ARTICLE IN PRESS

Applied Ergonomics 36 (2005) 185–192

www.elsevier.com/locate/apergo

QFD: a methodological tool for integration of ergonomics

at the design stage
Jacques Marsot
Working Equipment Engineering Department, French National Research and Safety Institute (INRS), B.P. 27– 54 501,
VANDOEUVRE LES NANCY CEDEX, France
Received 13 February 2002; accepted 27 October 2004

Abstract

As a marked increase in the number of musculoskeletal disorders was noted in many industrialized countries and more specifically
in companies that require the use of hand tools, the French National Research and Safety Institute launched in 1999 a research
program on the topic of integrating ergonomics into hand tool design.
After a brief review of the problems of integrating ergonomics at the design stage, the paper shows how the ‘‘Quality Function
Deployment’’ method has been applied to the design of a boning knife and it highlights the difficulties encountered. Then, it
demonstrates how this method can be a methodological tool geared to greater ergonomics consideration in product design.
r 2004 Elsevier Ltd. All rights reserved.

Keyword: Ergonomic; Design; Method; QFD; Knife

1. Introduction the topic of integrating ergonomics into hand tool

For a number of years, hand tool design has been a
focus of attention on the part of users, manufacturers
(Sandwick, 1995; Atlas Copco, 1998) and researchers
(Eurohandtools, 1997). Concern for imposing the least
possible demand on the user and thus a more
‘‘ergonomic’’ design (do the job harmlessly, effortlessly,
comfortably) has now complemented the initial aim
involving a concern for performance-related efficiency
(do the job better and quicker than by hand). Since the
early 1980s, we note a marked increase in the number of
musculoskeletal disorders in many industrialized coun-
tries and more specifically in companies that require the
use of hand tools: in the food industry (Armstrong et al.,
1982), in the car industry (Vanbergeijk, 1996), in
electronics (Tichauer and Gage, 1977) and in the
assembly of household appliances (Aptel, 1993).
Early in 1999, the French National Research and
Safety Institute (INRS) launched a research program on
Tel.: +33 383 502 139; fax: +33 383 502 115.
E-mail address: jacques.marsot@inrs.fr (J. Marsot).
design, within the scope of a multidisciplinary project
entitled CEROM1. This project hinges around industrial
problems applied to the meat boning and carving trades.
In this paper, we demonstrate how a specific design
method, the quality function deployment (QFD), can be
a vector for integrating ergonomics into hand tool
design and, more generally, occupational risk preven-
tion into work equipment design.
2. Problematics of integrating ergonomics into design
According to the view taken by the scientific commu-
nity, we consider design as the transformation of a
concept into a product with the aim of satisfying user
needs whilst ensuring respect for the environment,
legislation and corporate profitability (Duchamp,
1999). The initial stage of a design process therefore
involves identifying and formalizing various expectations
1
Acronym for ‘‘Conception ERgonomique d’Outils à Main’’
(ergonomic design of hand tools).

