ARTICLE IN PRESS

International Journal of Industrial Ergonomics 38 (2008) 900–909

www.elsevier.com/locate/ergon

Proposal of parameters to implement a workstation rotation system

to protect against MSDs
Michel Aptel, Franc- ois Cail, Anne Gerling, Olivier Louis
Biomechanics and Ergonomics Laboratory, Working Life Department, Institut National de Recherche et de Sécurité,
Avenue de Bourgogne, 54500 Vandoeuvre, France
Received 24 April 2006; received in revised form 12 December 2007; accepted 4 February 2008
Available online 1 April 2008

Abstract

The aim of this study was to propose parameters to be taken into account for implementing a workstation rotation system, which
reduces the risk of musculoskeletal disorders (MSDs). The case study has shown that the benefit provided by workstation rotation has a
positive impact on psychosocial factors, which partly explains why most operators are either not at all or only slightly stressed.
Workstation rotation is found to be relatively ineffective in relation to MSD prevention, notably on account of the lack of variability and
the intensity of the biomechanical demands. The assessment of a workstation rotation system intended to prevent MSDs must encompass
all the dimensions of the work. To achieve this, it is necessary to describe the rotation system in detail, ascertain the characteristics of the
population concerned, assess its experience of the work, and measure the degree of biomechanical demand of the workstations. A logic
diagram is proposed which can assist in arriving at a better identification of the conditions of success of a workstation rotation system. It
is based on four complementary dimensions: ergonomic study of the context, integration of scientific knowledge, mastery of the
implementation of the rotation system and evaluation and follow-up of the results.
Relevance to industry

This study provides information for those responsible for implementing a workstation rotation system in their enterprise. Parameters
and a logic diagram are proposed to assist them in their task.
r 2008 Elsevier B.V. All rights reserved.

Keywords: Workstation rotation; Parameters; MSDs; Risk factors; Multiskilling

1. Introduction (Rérat, 1985; MacLeod and Kennedy, 1993). From the

In meeting flexibility objectives, companies today seek
greater leeway than before in terms of labour deployment
and task distribution (St-Jacques et al., 2003). Multiskilling
and workstation rotation enable them to achieve this. The
former requires extending ranges of operator skills and is
reflected, particularly in organisational terms, by work-
station rotation. As a human resource tool, multiskilling is
used to ensure possible replacements (for sickness or
holidays) and to cope with increased workload, which
depends on the company’s orders and its sector of activity
Corresponding authors. Tel.: +33 383502194; fax: +33 383508706.
E-mail addresses: michel.aptel@inrs.fr (M. Aptel),
francois.cail@inrs.fr (F. Cail).
operator’s standpoint, workstation rotation implies regular
station changeover according to a cyclical order and a pre-
established work rate (Falardeau and Vézina, 2002). The
definition of workstation rotation embraces a number of
characteristics, such as the number of workstations
included, rotation frequency and order.
At present, application of this organisational device is
becoming increasingly popular in companies as a response
to the growth in musculoskeletal disorders (MSDs)
(Axelsson and Ponten, 1990; Henderson, 1992; Hinnen
et al., 1992; Ellis, 1999; Kuijer et al., 1999), which together
constitute the largest recognised occupational disease.
According to these authors, changing workstation in fact
diversifies the operator’s physical and psychological de-
mands. It therefore has an impact on both biomechanical

