Increasing the degree of automation in a production system: Consequences for the physical workload

https://doi.org/10.1016/j.ergon.2004.09.009Get rights and content

Abstract

In spite of the continuous development of production systems work-related musculoskeletal disorders is still a large problem. One reason might be the difficulties in quantifying the ergonomic effects of interventions. In this paper ergonomic consequences of technical and organisational changes were quantified in a plant for producing slats for parquet flooring. Muscle activity, work postures and movements were assessed for neck/shoulders and upper limb by direct technical measurements at three generations of production lines. The physical workload for 31 operators at the manual, semi-automated and automated line was derived based on all existing work tasks. The work was characterised by moderate muscular loads, the 50th percentiles being 1.2–3.8%MVE for the neck/shoulder muscles, high repetitiveness and constant movements of the hands and a high prevalence of neck/shoulder disorders. There were statistically significant differences considering exposure levels between the work tasks within each line as well as between the lines. The semi-automated line implied reduced muscular load for all muscles registered but more constrained work postures. The automated line, on the other hand, required higher skills, offered less constrained postures, lower loads and repetitiveness for the hands as well as frequent changes between different physical load levels.

Relevance to industry

When designing or redesigning production systems, technical solutions are often used to obtain ‘beneficial’ ergonomic conditions, implying a strive for low physical workloads. However, there is a risk that this might lead to more constrained physical activities. Possibilities for job enlargement must be thoroughly utilised.

Introduction

During the last decades, various efforts have been made to improve working conditions, as well as efficiency in the production systems. In spite of such efforts, work-related musculoskeletal disorders are still very common (Bernard, 1997; Official Statistics of Sweden, 2000; Coury, 1999; NRC, 2001). Such disorders often lead to prolonged sick-leave periods, disability pensions and decreased life quality (Pålsson et al., 1998). The economical consequences are huge, e.g. 0.5–2% of the gross national products in the different Nordic countries (Toomingas, 1998; Johansson, 2000; Norlund et al., 2000). In the United States the annual cost for the society amounts to 13 billion USD (NIOSH, 1996).

The lack of success of ergonomic interventions is probably due to the shortage of knowledge concerning ergonomic consequences of changes in production systems (Westgaard and Winkel, 1997). For example, one obstacle has been the lack of suitable methods to measure physical exposure as a consequence of the interventions. However, several ‘field methods’ are now available (Hansson et al., 1997, Hansson et al., 2000a; Stål et al., 2000; Unge Byström et al., 2002; Juul-Kristensen et al., 2002). Hence, case studies of ergonomic interventions should now be possible to carry out in a more conclusive and feasible way (Cole et al., 2003). The need of such studies are obvious—especially such studies that in some ways also relate to various types of tangible technical dimensions. For example, these can be choice of layouts and materials feeding techniques, work group size and organisation.

When investigating the infliction of ergonomic interventions on the human being, it is essential that the achieved data are in a suitable form to be able to link them to the actual production system design. Especially so, concerning the fact that there is an inarticulate zone as to what is considered as ‘healthy’ or ‘unhealthy’ work within the manufacturing industry.

During recent years, a number of ergonomic recommendations and regulations, national (e.g. Kilbom, 1994; AFS, 1998:1; ACGIH, 2001), as well as international (e.g. European Committee for standardization, 1996; ISO11226, 2000), have been introduced. However, these recommendations and regulations are not fully implemented and are in practice seldom used as a tool for the actual design of production systems. A more proactive and technically based type of ergonomic interventions are actually called for. This article will to some extent illuminate these issues.

This case study was conducted at a plant processing wooden boards for parquet flooring in the south of Sweden. There is, for several years, an ongoing stepwise mechanisation/automation of the manufacturing process aiming at improved manufacturing efficiency and product quality as well as improved ergonomic conditions. As a result, three different production lines are used simultaneously, each of them representing the state of art of mechanisation/automation at the time when they were designed and introduced. Consequently, they comprise different technical and ergonomic measures for efficiency and work conditions. In fact, these three generations of production lines represent different stages of an evolution from a manual towards an automated manufacturing process. Thus, this case study offers an unique possibility to assess the ergonomic effects of tangible technical evolution.

Accordingly, the main aim of this case study is to quantify and evaluate eventual change in physical workload in the form of muscular load as well as work postures and movements in the neck and the upper extremities, as a consequence of the stepwise technical development of the three generations of production system designs.

Section snippets

Production system design

On an overarching level, the production systems in the company process wooden boards for parquet flooring. The production systems studied consisted of three production lines; (1) the manual line, (2) the semi-automated line and (3) the automated line.

The manual line is the oldest production line (Table 1; Fig. 1A). The work is organised along two parallel conveyors with a production rate of 2 slats/s each. Four workstations are organised along each conveyor, thus attended by in total eight

Study design

The study base comprised all 152 currant female operators at the Hardwood Grading Department. They were all interviewed and had a physical examination of neck and upper limb.

For quantifying the physical workload direct technical measurements were performed. Five unique work tasks were chosen, i.e. for the manual line ‘inspection and sorting by surface’, ‘inspection and sorting by edges’ and ‘piling’ and for the semi-automated line ‘inspection and sorting by surface’ and ‘inspection and sorting

Operators

The mean age for the 152 operators was 38.5 (range 20–61) years and the mean employment time was 9.8 (0.25–38) years. Their work schedules were based on two- and three shift as well as work only night time.