0003-6870/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apergo.2004.10.005
 ARTICLE IN PRESS
186 J. Marsot / Applied Ergonomics 36 (2005) 185–192
with regard to the product to be designed, amongst are
which those relating to ergonomics features either
explicitly or implicitly. Although much work is still
required in this area, methods for fulfilling this stage are
already available such as needs functional analysis
(AFAV, 1989) and market identification techniques
(Daetz et al., 1995; Shiba et al., 1995).
Even if user expectations and ergonomics needs have
been identified and prioritized, this does not mean that
the designed object will in fact satisfy all of them.
Subsequent stages of the design process (overall design,
detailed design and manufacturing) are, in fact, usually
managed by engineering specialists. The inevitable
problems that arise during these stages, combined with
the difficulty or even absence of communication between
the engineering specialists and those representing
different disciplines such as ergonomics, can produce
an adverse and/or unpredictable impact on satisfying
such needs, especially those associated with occupa-
tional risk prevention. These needs are indeed often
perceived as design constraints (Didelot et al., 2000) and
are consequently addressed only at the end of the design
process through the adoption of remedial measures
embodying compromises, which can subsequently turn
out to contradict operational needs. For example, the
diameter of a power tool handle will often depend on
technical constraints involving power supply (electric or
pneumatic motor, electric batteries, etc.) to the detri-
ment of ergonomic criteria. Another ergonomics inte-
gration difficulty relates to what we call the paradox of
design ergonomics: ‘‘to express something effectively
based on a work situation, we must wait until it is fully
designed, yet then it will be too late to intervene in its
design’’ (Theureau and Pinsky, 1984). This problem can
only be overcome by an iterative design process that
allows the validation of each product development
phase by all design actors (Pomian et al., 1997).
QFD appears to be a methodological tool geared to
supporting greater ergonomics consideration in product
design because its aims are to safeguard customer needs
throughout the design process, to promote communica-
tion between design actors (engineers, ergonomists,
users, etc.) and to highlight possible contradictions
between the various design parameters.
3. The QFD method
In the 1970s, the drive for perpetual product
improvement led Japanese companies to seek optimiza-
tion and rationalization in the design of their products
and processes. It is in this context that the QFD method
was created and developed. One of its founders, Dr Yoji
Akao, defined the concept as follows: ‘‘QFD provides
specific methods for guaranteeing quality at each stage of
the product development process, starting with design. In
other words, it is a method for introducing quality right
from design stage to satisfy the customer and to transform
customer requirements into design objectives and key
points that will be required to ensure quality at production
stage’’ (Akao, 1993).
Introduced first in North America then in Europe in
the 1980s, this method has since experienced strong
development in the majority of industrialized countries.
As its name suggests, the QFD approach is based on
deploying user expectations (the ‘‘Whats’’) in terms of
design and production-related parameters (the ‘‘Hows’’)
for the new product. This process is represented by a
succession of double entry ‘‘Whats/Hows’’ tables allow-
ing the correlations between entries to be identified and
prioritized. The first matrix, also referred to as ‘‘the
house of quality’’ (HoQ) because of its shape (Fig. 1), is
the most recognised form of QFD. In addition to
‘‘Whats/Hows’’ correlations, this matrix allows the
integration of elements related to analyzing product
competition and to identifying synergies and/or contra-
dictions between different product characteristics. Thus,
this matrix offers the twin advantage of facilitating the
transition between the world of the user and that of the
designer, and of combining in the same document all
effective data for decision-making in relation to product
development. For further details of the QFD method
and its application, we recommend the reader to refer to
the bibliographical references (Akao, 1993; Daetz et al.,
1995; Madu, 2000).
4. Application of the house of quality
In the remainder of this paper, we present the work
undertaken within the scope of the CEROM project on
redesigning a boning knife (Fig. 2) to illustrate the
application of the house of quality matrix and to
demonstrate its advantage for improved integration of
ergonomics in design.
First of all, the QFD is not an occupation-specific tool
only used by specialists (which is the case with CAD/
CAM2 for example). On the contrary, this is a multi-
disciplinary method that has to be implemented within a
working group combining different design players. In
our case study, this work was carried out by a team
combining engineering and ergonomics specialists, bon-
ing knife users (deboners and sharpeners) and manu-
facturers.
4.1. Drawing up the ‘‘whats’’ list
As mentioned above the initial stage of a design
process involves identifying and formalizing various
2
CAD-CAM: Computer-Aided Design–Computer-Aided Mechan-
ical.
 ARTICLE IN PRESS
J. Marsot / Applied Ergonomics 36 (2005) 185–192 187

« Hows
vs
Hows »

Product characteristics
« Hows »

Engineer « Wolrd »
Customer Relationships between
needs and Competition
needs and characteristic s
expectations Analysis
« Whats vs Hows »
« Whats »

User « World »

Importance weights

Target values for

characteristics

Achieved values for

characteristics

Concept Evaluation

Fig. 1. Illustration of main parts of the HoQ.

problems for the users at the workplace. To reach this

goal, an ergonomic study was made up of:

interviews with managers of slaughterhouse and meat

packing plants to know more about their main choice
criteria (price, effectiveness, safety, hygiene, etc),

interviews with operators (a total of 196) to find out
their experience and viewpoint on the tools used
(Table 1),

a direct and/or video observation of the conditions in
Fig. 2. Example of a current boning knife. which knives were used.