0169-8141/$ - see front matter r 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.ergon.2008.02.006
 ARTICLE IN PRESS
M. Aptel et al. / International Journal of Industrial Ergonomics 38 (2008) 900–909
and psychosocial risk factors influencing MSD appearance.
In the case of precision tasks, it has been shown that
workstation rotation is necessary to allow physiological
recovery associated with visual and muscular fatigue
(Mikami et al., 1987). Moreover, some authors (Rissen
et al., 2002) agree that workstation rotation has a positive
impact on the operator’s subjective perception of work. In
this respect, a comparative study of multiskilled employees
has revealed that multiskilled operators experience a feeling
of work satisfaction linked to skills development, lack of
monotony, more widespread work relationships and a
feeling of greater responsibility (Thomas et al., 1994). The
operator feeling of fatigue decreases, when light and heavy
workstations are alternated, because this change allows
him/her to recovery physically (Kuijer et al., 1999).
Workstation rotation, to have a positive effect on
reducing MSD risk factors, requires a real willingness
and participation on the part of operators in both
developing new skills and setting up the rotation system
(Baggerman, 1993; Ellis, 1999; Vézina et al., 2003).
Introducing this organisational device to unwilling opera-
tors could trigger the appearance of stress symptoms in
contradiction with the pursued objectives (Thomas et al.,
1994). Prior employee training is also necessary. In this
respect, Christmansson et al. (1999) showed that, in
repetitive manual assembly tasks, reorganisation of the
work resulted in a increase in the number of postural pains,
primarily on account of the operators not being skilled
enough to take on a wider variety of tasks. Moreover,
different demands at the various workstations must be
guaranteed to offer the operator the opportunity of
alternating work-related demands (Kuijer et al., 1999;
Roquelaure et al., 1997). The workstations occupied during
rotation must allow a variation in biomechanical demands
(St-Vincent et al., 2003). Demand diversification requires a
preliminary assessment of these different biomechanical
demands. This can be carried out on the basis of work
activity observation to identify the joint segment subjected
to the greatest demand at the workstation concerned (Van
Velzer, 1992). The order of the rotation system is then
defined on the basis of alternating low-demand and high-
demand workstations. For example, based on work activity
observation, Henderson (1992) allocates a value of between
1 and 3 to the level of demand sustained by each back and
upper limb joint. Then he suggests organising the rotation
system such that each workstation with a high level of
demand is followed by a low-demand workstation, whilst
allocating a maximum of three high-demand workstations
per day. Frazer et al. (2003) estimate that it is still difficult
to find proof in the literature concerning the advantages of
workstation rotation in relation to MSD prevention.
Moller et al. (2004) are even more careful, considering
that greater exposure variability is not necessarily sufficient
for reducing MSD risks. On the other hand, delegates at
the 2003 Montreal Symposium on Workstation Rotation
concluded that this technique can prevent MSDs by
curtailing excessive muscle and tendon demand, and that
901
it can also have a positive impact on work satisfaction by
breaking the monotony of repeating the same work
movements (Vézina, 2003). However, this twofold positive
impact remains subject to compliance with a number of
principles involving adherence to rules whose outlines
require further detailing.
The aim of this study is to suggest an approach for
defining a workstation rotation system, which reduces
MSD risk factors, and to list parameters enabling an
efficient system to be set up. To achieve this, an ergonomic
study was conducted on two household electrical appliance
assembly lines subject to workstation rotation.
2. Methodology
The two production lines (A and B) are dedicated to
assembling a range of ovens. Workstation type and
distribution as well as psychosocial context are similar on
both lines. Two workstation rotation systems can be
observed on each line: a so-called large system embracing
the assembly stations and a second one called the shell
system. This rotation had been set up several months prior
to the study. There are 38 operators on the two lines. They
remain approximately 2 h at each station and change
stations during work breaks.
The methodology included the use of a questionnaire
during an interview conducted among 38 employees as well
as a detailed description of the rotation of line B and a
study of the biomechanical risk factors and the work
actions of each of the workstations comprising it.
2.1. Interviews
Firstly, all the operators on the two assembly lines were
asked to complete a questionnaire comprising 52 questions
taken mainly from INRS’s MSD questionnaire (Cail et al.,
2000). The researchers carried this out during a personal
interview with the operators.
This questionnaire allows the target population to be
analysed in terms of age, length of service, type of contract
and level of multiskilling.
The questionnaire also addresses state of health by
means of 11 closed questions taken from the Nordic
questionnaire (Kuorinka et al., 1987). These questions
enable the health problems experienced by employees in
nine anatomical regions (neck, upper back, lower back,
right shoulder, left shoulder, right elbow, left elbow, right
hand/wrist, left hand/wrist) to be researched.
State of psychological stress is assessed through a single
question, namely ‘‘Do you feel stressed at present?’’, with
three response options: ‘‘not stressed’’, ‘‘moderately
stressed’’ and ‘‘very stressed’’.
The questionnaire also focuses on perception of psycho-
social factors represented by workload (three questions),
pressure of work (two questions), participation (three
questions), monotony (two questions) and work satisfac-
tion (one question). The operator has five options of
 ARTICLE IN PRESS
902 M. Aptel et al. / International Journal of Industrial Ergonomics 38 (2008) 900–909
responding to each question, namely ‘‘not at all’’, ‘‘a little’’,
‘‘moderately’’, ‘‘very’’ and ‘‘enormously’’. The answers are
then consolidated into three categories. ‘‘not at all’’ and ‘‘a
little’’ answers are considered to represent a positive
feeling, the ‘‘moderately’’ answer is considered neutral,
and ‘‘very’’ and ‘‘enormously’’ answers are considered to
reflect a negative feeling in relation to the variable assessed.
The interviews of line A and B operators also included
the following questions proposed by Falardeau and Vézina
(2004) related to rotation systems:
Are you favourable to the workstation rotation?
Were you present during the last implementation of the
rotation system?
Took part in this implementation?
2.2. Description and biomechanical demands
Line B was chosen as the management of the company
wished to integrate new rotation systems into this line.
A description of the workstations of this line and of the
main actions carried out at each of the workstations was
drawn up. An OREGE1 rating was applied to this line.
Ratings were allocated to each workstation on line B.
OREGE is a movement identification and evaluation aid
whose purpose is to quantify biomechanical stresses
represented by forces, constraining postures and movement
repetitiveness. It has been developed in France by the
Institut National de Recherche et de Sécurité (INRS)
(Aptel et al., 2000). This pencil-and-paper aid is based on
observation of the operator, his or her perception of the
demands, and indications. Unlike other aids, like OSHA,
OREGE is based on dialogue between the researcher and
the operator (Apostoli et al., 2004).
For each action considered, force is allocated a value of
between 0 and 10 on an evaluation chart, which the
researcher and operator both complete separately. Effec-
tive force evaluation then results from a synthesis based on
the values allocated by the operator and the researcher,
who then establish a force evaluation following this
