Content uploaded by Monica Bordegoni
Author content
All content in this area was uploaded by Monica Bordegoni on Jul 15, 2016
Content may be subject to copyright.
1 23
Journal of Intelligent Manufacturing
ISSN 0956-5515
J Intell Manuf
DOI 10.1007/s10845-012-0662-5
An approach for capturing the Voice of the
Customer based on Virtual Prototyping
Marina Carulli, Monica Bordegoni &
Umberto Cugini
1 23
Your article is protected by copyright and
all rights are held exclusively by Springer
Science+Business Media, LLC. This e-offprint
is for personal use only and shall not be self-
archived in electronic repositories. If you
wish to self-archive your work, please use the
accepted author’s version for posting to your
own website or your institution’s repository.
You may further deposit the accepted author’s
version on a funder’s repository at a funder’s
request, provided it is not made publicly
available until 12 months after publication.
J Intell Manuf
DOI 10.1007/s10845-012-0662-5
An approach for capturing the Voice of the Customer based
on Virtual Prototyping
Marina Carulli ·Monica Bordegoni ·Umberto Cugini
Received: 16 September 2011 / Accepted: 17 May 2012
© Springer Science+Business Media, LLC 2012
Abstract The need for companies to improve their
competitiveness may lead to innovation and the reconcep-
tualization of traditional products and processes, with com-
panies making an effort to enhance product elements related
to functionality, attractiveness, technology and sustainability,
and implementing mass-customisation concepts. Mass-cus-
tomised products are developed to satisfy specific customer
needs, in line with increasing demand for product variety
and customisation. The analysis of what customers really
want, capturing the Voice of the Customer (VOC), is one
of the strategies used to establish effective product develop-
ment processes. Using a VOC survey, it is possible to trans-
form customer needs into the functional and psychological
requirements of the product. This paper presents a method-
ology based on Virtual Reality (VR) technologies to sup-
port the capturing of the VOC in regard to the visual, haptic
and auditory characteristics of products. This method can be
applied to the beginning of the product development process,
to allow companies to deduce from the data the requirements
of new industrial customised products. A flexible and inter-
active Virtual Prototype (VP) of a product category is then
developed as a product platform in a draft version by design-
ers and configured according to customer needs, using an
immersive VR environment. This method, based on the use
of VP, reduces the number of physical prototypes that need
to be manufactured during the product development process,
thus reducing overall costs. In addition, the VP based method
supports the mass-customisation process of products through
the real-time integration and collection of data for product
configuration preferences, involving as many users as possi-
ble representative of the target users of the new products.
M. Carulli (B
)·M. Bordegoni ·U. Cugini
Dipartimento di Meccanica, Politecnico di Milano, Via La Masa 1,
Milan, Italy
e-mail: marina.carulli@polimi.it
To demonstrate this process a case study concerning the
development of the VP for a washing machine, a sum-
mary of test sessions with users and results are presented.
Specifically, the results presented in this paper are related
to improvements in capturing the VOC and reductions in
Virtual Prototyping cost and time.
Keywords Voice of the Customer ·Product customisation ·
Virtual Prototyping
Introduction
In the changing scenario of the global market and in a
period of increasing worldwide competition in both devel-
oped and developing countries, companies need to remain
competitive, consolidate their position in the market and
strengthen their leadership. As a consequence, companies
have enhanced and improved elements related to func-
tionality, attractiveness, affordability, technology and sus-
tainability in traditional industrial products. Additionally,
companies have implemented the concept of mass-customi-
sation to provide customised products by developing more
flexible and efficient processes in relation to volume, quality
and cost (Da Silveira et al. 2001). With the aim of distin-
guishing themselves from industries that are located in low-
labour-cost countries, western industries use the design and
the development of best-in-class products, which are cus-
tomisable for individual customers, as one of their more
powerful strategies. These customised products, which are
usually based on a common platform and then configured
with different modules, are developed to maximise individ-
ual customer satisfaction at a cost nearly equivalent to the
cost of mass-produced items (Hart 1995).
The complete satisfaction of customers depends on the
fulfilment of the specific needs they may have, and these
123
Author's personal copy
J Intell Manuf
needs are in line with increasing demand for product variety
and customisation (Kotler 1989). The main objective of com-
panies is to create products that can represent value for indi-
vidual customers and attract and convince them to purchase
these products. In this scenario, capturing the customer needs
assumes a fundamental role in driving the product develop-
ment process.
Currently, the mass customisation concept is mainly
applied to wearable or individual products, for example,
shoes, clothes, cars, watches, luxury products and so on. In
these cases, companies usually offer customers the possi-
bility of personalising some of the technical features of the
products or other features that are linked to their aesthetic
appearance, such as colour, covering, finishing, etc. Never-
theless, expanding industrial competition has increased the
need for production strategies focused on individual custom-
ers in other product categories as well (Da Silveira et al.
2001), for instance, information appliances, mobile phones,
home appliances and so on. Specifically in these cases, and
in addition to the aesthetic appearance and technical features,
the interaction modalities are fundamental elements of dif-
ferentiation and of value for the customers. Consequently,
the possibility of customising the interface of these products
is becoming an increasingly key element of business strategy.
The analysis and mining of what customers really want,
both in terms of explicit needs and unspoken wishes, for
instance, capturing the Voice of the Customer (VOC), is a
strategy used to obtain the foundation of an effective prod-
uct development process focused on customisation. Start-
ing from a survey of the VOC, it is possible to transform
customer needs into functional, psychological, technologi-
cal, emotional and communicative product requirements and
characteristic components.
However, while capturing the VOC regarding the aesthetic
appearance of the products is a common practice, capturing
the VOC concerning the haptic characteristics of products
and of their interfaces is still a rare phenomenon, and the
results are not usually fully reliable. This difference origi-
nates from the characteristics of the human perceptive sys-
tem. In fact, it is our experience that during comparative test-
ing sessions in which several real prototypes are used and
evaluated over several time intervals (necessary to carry out
the modifications requested by customers), customers are not
able to express an objective assessment about the haptic char-
acteristics of product components, such as force feedback,
size identification, shape recognition and so on.
This paper presents an innovative Virtual Reality (VR)
Virtual Prototyping (VP) approach for capturing the VOC
concerning visual, haptic and auditory characteristics of
products at the beginning of a product development process.