After that, all these needs were then characterized

expectations with regard to the product to be designed
(‘‘Whats’’ list). Within the scope of the CEROM project,
this ‘‘Whats’’ list was obtained following a needs
functional analysis approach. The research resulted in
a list of 8 needs (Fig. 3). Application of this method
enabled a functional specification to be drafted for the
knife to be designed. A step that cannot be ignored in
such an approach is to gather all information on the use
of the existing tool and particularly to record the
and prioritized. In order to obtain a consensus
decision within the group, we used the analytical
hierarchy process (Saaty, 1987). It is based on pairwise
comparison between needs. It reduces the number of
needs items that each team member has to consider at
any one time to be able to assign a relative importance
weight.
With respect to these expectations, the QFD method
then allows assessment of existing competitive products
and establishes aims to be achieved for the product to be
 ARTICLE IN PRESS
188 J. Marsot / Applied Ergonomics 36 (2005) 185–192

Fig. 3. HoQ for boning knife.

designed. The field survey revealed that users did not
consider a particular brand to be significant. Then, it
was decided to summarize the nettings in the form of a
score (min ¼ 1, max ¼ 5) attributed to all current
boning knives and to the knife to be designed (Fig. 3).
For example, for the need ‘‘Not cause pain’’ the work
team evaluates that the current boning knives is doing
fairly (score 3) and targets a score of 5 for the future
knife.
In respect of our perticular case, expectations requir-
ing greater satisfaction are obviously those involving
occupational risk prevention (n1 3 and n1 4). Moreover,
the field survey revealed user dissatisfaction with respect
to cutting performance and hand grip (Table 1). Then,
 ARTICLE IN PRESS
J. Marsot / Applied Ergonomics 36 (2005) 185–192 189

Table 1
User knives experience (Claudon, 2000)

Part of the tool Viewpoint on Number of

operators (%)