discussion.
For each action concerned, the researcher observes the
operator’s upper limb movements at the neck, shoulder,
elbow and wrist joints. He allocates a value 1 (acceptable),
2 (not recommended) or 3 (to be avoided) to each joint on
both the left and right sides. Evaluation is assisted by
figures, in which angular classes corresponding to these
three values are represented.
Upper limb movement repetitiveness is evaluated from 0
to 10 over a period that depends on work cycle time. In the
same way as for force, the researcher and operator both
complete an evaluation chart separately, and the subse-
quent procedure is the same as for force evaluation.
Finally, from one force, angular and repetitiveness value
1 2
OREGE: Outil de Repérage et d’Evaluation des Gestes (movement
identification and evaluation aid).
for each action analysed, the researcher assess whether
this action is acceptable, not recommended or to be
avoided.
3. Results
3.1. Results of interviews
3.1.1. Population
The average age of the 38 operators on the two assembly
lines is 44.778.6 years. Gender distribution is 25 women
and 13 men. The average length of service of these
operators is 20710 years and 35 of them are employed
on a permanent contract.
Operator distribution within the two lines is gender-
related. Men occupy only workstations within the shell
system, the reserved packaging workstation, palletization
and oven handling. Women occupy workstations within
the large system and the reserved cleaning workstation.
Operator allocation to a rotation system according to
gender is a company decision, which is motivated by the
physical workload of the station. According to the
company, the shell stations are subject to a heavier
workload than the other assembly stations.
The questionnaire results show that operators who claim
to be multiskilled represent 79% of the study population,
against 21% of single-skill operators (eight persons). All
the multiskilled operators are rotated between at least two
workstations. Multiskilling of operators allocated to the
large system is greater than that of operators allocated to
the shell system. The women are therefore more multis-
killing than the men. Single-still operators comprise five
persons subject to medical restrictions,2 two warehousemen
and one pallet packer. The operators subject to medical
restrictions are allocated to the reserved workstations
whose workload is considered light. The single-skill status
of the warehousemen and the pallet packer is inherent to
their work activity.
Ten of the 38 operators are subject to medical restric-
tions. Their average age is 47.378.3 years, and is therefore
not significantly different from that of the other 28
operators. The population subject to medical restrictions
comprises six women and four men. Half of this population
is single-skilled and is allocated to reserved workstations.
The other half does not benefit from a reserved workstation
and is integrated into the two station rotation systems.
3.1.2. MSD complaints
MSD complaints concern mainly the wrists/hands,
especially the right wrist/hand, the lower back, the neck
and the shoulders. The 38 employees on the two assembly
lines refer to at least one painful location and more than
50% of them have six painful locations. The average
number of pains referred to by an operator is three. All
In France, the occupational physician must evaluate the medical health
status of employees to fit the job to the workers.
 ARTICLE IN PRESS
M. Aptel et al. / International Journal of Industrial Ergonomics 38 (2008) 900–909
operators allocated to the assembly lines therefore were
concerned by MSD pains, irrespective of their age, gender
or level of multiskilling.
3.1.3. Stress and psychosocial factors
In relation to employee state of stress, the questionnaire
shows that 71% of the operators claim not to be stressed,
while 13% claim to be very stressed.
Fig. 1 illustrates the results for the questions on
psychosocial factors from questionnaire.
The answers to two of the three questions concerning
perceived workload reveal that most operators have a
negative perception on this issue. According to them, the
work rate and the number of actions to be performed at the
workstations are both excessive. Similarly, the answers to
two of the three questions concerning participation in work
organisation reveal that most operators have a negative
feeling on this issue. These operators consider they play no
part in decisions concerning their work activity.
Conversely, most operators have a positive or neutral
perception of work pressure. These operators in fact
declare that they are never, or only very rarely, late in
their work because they are very familiar with the tasks and
because of efficient management of their time at the
stations occupied. Most operators claim to be satisfied with
their work activity. Furthermore, half the operators
questioned consider they did not have a monotonous job
and recognise the positive contribution of workstation
rotation to reducing monotony. Consequently, despite a
workload considered heavy and a lack of participation in
work organisation, the operators are satisfied, estimate
they can ensure their work rate and are convinced of the
positive impact of workstation rotation on reducing
monotony.
3.1.4. Experience of workstation rotation
Eighty-two percent of the 38 operators questioned are in
favour of workstation rotation, but only 5% consider
having contributed to the setting up of the workstation
100%
80%
60%
40%
20%
0%
workload pressure participation monotony
of work
satisfaction
Fig. 1. Perception of psychosocial factors. Note: +, positive; 0, neutral;
, negative.
903
rotation on the two assembly lines. Their contribution
involves mainly rotation characteristics (workstation types,
order of rotation, etc.). This result agrees with the
questionnaire data on psychosocial factors, according
to which operator participation in work organisation is
very low.
The operators also stated the advantages and drawbacks
of workstation rotation. These comments can be classified
under four headings: biomechanical demands, psychosocial
factors, productivity and experience.
The advantages most often referred to by employees are
a wider variety of movements, along with a reduction in
pains and task monotony.
The major drawback of workstation rotation is increased
work rate, although this is referred to by only three
employees.
To conclude, the 38 operators questioned consider
workstation rotation to create mainly advantages, both
biomechanical and psychosocial. Few drawbacks are
mentioned.
3.2. Description and evaluation of the biomechanical
demands of line B
3.2.1. Description of line B
This line is structured around 17 workstations and has
17 on-line operators plus two warehousemen, a supervisor
and a repairer. The ovens are placed on rotating platens,
which move from station to station by means of automa-
tion. At each station, the operator actuates oven discharge
using a pedal or pushbutton.
On line B, workstation distribution is as follows (Fig. 2):
 a shell system where the internal shell of the oven is
assembled and clipped together (P14, P15, P16),
 an ‘‘assembly stations’’ unit, where the oven peripherals
(thermostat, front, hood, feet, window, etc.) are
mounted (P1-P10),
 an ‘‘inspection station’’, where the operation of the
finished oven is checked (P11),
 a ‘‘cleaning and final inspection’’ station (P12),
 a ‘‘packaging station’’, where the oven is packed in a
cardboard box (P13) and
 a ‘‘palletization station’’, where the finished products
are packed on a pallet then transferred to a storage
area (P17).
The large system includes the ‘‘assembly stations’’ unit and
the inspection station (P11). Stations P12 and P13 are set
aside for low physical workload. They are excluded from
the large system and are reserved for operators subject to
medical restrictions.
Observation of the actions performed at the work-
stations of line B shows that oven assembly involves three
actions, namely screwing, wiring (electrical connections)
and clipping. These actions can be found in the operating
 ARTICLE IN PRESS
904 M. Aptel et al. / International Journal of Industrial Ergonomics 38 (2008) 900–909