The VP approach can be used to collect data and deduce
the requirements of new customised industrial products. In
the VP approach, product designers and engineers create a
flexible and configurable Virtual Prototype (VP) of a prod-
uct category, such as home appliances, information devices,
etc., as a product platform in a draft version. The VP is subse-
quently configured according to consumer needs using an im-
mersive VR environment. The improvements and the accom-
plishments that the use of the VP approach could bring to the
product development process can be summarised as follows:
– the VP approach avoids the need to manufacture several
real prototypes to capture the VOC, which reduces the
time and cost of the product development process and
enables companies to deliver new products to the market
quickly;
– the VP approach to product design is intrinsically collab-
orative because it allows the integration of the needs of
the customers, the product designers and the engineers;
– the use of the VP approach to capture the VOC during the
early stages of the product development process allows
for quick testing of several product configuration varia-
tions and supports capturing unspoken customer needs;
– the involvement of the customers in the early phases of
product development supports the personalisation of the
product design process starting from the product plat-
form and allows for the definition of customised prod-
ucts.
On the opposite side, a limitation of the VP approach is that it
is mainly used to create customised products for small groups
of customers, while other approaches are more suitable for
developing mass-customised products for large groups of
users.
This paper is organised as follows. Section “Research
background” reviews the existing literature related to the
research areas being analysed, specifically, reviews of Voice
of the Customer survey methods, the most-well known
methods to translate customer needs into product require-
ments, and Virtual Prototyping techniques to support prod-
uct development processes. Section “Research background”
also introduces some open issues in these areas that must
be overcome. In “The Virtual Prototyping approach”, the
Virtual Prototyping (VP) approach is described, with its key
elements. Section “A case study” describes an experimental
case study using the VP approach, presents an analysis of the
data collected and introduces possible benefits and applica-
tions. In “Conclusion”, the authors make some concluding
remarks and foresee possible future work.
Research background
Voice of the Customer (VOC) survey methods
The main objective of a VOC survey is to capture all the
features of a product that could be relevant to customers
123
Author's personal copy
J Intell Manuf
and to obtain information useful for defining the functional
requirements, design specifications, and psychological and
emotional aspects of the product. There are several types
of data that could be captured using a VOC survey. Ulwick
(2005) classified VOC survey data according to usefulness in
deriving the functional requirements and design parameters
of a new product. ‘Functions’, ‘outcomes’ and ‘constraints’
are useful. ‘Solutions’, ‘opinions’, ‘needs’ and ‘benefits’ are
not useful. They can sidetrack designers. While traditionally
VOC surveys are carried out by the marketing departments
of companies, generating distortions in translating the VOC
into functional requirements at the present moment, there
is a move to involve the engineers and the product design-
ers more in researching the customer needs. Nevertheless,
many studies have shown that VOC surveys based on custom-
ers’ statements do not express real customer needs (Wicker
1969;Loftus and Wells 1984). Currently, to overcome these
problems, ethnographic methods, which study the behaviours
of a small number of testers when they are using products,
are used. However, despite attempts to improve VOC sur-
vey methods, some issues still remain. These issues relate to
the difficulty in collecting data about ‘unspoken’ customer
needs, which could represent the most important input for
defining product requirements.
Once the data about customer needs are gathered, the data
have to be analysed and organised into a well-defined set of
aggregated customer needs. Statistical techniques are used
to detect and define metrics that compare numerical values.
Furthermore, the data in ‘word and notes’ form are analysed
to discover and identify their meanings, usually by using the
affinity diagram method (Kawakita 1991;Spool 2004). The
affinity diagram method is a four-step process for organising
the data from VOC surveys by sorting into logical categories
(ideas, issues, solutions, problems) so that it is possible to
understand the essence of a problem or solution.
Capturing the VOC is considered the first step in both the
classical product development process and the mass-custom-
ised product development process, which usually consist of
the steps shown in Fig. 1. These main phases are common to
many different product development process models (Jones
1992;Kroll et al. 2001;Shaw 2001;Ulman 2003;Risdiyono
and Koomsap 2011). These phases are described on the basis
of different approaches that influence the structure of the
product development process model.
From customer needs to product functional requirements
After capturing the VOC, aggregated customer needs that
have been identified are transformed into functional require-
ments that can be acted on by the engineering team. Many
methodologies have been developed to support this transfor-
mation process. Pahl et al. (2007) present a procedure to fol-
low for setting up, refining and extending a requirements list.
Fig. 1 The product development process and the mass-customised
product development process
Also, if these authors do not suggest a systematic method to
identify the functional requirements, they propose that, start-
ing from the comprehension of customer needs, three types of
specific functional requirements can be formulated. These are
‘basic requirements’, ‘technical performance requirements’,
and ‘attractiveness requirements’, which have subsequently
to be ranked and arranged in a clear order to be used in the
next phases of the product development process. More, the
Quality Function Deployment (QFD) methodology (Akao
1997) can be used to support the process of transforming the
VOC data into functional requirements. The data captured
using the VOC survey (customers’ statements about qualita-
tive characteristics of products) have to be listed as needs, and
the list represents the input for a relationship diagram called
QFD House of Quality (Cohen 1995;Clausing 1994). From
this list the product functional requirements are derived and
constantly verified during the product development process.
The approach described above is usually implemented in
the case of product development processes focused on devel-
oping mass-customised products. Specifically, the mass cus-
tomisation level ‘collaborative’ (according to the Gilmore
and Pine’s 1997 classification of mass customisation lev-
els) foresees designers having a dialogue with customers
and referring to specific projects. This can also be extended
to the levels of named ‘fabrication’ and ‘assembly’, which
foresee the capturing of customer needs at different stage of
the product development process to develop mass-custom-
ised products (Da Silveira et al. 2001). For these levels of
mass customisation, the capturing of the VOC is a fundamen-
tal activity during the entire product development process,
which implements a User-Centered Design (UCD) approach
123
Author's personal copy
J Intell Manuf
(Kramer et al. 2000). The UCD is a design theory introduced
by Norman and Draper (1986), which gives extensive atten-
tion to customer needs to optimise the design solutions for
new products at every stage of the product development pro-
cess. In the case of the implementation of the UCD approach
within a mass customisation process, designers must identify
the needs of several categories of customers and translate
these needs into products constituted by a common platform
and a variety of modules (components) (Kramer et al. 2000).
Additionally, in the mass customisation and personalisation
area, Risdiyono and Koomsap (2011) introduce the Design by
Customers (DBC) concept. Conversely, the DBC focuses on
an approach that integrate the customer involvement decou-
pling point (CIDP), which is the point where the customer
can be involved in the creation of product value, tradition-
ally placed at the end of the process, in various points of
the design and manufacturing processes to define his or her
desired product specifications according to his or her indi-
vidual needs.
Virtual Prototyping techniques for supporting the product
development process
Several techniques have been developed to produce proto-
types of design solutions in development, which are useful
for supporting the integration of customer needs within the
product development process. Virtual Prototyping is, today, a
practice commonly used to evaluate design solutions before
building a physical prototype of the product.