Handle Inadequate shape 22

Inadequate length 12
Inadequate diameter 16
Inadequate grip 45
Inadequate comfort 31
Inadequate hardness 23 Fig. 4. Illustration of removable blade and bimaterial handle design.
Handle guarding Inadequate height 31
Insufficient protection 25
Blade Inadequate length 33 satisfied. They can then be prioritized4 and target (‘‘X’’
Inadequate height 30
symbol in the evolution line of the HoQ) or limiting
Inadequate thickness 17
Inadequate stiffness 20 values (‘‘k’’ or ‘‘m’’ symbols) can subsequently be fixed
Inadequate curvature 14 for them. Matrix presentation very easily allows on the
Bad sharpness 42 one hand, assessment of the impact of ergonomics
Too short duration of life 51 criteria on other expectations and, on the other hand,
prediction of the consequences of modifying one or
more design parameters on ergonomic expectations.
Thus, in terms of our boning knife redesign problem,
corresponding expectations (n1 1 and n1 6 for cutting
all the target values for the ergonomic criteria were
performances and n1 5 for hand grip) must therefore
defined by following an activity analysis of operators,
also be better satisfied.
experimental results and also a bibliographic study
The second stage of a design process is the pre-
(Claudon, 2002). One of the main criteria identified with
liminary design phase in which the design parameters
respect to the ‘‘not cause pain’’ expectation is the shape
and conceptual diagrams are drawn up and analyzed.
of the knife handle (strong correlation). Moreover, this
Within the HoQ process, this stage corresponds to
parameter correlates with several other expectations
drawing up the ‘‘Hows’’ list, establishing ‘‘Whats’’/
and, as a result, it turns out to be one of the most
‘‘Hows’’ correlations, and seeking interactions between
important knife criteria (Fig. 4). If one considers that in
‘‘Hows’’.
ergonomics there is no average individual (Roussel,
1996), ideally this parameter should not be given a single
4.2. Drawing up the ‘‘Hows’’ list and establishing target value but several and even a specific value for
‘‘Whats’’/‘‘Hows’’ correlations each individual to take into account the variation in
hand anthropometry. Consequently, for this parameter,
Drawing up the ‘‘Hows’’ involves listing product and also for the ‘‘Blade shape’’ we retain as the limiting
characteristics that will enable the previously identified value a user satisfaction rate higher than to the rate
needs to be satisfied. These characteristics could be obtained during the activity analysis.
engineering and/or ergonomics characteristics (handle
shape for example) and they must be measurable (Target
versus Achieved values – Fig. 1). Then the relationship 4.3. Seeking interactions between ‘‘Hows’’
between needs and product characteristics could be
determined. Based on a consensus decision, the team This stage of the HoQ process involves a paired
assigns a ‘‘strong’’, ‘‘medium’’, ‘‘weak or no’’ relationship comparison of the different design characteristics. We
value to each specific ‘‘what/how’’ pairing. For this an thereby obtain a ‘‘roof’’-shaped half-matrix allowing
appropriate scale3 (9–3–1) is applied, illustrated using design characteristics to be identified in terms of
symbols (K–J–X). This stage is one of the key synergies (+sign) and opposites (sign).
elements of the QFD method because it is the one With respect to our specific problem, the advantage of
which permits the transition to be made between what this triangular matrix is that it highlights potential
the user wants and what the designer offers. conflicts between ergonomics criteria and other design
It is at this stage of HoQ establishment that parameters from the outset in the product development
ergonomics criteria are effectively integrated into the 4
The importance weighting is obtained by multiplying numerical
product design process. These criteria are the para- values of the relationships (1-3-9) by the weighting for the relevant
meters that allow ergonomics-related expectations to be needs, and the results added for the matrix column associated with
each design requirement. The total is displayed in the ‘‘Importance’’
3
These values are those usually found in literature (Akao, 1993). line of the HoQ.
 ARTICLE IN PRESS
190 J. Marsot / Applied Ergonomics 36 (2005) 185–192
process. Resolving these conflicts using creative techni-
ques will avoid implementing remedial measures that are
usually compromise-based and therefore not totally
satisfactory, at the end of the design process.
For example, analyzing our boning knife HoQ ‘‘roof’’
reveals a negative interaction between the surface
hardness of the handle, which must not be high (‘‘k’’
symbol in the evolution line) to prevent causing pain,
and the rigidity of the blade/handle connection, which
must be as high as possible (‘‘m’’ symbol) in order to
satisfy usability needs (Fig. 3).
4.4. Assessing concepts
In our case study, the working team generated 5
different boning knife concepts. On the lower side of the
house of quality, always by using symbols, the working
team evaluates how these concepts are doing with the
product characteristics and then it is possible to choose
which one will best meet user expectations.
The concept selected by the team is a design featuring
a removable blade associated with a bimaterial handle
(Fig. 4). Attractions of this concept are on the one hand
resolving the conflicts previously described and on the
other hand favoring handle adaptation to hand anthro-
pometric characteristics and the facilitating of blade
choice in relation to the tasks to be performed. Other
advantages are also expected:

greater accuracy in blade sharpening activities be-
cause the handle and especially the guard on current
single piece knives hampers accurate blade position-
ing,

easier recycling,

easier knife personalization and identification through
handle color and/or marking.
Currently, prototypes of knives based on this concept
that incorporated the proposed ergonomics criteria
(Claudon, 2002) are currently made in order to test
them in real working conditions by the users and then to
validate their usability. At the time of writing this paper,
for reasons of industrial ownership (registration of
patent in progress) we cannot describe in details, the
proposed technical solution.
5. Conclusion
The validity of the QFD method for linking customer
expectations and product definition is increasingly
recognized today and it can be asserted that this
methodological tool is indeed capable of providing a
solution to the problems of integrating ergonomics into
product design (Haapalainen et al., 2000). Starting from
an activity analysis, inside the work team the ergonomist
will be in a position to:

integrate ergonomics-related expectations into the
‘‘Whats’’ list

take part in drawing up the ‘‘Hows’’ list by integrating
the necessary ergonomics criteria

contribute to obtain a consensus decision within the
group for the various degrees of correlation for
expectations and/or ergonomics criteria

identify possible contradictions between these ergo-
nomics criteria and other design parameters (‘‘the
HoQ roof’’)

