International Journal of Industrial Ergonomics xxx (2015) 1e11

Contents lists available at ScienceDirect

International Journal of Industrial Ergonomics

journal homepage: www.elsevier.com/locate/ergon

Participatory ergonomics intervention for improving human and

production outcomes of a Brazilian furniture company
~es*, Michel Jose Anzanello, Jose Luis Duarte Ribeiro,
Lia Buarque de Macedo Guimara
Tarcisio Abreu Saurin
Graduate Program in Industrial Engineering, Federal University of Rio Grande do Sul (UFRGS), Av. Osvaldo Aranha, 99, 5
andar, 90035-190 Porto Alegre,
RS, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: This article presents a participatory intervention in a furniture manufacturing company in Southern
Received 8 August 2013 Brazil aiming to improve both ergonomic and production outcomes. The existing Tayloristic model was
Received in revised form replaced by a cellular teamwork model. Work enlargement and enrichment, and the improvements in
24 January 2015
workstation design and process flow increased worker satisfaction and reduced postural risk, fatigue,
Accepted 2 February 2015
Available online xxx
body pain and production waste. Workload was reduced by 42% and productivity increased by 46% (25%
being attributable to unnecessary load handling, waiting and transportation, and 21% attributable
directly to manufacturing times). Workers' participation in the stages of problems identification, design
Keywords:
Participatory ergonomics
and evaluation of solutions played a major role in these outcomes.
Furniture manufacturer Relevance to industry: This study indicates that it is possible to balance ergonomics and production
Functional work design demands, and that it is necessary to make it clear to management. The integration of macroergonomics
Cellular work design and production management principles increases both worker well-being and productivity levels, thus
leading to a more sustainable system.
© 2015 Published by Elsevier B.V.

1. Introduction
Furniture manufacturing is an important industry in many
countries (Mirka et al., 2002a; Gauthier et al., 2012; Ratnasingam
et al., 2012). In Brazil, 18.7 thousand furniture companies
employed 328.6 thousand people in 2013, which is the equivalent
to 3.3% of the employments in the industrial sector (IEMI, 2014).
However, workers in furniture manufacturing are often exposed to
hazards such as dust (Goldsmith and Shy, 1988; Vinzents, 1988;
Pisaniello et al., 1991, 1992; Scheeper et al., 1995; Demers et al.,
1997; Holcroft and Punnett, 2009; Ratnasingam et al., 2010,
2011), chemicals (Goldsmith and Shy, 1988; Voog and Jansson,
1992; Vinzents and Laursen, 1993; Estill and Spencer, 1996;
Holcroft and Punnett, 2009; Ratnasingam et al., 2010, 2011), noise
(Vinzents and Laursen, 1993; Holcroft and Punnett, 2009;
Ratnasingam et al., 2010, 2011), tool vibration (Gauthier et al.,
2012), hazardous machinery, vehicle traffic, fires and explosions
(Holcroft and Punnett, 2009).
* Corresponding author. Tel.: þ55 51 3308 3948; fax: þ55 51 3308 4007.
E-mail addresses: liabmg@gmail.com, lia@producao.ufrgs.br (L.B.M. Guimar~
aes).
Work in the furniture industry has been described as heavy and
repetitive, involving frequent lifting, pushing and pulling of heavy
loads, and the adoption of awkward static postures, like bending
and twisting (Holcroft and Punnett, 2009; Mirka et al., 2002a,b;
Ratnasingam et al., 2010, 2011). Such factors offer occupational
risk for work-related musculoskeletal disorders (WMSD)
(Christensen et al., 1995; Mirka et al., 2002a,b) and injuries
(Aaltonen, 1996).
Workstation improvements, such as sit-stand seating (Urlings
et al., 1990), height-adjustable tables, lift-assisting devices (Mirka
et al., 2002a) and better hand tools (Mirka et al., 2002b) have
been proposed in order to mitigate WMSD and accidents in furni-
ture manufacturing. However, it is known that improvement in
work design by job enlargement and enrichment plays a major role
in WMSD risk prevention (Carayon et al., 1999; Rivilis et al., 2006,
2008) and has proved to reduce the high prevalence of WMSD in
the furniture industry (Christensen et al., 1995). For example,
Hunter (2008) enlarged the work using a cellular design in a
furniture assembly unit: cell workers walked from one workstation
to another on rubber mats to reduce fatigue, and moved hardware
assembly fixtures along roller conveyors. Such improvements
reduced musculoskeletal disorders risk and increased productivity

http://dx.doi.org/10.1016/j.ergon.2015.02.002
0169-8141/© 2015 Published by Elsevier B.V.