Over the last few years, visualisation technologies have
been extensively improved, and the quality of the informa-
tion they can reproduce is high (Hainich and Bimber 2011;De
Fanti et al. 2011). Indeed, in recent years, most of the product
evaluation tests performed on Virtual Prototypes have been
limited to visualisation aspects (i.e. Santos et al. 2007). These
environments are appropriate for the evaluation of aesthetic
aspects of products, but not for the evaluation of other fea-
tures, especially those related to user interaction, for which
the possibility of physically touching and interacting is an
important issue. In fact, the evaluation of the physical inter-
action by means of ‘non-physical’ components is definitely
poor and does not provide useful results.
Simple haptic control devices, such as knobs and but-
tons, have been developed for testing user interaction with
the interface components of consumer products (Bordegoni
et al. 2006;Kim et al. 2008). The users are able to evaluate
different pre-defined haptic behaviours of knobs to choose
their favourite behaviours. Limitations of these approaches
are related to the fact that, if the fidelities of the representa-
tions of the knobs’ haptic behaviours are high, the tests of the
users’ interactions with the products are limited to the knobs,
and the approaches do not use stereoscopic visualisation,
tracking, and sound. Actually, rendering visual information
in stereoscopy and including an active exploration that is
allowed by a tracking system, as well as adding sound, are
very important because the results of the tests can be altered
if a part of the information is missing or incomplete.
An interesting contribution in the field of sound is Sonic
Interaction Design (Rocchesso and Serafin 2009), which is
based on the concept of Interaction Design (Buxton 2007),
and which concerns the use of tangible interfaces and inter-
active sound simulations in prototyping the sound effects
produced during object manipulation.Multimodal Virtual
Prototyping environments, in which visual, haptic and sound
modalities are fused together, have been developed for sup-
porting designers in activities of the product development
process that differ from the ‘Customer Evaluation’ one. Bor-
degoni et al. (2008) present a multimodal VirtualPrototyping
system for supporting designers in the development, the mod-
ification and the evaluation of product digital shapes through
free-hand interaction. Gupta et al. (1997) present a physically
based multimodal virtual environment to be used in Design
for Assembly analysis and prototyping designs. Also, the
multimodal Virtual Prototyping environments are developed
for commercial applications, such as in the case of Kanno
et al. (2006), who present a multimodal output device made
up of a rotary motion-platform (a pivot chair) information
appliance, in which haptic, visual and audio displays are
dynamically integrated and aligned in virtual spaces. This
device can be used in both stand-alone and networked appli-
cations, including simulators, games, and 360 browsers, thus
providing sensory-integrated multimodal applications.
Virtual Prototyping has proven to be effective because it
allows designers and engineers to determine various design
problems and errors early in the product development pro-
cess (Park et al. 2008;Bordegoni et al. 2011). In particular,
the use of these techniques allows the creation of an inter-
active environment where it is possible to carry out tests
with customers to validate the most important aesthetic and
ergonomic characteristics of the design solutions. Various
Virtual Prototyping techniques and applications have been
proposed as a means for the validation of different typolo-
gies of products, such as information appliances (Bordegoni
et al. 2010;Lee et al. 2009;Nam 2005), which helps keep the
cost of prototyping a traditional product economical. Addi-
tionally, alternative approaches, such as Rapid Prototyping
techniques, have been applied to reduce the costs and the time
associated with prototyping activities (de Beer et al. 2009)
and to support the UCD approach to the product development
process.
Furthermore, today, Virtual Prototyping is used exten-
sively for the personalisation of products, and specifically, it
is widely exploited to allow customers to personalise their
products starting from a product platform and some pre-
defined product variants (Tseng et al. 1997). In this case,
Virtual Prototyping is used to support the visual evaluation of
123
Author's personal copy
J Intell Manuf
product variants, and its use facilitates the product selection
process and the contemporary negotiation process between
sales and customers. Additionally, this approach has been
extended to web-based applications where customers can
interact (we have several examples currently on the market,
such as Nike at store.nike.com and Smart tailor made cars at
www.smartbrabustailormade.com).
Finally, the application of Virtual Prototyping techniques
in the field of capturing the VOC is still a new and a rare
phenomenon. Some experimental activities have been suc-
cessfully carried out (Berneburg 2007) to demonstrate the
usability of interactive three-dimensional environments in a
choice-based conjoint study. In this case, the set up of the
virtual environment enables the three-dimensional projec-
tion of objects, but does not support direct haptic interaction
with the objects. Indeed, the tests regarding the haptic charac-
teristics of the objects are carried out using life-sized human
models. Moreover, Jiao et al. (2007) propose a framework
in which the affective needs of customers can be elicited
and analysed through the use of Virtual Reality (VR) and
Augmented Reality (AR) techniques. Specifically, in the AR
system some expected customisable features of the prod-
uct in development are superimposed onto the real model of
the actual product by using a head-mounted display device
that the customer wears. In this case, no haptic interfaces
are integrated in the AR system, and the user can haptical-
ly interact only with some real model product components.
Also, Ren and Papalambros (2011) introduce a system for
eliciting design preference based on the collection and the
management of individual design preferences regarding the
aesthetic characteristics of products using three-dimensional
virtual models. A case study in their work develops a car
exterior using a three-dimensional parametric model, which
allows for the modification of the car’s shape according to
the customer preferences. However, the virtual models of the
car can only be experienced by the user visually using a web
browser, and the system does not include haptic interaction
with the virtual models. Finally, Luh et al. (2012) present
a framework for the customisation of footwear for children.
By using the framework, customers can take part in the prod-
uct development process by choosing among various design
attributes (colour, texture, shape, embroidery, carving style)
of some identified modular parts of the shoes (shoe surface,
shoe bottom, shoe cloth and accessory). Based on the frame-
work, the authors implement a prototype system made up of
a database of the geometric models of all variants of the shoe
modules, an AR customisation module and an AR virtual
try-on module. Specifically, by using the AR customisation
module, in which the 3D models are visually placed on a
real shoe frame, customers can design their customised shoe.
Then, the AR virtual try-on module allows customers to vir-
tually dress the selected customised shoe model on his/her
foot and visually evaluate its design.
Open issues
The different methods and approaches described above,
mainly concern the ‘Customer Evaluation’ phase of the prod-
uct development process, and present common issues, which
have limited their full implementation in daily design prac-
tice.