Large system Shell system

Palletization

P
17 Assembly stations
P
Packaging Cleaning 15

P P P P P P P P P P P P P
P P
13 12 11 10 9 8 7 6 5 4 3 2 1
16 14
Inspection
Reserved stations

Key :

Station change direction

P
Workstation
x

Fig. 2. Workstation rotation system on line B according to order of rotation P11–P2.

Table 1 Examination of these results for each workstation leads

Summary of demands at workstations within the large system
Workstation P11 P9 P7 P5 P3 P10 P8 P6 P4 P2
Number of actions ‘‘to be avoided’’ 2 3 2 2 2
Screwing action
Wiring action
Clipping action
Posture (sitting/standing) Si St Si Si Si Si
Note: Si: sitting; St: standing.
procedures at the different workstations on assembly line B
(Table 1).
A succession of three wiring stations (P7, P5, P3) and
three clipping stations (P10, P8, P6) is observed. Further-
more, no station offers more than two types of action.
3.2.2. OREGE
The OREGE tool was used to evaluate the biomecha-
nical demands associated with operator actions. Of the 102
actions listed at the 15 line B workstations, the biomecha-
nical demand evaluations concerned 65 actions considered
to involve a potential MSD risks by the research team.
Fig. 3 illustrates the risk diagnosis results. Both the total
number of actions required at each workstation and the
OREGE-rated actions are shown in this figure. Work-
station P1 was not occupied during these evaluations.
The OREGE-based evaluations led us to consider 27
actions as ‘‘to be avoided’’, 32 actions as ‘‘not recom-
mended’’ and six actions as ‘‘acceptable’’. Thus, nearly
40% of the 65 actions evaluated involve MSD risks.
A small percentage (9%) of the actions appear not to
involve a risk.
to the following findings:
 For the large system, OREGE was applied to analyse 45
1 4 1 0 1 actions at 10 workstations within the large system
(station P1 was not studied). This large system includes
18 actions rated as ‘‘to be avoided’’, 24 actions rated as
St St Si St
‘‘not recommended’’ and three actions considered
‘‘acceptable’’.
 For workstations within the shell system, 12 actions
were evaluated at the three workstations within the shell
system. seven actions are rated ‘‘to be avoided’’, four
actions are ‘‘not recommended’’ and one action is
‘‘acceptable’’.
 For workstations reserved for employees subject to
medical restrictions, eight actions were also rated on the
two reserved stations. Station P13 is the most demand-
ing with two actions out of four rated ‘‘to be avoided’’.
Station P12 includes two actions rated ‘‘not recom-
mended’’ and two actions rated ‘‘acceptable’’.
To conclude, repetitiveness is the most infrequent cause
associated with actions ‘‘to be avoided’’, as it appears only
6 times. There are 10 actions ‘‘to be avoided’’ because of
force, and the most frequent causes associated with actions
‘‘to be avoided’’ are those involving posture. Indeed, 21 out
of the 22 actions ‘‘to be avoided’’ within the large system
and all actions ‘‘to be avoided’’ within the shell system are
so rated because of postural demand, especially for the
shoulder, by reason of the height of the component parts.
For example, the four actions at station P8 are considered
at risk; both screwing actions require a forearm extension,
which should be avoided. Hood pick-up from the supply
 ARTICLE IN PRESS
M. Aptel et al. / International Journal of Industrial Ergonomics 38 (2008) 900–909 905

Shell system
F F
P P P
Reserved stations Large system

F F F
P F P P P PP P
F F F
PPPP
F
P
F
P P
R
P
R
P PP
R
P P P15 W
10 actions C
R RR
P PP

P13 P12 P11 P10 P9 P8 P7 P6 P5 P4 P3 P2 P1 P16

9 actions 9 actions 5 actions 8 actions 6 actions 4 actions 5 actions 8 actions 5 actions 4 actions 4 actions 6 actions 9 actions