~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002
2 ~es et al. / International Journal of Industrial Ergonomics xxx (2015) 1e11
L.B.M. Guimara
by 11.2% (Hunter, 2008). Cell manufacturing is indeed widely
known as a strategy for improving production management by
shortening lead times, improving product quality, and increasing
flexibility (Sheridan, 1990).
Empirical studies rarely explicitly link practices and principles
from ergonomics and production management regardless of the
synergistic relationship between both disciplines. This is an
important drawback as positive impacts of work organization in-
terventions are likely to arise from ergonomics and production-
oriented actions rather than isolated initiatives.
The goal of this study was to re-design a production system to
address both human and production demands according to mac-
roergonomics, i.e., relying on participatory ergonomics and
focusing on the process rather than the workstation. The study is
aligned with a trend in ergonomics that argues for the balance
between production demands and ergonomics to achieve a sus-
tainable production system (Hendrick, 1997; Dul and Neumann,
2009; Thun et al., 2011; Westgaard and Winkel, 2011). This study
also aims to contribute to the literature on participatory ergo-
nomics (Noro and Imada, 1991), which seldom focus on the social
relations in workplaces (Dixon and Theberge, 2011) since the focus
is mainly on the importance of worker participation on the
appraisal stage for hazard identification (Wilson et al., 2005; Cann
et al., 2008) and its effects on health outcomes (Brown, 1993;
Kuorinka and Patry, 1995; Halpern and Dawson, 1997; Haims and
Carayon, 1998; Moore and Garg, 1998; Hignett et al., 2005; Laing
et al., 2005; Haukka et al., 2008; Rivilis et al., 2006, 2008)
although considering productivity is also relevant (Brown, 1993;
Nagamachi, 1996; Looze et al., 2005; Rosecrance et al., 2005; Vink
et al., 2008). This study details how workers and managerial staff
endured the appraisal, proposal of solutions and testing stages
aimed at improving work quality and productivity; it also presents
the impacts of the intervention on both social and organizational
climate.
The study resulted from a partnership between the University in
charge of this study and a large Brazilian furniture company, which
needed a model that would balance production and ergonomic
issues in the new facilities to accommodate for the expansion of the
sofas manufacturing sector.
2. Method
2.1. Overview of the company and project team composition
The company, established in 1970 in the State of Rio Grande do
Sul in Brazil, employs more than 4000 workers in the furniture,
mattress and chemical sectors. Around 90% of the furniture is
manufactured for the internal market. Sofa manufacturing requires
290 workers (166 men and 124 women) grouped in 16 sectors. Only
four out of six sectors related to assembly/transformation (wooden
frame assembly, straps placement, foam gluing, and upholstery),
encompassing 81 workers, were supposed to occupy the new fa-
cilities. Therefore only those sectors were focused on this study. The
project team was composed of four ergonomists (two engineers,
two physiotherapists), three industrial engineers from the Univer-
sity, two managers and the workers from the mentioned sectors.
2.2. Ergonomic evaluation
Ergonomic evaluation followed the participatory Macro-
ergonomic Work Analysis (MA) method (Guimara ~es, 1999)
assuming that worker participation in an ergonomic intervention
assures better acceptance of new ideas (Hendrick, 1990; Brown,
1995; Nagamachi, 1995, 1996). In the MA there are discussion
meetings (named gates) held between each stage of the ergonomic
~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002
intervention (appraisal, diagnosis, proposal of solutions, prototyp-
ing, and validation), which engage workers and keeps them aware
of project developments. A total of 77 workers from four sectors of
the company joined the appraisal stage, and 11 volunteer workers
from these sectors took part in the prototyping, testing and vali-
dation stages. Diagnosis was mainly developed by the University's
experts, while proposals for solutions, testing and validation stages
involved these experts along with 2 managers and the 11 volunteer
workers.
The appraisal stage comprises the identification of worker de-
mands and experts' observations, and follows the three first steps
of the Macroergonomic Design (MD) method (Guimara ~es and
Fogliatto, 2000):
1) interviews for gathering information on user ergonomic de-
mands or ergonomic demand items (EDIs);
2) identification of an importance score for the EDIs based on how
often they are mentioned by respondents in the interview, and
on the order in which items are mentioned. Items are scored, for
each respondent, according to the order of their mentioning (i.e.
first mentioned EDI scores 1, second scores 2, third scores 3, and
so on). The item importance weight is then given by its inverse
order, therefore first mentioned items received higher scores: 1/
1 ¼ 1.00; 1/2 ¼ 0.50, 1/3 ¼ 0.33 etc. Items from all interviews are
reviewed and grouped to avoid replication due to how re-
spondents may refer to the same EDI by different names. The
final importance score of each EDI is obtained summing up the
scores from all respondents;
3) incorporation of experts' opinion for the inclusion of EDIs not
identified by workers but deemed important by the experts.
Steps 1 and 2 rely on semi-structured interviews with approx-
imately 30% of the workforce, and stage 3 is performed by the er-
gonomists through observation and analysis. These steps provide
qualitative data to design a questionnaire to be answered by 100%
of the workforce through the marking of a 15 cm continuous scale
(Stone et al., 1974) with two anchors (i.e., not satisfied/very satisfied
or none/a lot). Therefore, quantitatively, the value of an EDI varies
from 0 to 15 and the weight of each EDI is generated by the
arithmetic average of the results from all respondents.
The appraisal gate has three major goals: 1) confirm the most
important EDIs to be fully analyzed in the diagnostic stage and to be
considered in the proposal of solutions; 2) plan for the prototyping
stage by defining how and when prototyping will take place; and 3)
call for volunteers for the prototyping stage. The results from the
questionnaire and ergonomist observations lead to the diagnosis
and subsequent proposal of solutions, testing and validation stages.
Two ergonomists interviewed 26 volunteer workers (32% of the
population working in the four sectors) individually or in groups
depending on worker preference and practical constraints in the
appraisal stage. Interviews lasted 10e40 min and were based on
generic questions such as: what do you think about your work?
What is good? What could be better? What suggestions for
improvement do you have? Interviews were tape recorded for
further analysis with worker permission. EDIs mentioned by all
groups in addition to ergonomist questions of interest resulted in
four different questionnaires to accommodate particularities of the
four sectors. Questionnaires encompassed a total of 47 questions
grouped in six constructs: physical environment (four questions),
work content (13 questions); pain/discomfort (eight questions);
workstation (eight questions), work design (eight questions); and
company (six questions). Workers received the questionnaires in
the morning and handed them out on the afternoon of the same
day. 95% of the population of the four sectors (77 workers)
answered the questionnaires. Questions (EDIs) from the six
 ~es et al. / International Journal of Industrial Ergonomics xxx (2015) 1e11
L.B.M. Guimara
constructs of the 4 questionnaires showed acceptable consistency
(Cronbach's alpha ¼ 0.615). Next, the non-parametric Krus-
kaleWallis (KW) test was used to assess statistical differences on
the sample means. The option for the KW test is justified by the fact
that the sample is small, and there is no certainty regarding normal
distribution of data. The KW test does not require a normal distri-
bution of the residuals as the analogous one-way ANOVA does
(Corder and Foreman, 2009). Tukey's test was used to point out the
sectors that are significantly different from the others. Questions
related to workstation, work organization and company constructs
reflected EDIs particular to each sector, do not enabling direct
comparison among them. Comparative results presented in this
article were therefore drawn based on EDIs from physical envi-
ronment, work content and discomfort/pain constructs that were
common to all four evaluated sectors. Managers and supervisors
were not interviewed and did not answer questionnaires, therefore
their opinions regarding work performed in the sectors as well as
the possible solutions for improvement were obtained through
dialogues and discussions held on many occasions during the
interview process.
Diagnosis was split into ergonomic and production data anal-
ysis. Ergonomics diagnosis mainly comprised the analysis of work
carried out in the four evaluated sectors and an in depth analysis of
the problems raised in the appraisal stage. Two ergonomists per-
formed non-systematic observations during a two-week period to
understand the work in each sector and their interactions. Work
was videotaped, and the cycle times required to perform different
tasks were recorded during different times of the journey. This data
allowed a posteriori assessment of how work was performed, a
more detailed time/motion analysis, and a postural risk evaluation
based on the computerized OWAS (WinOwas) method (Kivi and
Matilla, 1991).
Analysis of production outcomes included: 1) one year of his-
torical data from customer orders, product costs, commercial and
technical information of products with highest impact on Com-
pany's revenue that was obtained from the commercial department
e this data was useful to select the product models to be analyzed;
and 2) cycle times and lead times, which was used to evaluate
production outcomes before and after the intervention; 3) opera-
tion, waiting and transportation times on different tasks were
measured from the first (wooden frame assembly) until the last
(product expedition) operation of five highly demanded sofa
models. Data mentioned in 2) and 3) were collected by two in-
dustrial engineers at different times of the day during the first two
weeks of the diagnosis stage. Data collection focused on two
courses of action: (i) counting manufacturing parts on all work-
stations related to sofa manufacturing (enabling a snapshot of
process material accumulation), and (ii) collecting a minimum
sample of 10 replications of all times required for process
manufacturing in each workstation. Work-in-progress (WIP) data
was collected by counting the number of units waiting for process
on each workstation; the number of delayed days was provided by
the company personnel responsible for monitoring productive
levels.
Based on diagnosis and worker proposals raised in the appraisal
stage (interviews) the experts' team developed alternative solu-
tions that focused on the design of the new work system, although
workstation and environmental solutions were also proposed. One
of the alternatives was selected for prototyping and testing after
discussions between the experts' team and two managers. Some
changes in the work system and workstations were carried out and
tested based on feedback from 11 volunteer workers involved in the
prototyping stage and from the analysis from the experts' team.
Afterwards, the new design was qualitatively validated by the same
eleven workers, two managers and experts' team.
~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002
3
Both ergonomic and production data were gathered in the
second week of the diagnosis stage, and 10 weeks late during the
second week of the validation stage. Therefore, the experts' team
performed a quantitative validation based on pre and post inter-
vention data comparison by considering:
1) general constraints based on the interviews/questionnaires
from the appraisal stage and spontaneous statement from the
workers and managers during the prototyping stage;
2) OWAS postural risk (WinOWAS®), comparing the analysis of the
films taken from workers from the four sectors during the two
weeks before the intervention and the ones taken during the
two weeks of experiment. Pre and post evaluation were done by
the same two ergonomists that based themselves on videos
taken on Tuesday of the second week of the diagnosis and
validation stages. Tuesday is usually the day of the week of
maximum productivity (Parker and Oglesby, 1972).
3) OWAS analysis was done at the recommended time interval of
30 s, in the second, third, and fourth hour in the morning and
the afternoon. The first and last hours of the journey were ruled
out assuming workers were not in their full capacity in the
beginning (because they are warming up) and at the end
(because they are tired) of the journey. Therefore 360 images
were evaluated for each of the 4 tasks: frame assembly, straps
placement, foam gluing and upholstery. Workers were picked at
random in the diagnosis stage (pre-evaluation) because the
analysis considered the worker on videotape at the selected
time. However, the same eleven workers were on tape and
evaluated in the validation stage (post evaluation).
4) estimated workload involved, which was assumed as the sum of
three indices: handling effort, work repetitiveness and work
content, as evaluated by the researchers, on a continuous scale
varying from 1 (minimum cost) to 3 (maximum cost). This
estimation was done, considering that OWAS gives an estima-
tion of postural risk, rather than an effort evaluation.
5) motion and time spent in the two work systems. The collection
of 10 repetitions for each layout and sofa models, enabled the
computation of averages and standard deviations (SD). Student-
t test with 95% of confidence level was used to evaluate statis-
tical differences on the times yielded by the traditional (func-
tional) and proposed model.
Table 1 shows the goals, period of the intervention stages, and
participants' role in each stage.
3. Results
3.1. Analysis of the system prior to redesign
3.1.1. Demographic and work characteristics
Sectors in charge of furniture manufacturing, including the four
sectors focused on this study, work the normal shift (from 7:00 AM
to 5:30 PM with 1 h for lunch break). Men are the majority in these
sectors because tasks demand high physical effort, while women
work in sectors demanding fine motor skills (such as sewing). Most
workers have low level of schooling. Table 2 summarizes de-
mographic data and the tasks performed in each sector.
3.1.2. Ergonomic appraisal
Sofas manufacturing was organized according to a functional
model with workstations operated by specialized workers (Tay-
loristic model). Production starts with the assembly of the wooden
frame, straps placement and foam gluing, which can be considered
as part of a mini-plant (MP1). After foam gluing, the product heads
to the upholstery and final assembly sectors seen as a second mini-
4 ~es et al. / International Journal of Industrial Ergonomics xxx (2015) 1e11
L.B.M. Guimara