One issue is related to the problematic involvement of
customers in the evaluation of the interaction with the design
solutions for a new product. Many approaches perform cus-
tomer design evaluations with real product prototypes. This
takes place later in the product development process, when
only small modifications can be made, due to the cost of the
modifications. On the other hand, in the case of testing ses-
sions carried out early in the product development process in
relation to design solutions, these sessions are based mainly
on the assessment of the visual appearance of the design solu-
tion concepts with figures, pictures etc. As previously noted,
in the first case, the comparison of haptic interaction with
a product among several design solutions is possible, but it
is not reliable. In the second case, the haptic assessment is
completely unfeasible. At the present moment, the testing
sessions that involve the sense of touch are based on real
prototypes (Berneburg 2007;Jiao et al. 2007) and are mainly
focused on the ergonomic and usability features of the prod-
ucts and of their interfaces. Also, capturing the VOC applied
to the haptic needs of customers is still a rare activity. How-
ever, it also depends on customers’ typology (such as in more
or less haptically oriented consumers), and the information
that customers can derive from the haptic interaction with
products, which deeply influences their judgments about the
product (Peck and Childers 2003;Spence and Gallace 2001).
Finally, as previously introduced, multimodal Virtual Pro-
totyping environments are today used in several aspects of
the product development process, such as the development of
product shapes and the Design for Assembly analysis, while
no works use multimodal Virtual Prototyping environments
for the capturing of the VOC regarding design solutions of
products in development.
Consequently, the aim of the VP approach is to support a
holistic survey of customers’ perspectives, which allows for
the exploration of the visual, haptic and auditory character-
istics of the design solutions of products in development by
using a multimodal Virtual Prototyping environment, and to
explore in which way these characteristics could influence
the subjective ranking in a customer’s mind.
The Virtual Prototyping approach
Starting from the observations in the research presented
above, an application of Virtual Reality technologies has
been developed by the authors with the goal of improving the
123
Author's personal copy
J Intell Manuf
design process of mass customised products. This research
work has focused on the development and validation of flex-
ible and configurable Virtual Prototypes of products that
include a user interface. Examples are consumer products,
such as domestic appliances, electronic devices, video tools,
control panels, etc. The main characteristic of the Virtual
Prototyping (VP) approach is that Virtual Prototypes can
be developed in a draft version as a platform with differ-
ent configurations, and subsequently, can be easily config-
ured according to consumer needs. The interactive Virtual
Prototype of a specific product category is developed by
product designers and engineers starting from the geomet-
ric model of a product by adding aesthetic, behavioural and
functional parameters and is then validated using a multi-
modal interface based on visual, haptic and auditory modal-
ities. As previously described, the implementation of the
multimodal environment that integrates all of these inter-
active modalities is quite innovative, especially in the area
of VOC capturing, where no other studies present a similar
multimodal approach. This collaborative approach to prod-
uct design allows for the integration of the needs of prod-
uct designers, engineers and, above all, customers who can
express their needs and can collaborate in the development
process of the product. Specifically, the VP is implemented
in the first stages of the process to capture the customer needs
about both the product platform and its modular components.
Then, the customers express their needs and personalise the
product, while product designers and engineers can, if nec-
essary, resolve conflicts in real time. From the analysis of
the collected data, which are organised according to market
segments, it is possible to derive the design requirements on
which the product development process is based.
The VP approach can be considered as a design approach,
which can be used at the beginning of the design process, to
improve the creative process as a whole and reduce mistakes
and, consequently, process cost and time.
While Virtual Prototyping techniques are in fact com-
monly applied to the evaluation of design solutions with cus-
tomers (the ‘Customer Evaluation’ step in Fig. 2below), the
VP approach represents a completely innovative approach
for capturing the VOC. Moreover, the VP approach supports
the mass-customisation and the personalisation processes of
products. In fact, thanks to the possibility of configuring the
VP in real time and according to customer needs, it is pos-
sible to collect the needs of customers (or small groups of
customers) that are representative of the new product target
customers or to develop products personalised for a specific
target customer (Fig. 3).
The integration of the VP approach in a traditional prod-
uct development process can allow for an initial verification
of design solutions and furthermore enable the development
of new design solutions according to customer needs and is
in line with the UCD approach presented above. In fact, the
Fig. 2 The modification of the product development process for both
the standard and customised products based on the use of the Virtual
Prototyping (VP) approach
integration of capturing the VOC by using the VP approach
in the early stages of the traditional product development pro-
cess allows for a quick test of several product variants that
are under development. The VP approach supports the col-
lection of comparative data regarding customer needs about
one or more specific configurations, such as visual, haptic
and sound, and then compares the experimental results to
capture unspoken needs.
The VP approach can be applied mainly to those indus-
trial products that present an interactive/intelligent interface.
The VP approach can be useful not only to collect customer
needs about a product’s shape and aesthetic appearance, but
also (and in particular) about the interactive parts of the prod-
uct. These elements, for specific categories of products, such
as information devices and home appliances, can, in fact,
improve the value of the product for the customers who intend
to buy them, because they are the most important elements
through which the customers interact with the device. The
implementation of the VP is based on an immersive Virtual
Reality environment that supports customer testing and val-
idation of product interface components, such as knobs and
buttons, and their characteristics.
The environment is multimodal and is equipped with a
visual, a haptic and an auditory interface that enables the
interaction between customers and the virtual representation
of the product with the combination of all of these senses.
The customers can see the product and its elements in three-
dimensions and from different points of view; they can touch
and interact with the knobs and buttons using their hands
123
Author's personal copy
J Intell Manuf
Fig. 3 A detailed schematic of
the use of the Virtual
Prototyping (VP) approach in
the product development
process for capturing the Voice
of the Customers
(feeling the degree of grip, friction, and stiffness), and they
can hear the auditory feedbacks of their actions as well.
The most important benefits of the VP approach overcome
the actual approaches presented above and have the following
possibilities:
– developing at the beginning of the product development
process a draft prototype (VP) that represents an incom-
plete design solution (the VP is easily configurable with
different set-ups) of a product platform and its main com-
ponents (modules), which will lead to the support of the
collection of customer needs and feelings about abstract
ideas of new customisable products as well as current
product modifications to support both radical and incre-
mental innovations;
– supporting the communication of ideas and the collab-
oration between product designers, engineers and cus-
tomers in the product development process;
– continuously and easily modifying the VPs during the
entire duration of the design process to allow for record-
ing modifications, integrating and mixing some of their
elements and developing several variants of product con-
figurations; and
– supporting the analysis of haptic and sound needs of cus-
tomers about product components using an easy-to-use
and effective immersive virtual multimodal environment
in which the customer experience is particularly realistic,
with immediate possibilities for modifications.
A case study
In this section, a case study based on the VP approach is pre-
sented. The methodology used to carrying out the case study
is structured as follows:
– definition of the product typology;
– first conception of the product platform and main mod-
ules with a particular focus on the product interface, its
components and their physical characteristics;
– implementation of a VP of the product;
– testing sessions with customers about the VP quality and
its level of fidelity;
– testing sessions with customers about the use of the VP
for specific product development.