C,S S C,S Sc,C,S Sc,C W,C,S C Sc,W,S Sc,W,S Sc,W,S W,C Sc,W,S Sc P14
10 actions
:
Key W,C
PX : workstation
Sc : screwing at station
Acceptable action W : wiring at station
C : clipping at station

Action not recommended S : seated station

R : repetitiveness rating
Action to be avoided F : force rating
P : upper limb posture rating

Fig. 3. Results of the OREGE ratings at line B workstations.

pallet creates difficult postures because of the pick-up sented in Fig. 4. It sets out the conditions to be fulfilled to
height, which varies between 30 and 160 cm above the ensure the effectiveness of rotation as a means of
floor; finally, hood placement requires a large force and preventing MSDs. This logic diagram is organised around
sometimes an extreme upper limb posture. several dimensions, namely context, knowledge, implemen-
Furthermore, the OREGE-based evaluation of the tation of the rotation system and assessment of its
biomechanical demands at the workstations of line B effectiveness.
allows us to appreciate the strenuousness of the large
system. This system includes mainly stations involving two 4.1. Context
actions ‘‘to be avoided’’. Station P8 features the most
actions ‘‘to be avoided’’ and its level of demand is therefore Before implementing a rotation system, the existing
the highest in the line. Conversely, the MSD risk levels of situation should be analysed. This corresponds to the first
station P4 is low because it features no action ‘‘to be of the four conditions proposed by Ouellet et al. (2003).
avoided’’. This ergonomic analysis is intended to determine the MSD
Table 1 provides a summary chart of station demand risk factors and to understand the modalities of production
within the large system. This representation allows us to and the management of resources. In this respect, an
better understand where the demands are in space and approach employing interviews yielded an understanding
time. of the psychosocial factors and the conditions that
determine the work of the operators. It is associated with
4. Discussion a study of the workstations comprising the rotation system
and an evaluation of the biomechanical risk factors.
Workstation rotation is often presented as a means of In this study, the rotation system set up on the two
preventing MSDs, and numerous ergonomic studies have assembly lines had a positive impact on the psychosocial
shown the benefits, but also the limits, of this solution factors. Despite a heavy workload and lack of participation
(Vézina, 2003; Moller et al., 2004). It is therefore necessary in work organisation, the operators on these two lines are,
to provide parameters to define rigorously the conditions in fact, very satisfied with the workstation rotation system.
of the effectiveness of this means of prevention. Coutarel The management’s organisational decision is effectively
et al. (2003) and Ouellet et al. (2003) propose a set of supported by all the multiskilled operators and half of the
principles for ensuring the effectiveness of workstation single-skill operators. Those questioned stress the attrac-
rotation in preventing MSDs. For example, Ouellet et al. tiveness of this organisational system, one of the major
(2003) consider that the approach to setting up rotation advantages of which is the reduction in task monotony,
could comprise four stages, namely analysis of the in- which reflects the conclusions of several authors (Kuijer
company situation, consideration of human resource et al., 1999; Thomas et al., 1994). This positive psychoso-
management issues, reduction of workload at workstations cial effect is not negligible since, as Daniellou (2003) points
and training of workers. Conversion of these conditions out, ‘‘health is merely the health of the joints’’. On the
into operational parameters remains to be achieved if we other hand, the satisfaction of the operators did not
are to validate the hypothesis that workstation rotation is a eliminate complaints of MSDs. A medical study conducted
means of preventing MSDs. in the same sector of activity (Aublet-Cuvelier et al., 2006)
Based on these principles and on knowledge relative to corroborates this high prevalence of MSD complaints. This
safety management, we propose the logic diagram pre- result is similar to that obtained by Kuijer et al. (2005),
 ARTICLE IN PRESS
906 M. Aptel et al. / International Journal of Industrial Ergonomics 38 (2008) 900–909

CONTEXT KNOWLEDGE

Formalisation of the
parameters which define
a rotation system
Ergonomic
analysis of the Validation of the parameters
existing system to assess the rotation system
in conjunction with the
targeted objectives

IMPLEMENTATION

Development of a
specific rotation system

Implementation
and validation

ASSESSMENTS

Follow-up of impacts

on the on the risk of

experience of MSDs
the employees

Fig. 4. Parameters to take into account to design a rotation system aimed at preventing the risk of musculoskeletal disorders (MSDs).