By a structured interview, data on subjective complaints were registered for the last 12 months as well as for the last 7 days (Kuorinka et al., 1987). Further, a standardised physical examination of the neck and upper extremity was performed (Ohlsson et al., 1994a). Diagnoses were decided on

General

There were no systematic differences between the right and left side comparing the different tasks between the lines. Hence, only the results from the right side are presented. However, all data are available on our web http://www.ymed.lu.se/papers/appendix/Increasingautomation.pdf.

The work, when producing slats for parquet flooring was, in general, characterised by moderate muscular loads (Table 2, Table 3, Table 4). Most of the work tasks were performed with moderate flexion of the head, with

General

Producing slats for parquet flooring implied moderate muscular loads, moderate head flexion, somewhat elevated upper arms and constant hand movements, yet these exposures involved high prevalence of musculoskeletal disorders. The exposure circumstances were most pronounced for the manual line. By the stepwise automation the physical exposure changed significantly. Thus, the semi-automated line displayed an overall decreased muscular activity. Moreover, the technical development resulted in

Acknowledgements

Ms. Lothy Granqvist, Christina Glans and Jeanette Unge gave skilful technical assistance. The case study was supported by grants from the Swedish Council for Work life Research (including the program Change@Work), the Swedish Medical Research Council, the Swedish National Institute of Working Life (Co-operative for Optimization of industrial production lines regarding Productivity and Ergonomics; COPE), the Swedish Council for Planning and Coordination of Research, the Medical Faculty of Lund

References (51)

  • R.H. Westgaard et al.

    Guidelines for occupational musculoskeletal load as a basis for intervention: a critical review

    Applied Ergonomics

    (1996)
  • ACGIH, 2001. Documentation of the threshold limit values for physical agents. Ergonomics. In: Proceedings of the...
  • Ergonomics for the Prevention of Musculoskeletal Disorders

    (1998:1)
  • I. Åkesson et al.

    Quantifying workload in neck shoulders and wrists in female dentists

    International Archives of Occupational and Environmental Health

    (1997)
  • I. Balogh et al.

    Interindividual variation of physical load in a work task

    Scandinavian Journal of Work, Environment & Health

    (1999)
  • D.C. Cole et al.

    Methodological issues in evaluating workplace interventions to reduce work-related musculoskeletal disorders through mechanical exposure reduction

    Scandinavian Journal of Work, Environment & Health

    (2003)
  • H.J.C.G. Coury

    The effects of production changes on the musculoskeletal disorders in Brazil and South America

    International Journal of Industrial Ergonomics

    (1999)
  • Safety of Machinery—Human Physical Performance—Part 4: Evaluation of Working Postures in Relation to Machinery

    (1996)
  • N. Fallentin et al.

    Physical exposure assessment in monotonous repetitive work—the PRIM study

    Scandinavian Journal of Work, Environment & Health

    (2001)
  • N. Fallentin et al.

    Evaluation of physical workload standards and guidelines from a Nordic perspective

    Scandinavian Journal of Work, Environment & Health

    (2001)
  • A.H. Garde et al.

    The influence of production systems on physiological responses measured in urine and saliva

    Stress and Health

    (2003)
  • G.-Å. Hansson et al.

    Kinematic evaluation of occupational work

    Advances in Occupational Medicine and Rehabilitation

    (1997)
  • Hansson, G.-Å., Asterland, P., Skerfving, S., 1997. Acquisition and analysis of whole-day electromyographic field...
  • G.-Å. Hansson et al.

    Questionnaire versus direct technical measurements in assessing postures and movements of the head, upper back, arms and hands

    Scandinavian Journal of Work, Environment & Health

    (2001)
  • Cited by (50)

    • Biomechanical exposure of industrial workers – Influence of automation process

      2018, International Journal of Industrial Ergonomics
      Citation Excerpt :

      While the operators on the manual line had to manually handle the clutch discs, position the component parts, and manually activate the machinery of the production system, the operators on the semi-automated line only performed the manual positioning of the component parts. This is in accordance with Balogh et al. (2006) and Arvidsson et al. (2012) who also found higher angular velocities for head and upper arms on the most manual system when comparing different automated processes in parquet flooring and meat cut industries, respectively. However, in both the study of Balogh et al. (2006) and Arvidsson et al. (2012) a higher muscular activity of the trapezius muscles was also found on the most manual system, which was not observed in our study.

    • Job rotation designed to prevent musculoskeletal disorders and control risk in manufacturing industries: A systematic review

      2017, Applied Ergonomics
      Citation Excerpt :

      Of the final 14 studies, two were non-randomized control trials (Guimaraes et al., 2012; Fredriksson et al., 2001), one had case-control design (Roquelaure et al., 1997), and 11 were of cross-sectional designs (Table 2). In terms of outcomes, five studies targeted MSD prevention (outcomes included measures of disability, disorders, pain, discomfort) (Balogh et al., 2006; Guimaraes et al., 2012; Fredriksson et al., 2001; Roquelaure et al., 1997; Sato and Coury, 2009), 11 studies targeted changes in exposure to physical risk factors (biomechanics, repetition, fatigue, effort exertion) (Asensio-Cuesta et al., 2012a, 2012b; Balogh et al., 2006; Carnahan et al., 2000; Diego-Mas et al., 2009; Frazer et al., 2003; Fredriksson et al., 2001; Filus and Okimorto, 2012; Roquelaure et al., 1997; Sato and Coury, 2009; Tharmmaphornphilas and Norman, 2007), and three targeted the job-rotation effect on psychosocial factors (job satisfaction) (Dawal et al., 2009; Dawal and Taha, 2007; Guimaraes et al., 2012). The indicators and tools used to evaluate the outcomes varied widely in the studies.

    View all citing articles on Scopus
    View full text