The product platform and main components
The VP approach was used to capture VOC for a Whirl-
pool washing machine (www.whirlpool.com). Whirlpool
provided a washing machine. The company received a copy
of the case study results, and provided feedback concern-
ing the VP approach and the case study results. The washing
machine was selected for several reasons. Washing machines
are mature industrial products (Kotler and Keller 2006). They
have similar design configurations: front-door, top-door, pre-
defined washing programs, knobs, buttons, touch-screens.
For this type of industrial product, the possibilities of inno-
vation are quite limited. In the last few years, we observed
incremental innovations that have concerned improvement
of technical performances, increased on-board intelligence,
measures to save energy and water consumption and eas-
ier methods for customers to control machine features. This
control is carried out using the main interface of the washing
machine, which can have different configuration consisting
of knobs and buttons, a touch-screen display, or knobs, but-
tons and a touch-screen display.
In the case where there are knobs and buttons, their char-
acteristics constitute one of the most important elements for
the customers during the purchasing moment. According to
the manufacturers, customers are used to evaluating the qual-
ity of products by means of direct interaction with some of
these elements. Specifically, washing machines are lined up
in shops without the possibility to plug them in and to test
their performances. According to Kotler and Keller (2006),
if standing in front of a variety of products of any cate-
gory produced by different manufacturers and with similar
declared performances and prices, customers need to evalu-
ate the products and associate with them a perceived quality
to motivate and give reason to their purchasing decisions.
In the case of the washing machine, the mechanical knobs,
the buttons, the cleaner drawer and the door, as compared
with characterising the aesthetic shape of the whole product,
are the only elements that the customers can interact with as
well as test, open, close, touch, rotate, push and so on. In the
case of the washing machine, this interaction allows custom-
ers to associate the characteristics of these elements with the
overall value of the product.
Consequently, the identification of the customer needs
improves the design activities that are concerned with the
components of the washing machine interface for which a
123
Author's personal copy
J Intell Manuf
Tabl e 1 The table correlates the design parameters defined for each component and the possible correspondent design solutions
Components Design parameters Design solutions
Front door Weight Heavier or lighter door (up to 10 N—due to the limits of the haptic device)
Rotational limits Longer or shorter stroke (from 110◦up to 170◦)
Closing and opening effects Different kind of clicks (sound of real washing machine, tiny-force sound, hermetic sound)
Weaker or stronger reaction force (up to 10 N)
Drawer Weight Heavier or lighter drawer (up to 10N)
Translational limits Longer or shorter stroke (up to 200mm)
Closing and opening effects Different kind of clicks (sound of real washing machine, tiny-force sound, hermetic sound)
Weaker or stronger reaction force (up to 10 N)
Knob Torque modalities Weaker or stronger reaction force (up to 1 Nm)
Sound effects Different kind of clicks (sound of real washing machine, tiny-force sound, hermetic sound)
mass customisation approach can be proposed and acquires
particular importance for the success of the product on the
market. In fact, customers are usually not aware of their needs
about the shape and behaviour of the knob, the buttons and
the door of a washing machine, but these characteristics sub-
consciously influence their evaluation. Furthermore, the cus-
tomisation of this product’s interface, by selecting different
pre-defined layouts (zero, one or more knobs, with or without
display,one or morebuttons,etc.),couldimprovethefamiliar-
ity and the usability of this product for the customers. In this
case,theimplementationofthe VP approach could supportthe
collection of the customer needs and their integration into the
product at the beginning of the product development process.
Thepossibility to inquire andthenintegratethese types ofcus-
tomer needs in the design solution for a new product and its
configurations is the distinctive element of the VP approach.
As the second step in the case study, several design solu-
tions for each component of the washing machine interface
have been sketched (Table 1). These design solutions mainly
refer to the technical haptic characteristics of the interface
components, which the authors suppose are strictly linked to
the customer needs.
Implementation of the VP
The components of the washing machine selected to be eval-
uated are the front door, the drawer for the detergents, the
knob, the buttons and the display (for feedback from the
other interactions that are controlled by rotating and pushing
the knob).
First, the VP of the washing machine (including the com-
ponents to test with the customers), which is a draft digi-
tal model of the product platform with different components
configurations, has been developed by product designers and
engineers (Fig. 4).
Then, the immersive environment (Fig. 5) has been devel-
oped using Virtual Reality technologies, as described in
details in (Ferrise et al. 2010;Bordegoni et al. 2011), and is
based on the following Virtual Reality technologies (Fig. 6):
– Haption Virtuose, a haptic device with 6 DOF to render
forces and torques (www.haption.com), which is used
to simulate the haptic interaction with the components
of the washing machine (the door, the knob, the buttons
and the drawer) through an ergonomic handling tool. Its
working space is a cube of 450 mm in size, and it is capa-
ble of rendering 10 N of continuous force and 1Nm of
continuous torque. These technical features are particu-
larly suitable for simulating the effects of the components
of the washing machine because they cover an adequate
range of force and torque effects and avoid possible lim-
itations in the users’ experience;
– a rear-projected wall display, Cyviz, for the stereoscopic
visualisation and rendering of a scale model of the appli-
ance (www.cyviz.com); the display is based on two
projectors and linear polarisers mounted first on the pro-
jectors and also on lightweight glasses worn by the users;
– an optical tracking system, made up of three AR-Track-
ing cameras for the detection of the user’s point of view
(www.ar-tracking.de), which is the position and orienta-
tion, in real time, which is necessary to allow the user to
visually explore the VP;
– a speaker positioned in the space behind the virtual wash-
ing machine. During the development of the applica-
tion, several solutions have been tested: headsets, two
speakers and only one speaker. Wireless headsets, while
useful for reducing external noises and avoiding user’s
distraction, have been substituted first by two speakers
and then by a single speaker located around the virtual
sound source. This is one of the trends in 3D sound ren-
dering as described in (Farnell 2010);
– the 3DVIA VirTools, which is a development environ-
ment (www.3dvia.com) selected among others for its
ease of integration with the 3DVIA CAD tools, was used
to implement the digital models.
The application consists of three integrated modalities:
visual, haptic and auditory (Fig. 6). First, the VP con-
sists of the stereoscopic visualisation of a highly realistic
123
Author's personal copy
J Intell Manuf
Fig. 4 Identification of those components of a real washing machine, which users can interact with at the moment of purchase or during daily
usage, and their virtual replica
Fig. 5 A concept of the virtual environment, which contains the Virtual Prototype used for capturing the VOC
representation of the washing machine and of its compo-
nents. The point of view of the user is detected by tracking
the user’s head, and the point of view changes according
to the user’s movements within the tracking space. Further-
more, the haptic interaction between the user and the virtual
model has been developed by adding force feedback to the
interactive components of the washing machine being tested.