Table 1
Overview of the intervention stages, goals and participants.

Stage Goal Period Participants

Appraisal Month 1 (2 weeks)

Qualitative Raise problems (EDIs) and solutions 2 Experts (ergonomists)
based on interviews and 26 Workers from 4 sectors
Direct and indirect (videotape) observations 2 Experts (ergonomists)
Quantitative Measure EDIs importance based on questionnaires 2 Experts (ergonomists)
77 Workers from 4 sectors
Appraisal gate Feedback and discussion with workers and 2 Experts (ergonomists)
management All workers (81) from 4 sectors
Invite volunteers for participating in next stages

Diagnosis Understand work system Month 1e2 (3 weeks) 2 Experts (ergonomists)

Analyze problems (work system, workstation, 4 Experts (2 ergonomists, 2 industrial engineers)
environment) 2 Experts (industrial engineers)
Collect and analyze production data
Diagnosis gate Feedback and discussion with workers and 4 Experts (2 ergonomists, 2 industrial engineers)
management All workers (81) from 4 sectors
2 Managers from the industrial sector

Proposal of solutions Draw alternative solutions based on workers and Month 2 (3 weeks) 4 Experts (2 ergonomists, 2 industrial engineers)
managers suggestions (from the interviews) 2 Managers from the industrial sector
and experts knowledge 11 Volunteer workers from 4 sectors

Prototyping Prototype and test work system and workstations Month 3 (4 weeks) 4 Experts (2 ergonomists, 2 industrial engineers)
Make changes 11 Volunteer workers from 4 sectors
Final test 2 Managers from the industrial sector

Validation Qualitative evaluation of tested work system Month 4 (4 weeks) 4 Experts (2 ergonomists, 2 industrial engineers)
and workstations 2 Managers from the industrial sector
11 Volunteer workers from 4 sectors
Quantitative evaluation of tested work system and workstations 4 Experts (2 ergonomists, 2 industrial engineers)
Validation gate Feedback and discussion with workers, management and directors 4 Experts (2 ergonomists, 2 industrial engineers)
2 Directors
2 Managers from the industrial sector
RH management
Workers representative
Final feedback 4 Experts (2 ergonomists, 2 industrial engineers)
2 Managers from the industrial sector
All workers (81) from 4 sectors

plant (MP2). Carpentry and foam sectors provide the materials for
MP1, and the sewing sectors provides for MP2. Expedition is the
final stage. An illustration of the sofas manufacturing process is
depicted in Fig. 1.
Table 3 presents the results from the questionnaires. EDIs
temperature and ventilation are perceived as unsatisfactory by
workers from all four sectors, suggesting that the new facilities
should have a better ventilation and insulation. Noise significantly
(p ¼ 0.025) bothers the strap placement workers, because they are
placed between two noisy sectors: frame assembly (most noise
deriving from hammering, screw driving and wood handling op-
erations) and foam gluing (major source of noise being the
exhaustion system).
There is no statistical difference among sectors for most EDIs of
the work content construct. Based on the averages in Table 3, work
is repetitive and involves high levels of physical and mental effort,
although is perceived as dynamic because there is the need for
moving frequently in order to get materials. The exception is the
strap placement sector (p ¼ 0.048), whose workers tend to perform
the job in a more restricted space. With the exception (p ¼ 0.042) of
the frame assembly sector, work is not perceived as stimulating
despite some creativity in the manufacturing of different sofa
models. Workers from foam gluing and upholstery sectors feel less
valued by the Company (p < 0.01) in comparison with their col-
leagues. Frame assembly workers recognize their work as more
important than the others since they are responsible for building

Table 2
Description of tasks and demographics of the four evaluated sectors of the furniture plant.

Sector Task Workers Gender Age (years) Average Average time in Time in company
schooling company (years) (SD) (years)
Male Female
(years)

Frame assembly 1 Stitching; 9 9 0 21e48 7.6 4.5 2.3

2 Gluing;
3 Assemblage of the basic structure of the products
Straps placement 1 Straps and springs placement in the frame of the 10 9 1 27e50 6.1 2.4 1.4
products coming from the frame assembly sector
Foam gluing 1 Glue spreading; 14 10 4 19e46 7.2 3.4 1.7
2 Foam placement
Upholstery 1 Gluing and clamping of the capes coming from 48 48 0 16e48 9.0 2.9 1.2
the Sewing sector, in the frame already foamed

~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002
 ~es et al. / International Journal of Industrial Ergonomics xxx (2015) 1e11
L.B.M. Guimara 5