who noted that workstation rotation has no positive effect
on complaints of low-back pain. The prevalence of MSD
complaints is high in employees on both production lines,
who describe at least one painful anatomical site. More-
over, this population has been exposed to MSD risk factors
for over 10 years, as a study carried out in this enterprise by
one of the authors shows (Aptel et al., 1993). It is therefore
difficult to determine the impact of the rotation system
established in the case studied, given that the employees
have a known previous history of MSDs. This being said,
the perception that the operators have of their work
context probably has an influence on the determination of
MSDs, as Devereux et al. (2004) consider is the case.
The evaluation of the rotation system in itself has
allowed an understanding of the biomechanical demands
that it generates. The large system on line B is demanding
on the upper limbs. This large system in fact includes only
one workstation with no action ‘‘to be avoided’’ (P4), while
it includes eight workstations with one or two actions ‘‘to
be avoided’’ and two workstations with three or four
actions ‘‘to be avoided’’. The biomechanical demand
analysis reveals that the amplitudes at the postures are
the main MSD risk factors. Most actions to be avoided are
determined by supply areas or worktops unsuited to the
task. This observation is identical for the shell work-
stations. Each of the two line B rotation systems are
structured around a set of workstations with characteristics
designed around the same rules. Process engineering
department replication of design principles at each work-
station is probably the reason for this. Medical restrictions
and gender-based rotation system division may also be
reasons explaining the MSD complaints level. Operators
subject to medical restrictions are allocated to reserved
workstations considered fairly undemanding by the com-
pany. In fact, only station P12 meets this requirement on
line B. Station P13 involves two actions ‘‘to be avoided’’,
yet is considered fairly undemanding by the company. This
station therefore creates MSD risks in the same way as
most of the other line B stations. Furthermore, allocation
to a reserved workstation testifies to a health problem on
the part of the operator, who considers it does little to
enhance his/her self-esteem. This perception can contribute
 ARTICLE IN PRESS
M. Aptel et al. / International Journal of Industrial Ergonomics 38 (2008) 900–909
to his/her ill being at work. That being the case, employees
subject to a medical restriction can logically continue to be
at risk because they remain exposed to MSD biomechani-
cal and psychosocial risk factors. Finally, the latter
hypothesis, potentially explaining the high level of MSD
complaints, stems from the fact that work organisation
results in men working only within the shell stations and
never at the assembly stations. The shell stations comprise
three demanding workstations, as shown by the OREGE-
based evaluations, and thus the workstation rotation-
related benefits are modest for the male operators.
Workstation rotation can therefore be considered from
two complementary standpoints. One, physical, concerns
the physical system itself (nature of the tasks, intensity and
diversity of the biomechanical demands, order of work-
station changeover, number of workstations in the rota-
tion, etc.). From the point of view of MSD risk prevention,
this physical dimension is essential and must result in a
reduction in biomechanical demands. This is an essential
condition. The second aspect concerns the perception that
the operators have of the versatility stemming from the
rotation. The operators can perceive the implementation of
a rotation system positively when it enriches their work,
even if it does not reduce the biomechanical demands. Its
impact on risk will, in this case, be low but not necessarily
zero. On the other hand, certain employees can perceive the
implementation of a rotation system negatively if the
number of work situations that they must master is too
great, even if it reduces the biomechanical demands. The
present study confirms this analysis. In addition, the links
between the physical data relative to the organisation of the
rotation and those relative to the perception of the work
(stress and psychosocial factors) are complex, even contra-
dictory. Thus, there are divergences concerning the
biomechanical demands. In fact, increased movement
variability forms part of the workstation rotation advan-
tages perceived by the employees, but the OREGE-based
biomechanical demand evaluation shows, on the contrary,
that the demands are very similar. These divergences also
concern complaints of MSDs. According to the Nordic
questionnaire results, very many employees complain of
MSD problems; when questioned on workstation rotation
advantages only 21% of employees deem that rotation
reduces their pains. The divergences merely strengthen the
need for a comprehensive ergonomic approach (question-
naire, work observation, light movement analysis, etc.) as
this alone allows an understanding of the different work
dimensions of the operators.
4.2. Knowledge
Although a prior ergonomic study is necessary as it is the
only way to understand the existing stations, it is not
enough to design an effective rotation system. Other
conditions stemming from scientific knowledge are neces-
sary. Many researchers suggest the need for employee
participation and the introduction of appropriate training
907
(Ouellet et al., 2003). The favourable impact of rotation on
the psychosocial sphere found in our study is coherent with
the bibliographic data. On the other hand, although
researchers agree on the importance of reducing biome-
chanical demands and the introduction of workstations
favouring the diversity of occupational gestures, the
knowledge allowing this to be achieved is lacking. More-