The interactive components could be haptically tested, which
means that they could be touched and manipulated, by the
user through the use of the Haption device. The magnitude
of the feedback forces has been roughly computed from the
simplified mechanical laws derived from the CAD (Com-
puter Aided Design) model and empirically adapted to match
the real forces. Furthermore, sound has been added in the vir-
tual environment to the interactive components. Beginning
by recording the collisions among the product components
in reality (the door with the gasket, the drawer with the main
body of the washing machine and so on), these soundtracks
have been manipulated with a sound tool to reproduce dif-
ferent effects. Then, the possible sound effects are all of the
123
Author's personal copy
J Intell Manuf
Fig. 6 Multimodal virtual
environment hardware setup
sounds of the real washing machine: a tiny-force sound (sim-
ulated with the equaliser) and a hermetic sound (simulated
with the compressor and some reverbs).
To use the customer needs in defining the requirements
of the product, we needed to convert the interaction model
parameters of the VP into real product parameters. To do
that, the system has been implemented on the basis of a table
correlating the real behaviour of a component with its simu-
lated effect, which is obtained by setting and tuning a set of
parameters (Table 2). For example, several behaviours of the
knob are available. It is possible to set-up weaker or stron-
ger reaction forces when the user is rotating the knob. The
product designers and engineers, who have developed the
draft VP, can change the torque parameter of the knob model
(interaction model parameter) so it changes the force returned
(effect). When the user confirms that he likes the effect, the
engineer can derive the corresponding value of the dynamic
friction that, in the real knob, determines that specific effect
(real product parameter). In this way, the design changes can
be modified by integrating the customer needs with the engi-
neering requirements of the washing machine knob.
Evaluation of the VP quality
Some tests have been carried out in previous studies (Ferrise
et al. 2010;Bordegoni et al. 2011) to evaluate the quality of
the washing machine VP in comparison with the real appli-
ance. The evaluation was based on the following parameters:
the level of fidelity, the completeness, the flexibility in modi-
fication and the development complexity of the visual, haptic
and auditory interfaces.
The described interactive washing machine Virtual Pro-
totype was evaluated by ten users, who represented a small
group of customers (Fig. 7). Specifically, the group of users
was made up of eight males and two females with ages of
20–35years who were mechanical engineering and product
design students. They had never experienced an interaction
with an immersive multimodal Virtual Reality environment
based on force-feedback haptic devices, while most of them
(eight) had previous experience with gaming technology,
e.g., the Nintendo Wiimote, which only returns vibratory
cues to the user’s hand (www.nintendo.com). Today, these
previous experiences are common to many people but differ
greatly in the level of immersion compared to the multimodal
environment used in this work. Then, starting from the impor-
tant differences between the previous experiences and those
experienced using the multimodal environment proposed in
this work, we considered the previous experiences as not
influencing and impacting on the tests.
The results of the tests are detailed in previous studies
(Ferrise et al. 2010;Bordegoni et al. 2011). Specifically, the
testing session has been set-up in the following way. After
explaining the test purpose, the users were asked to use the
virtual system for 10min to become familiar with it. Then,
the users had to interact with the components of a real wash-
ing machine (door, drawer, knob and buttons) for 5min and,
subsequently, with the same components of the virtual wash-
ing machine for an additional 5 min. The users were invited to
express their comments verbally. These comments were col-
lected and analysed. The users had to compare the following
effects while taking into account the realism perceived:
– haptic response of the buttons;
– knob click-effect;
– knob torque;
– door weight;
– door click-effect when closing;
– drawer weight;
– drawer click-effect when closing.
123
Author's personal copy
J Intell Manuf
Tabl e 2 Real behaviour and simulated effects of the washing machine knob
Component Real product parameter Interaction model parameters Effects Customer needs
Knob Dynamic friction Torque Force returned when turning Weaker or stronger reaction force
Collisions Localized efforts Physical click More/less number of clicks
Sound clicks Sound clicks Type of sound
Fig. 7 Testing sessions with the users. Users were asked to evaluate the fidelity of the virtual replica of the four interactive elements (the buttons,
knob, drawer and door) by alternatively using the real and the virtual element
At the end of the testing session, the users were invited to fill
out a questionnaire to express their assessment (on a scale
of 1= not realistic to 6=realistic) for each aspect being ana-
lysed. Specifically, they were asked to express their feedback
concerning each component of the virtual washing machine’s
degree of realism compared to the real product (Table 3).
The data collected and analysed in previous work (Ferrise
et al. 2010;Bordegoni et al. 2011) showed that the level of
fidelity of the VP perceived by the users compared with the
real washing machine is generally evaluated as good, and the
level of fidelity varies depending on the typology of the inter-
faces being analysed. Specifically, while the level of fidelity
is considered high in the case of the visual and auditory inter-
faces, in the case of the haptic interface, the level of fidelity
depends on the component with which the users have inter-
acted and on the simulated interaction effect. For example,
while the users expressed high scores when using the haptic
interfaces of the door and the drawer, the scores assigned to
the buttons were lower. Finally, with regards to the washing
machine knob, the click effect is perceived as being similar
to the real one, while the torque effect is perceived to have
a lower level of fidelity. Moreover, concerning the collected
results about the evaluation of the haptic interface, previous
work (Ferrise et al. 2010;Bordegoni et al. 2011) highlighted
that some haptic effects could be considered more realis-
tic, while others are less realistic and still not good enough
for substituting a test performed with a real product. This is
related to the type of the haptic device used and the shape
of its end effector, which is comparable to a handle. There-
fore, to obtain more effective tests, the haptic device could
be equipped with an end effector that has the exact shape
of a real washing machine component (or a set of variants
of the component), which can be constructed using a rapid
prototyping technique and mounted on the haptic device.
123
Author's personal copy
J Intell Manuf
Tabl e 3 Assessment of the comparison of real and virtual perceived effects on the VP of the washing machine
Element Average Median Min Max SD Variance Exp. average Ztest
Buttons 3.1 3 2 5 0.88 0.69 3.5 0.926
Knob clicks 4.4 4,5 3 5 0.70 0.44 3.5 0.000
Knob torque 3.6 3,5 2 5 0.97 0.84 3.5 0.372
Door weight 4.0 4 3 5 0.67 0.40 3.5 0.009
Door clicks 4.6 5 4 5 0.52 0.24 3.5 0.000
Drawer weight 4.3 4 3 5 0.67 0.41 3.5 0.000
Drawer clicks 4.6 5 3 6 0.84 0.64 3.5 0.000
The SD show that the differences in the averages are not statistically significant in the case of the buttons and the knob
Use of the VP for product personalisation
In this study, we performed new tests with the goal of dem-
onstrating the applicability of the VP approach in the devel-
opment of industrial products and introducing the use of the
VP in the first phases of the product development process.