Ergonomists' analysis (through direct and indirect observation,

Carpentry Sewing
Frame Strap
Foam gluing Upholstery
assembly placement
Mini Plant 1 Mini Plant 2
Fig. 1. Process fluxogram with the mini-plants 1 and 2; carpentry and sewing sectors
are suppliers for MP1 and MP2, respectively.
the sofas, while the foam gluing workers understand they can do
more than just gluing. Level of autonomy, stress and pressure for
production is in the medium range of the scale for all sectors.
With the exception of the strap placement sector (p < 0.01),
workers feel discomfort/pain in the back. This result was expected
since the allocation of the straps does not impose handling loads or
the adoption of awkward postures, as it happens in the other sec-
tors. Workers from other sectors also complain about non-
adjustable benches. Pain in arms and hands is high (p < 0.01) for
upholstery workers who use force in these parts of the body in
order to conform the capes in the sofas frames. Pain in head and
stomach is evaluated as low.
3.1.3. Ergonomic diagnosis
Workers evaluation of physical effort and pain/discomfort are in
accordance to the postural risk evaluation made by the ergono-
mists. Postures were classified into OWAS categories 4 (frame as-
sembly, foam gluing and upholstery) and 3 (straps placement),
indicating immediate need for correction. Postural risk is higher for
the back and upper limbs, mainly in the frame assembly and foam
gluing sectors, due to the nature of movements and load handling.
These results, as well as the repetitive characteristic of the tasks
carried out in the sectors, demand revision of the work organiza-
tion, redesign of workstations and use of proper devices for load
handling.
Expedition
workload evaluation and time/motion analysis) pointed out the
following aspects as having major negative impact on the current
Final work design: (i) adoption of awkward postures and heavy load
assembly
handling; (ii) unsatisfactory environmental conditions; (iii) unsat-
isfactory use of the physical space; (iv) unnecessary transportation
and long distances to be covered by people and products; and (v)
lack of diversified tasks and teamwork. Each worker, responsible for
a single task, has restricted autonomy, lacks knowledge of the
entire system, and is not responsible for the final product quality.
Although the literature emphasizes the presence of dust in
furniture manufacturing, it is restricted to the carpentry sector,
which is intended to be placed in the new facilities. The experts'
team oriented the company to address the problem. Toxic chem-
icals are used in the foam-gluing sector, and despite the exhaustion
system in the workstation, there is the need for use of masks, and
residuals might be found in the nearby workstations of other
sectors.
3.1.4. Production diagnosis
Analysis of production data indicated several inefficiencies
which included: (i) high levels of WIP that contribute for isolating
the workstations; (ii) confusing production flow and poor visibility
of the process, thus hindering collaboration between workers from
different sectors; and (iii) unbalanced cycle times, yielding high
WIP levels, long lead times and delivery delays. Estimated average
WIP (units) in each sector is displayed in Table 4. Average delayed
days were: 2.5 (SD ¼ 1.2) in frame assembly, 1.3 (SD ¼ 0.5) in strap
placement, 1.2 (SD ¼ 0.4) in foam gluing, and 2.1 (SD ¼ 0.7) in
upholstery.
3.2. Proposal of solutions: design of the new work system
3.2.1. Basis for redesign
Based on the literature review and experts' diagnosis, eight er-
gonomic parameters were deemed important in order to assure

Table 3
Results from the questionnaires used in the appraisal stage.

Sector Frame assembly Strap placement Foam gluing Upholstery

N¼8 N ¼ 10 N ¼ 14 N ¼ 45

IDE Average SD Average SD Average SD Average SD

Environment Noisea 8.7 A 4.3 3.2 B 3.3 6.6 A 3.7 7.1 A 4.2
Temp 2.0 A 2.3 1.9 A 2.4 2.6 A 2.7 2.4 A 3.4
Vent 3.1 A 3.0 1.3 A 1.7 1.6 A 0.0 2.9 A 3.5
Ilumin 10.2 A 3.8 13.7 A 0.4 10.7 A 3.2 9.8 A 4.2
Work content Phys. Effort 10.8 A 3.7 11.2 A 3.6 11.7 A 3.7 12.6 A 3.1
Mental Effort 11.2 A 2.8 10.4 A 3.4 10.3 A 2.8 9.9 A 3.5
Monotony 7.9 A 2.8 6.7 A 3.3 7.2 A 4.7 8.0 A 4.1
Limited 9.5 A 3.2 7.0 A 5.7 8.6 A 4.6 8.7 A 4.0
Creative 10.5 A 3.2 7.6 A 4.0 9.0 A 3.6 8.8 A 4.4
Dinamica 11.5 A 3.7 6.8 B 4.7 9.9 A 4.1 9.5 A 3.3
Stimulatinga 10.4 B 3.1 8.2 A 3.7 7.0 A 3.3 6.2 A 4.2
Repetitive 11.4 A 3.8 10.4 A 5.3 12.1 A 4.4 11.9 A 4.7
Responsab. 13.6 A 3.4 14.8 A 2.4 13.8 A 4.0 14.8 A 1.8
Valueda 12.4 C 3.6 9.6 B 7.0 3.0 A 3.5 5.7 A 5.4
Autonomy 11.0 A 3.4 8.1 A 5.1 9.0 A 4.2 8.0 A 4.4
Stress 8.3 A 5.8 6.6 A 3.3 9.1 A 5.6 10.7 A 5.1
Pressure 6.6 A 5.7 5.8 A 5.9 8.7 A 5.0 10.8 A 4.9
Pain/discomfort Arma 6.7 A 3.3 6.3 A 4.7 7.9 A 4.0 11.0 B 3.0
Handa 6.1 A 3.6 6.6 A 2.9 5.4 A 3.8 10.2 B 3.6
Leg 6.6 A 5.0 7.4 A 5.1 8.6 A 4.5 9.1 A 4.3
Feet 5.9 A 3.4 5.9 A 4.7 8.0 A 5.1 7.9 A 4.4
Backa 9.5 A 4.2 5.2 B 5.0 11.1 A 4.4 9.7 A 4.1
Neck 5.8 A 5.4 4.9 A 5.3 7.8 A 5.1 7.0 A 5.2
Head 4.2 A 3.6 5.8 A 6.0 4.7 A 4.6 5.3 A 4.8
Stomach 2.5 A 3.4 1.4 A 2.5 3.6 A 4.8 2.5 A 4.2
a
Statistically different at 95% confidence level.

~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002
6 ~es et al. / International Journal of Industrial Ergonomics xxx (2015) 1e11
L.B.M. Guimara

comfort and safety for the workforce in the new model: 1) work
should have a certain degree of variation; 2) work should be
organized at the group level; 3) workstations should face South to
maximize thermal comfort, mainly in the summer; 4) workstations
should be positioned near sources of natural light (windows and
doors) and artificial lighting should be replaced by a more efficient
one, to enhance visibility of work flow; 5) workstations, tools and
equipment should imply a minimum of noise, chemical emissions,
and load handling; 6) workstations dimensions should guarantee
comfort, safety, and maximum freedom of body motion (specially
arms, hands and back); 7) workstations, materials and equipment
should be positioned near to workers in order to reduce noise,
transportation, handling and waiting; 8) circulation of both work-
force and items inside the cell must ensure workers safety and
reduction of transportation times.
Parameters (7) and (8) attend both ergonomics and production
demands, as they reduce exposition of workers to hazards and
process waste. In order to improve production management, three
other parameters, aligned with the lean production paradigm
(Ohno, 1988; Shingo, 1981) were adopted: (i) just-in-time (JIT)
delivery of raw materials and components, which implied the ex-
istence of WIP caps (Hopp and Spearman, 1996); (ii) reduction of
finished stocks; and (iii) flexibility to enable the production of a
larger portfolio of sofas with minimum setup.
These eleven parameters guided the design team through the
conception of six cellular alternatives. All of them assume team-
work in order to enlarge (enable work to demand the use of
different muscles) and enrich (make work content richer) work and
to minimize idle times, load handling, unnecessary transportation
and rework. Although work design is the same, alternatives differ in
the type of aid used. Some of them rely on technology for material
transportation (e.g., a device or jig that moves among the cells and
throughout the factory) while others consider the addition of
simple transportation devices (i.e., conveyor belts, rollers and carts)
to the basic equipment and tools currently in use. Therefore,
handling effort is the only index that differentiates workload
among the alternatives.
Considering that alternatives demand multi-skilled workers,
training cost was assumed to be the same for all of them; therefore,
monetary cost of alternatives implementation increases with the
level of technology involved. However, this cost tends to be
absorbed after a short time period, since it is compensated by its
social and productivity gains.
Managers decided on one of the most conservative alternatives,
which was therefore prototyped. There are tables/shelves and carts
for keeping material in use and a roller system for transporting
components from assembly to gluing and to upholstery. Although
this alternative eliminates most manual transportation, it requires
manual motion of assembled parts to the roller system. Capes for
upholstery are kept above the table or hanged over it. Product is
then sent by cart for final assembly (placement of feet, plastic
covers, etc.), which is done outside the cell (at the Expedition).
The alternative employs eleven workers for producing two sofas
simultaneously: four assemblers at MP1, who are not only
responsible for wooden frame assembly, but also for placing straps
(tasks that were performed by different workers in the functional
model). Furthermore, the worker previously responsible for only
spreading glue on wooden pieces now glues wooden pieces and
foam. The idea was to eliminate the need for this specialized worker
by promoting the use of non-toxic water-based glue. Under a harm
free environment, any worker could perform the task with no
personal protective equipment (such as a mask) and there is no
need to “buy” workers' health and safety risk by paying insalubrity
premium. However, because at that time the technology was not
fully available, a customized workstation with exhaustion for
chemical substances was designed for the specialized gluing
operation. Six workers at MP2 became responsible for the whole
upholstery task. By training for all tasks, workers could act at all
MP1 and MP2 workstations, including the evaluation of the quality
of the product (Fig. 2).
3.2.2. Prototyping the new system
Prototyping of the selected alternative took place at the new
facilities (next to the building that accommodates the studied
sectors), which was at its final construction stage. Sources of light,
tables, rollers, carts, tools and other equipment used were available
in the factory and installed for the experiment. Prototyping
involved eleven volunteers, all male between 21 and 50 years old,
with 7 months to 9 years of work experience, therefore being
representative of the workforce.
Workers helped arranging the new system and trained for one
day with the University team before testing. They were informed
that the system was being tested, time and motion were being
recorded for further adjustments and that they should work as a
team and at their own rhythm. Besides, following the participatory
approach, they should criticize the model and interrupt the process
at any time if adjustments were necessary.
Additional adjustments were required in the prototype to bal-
ance cycle times. They included a better positioning of stands of
material supply and the introduction of carts for material