over, the biomechanical data of the study show how many
workstations of the rotation system are comparable and
arduous from a biomechanical point of view. This reality
may explain the permanence of MSD complaints, and the
relative ineffectiveness of workstation rotation in prevent-
ing MSDs. In these conditions, it is necessary to extend
both physiological and biomechanical research on this
essential physical dimension of the problem of workstation
rotation as a means of preventing MSDs. Only evidential
data in this domain will allow us to propose parameters
relevant to validating the solutions proposed. This condi-
tion is insufficiently fulfilled in this case at the line B
workstations. Subsequently, it is not possible at this time to
categorise the workstations and the order of rotation, as
recommended by Henderson (1992) and Van Velzer (1992).
Thus, there is a need for research concerning workstation
risk level assessment, workstation categorisation, order of
rotation, frequency of rotation and optimal number of
workstations integrated into the rotation system. Propos-
ing validated parameters must constitute a research
priority as these parameters are required when structuring
a rotation system to ensure protection against MSD risk.
4.3. Implementing a rotation system and assessing its
efficiency
Available scientific data cannot be applied directly in
factories. Indeed, each workshop is specific and the
solutions retained are, by their nature, compromises
between the available scientific data and those stemming
from the ergonomic analysis. Furthermore, these compro-
mises can interfere with the expected efficiency of the
solutions proposed. It is therefore necessary for the
enterprise to set objectives as well as indicators, which
must be monitored over time. The complexity of MSDs
and the diversity of the solutions proposed justify that the
assessments, and therefore the indicators, are not reduced
solely to the reduction of MSDs. Indeed, our study shows
that the results are more equivocal. For example, the
satisfaction of the employees with respect to the psycho-
social conditions created by the rotation is an advance,
whereas the high level of MSD complaints is a disadvan-
tage. It is therefore a process, which should be evaluated in
all its dimensions. In this sense, the indicators must be
numerous and take into account different elements of the
context and their possible interactions. These indicators
can be reactive (number of MSD complaints) or proactive
(operator satisfaction). This the point of view of Berthelette
and Baril (2002), who estimate ‘‘the absence of effect can
be attributed to an inadequate underlying theory and,
 ARTICLE IN PRESS
908 M. Aptel et al. / International Journal of Industrial Ergonomics 38 (2008) 900–909
consequently, to an inefficient intervention, to the presence
of divergences between the characteristics of the recom-
mended programme and those of that implemented by
those involved, which can degrade its effectiveness, or
again to the methodological limits of the assessment’’.
Finally, one final result is worthy of note; this concerns
the consequences, for the collective formed by multiskilled
operators, of the fact that operators subject to medical
restrictions are allocated to reserved workstations. This in
fact makes the rotation system rigid and curtails the
expected benefits to all employees who are still fit, because
they will work at more arduous workstations. Conversely,
this situation can create a feeling of exclusion for operators
allocated to reserved workstations. It is therefore abso-
lutely essential that workers be involved in organising a
workstation rotation system (Ouellet et al., 2003); this was
not the case for the employees used in this study, and they
did indeed complain about this. Taking into consideration
the health reality of the employees in the organisation of
the rotation system is an additional dimension that should
not be ignored while the prevalence of MSDs is high and
often old. This clearly makes the construction of solutions
more difficult.
5. Conclusion
This study shows that the workstation rotation studied
had a positive impact on psychosocial factors but is
relatively ineffective in relation to the prevention of MSDs,
notably on account of the lack of variability in the
biomechanical demands. The assessment of a rotation
system aimed at preventing MSDs must encompass all the
dimensions of the work. To achieve this, it is necessary to
describe the system in detail, to know the characteristics of
the population concerned, to assess its experience of the
work and to measure the degree of the biomechanical
demands of the workstations. It is also necessary to further
develop research to assess the biomechanical demands.
Finally, implementing a rotation system requires regular
assessments of the results based on scientifically evidential
parameters adapted to the context of the factories. It is on
these conditions alone that rotation will be an effective
MSD prevention system.
References
Apostoli, P., Sala, E., Gllino, A., Romano, C., 2004. Comparative analysis
of the use of 4 methods in the evaluation of the biomechanical risk to
upper limb. Giornale Italiano di Medicina del Lavoro ed Ergonomia
26, 223–241.
Aptel, M., Cail, F., Cnockaert, J.C., et al., 1993. Etude dans une entreprise
de montage d’appareils électroménagers des facteurs de risques
professionnels du syndrome du canal carpien. Documents pour le
médecin du travail 54, 149–164.
Aptel, M., Lahaye, S., Gerling, A., 2000. Un outil d’évaluation des
facteurs de risques biomécaniques: OREGE (Outil de Repérage et
d’Evaluation des Gestes). Document pour le médecin du travail 83,
217–223.
Aublet-Cuvelier, A., Aptel, M., Weber, H., 2006. The dynamic course
of musculoskeletal disorders in an assembly line factory. Internat-
ional Archives of Occupational and Environmental Health 79,
578–584.