These new tests, specifically performed in the current study,
were carried out using the same layout of the application (the
immersive environment) and the VP of the washing machine
previously developed (Ferrise et al. 2010;Bordegoni et al.
2011) but with a different purpose. Specifically, these tests
aim to evaluate the following for the first time:
– the possibility of using a VP of the product for experi-
menting, evaluating and determining component behav-
iours directly by the users;
– the subsequent usefulness of using a VP in the collection
of users needs; and
– the effectiveness of the real-time integration of the col-
lected data with the engineering requirements within the
VP.
The testing session, carried out by the same users who par-
ticipated in the first test session, was set-up in the following
manner. Each user wore a pair of stereo glasses, stood in front
of the wall display and handled the haptic device (as shown
in Fig. 7). He or she looked at the washing machine from
different points of view with the aid of the tracking system
connected to the stereo glasses.
The users were asked to perform the following tasks. First,
use the haptic device to interact with the interactive com-
ponents of the washing machine: turn the knob, push the
buttons, open and close the drawer. Visual, haptic and sound
feedback is perceived by the users in response to their actions.
For example, when the users turn and click the knob, the visu-
alisation of the component changes accordingly, and a virtual
hand, which is a virtual representation of a human hand inte-
grated in the environment, changes its shape (in this case,
it takes a two-finger grasping shape) to provide visual feed-
back of the action. Moreover, the users perceive a force effect
returned to the hand. When the users turn the knob, they per-
ceive a continuous torque, which replicates the real dynamic
friction on the hinge, while when the users click the knob,
they perceive some forces that are in an opposite direction
with respect to the motion that replicates the clicks. If the
users perform both actions at once, the force feedback is a
sum of the forces presented above. At the same time, the
users hear some clicks that change according to the action
performed.
The users were also asked to express when they did not like
the haptic or the auditory responses provided by the compo-
nents and how they would like to change them. The users can
express their needs with simple and natural sentences, such
as, “I would like the door to be less resistant when opening.”
According to the users’ requests, some parameters of the VP
interaction model were changed to modify the effect accord-
ingly. For example, regarding the knob (Table 2), the users
asked to change the effects with simple expressions, such as,
”I would like a stronger reaction when turning the knob,” or,
“I would like fewer clicks for the knob when selecting the
washing programs and a different click sound.”
The sequences of users’ requests and the real-time modi-
fications of the VP represented the input of a product devel-
opment process in which product designers, engineers and
users collaborate. Specifically, each user, through his or her
requests, produced a customised washing machine by mod-
ifying the washing machine’s standard platform by choos-
ing different design solutions for each design parameter
(Table 1). The sequences of two of these experimental activ-
ities in which some design changes were made starting
from the same washing machine VP and according to users’
requests are shown in Figs. 8and 9below.
Moreover, Table 4shows the results of the testing
sessions quantitatively summarised by percentages. Specif-
ically, starting from Table 1, for each component of the
washing machine and the related design parameters, the per-
centage of selection of each design solution is presented.
Moreover, after the tests, the users were asked to express
their remarks concerning the use of the VP of the wash-
ing machine for evaluating and comparing different interface
123
Author's personal copy
J Intell Manuf
Fig. 8 Graphic representation
of the sequence of a user’s
requests for testing and
choosing among the possible
design solutions
Fig. 9 Graphic representation
of the sequence of a user’s
requests for testing and
choosing among the possible
design solutions
component design solutions. The data and feedback collected
and analysed, in reference to the testing session goals, accom-
plish the following:
– the use of the product VP (the washing machine) for
experimenting, evaluating and determining component
behaviours directly by the users is effective and favour-
able. Indeed, the users generally appreciated the quality
of the visualisation (70% of the users strongly appreci-
ated and 20% appreciated), the possibility to haptical-
ly interact with the VP and the level of immersiveness,
which is improved by the auditory feedback. Addition-
ally, all of the users considered the interaction with the
VP, achieved through the immersive environment, effec-
tive and quite similar to an experience with a real mock-
up of a washing machine;
– users appreciated the VP approach for collecting cus-
tomer needs. Specifically, 80% of the users reported a
strong appreciation of the experience and considered the
experience engaging and with a high degree of involve-
ment. Additionally, from an emotional point of view,
which is usually not experienced in the interaction with
a real washing machine, there was more attention paid to
the product and to the different design solutions available
to evaluate;
– the real-time integration of the collected data about the
customer needs of the engineering requirements within
the VP is feasible and appreciated by the users. In fact,
70% of the users considered the possibility of express-
ing their preferences about the behaviours of the product
components and the possibility to test the components
and the derived correlations in real time very valuable
features of the VP approach, which are not possible when
123
Author's personal copy
J Intell Manuf
using real washing machines or actual physical mock-ups
of a product.
Discussion of the VP approach
At the end of the experimental testing sessions, the data col-
lected regarding the customer needs can be used in two ways.
The first use is that the customer needs can be used to ‘medi-
ate’ design solutions. Design solutions can be developed
from a product platform with product components config-
ured in different ways, to satisfy specific market segments.
Conversely, personalised design solutions can be developed
from a basic product for each customer.
The VP can be used as a configuration tool that gives
us the possibility to set some product features according
to customer needs and allows designers to optimise prod-
ucts according to customer demands (Hvam and Mortensen
2007). To guarantee the manufacturability of a customised
product at mass production costs, the VP can propose similar
variants of the same mass-produced product. Consequently,
the VP approach can be implemented in the design processes
oriented to develop mass-customised products and to collect
data regarding the needs of customers (meant as small groups
of users, that are representative of the target customers) as
input for the definition of design parameters.
The benefits of the VP approach, when compared with
the other methods currently used to capture the VOC, are
the reduction of costs and times associated with prototyping
activities. In fact, while it is currently necessary to manufac-
ture several real prototypes with different product character-
istics and layouts, by using the VP, only one virtual prototype
that is able to provide different configurations and behav-
iours has to be developed. Moreover, a virtual prototype can
be used for many different configurations and modifications.
This means that the same virtual prototype of a platform can
be used for several VOC capturing sessions. Additionally, the
VP foresees a strict correlation between the real behaviour
of the components and their simulated effects. Consequently,
after the testing sessions, the customer needs can be directly
integrated in the design solutions as engineering parameters.