identify the solution concept from those proposed
that best responds to ergonomics-related expectations

anticipate the consequences of modifying a specific
design parameter on ergonomics-related expectations.
Furthermore, by ensuring a visual grouping of all
effective data for decision-making in relation to product
design, QFD matrices can be considered as common
frames of reference encouraging communication be-
tween different design players. As a result, the QFD
method integrates perfectly into a concurent engineering
approach (Tsuda, 1995; Sivaloganathan et al., 1995).
This concept, which appeared in the 1980s with the main
aim of reducing new product development time, is
recognized today as an organizational system encoura-
ging the integration of ergonomics into design (Jouffroy
et al., 1998; Roussel, 1996). Moreover, the HoQ can be
recognized as a graphical model of the product allowing
validation by all actors involved in the needs expression
and specification stages (Fig. 5). It plays a similar role to
digital models and/or prototypes in an iterative design
process which was also called ‘‘ergonomic development
process’’ (Laprie, 1996).
Finally, we should stress that recent research work
(Terninko, 1998; Wu, 1997; Leon and Aguayo, 1998;
Martin and Bocquet, 1999) is attempting to integrate
QFD into a logical sequence of methodological tools to
thereby make formalization of the overall design process
possible.
Drawing up QFD matrices does, nevertheless, raise
certain difficulties. The main one involves the creation of
enormous matrices, which subsequently become unusa-
ble, with a view to being exhaustive. It is therefore very
important to establish priorities for the elements to be
deployed. For example, needs such as ‘‘Not cause
pressure points’’ or ‘‘Not remain in a dangerous position
when posed’’ appeared in early versions of the list
‘‘Whats’’ list. They were deleted because they are, in
fact, ways of satisfying actual needs, which are not to
cause pain or injury to the operator.
A second difficulty is associated with manipulating
the matrices combined with the need to keep them up to
date. As a communication support, they must, in effect,
 ARTICLE IN PRESS
J. Marsot / Applied Ergonomics 36 (2005) 185–192 191

Concept

Expression of needs
Needs analysis
Specifications
Modelling of
Overall design concepts

Modelling of
Detailed design functionalities model

Prototyping of Functional
functions prototype

Pilot production
un
Production
Checking of product in its
sociotechnical environment
Development phases Real-life situations
Result

Fig. 5. Ergonomic development process (Laprie, 1996).