Table 4
Comparison of production outcomes from the functional and cellular work organization on five highly demanded sofa models.

Sofa model Frame assembly and placement of straps Application of glue and foam gluing

Functional (min) Cellular (min) Average Lower-tail Functional (min) Cellular (min) Average Lower-tail
improvement Student improvement Student t-test
Average time SD Average SD Average time SD Average time SD
(%) t-test (%)
time

A1 47.58 2.14 22.33 1.78 53.1 13.6a 26.95 2.15 20.02 1.45 25.7 3.65a
A2 60.66 2.98 45.37 1.97 25.2 7.81a 18.31 1.89 14.76 1.32 19.4 2.87a
A3 42.28 2.40 37.5 2.19 11.2 1.01 20.48 1.78 16.28 1.37 20.5 3.98a
A4 43.6 3.78 24.82 1.65 43.1 9.78a 19.91 2.04 10.1 0.67 49.3 5.74a
A5 61.62 5.13 35.3 1.74 42.6 8.47a 15.17 0.87 8.18 0.78 46.1 5.67a

Average 51.14 33.08 35.0 20.16 13.86 32.2

time (min)
Average 11.2 Frame 2.1 Frame 7.5 1.5 33.0 6.9 1.2 5.3 0.9 23.2
WIP (units) Frame Frame Frame Gluing Gluing Gluing Gluing Gluing
9.4 2.5 7.6 2.1 19,1
Strap Strap Strap Strap Strap
a
Statistically different at 95% confidence level.

~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002
 ~es et al. / International Journal of Industrial Ergonomics xxx (2015) 1e11
L.B.M. Guimara
transportation. During the three weeks period of prototype testing,
problems not foreseen by the experts were detected by the
workers, therefore improvements were also carried out based on
their suggestions: new positioning of the supplier cart, better or-
ganization of carpentry material, installation of auxiliary tables and
support for materials like clamps, glue and tools, and installation of
a latch in the table to make straps placement easier. In addition,
gluing was now performed on a turntable with height adjustment
to avoid uncomfortable movements, and a round table with 1.20 m
of diameter replaced the traditional rectangular one. A lighter
stapler was tested on the wood assembly operation aimed at
reducing static effort of the upper limbs.
3.3. Comparative results
Comparative analysis was performed based on data from the
second week of prototype testing assuming that they would be
more reliable. After a week, most adjustments were already made
and it could be expected that the 11 volunteers were reasonably
adapted to the new work model and research method. Upholstery
at MP2 of the prototype system could not be evaluated due to
problems on cape delivery.
In spite of workers' interest and support, the immediate su-
pervisor resisted the project and imposed many obstacles to run
the prototype, mainly in MP2. Supervisors and managers resistance
to change is common in ergonomic interventions either because
they do not see any reason for change (Vink et al., 2008), or because
it challenges the status quo (Guimara ~es, 2009; Scott, 2009) and
changes are culturally costly (Guimara ~es et al., 2012). In this
particular case, the immediate supervisor understood that accept-
ing the project was admitting he failed supervising the plant for the
last 30 years.
Table 4 presents manufacturing times (in minutes, without
considering waiting, handling and transportation) required for the
functional model and the prototyped cell for manufacturing five
types of highly demanded sofas in MP1, by three workers: two for
wood assembly/straps placement and one for foam gluing.
Fig. 2. Characteristics of the traditional (functional) and the prototyped cellular system.
~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002
7
Differences in manufacturing times measured for each sofa
model, either in the functional or the cellular design, are due to the
complexity of each model (e.g. number and diversity of compo-
nents). Nevertheless, the cellular design favored the manufacturing
of all models. Wood assembly and straps placement required, on
average, 35% less time to complete a sofa unit. As for the foam
gluing, the average cycle time in the new system was 32.2% smaller.
Adding all tasks performed in MP1, the average manufacturing time
in the proposed system is 34% smaller than the functional one.
Average WIP was reduced in 33% in the frame assembly, 19.1% in the
strap placement and 23.2% in the gluing sector, yielding average
25.1% less parts in process.
For both designs, workload and times involved in the production
of a class A sofa model (a highly demanded model) were estimated.
Workload involved in each proposed alternative was assumed as
the sum of handling effort, work repetitiveness and work content,
as evaluated by two ergonomists, on a continuous scale varying
from 1 (minimum cost, as in the case of Alternative 1) to 3
(maximum cost assumed for the functional model). By evaluating
the lightest (alternative 1 with jig and rail) and the heaviest
(functional model) work conditions, the researchers estimated that
the former implies a workload reduction of around 50% compared
to the latter (0% reduction). Based on that, the proportional work-
load reduction associated to the prototyped cellular design was
estimated as 42% less than the functional design (column 12 in
Table 4).
OWAS postural risk level reduced from 4 (frame assembly), 3
(straps placement) and 4 (gluing) to level 2 in all tasks, meaning
that instead of “immediate need for correction”, “corrective mea-
sures are necessary in the near future”. These improvements can be
attributable to the reduction of static effort of the upper limbs and
back, awkward/unnecessary posture, manual motion and load
handling, and elimination of manual transportation. Although risk
was not eliminated, progress to level 2 is a considerable reduction
of postural risk. No hazards were introduced as a result of the new
design, but better results could be achieved if workers did not need
to handle any load, as happens when they move parts from one
8 ~es et al. / International Journal of Industrial Ergonomics xxx (2015) 1e11
L.B.M. Guimara

workstation to another. OWAS is a suitable tool for evaluating dy-

and foam gluing

namic work, but it was not sensitive enough to evaluate minor

OWAS level
differences in the risk imposed by the two model designs. Some

application
limitations of the tool (such as no separation of right and left ex-

of glue
tremities; broad coding for shoulders and back, no assessment of

2
4
neck, wrists elbows and no consideration of repetition or duration
of postures) did not allow for a more rigorous evaluation.

frame assembly/
Table 5 details the ergonomics (including the estimated work-

OWAS level

placement
load and OWAS postural risk evaluation) and production results

of straps
obtained for the functional and cellular work designs, for
manufacturing one sofa at MP1. There is a substantial increase in

2
4/3
average productive level (þ46%) when comparing both layouts; in

reduction (%)
addition, process variability also reduced from 9.9 to 4.8, suggesting

Workload
that the cellular layout leads to more robust and reliable productive
processes.