Axelsson, S., Ponten, B., 1990. New ergonomic problems in mechanized
logging operations. International Journal of Ergonomics 5, 267–273.
Baggerman, M., 1993. Macro-Ergonomics aid in meeting OSHA
Standards. Apparel Industry Magazine August, 56–59.
Berthelette, D., Baril, R., 2002. Les dimensions des interventions
organisationnelles de maintien du lien d’emploi des travailleurs
victimes de lésions professionnelles. Pistes 4, 27. /http://www.pistes.
uqam.ca/v4n2/articles/v4n2a7.htmS.
Cail, F., Morel, O., Aptel, M., 2000. Un outil de recueil et d’analyse des
facteurs de risque: le questionnaire TMS. Document pour le médecin
du travail 83, 199–216.
Christmansson, M., Friden, J., Sollerman, C., 1999. Task design, psycho-
social work climate and upper extremity pain disorders—effects of an
organisational redesign on manual repetitive assembly jobs. Applied
Ergonomics 30, 463–472.
Coutarel, F., Daniellou, F., Dugué, B., 2003. Interroger l’organisation du
travail au regard des marges de manœuvre en conception et en
fonctionnement—la rotation est-elle une solution aux TMS? Pistes 5,
27. /http://www.pistes.uqam.ca/v5n2/articles/v5n2a2.htmS.
Daniellou, F., 2003. De la rotation sur les postes à la santé au travail.
Synthèse du colloque ‘‘La rotation, est-ce une solution?’’. Pistes 5, 7.
/http://www.pistes.uqam.ca/v5n2/articles/v5n2a12.htmS.
Devereux, J., Rydstedt, L., Kelly, V., Weston, P., Buckle, P., 2004. The
role of work stress and psychological factors in the development of
musculoskeletal disorders. Research Report 273, Robens Centre for
Health Ergonomics, University of Surrey, 139pp.
Ellis, T., 1999. Implementing job rotation. Occupational Health and
Safety 68, 82–84.
Falardeau, A., Vézina, N., 2002. Rotation de postes, assignation
temporaire et impact des absences dans une usine d’abattage et de
transformation du porc. Pistes 4, 25. /http://www.pistes.uqam.ca/
v4n2/articles/v4n2a9.htmS.
Falardeau, A., Vézina, N., 2004. Apport de différentes sources de données
à la description comparée des contraintes et à l’amélioration d’un
groupe de postes occupés en rotation. Pistes 6, 22p. /http://
www.pistes.uqam.ca/v6n1/pdf/v6n1a7.pdfS.
Frazer, M., Norman, R., Wells, R., Neumann, W., 2003. The effects
of job rotation on the risk of reporting low back pain. Ergonomics 46,
904–919.
Henderson, C., 1992. Ergonomic Job Rotation in Poultry Processing.
Advances in Industrial Ergonomics and Safety IV. Taylor & Francis,
London, pp. 443–448.
Hinnen, U., Läubli, T., Guggenbühl, U., Krueger, H., 1992. Design
of check-out systems including laser scanners for sitting work
posture. Scandinavia Journal of Work Environment & Health 18,
186–194.
Kuijer, P., Van der Beek, A., Van Dieën, J., Visser, B., Frings-Dresen, M.,
2005. Effect of job rotation on need for recovery, musculoskeletal
complaints, and sick leave due to musculoskeletal complaints: a
prospective study among refuse collectors. American Journal of
Industrial Medicine 47, 394–402.
Kuijer, P.P.F.M., Visser, B., Kemper Han, C., 1999. Job rotation as a
factor in reducing physical workload at a refuse collecting department.
Ergonomics 42, 1167–1178.
Kuorinka, I., Jonsson, B., Kilbom, A.etal., 1987. Standardised Nordic
questionnaires for the analysis of musculoskeletal symptoms. Applied
Ergonomics 18, 233–237.
MacLeod, D., Kennedy, E., 1993. Job rotation system. Report to XYZ
CO, 12pp. /http://www.lni.wa.gov/wisha/ergoideas/DocumentsFiles/
idea719.pdfS.
Mikami, K., Kumashiro, M., Izumi, S., 1987. Work load of repetitive,
precision solder work. New methods in applied ergonomics. In:
Proceeding of the Second International Occupational Ergonomics
Symposium, Zadar, Yougoslavie, Avril. Taylor & Francis, London.
 ARTICLE IN PRESS
M. Aptel et al. / International Journal of Industrial Ergonomics 38 (2008) 900–909
Moller, T., Mathiassen, S., Franzon, H., Kihlberg, S., 2004. Job
enlargement and mechanical exposure variability in cyclic assembly
work. Ergonomics 47, 19–40.
Ouellet, S., Vézina, N., Chartrand, J., Perrier, P.-P., Malo, J.-L., 2003.
L’implantation de la rotation de postes: un exemple de démarche préalable.
Pistes 5, 27pp. /http://www.pistes.uqam.ca/v5n2/articles/v5n2a14.htmS.
Rérat, F., 1985. La polyvalence comme méthode d’organisation du travail.
Journées des économistes du travail de Grenoble. Dossier formation
emploi 14, 43–53.
Rissen, D., Melin, B., Sandsjos, L., Dohns, I., Lundberg, U., 2002.
Psychophysiological stress reactions, trapezius muscle activity, and
neck and shoulder pain among female cashiers before and after
introduction of job rotation. Work and Stress 16, 127–137.
Roquelaure, Y., Mechali, S., Dano, C., et al., 1997. Occupational &
personal risk factors for carpal tunnel syndrome in industrial workers.
Scandinavian Journal of Work Environment & Health 23, 364–369.
St-Jacques, Y., Vézina, N., Stock, N., Boucher, M., 2003. Le système
modulaire en vaut-il la ‘‘peine’’? In: La rotation est-ce une solution?
909
Actes du Forum d’échanges de la chaire GM en ergonomie de
l’UQAM, Université du Québec à Montréal, pp. 97–102.
St-Vincent, M., Vézina, N., Dufour, B., St-Jacques, Y., Cloutier, E., 2003.
La rotation des postes: ce qu’en pensent des travailleurs d’une usine
d’assemblage automobile. Pistes 5, 23. /http://www.pistes.uqam.ca/
v5n2/articles/v5n2a1.htmS.
Thomas, N., Thessier-Cotte, C., Christol, J., 1994. Polyvalence et
contrainte mentale. XXIIIes journées nationales de médecine du
travail, France, pp. 314–316.
Van Velzer, C., 1992. Economic ergonomic interventions: the benefits of
job rotation. In: 25th Annual Conference of the Human Factors
Association of Canada, pp. 207–212.
Vézina, N., 2003. Implantation de la rotation: quels sont les enjeux?
Quelles sont les balises, Pistes 5, 23.
Vézina, N., Saint-Vincent, M., Dufour, B., St-Jacques, Y., Cloutier, E.,
2003. La pratique de la rotation dans une usine d’assemblage
automobile: une étude exploratoire. Collection Etudes et Recherches,
IRSST, R-343, 88pp.