be regularly updated or else they will quickly become
obsolete. Use of computing tools is an effective aid in
this field.
References
Afav, 1989. Exprimer le Besoin—Application de la démarche
Fonctionnelle. Afnor, Paris, ISBN 2-12-476911-1, 372pp.
Akao, Y., 1993. Prendre en compte les besoins du client dans la
conception du produit. Afnor, Paris, ISBN 2-12-468311-X, 349pp.
Aptel, M., 1993. Etude dans une entreprise de montage d’appareils
électroménagers des facteurs de risque professionnels du syndrome
du canal carpien. Documents pour le Médecin du Travail 54,
149–164.
Armstrong, T.J., Foulke, J.A., Joseph, B.S., Goldstein, S.A., 1982.
Investigation of cumulative trauma disorders in a poultry proces-
sing plant. American Industrial Hygiene Association Journal 43,
103–116.
Atlas Copco, 1998. L’ergonomie des machines portatives. ISBN 91-
630-5217-2, 180pp.
Claudon, L., 2000. Ergonomic Hand Tool Design: Interview of users—
ERGON-AXIA’ 2000: Second International Conference on
Ergonomics and Safety for Global Business Quality and Produc-
tivity, Warsaw, Poland.
Claudon, L., 2002. Ergonomic Hand Tool Design Project. Final report
(in french), INRS, Nancy, France, 2002, 15pp.
Daetz, D., Barnard, B., Norman, R., 1995. Statistical methods for
rates and proportions. Wiley, New York, USA, ISBN 0-471-13277-
2, 300pp.
Didelot, A., Fadier, E., Ciccotelli, J., 2000. Contributions and
limitations of standardization with respect to automated system
design. Proceedings of ESREL 2000, Foresight and Precaution,
Cottam, Harvey, Pape and Tate (Eds.), Balkema, Rotterdam,
ISBN 90-5809-140-6, pp. 127–131.
Duchamp, R., 1999. La conception de produits nouveaux. Hermès
Editions–Paris, ISBN 2-7462-0045-7, 191pp.
Eurohandtool, 1997. Usability, ergonomics, quality and productivity
of non-powered hand tools. European Commission Brite Euram
Project BE96-3735, coordinated by Tampere University of
Technology, 1997–1999.
Haapalainen, M., Kivisto-Rahnasto, J., Mattila, M., 2000. Ergonomic
Design of non-powered hand tools: An application of quality
function deployment (QFD). Occupational Ergonomics 2 (3),
179–189.
Jouffroy, D., Ciccotelli, J., Martin, P., 1998. Contribution of a
distributed design method to the safety of numerically controlled
woodworking machines. IMS Europe 98, First International
Workshop on Intelligent Manufacturing Systems, Lausanne,
Switzerland, pp. 449–458.
Laprie, J.C., 1996. Guide de la sûreté de fonctionnement. ISBN
2.85428.382.1, CEPADUES éditions, Toulouse, 369pp.
Leon, L.R. and Aguayo, H., 1998. A new model of the design process
using QFD/FA/TRIZ. Proceedings 10th annual Quality Func-
tional Deployment Symposium, 8pp.
Madu, C.N., 2000. House of quality in a minute. CHI Publishers,
Fairfield CT 102pp.
Martin, C., Bocquet, J.C., 1999. Conception intégrée. Interopérativité
des méthodes: AF, QFD, AMDEC dans le cadre du projet
PIRAMID. Colloque national PRIMECA, pp. 247–254.
Pomian, J.L., Pradere, T., Gaillard, I., 1997. Ingénierie et ergonomie—
CEPADUES Editions. Toulouse, ISBN 2.85428.442.9, 259pp.
Roussel, B., 1996. Ergonomie en conception de produits: Proposition
d’une méthode centrée sur la formulation de principes de solutions
ergonomiques dans le processus interdisciplinaire de conception de
produits. Thèse de doctorat en génie industriel, sciences des
systèmes et des produits industriels, Ecole Nationale Supérieure
d’Arts et Métiers, Paris, 245pp.
Saaty, T.L., 1987. Rank generation, preservation and reversal in the
analytic hierarchy decision process. Decision Sciences Journal,
Atlanta 18, 157–177.
Sandwick, 1995. Ergonomie—Une perspective de recherche pour des
outils à mains ergonomiques. M-0015-FRE-95, 10pp.
Shiba, S., Jouslin de Norav, B., Morel, M., 1995. La Conception à
l’Ecoute du Marché. Editions INSEP, ISBN 2-901323-63-4.
 ARTICLE IN PRESS
192 J. Marsot / Applied Ergonomics 36 (2005) 185–192
Sivaloganathan, S., Evbuomwan, N.F.O., Jebb, A., 1995. The
development of a design system for Concurrent engineering.
Concurrent Eng.: Res Appli. 3 (4), 257–269.
Terninko, J., 1998. The QFD, TRIZ and Taguchi connection:
customer-driven robust innovation. http://www.triz-journal.com/
archives/98jan/article2/98jan-article2.htm.
Theureau, J., Pinsky, L., 1984. Paradoxe de l’ergonomie de conception
et logiciel informatique. Rev. Conditions Trav. 9, 25–31.
Tichauer, E.R., Gage, H., 1977. Ergonomic principles basic to hand
tool design. Am. Ind. Hyg. Assoc. J. 38, 622–634.
Tsuda, Y., 1995. QFD models for concurrent engineering development
processes of automobiles. Concurrent Eng.: Res. Appli. 3 (3),
213–220.
Vanbergeijk, E., 1996. Selection of power tools and mechanical assists,
Ergonomic considerations of hand-held power tools for control of
occupational hand and wrist injuries. In: Hand Tools: Ergonomics
Issues in Evaluation and Selection, University of Michigan, USA.
Wu, A., 1997. Integration of QFD, TRIZ, and Robust Design:
overview & luggage case study. American supplier Institute,
Livonia 29pp.