42
0
Manufacturing, waiting, handling, and transportation times
(columns 2, 3, 4 and 5, respectively, in Table 5) add up to the total

Work
index
time needed for producing the sofa (column 6 in Table 5). The ex-

4.0
9.0
pected productivity gain (column 7 in Table 5) was calculated

content
considering the percentage of time reduction in comparison to the

Work
total production time in the functional model. Average waiting time

1
3
in the functional model was 18.3 min, leading to a total time of

repetitiveness
118.1 min for producing a sofa at MP1. This means that 15% of
average total time is lost due to line unbalance, lack of JIT material
delivery and setup, what was practically eliminated when adopting

Work
Ergonomics data
lean production principles for the prototyped alternative. Average

1
3
gain of the prototyped cellular design over the functional model

Handling
was 46%, 21% attributable to manufacturing time reduction and 25%

effort
related to waiting, handling and transportation time reduction.

2
3
4. Discussion

productivity
gain (%)
Averg.
4.1. On the relationship between ergonomics and production
outcomes

46
0
64.3 (4.8)
118.1 (9.9)
(min) (SD)
Total time

This study indicated that, in order to promote changes in such a

traditional company, it is mandatory to make it clear to managers
that ergonomic and production gains are compatible. In fact, the
positive results deriving from the propositions were not expected
time (min) (SD)

by the managerial staff, which assumed that the functional model

Ergonomics and production outcomes of the functional and prototyped cellular models.

6.93 (0.4)
14.25 (0.9)

would be hailed more productive. It turned out that the prototyped

Transport

cellular system met both ergonomic and production demands by

reducing postural risk, perceived fatigue/muscular pain, shortening
lead times and lowering WIP. The obtained ergonomic and pro-
time (min) (SD)

duction gains can be mostly justified by:

10.44 (2.1)
14.24 (3.6)
Handling

(i) introduction of one-piece flow of sofas in the cells. Such

decision is beneficial in terms of ergonomics and production
since it creates a meaningful job (i.e. producing sofas rather
than components), and reduces idle times while keeping
time (min) (SD)

fairly long cycle times that provide variety of tasks;

(ii) flexibility to produce different models in the same cell,
18.3 (3.2)
Waiting

0 (0)

allowing production in smaller batches and closer to real

customer orders, while improving work content by
increasing task variety;
Production data

time (min) (SD)

Manufacturing

(iii) JIT delivery of materials and components, leading to elimi-

nation of unnecessary raw materials in shelves, WIP and
46.9 (4.3)
71.3 (8.7)

finished stocks. This couples the different production steps

more tightly, thus encouraging teamwork to solve problems
as these become more visible and urgent, while reducing
System model

