were also rated as highly straining. Next are items of work organization such as “shift-“ and “weekend-working”, followed by items typically associated with repetitive work such as “monotonous work”, “cycle-dependent work” and “time coping”.

There are similar studies regarding the occurrence of complaints and pains in the area of environment assembly tasks. The comparison can be seen in Table 2. It shows the percentile values of complaints and pains as well as the difference between them and this study (bracketed).

Table 2 Comparison of complaint- and pain-values of different body parts of three different studies
Full size table

The results from complaints and pains of the upper back are lower, but results from the lower back are higher in comparison. Furthermore, in Kotzab’s study (2014) the complaints in the neck and shoulders are smaller, but higher for the wrists. Again, the pains in the elbows are in the same range.

Wittig et al. (2013) did a survey on behalf of the Federal Institute for Occupational Safety and Health (BAuA) with 15,239 employees of the working population. The subjects were distributed over all working sectors with the proportion of interviewees from “manufacturing” at 32 %. Compared to the subjects of our study, who are working continuously in repetitive assembly tasks, complaints of the lower back are approximately the same, but pains in neck and shoulders, and in the arms and the hands/fingers are considerably less.

Another comparison comes from the data of the researches of Kotzab (2014) and Steinberg et al. (2011), who recorded complaints and pains of production staff. Therefore, the comparison to these two studies should even be more interesting.

Steinberg et al. (2011) interviewed 1293 employees, of which approximately half perform manual assembly tasks. The other half of the interviewees do office work. The prevalence of symptoms in the musculoskeletal system is specified for the entire study population. The complaints of the population are continuously lower than those of this study. Complaints centered around the elbows are in approximately the same range, but for the other body regions there are major differences. In this study, complaints and pains of the back (by 10 %) and neck (by 7 %) are considerably higher than for the interviewees of Steinberg et al. (2011). The difference for the shoulders, forearms and hands is even higher, as they differ by as much as 15 %.

Kotzab (2014) on the other hand studied three different assembly lines with cycle times of 25–84 s and interviewed 621 employees of these lines. Therefore, Kotzab’s study is most similar to this one and, therefore, the results mostly comparable. The results from complaints and pains of the upper back are lower, but results from the lower back are higher in comparison. Furthermore, in Kotzab’s study (2014) the complaints in the neck and shoulders are smaller, but higher for the wrists. Again, the pains in the elbows are in the same range.

In contrast with the results of this study and the comparison to other studies, the previous approach of the assessment methods focuses on the movement frequency in combination with force and grip conditions, as well as the joint position during the assembly tasks. Even though in some assessment methods the duration of the tasks and the breaks are already established, however there is more that should be done.

In future, repetitive tasks should be approached holistically. There are, by way of example, numerous work places where the employee has to be standing, which leads, as has been proven (amongst others by Garcia et al. 2015), to discomfort in the legs and back. But standing is only considered in EAWS, and when considering standing they are only considering the strain on the whole body, which lies in a different area than the one (Sect. 4) regarding repetitive loads. Moreover, the neck, which causes complaints in most of the interviewees, is not considered in any of the assessment methods. Those and other stressors of the whole body should be considered in the assessment of repetitive working tasks.

As a next step, all the selected working tasks will be analyzed in detail. They will be statistically evaluated as to which method is best for the assessment of manual repetitive working tasks. Thereupon suggestions will be worked out as to how to adapt this method to optimize the results.