lead times. Lower WIP also enabled the reduction of glue

Functional

consumption due to the grouping of glue spreading and foam

Cellular

gluing, which were previously performed by independent

Table 5

workers. In the functional system, workers used to add an

extra amount of glue to avoid glue to get completely dry

~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002
 ~es et al. / International Journal of Industrial Ergonomics xxx (2015) 1e11
L.B.M. Guimara
while waiting for the next operation. In addition, a less toxic
glue was suggested to avoid human and environmental
impacts.
(iv) work enlargement, enrichment and teamwork. From an er-
gonomics viewpoint this was useful to improve workers'
satisfaction and reduce monotony; from a production
perspective this provides labor flexibility do deal with vari-
ations in demand. Furthermore, job enlargement and
enrichment reduced idle times, handling and transportation,
since there was no need to move materials so often (from one
workstation to another, for example), and most waiting time
(e.g., time between spreading glue and gluing foam) was
eliminated. It is worth noting that the factors aforemen-
tioned are consistent with lean production principles (e.g.
Liker, 2004; Ohno, 1988; Shingo, 1981), thus providing
empirical evidence of synergistic relationships between lean
principles and ergonomics.
Although Hawthorne effect might have had a positive impact on
the results, it might have been counter balanced by the insufficient
training offered to workers in the prototyping stage. Because the
project was carried out in mere four months, there was not enough
time to train for all tasks, mainly for upholstering. Both frame as-
sembly and upholstery tasks require more skills, the reason why
workers in these sectors get about 10% higher wages.
The obtained gains are expected to be higher as workers become
adapted to the new work design. In the long run, when workers get
more training and become fully multiskilled (i.e., able to perform all
tasks in MP1 and MP2), the improved system should be more
efficient and workers' wages should be higher and be the same for
all workers. Some companies may resist training workers for
multiskilling assuming that it is cheaper to keep the one-task
worker. However, the gains derived by a better work system
employing skillful workers should be economically effective, even
considering the training costs and the higher wages these workers
deserve.
4.2. The role of participatory ergonomics
Implementing ergonomic changes is difficult: in general, people
are resistant to change, often do not like to embrace new ways of
doing things, old habits are difficult to break, therefore people often
resist to participate in the change process (Neumann et al., 1999).
Although participatory ergonomics is difficult to implement in
most companies around the world (Vink et al., 2008; Guimara ~es,
2009), the studied company adopted most of the introduced con-
cepts, and keeps an Ergonomics Committee that embraced mac-
roergonomic principles, which are applied in all sectors of all
factories hold by the Company. The new production system was
installed in the new facilities under the supervision of this Com-
mittee. As of 2012, the Company has trained multiskilled workers
who are being paid about 20% more than the one-task worker.
Participatory ergonomics also supported the creation of an
appropriate environment for workers freely and fully expressing
their concerns with the impacts of the changes. For example, in the
first three days, workers were confused about the way of working
in the just prototyped new system, and complained about feeling
more pain than the usual. The researchers explained that it was due
to the use of other muscular groups to perform the new tasks.
Although the delayed onset muscle soreness (DOMS) mechanism is
not completely understood yet, it is well accepted that pain results
from soreness of the temporary changes caused in muscles by
unaccustomed eccentric exercise (Nosaka, 2008). Actually, in the
second week, they expressed a reduced fatigue and discomfort/
pain, and showed increased motivation and satisfaction with the
~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002
9
new work design, what can be explained by job enlargement and
enrichment as well as teamwork (Carayon et al., 1999; Rivilis et al.,
2006, 2008).
Overall, workers' participation was crucial for the success of the
study. Besides their suggestions, involvement to make prototyping
viable and the contributions for enhancing the cell, the relationship
between managers and workers and overall organizational climate
improved, as expected in participatory ergonomics (e.g. Maciel,
1998; Wilson et al., 2005). Workers mentioned that, previously to
the intervention, the social climate lacked willingness, trust and
respect. For example, workers reported that the supervisor used to
assign the easiest and lighter sofa models to his “friends” while
overloading other workers. The democratic character of the pro-
posed system changed this situation since all workers, as a team,
have to produce any type of product as a function of the production
demand and no longer based on the immediate supervisor wish.
4.3. Weaknesses and limitations of the study
A limitation of this study was its short period time (4 months),
what affected both the alternatives selection and the validation
stages. Because there was not enough time for discussing better
alternatives with the managerial staff, the tested solution was too
conservative and improvements were kept at a minimum. Longer
projects leave room for better solutions, more discussion and
participation, therefore more robust results.
5. Conclusion
This article presented a four months study, which applied
participatory ergonomics and cellular manufacturing to re-design
the furniture production system of a Brazilian company. Partici-
patory ergonomics was used to identify workers' needs and to
design, prototype and test a new sustainable production system for
the wooden frame assembly, straps placement, foam gluing and
upholstery sectors. Interviews/questionnaires carried out before
the intervention indicated that the existing functional model of
work organization was stressful and tiring. Body pain (mainly in
back, arms, hands and legs) was associated with repetitive work,
inappropriate postures/motion and materials handling. Designed
alternatives focused on enhancing ergonomic aspects (larger and
richer teamwork, adoption of safe motion and postures, reduced
load handling and elimination of manual transportation), reducing
wastes and turning the process more flexible to facilitate
manufacturing of different product models in reduced setup.
The tested cellular model reduced workload in 42% and postural
risk decreased from OWAS level 4 or 3 to level 2. It required 35% less
time to assembly wooden frames and straps, and 32.2% less time to
perform foam-gluing operations. The average production gain of
the cellular system over the functional one was 46%, 25% attribut-
able to elimination of unnecessary load handling, transportation
and waiting (i.e. these are both production wastes and potentially
harmful activities), and 21% being related to manufacturing times.
The positive results are due to ergonomic improvements, mainly
work enlargement and enrichment, teamwork, better workstation
design and placement of manufacturing parts and production
materials. Under the production point of view, the use of
manufacturing cells enhanced process flexibility and products flow,
just-in-time delivery of materials and components, as well as in-
ventory control. These gains confirm that ergonomic and produc-
tion goals are not incompatible and that a sustainable design is
achievable.
Although the university team did not follow up the imple-
mentation of solutions stage, it was clear that the major impact of
the project was worker empowerment and engagement in the
10 ~es et al. / International Journal of Industrial Ergonomics xxx (2015) 1e11
L.B.M. Guimara
production system. Participatory ergonomics played a major role in
bridging the gap between shop floor workers and managerial staff.
Communication, trust and respect improved during appraisal,
proposal of solutions and testing stages, what was the key for
implementing macroergonomics in the whole company.
Acknowledgments
The authors wish to recognize the pioneering characteristic of
the furniture manufacturing Company and the participation of
managers and workers, specially the ones who take part in the
prototyping stage. The research was partially supported by CETA/
FINEP, the Coordination for the Improvement of Higher Education
Personnel/Ministry of Education (CAPES Foundation) and the Na-
tional Council for Scientific and Technological Development/Min-
istry of Science and Technology (CNPq).
References
Aaltonen, M.P., 1996. Occupational injuries in the Finnish furniture industry. Scand.
J. Work Environ. Health 22 (3), 197e203.
Brown Jr., O., 1993. On the relationship between participatory ergonomics, perfor-
mance and productivity in organisational systems. In: Marras, W.,
Karwowski, W., Smith, J., Pacholski, L. (Eds.), The Ergonomics of Manual Work.
Taylor and Francis, London, pp. 495e498.
Brown Jr., O., 1995. The development and domain of participatory ergonomics. In:
International Ergonomics Association World Conference 1995 and Brazilian
Ergonomics Congress, 7, 1995, Rio de Janeiro. Proceedings…. ABERGO, Rio de
Janeiro, pp. 28e31.
Cann, A.P., MacEachen, E., Vandervoort, A.A., 2008. Lay versus expert un-
derstandings of workplace risk in the food service industry: a multi-
dimensional model with implications for participatory ergonomics. Work J.
Prev. Assess. Rehabil. 30 (3), 219e228.
Carayon, P., Smith, M.J., Haims, M.C., 1999. Work organization, job stress, and work-
related musculoskeletal disorders. Hum. Factors 41, 644e663.
Christensen, H., Pedersen, M.B., Sjogaard, G., 1995. A national cross-sectional study
in the Danish wood and furniture industry on working postures and manual
materials handling. Ergonomics 38, 793e805.
Corder, Gregory W., Foreman, Dale I., 2009. Nonparametric Statistics for Non-Stat-
isticians. John Wiley & Sons, Hoboken, pp. 99e105.
Demers, P.A., Teschke, K., Kennedy, S.M., 1997. What to do about softwood? A re-
view of respiratory effects and recommendations regarding exposure limits.
Am. J. Ind. Med. 31 (4), 385e398.
Dixon, S.M., Theberge, N., 2011. Contextual factors affecting task distribution in two
participatory ergonomic interventions: a qualitative study. Ergonomics 54 (11),
1005e1016.
Dul, J., Neumann, W.P., 2009. Ergonomics contributions to company strategies. Appl.
Ergon. 40, 745e752.
Estill, C.F., Spencer, A.B., 1996. Case study: control of methylene chloride exposures
during furniture stripping. Am. Ind. Hyg. Assoc. J. 57 (1), 43e49.
Gauthier, F., Gelinas, D., Marcotte, P., 2012. Vibration of portable orbital sanders and
its impact on the development of work-related musculoskeletal disorders in the
furniture industry. Comput. Ind. Eng. 62, 762e769.
Goldsmith, D.F., Shy, C.M., 1988. An epidemiologic-study of respiratory health-
effects in a group of North Carolina furniture workers. J. Occup. Environ. Med.
30, 959e965.
Guimara ~es, L.B. de M., 1999. Ergonomic approach: the macroergonomic method. In:
Guimara ~es, L.B. de M. (Ed.), Ergonomics in Production, fourth ed., vol. 1. FEENG,
Porto Alegre, pp. 1e16 (Chapter 1.1).
Guimara ~es, L. B. de M., Ribeiro, J.L.D., Renner, J.S., 2012. Costebenefit analysis of a
socio-technical intervention in a Brazilian footwear company. Appl. Ergon. 43,
948e957.
Guimara ~es, L.B. de M., 2009. The practice of ergonomics in the south of Brazil from a
sociotechnical perspective. In: Scott, P.A. (Ed.), Ergonomics in Developing Re-
gions: Needs and Applications. CRC Press, Boca Raton, pp. 67e87.
Guimara ~es, L.B. de M., Fogliatto, F.S., 2000. Macroergonomic design: a new meth-
odology for ergonomic product design. In: Fourteenth Triennial Meeting of the
International Ergonomics Association. International Ergonomics Association,
San Diego, p. 328.
Haims, M., Carayon, P., 1998. Theory and practice for the implementation of ‘in-
house’, continuous improvement participatory ergonomics programs. Appl.
Ergon. 29, 461e472.
Halpern, C.A., Dawson, K.D., 1997. Design and implementation of a participatory
ergonomics program for machine sewing tasks. Int. J. Ind. Ergon. 20 (6),
429e440.
Haukka, E., et al., 2008. A randomised controlled trial on whether a participatory
ergonomics intervention could prevent musculoskeletal disorders. Occup. En-
viron. Med. 65, 849e856.
~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002
Hendrick, H.W., 1990. Macroergonomics: a system approach to integrating human
factors with organizational design and management. In: Annual Conference of
the Human Factors Association of Canada, 23, Ottawa, Canad a Proceedings….
HFAC, Ottawa.
Hendrick, H.W., 1997. Good ergonomics is good economics. Ergon. Des. 5, 1e15.
Hignett, S., Wilson, J.R., Morris, W., 2005. Finding ergonomic solutions e partici-
patory approaches. Occup. Med. 55 (3), 200e207.
Holcroft, C.A., Punnett, L., 2009. Work environment risk factors for injuries in wood
processing. J. Saf. Res. 40, 247e255.
Hopp, W., Spearman, M., 1996. Factory Physics: Foundations of Manufacturing
Management. McGraw-Hill, Boston.
Hunter, S.L., 2008. The Toyota production system applied to the upholstery furni-
ture manufacturing industry. Mater. Manuf. Process. 23, 629e634.
IEMI, 2014. Press release: IEMI furniture sector 2014 review. Available from: http://
www.iemi.com.br/press-release-iemi-lanca-relatorio-setorial-da-industria-de-
moveis-no-brasil/ (accessed 22.09.14.).
Kivi, P., Matilla, M., 1991. Analysis and improvement of work postures in the
building industry: application of the computerised OWAS method. Appl. Ergon.
22, 43e48.
Kuorinka, I., Patry, L., 1995. Participation as a means of promoting occupational
health. Int. J. Ind. Ergon. 15 (5), 365e370.
Laing, A.C., et al., 2005. Study of the effectiveness of a participatory ergonomics
intervention in reducing worker pain severity through physical exposure
pathways. Ergonomics 48 (2), 150e170.
Liker, J., 2004. The Toyota Way: 14 Management Principles from the World's
Greatest Manufacturer. McGraw-Hill, New York.
Looze, M.P. de, Rhijn, J.W. van, Schoenmaker, N., Grinten, M.P. van der, Deursen, J.
van, 2005. Productivity and discomfort in assembly work: the effects of an
ergonomic workplace adjustment at Philips DAP. In: Vink, P. (Ed.), Comfort and
Design: Principles and Good Practice. CRC Press, Boca Raton, pp. 129e136.
Maciel, R., 1998. Participatory ergonomics and organisational change. Int. J. Ind.
Ergon. 22, 319e325.
Mirka, G.A., Smith, C., Shivers, C., Taylor, J., 2002a. Ergonomic interventions for the
furniture manufacturing industry: part I e lift assist devices. Int. J. Ind. Ergon.
29, 263e273.
Mirka, G.A., Smith, C., Shivers, C., Taylor, J., 2002b. Ergonomic interventions for the
furniture manufacturing industry: part II e handtools. Int. J. Ind. Ergon. 29,
275e287.
Moore, J.S., Garg, A., 1998. The effectiveness of a participatory ergonomics program
in the red meat packing industrydevaluation of a corporation. Am. J. Ind. Med.
29 (4), 402e408.
Nagamachi, M., 1995. Requisites and practices of participatory ergonomics. Int. J.
Ind. Ergon. 15 (5), 371e377.
Nagamachi, M., 1996. Relationship between job design, macroergonomics and
productivity. Hum. Factors Ergon. Manuf. 6, 309e322.
Neumann, W.P., Wells, R., Norman, R., Jeans, B., Dubblestyne, D., Harvey, H., Peter, O.,
1999. Roles and Relationships for Making Ergonomics Change: Results of a 2-
Day Focus Session with Industry Personnel. Presented at: Association of Ca-
nadian Ergonomists Annual Conference, Hull, Quebec, 1999.
Noro, K., Imada, A. (Eds.), 1991. Participatory Ergonomics. Taylor & Francis, London.
Nosaka, K., 2008. Muscle soreness and damage and the repeated-bout effect. In:
Tiidus, Peter M. (Ed.), Skeletal Muscle Damage and Repair. Human Kinetics,
pp. 59e76.
Ohno, T., 1988. Toyota Production System: Beyond Large-scale Production. Pro-
ductivity Press, Cambridge.
Parker, H.W., Oglesby, C.H., 1972. Methods Improvement for Construction Managers.
McGraw-Hill, New York, NY.
Pisaniello, D.L., Connell, K.E., Muriale, L., 1991. Wood dust exposure during furniture
manufacture e results from an Australian survey and considerations for
threshold limit value development. Am. Ind. Hyg. Assoc. J. 52, 485e492.
Pisaniello, D.L., Tkaczuk, M.N., Owen, N., 1992. Occupational wood dust expo-
sures, lifestyle variables, and respiratory symptoms. J. Occup. Med. 34,
788e792.
Ratnasingam, J., Ioras, Abrudan, I.V., 2012. An evaluation of occupational accidents
in the wooden furniture industry e a regional study in South East Asia. Saf. Sci.
50, 1190e1195.
Ratnasingam, J., Ioras, F., Swan, T.T., Yoon, C.Y., Thanasegaran, G., 2011. Determinants
of occupational accidents in the woodworking sector: the case of the Malaysian
wooden furniture industry. J. Appl. Sci. 11, 561e566.
Ratnasingam, J., Natthondan, V., Ioras, F., McNulty, T., 2010. Dust, noise and chemical
exposure of workers in the wooden furniture industry in South East Asia.
J. Appl. Sci. 10, 1413e1420.
Rivilis, I., Cole, D.C., Frazer, M.B., Kerr, M.S., Wells, R.P., Ibrahim, S., 2006. Evaluation
of a participatory ergonomic intervention aimed at improving musculoskeletal
health. Am. J. Ind. Med. 49, 801e810.
Rivilis, I., Van Eerd, D., Cullen, K., Irvin, E., Tyson, J., Mahood, Q., 2008. Effectiveness
of participatory ergonomic interventions: a systematic review. Appl. Ergon. 39,
342e358.
Rosecrance, J., Dpuphrate, D., Cross, S., 2005. Integration of participatory ergo-
nomics and lean manufacturing: a model and case study. In: Carayon, P.,
Robertson, M., Kleiner, B., Hoonakker, P.L.T. (Eds.), Human Factors in Organi-
zational Design and Management e VIII. IEA Press, Santa Monica,
pp. 437e442.
Scheeper, B., Kromhout, H., Boleij, J.S., 1995. Wood-dust exposure during wood-
working processes. Ann. Occup. Hyg. 39, 141e154.
 ~es et al. / International Journal of Industrial Ergonomics xxx (2015) 1e11
L.B.M. Guimara 11

Scott, P.A., 2009. Sustainability: an ergonomics watchword for the twenty-first
century. In: Scott, P.A. (Ed.), Ergonomics in Developing Countries: Needs and
Applications, pp. 437e444 (Chapter 34).
Sheridan, J., 1990. World-class manufacturing. Ind. Week 239 (13), 36e46.
Shingo, S., 1981. The Toyota Production System. Japan Management Association,
Tokyo.
Stone, H., Sidel, J., Oliver, S., Woolsey, A., Singleton, R.C., 1974. Sensory evaluation by
quantitative descriptive analysis. Food Technol. 28, 24e34.
Thun, J.-H., Lehr, C.B., Bierwirth, M., 2011. Feel free to feel comfortabledan
empirical analysis of ergonomics in the German automotive industry. Int. J.
Prod. Econ. 133, 551e561.
Urlings, I.J.M., Nijboer, I.D., Dul, J., 1990. A method for changing the attitudes and
behaviour of management and employees to stimulate the implementation of
ergonomic improvement. Ergonomics 33, 629e637.
Vink, P., Imada, A.S., Zink, K.J., 2008. Defining stakeholder involvement in partici-
patory design processes. Appl. Ergon. 39, 519e526.
Vinzents, P., 1988. Personal sampling of total and inspirable dust e results from a
survey in the Danish wood industry and furniture. Ind. J. Aerosol Sci. 19,
1437e1439.
Vinzents, P., Laursen, B., 1993. A national cross-sectional study of the working
environment in the Danish wood and furniture industry e air-pollution and
noise. Ann. Occup. Hyg. 37, 25e34.
Voog, L., Jansson, B., 1992. Identification and control of contact-dermatitis from
polyfunctional acrylicmonomers in 5 Swedish furniture companies. J. Environ.
Sci. Health 27, 1925e1938.
Westgaard, R.H., Winkel, J., 2011. Occupational musculoskeletal and mental
health: significance of rationalization and opportunities to create sus-
tainable production systems e a systematic review. Appl. Ergon. 42,
261e296.
Wilson, J.R., Haines, H., Morris, W., 2005. Participatory ergonomics. In: Wilson, John
R., Corllett, Nigel (Eds.), Evaluation of Human Work. Taylor and Francis, London,
pp. 933e962.

~es, L.B.d.M., et al., Participatory ergonomics intervention for improving human and production
Please cite this article in press as: Guimara
outcomes of a Brazilian furniture company, International Journal of Industrial Ergonomics (2015), http://dx.doi.org/10.1016/j.ergon.2015.02.002