Second, the VP can also be developed in a draft version,
such as in a basic layout without several details, in the case
where the product idea is still in the conceptual stage, and
capturing the VOC is directed at identify the customer needs
about a specific element of the product. This type of testing
session could be particularly useful with brand new products
and technologies. In these cases, in fact, the possible area of
modifications is very wide, and the manufacturing of many
real prototypes can be a very expensive activity for compa-
nies.
Also, the benefits of the VP approach concern an improved
integration and a strengthened implementation of multi-
modal Virtual Prototyping environments within the product
development process. Indeed, at the present moment the use
of multimodal environments is mainly related to the ‘Cus-
tomer Evaluation’ phase, or to other activities concerning the
analysis of the product characteristics when these have been
already developed. On the contrary, the VP approach intro-
duces the use of multimodal Virtual Prototyping environment
in the early phases of the product development process and,
also if it does not present innovations from the technical point
of view, especially for what concerns the layout of the multi-
modal environment, it constitutes an innovative application
in the area of the capturing of the VOC.
Moreover, the VP approach supports capturing of cus-
tomer needs about elements and behaviours of product com-
ponents that are usually not investigated, such as the haptic
and auditory components. Specifically, the immersive mul-
timodal environment in which the VP is tested allows for the
verification of the visual, haptic and auditory characteristics
of the product components and supports specific verification
of each component one by one. The integration of these two
approaches allows the identification of the main character-
istics of these aspects and their influence on the customer
needs.
According to Whirlpool, the results of the case study are
very interesting and useful for capturing the VOC concern-
ing the characteristics of design solutions of products in
development, which result to be very important in influenc-
ing customer evaluation and ranking but, up to now, are not
extensively investigated. Consequently, Whirlpool is consid-
ering modifying their product development process to inte-
grate the VP approach, as show in Fig. 3.
Conclusion
The experimental results of the testing sessions used to deter-
mine the customer needs about the washing machine and the
components of its user interface demonstrate the possibility
of introducing the VP approach in everyday design practice
when considering mass-customised products. Specifically,
some remarks derived from the analysis of the collected data
and from the comparison with the theoretical basis of the
present research will be described hereafter.
Concerning the comparison with the traditional product
development process, it is possible to affirm that the level
of development and definition of the design solutions to be
tested using the VP approach is similar to that obtainable in
the concept phase of a traditional product development pro-
cess. In fact, in the described case study, the platform and the
main components of the washing machine have only been
defined in a rudimentary manner, and their behaviours have
been tested by customers and configured according to the
customer needs.
123
Author's personal copy
J Intell Manuf
Tabl e 4 The table statistically
summarises the results of the
testing sessions and correlates
each component of the washing
machine design parameters to a
design solution chosen by the
users (in percentage)
Components Design parameters Design solutions chosen by the users
Front door Weight Heavier door: 80% of the users
Lighter door: 20% of the users
Rotational limits Longer stroke: 40% of the users
Shorter stroke: 60% of the users
Closing and opening effects Door opening:
Sound of real washing machine: 30% of the users
Tiny-force sound: 20% of the users
Hermetic sound: 50% of the users
Weaker reaction force: 30 % of the users
Stronger reaction force: 70% of the users
Door closing:
Sound of real washing machine: 30% of the users
Tiny-force sound: 30% of the users
Hermetic sound: 40% of the users
Weaker reaction force: 10 % of the users
Stronger reaction force: 90% of the users
Drawer Weight Heavier drawer: 90% of the users
Lighter drawer: 10% of the users
Translational limits Longer stroke: 30% of the users
Shorter stroke: 70% of the users
Closing and opening effects Drawer opening:
Sound of real washing machine: 30% of the users
Tiny-force sound: 20% of the users
Hermetic sound: 50% of the users
Weaker reaction force: 20 % of the users
Stronger reaction force: 80% of the users
Drawer closing:
Sound of real washing machine: 40% of the users
Tiny-force sound: 20% of the users
Hermetic sound: 40% of the users
Weaker reaction force: 20 % of the users
Stronger reaction force: 80% of the users
Knob Torque modalities Weaker reaction force: 40 % of the users
Stronger reaction force: 60% of the users
Sound clicks Sound of real washing machine: 30% of the users
Tiny-force sound: 30% of the users
Hermetic sound: 40% of the users
Number of clicks Higher number of clicks: 40% of the users
Same number of clicks: 60% of the users
Furthermore, the development of a draft and configu-
rable VP that integrates several new product design solu-
tions can lead to reduced time and costs compared to design
approaches that use traditional VOC surveys and physical
design prototypes. This is particularly important in the case of
the product development process of mass-customised prod-
ucts based on the UCD approach. In these cases, currently,
several real draft mock-ups have to be produced to support
tests by users for different product configurations derived by
the integration of the product platform and its modules (com-
ponents), which is a time-consuming and expensive activity.
Among the benefits of the implementation a VP approach,
it is possible to enumerate the improvement of the commu-
nication of ideas between product designers and customers.
123
Author's personal copy
J Intell Manuf
In fact, during the testing sessions, the users, who may not be
technicians or expert users of a specific industrial product,
have described the components of the washing machine and
their actual and desired behaviours using the VP represen-
tation presented. Finally, mapping the correlation between
the VP parameters, which are the related effects that the cus-
tomers can test and the real product engineering parameters,
supports both the communication of ideas and the integration
of customer needs into the engineering parameters.
Concerning the possibility of collecting customer needs
about the customised products under development (in our
case study of the washing machine, we were concerned with
its user interface and front door) and the integration of them
with the previously defined engineering requirements, the
use of the VP supports all of these activities at the same time.
By interacting with the virtual environment, the customers
can evaluate the visual, haptic and auditory behaviours of
each component, can seek their modification (of one or more
parameters) and can express their needs. Specifically, in the
case study presented above, the haptic behaviours of the com-
ponents of the washing machine interface have been defined
a priori as design solutions for specific design parameters.
Then, using the immersive environment, product designers
and engineers manage the VP of the product by modifying
the characteristics of its components according to customer
needs, integrating them with the previously defined engineer-
ing requirements and developing the customised products in
collaboration with the customers.
Moreover, the collection of the customer needs by means
of the VP approach can lead to identifying unexpected design
requirements related to the haptic and auditory characteris-
tics more so than the visual requirements. For example, in
the case study presented above, the users had the opportu-
nity to test different design solutions and behaviours of the
components and were able to choose a preferred solution
from a visual, haptic and auditory point of view. The users
can ask for a weaker or stronger force effect behaviour for
the washing machine door or try different combinations of
continuous and localised forces until a desirable effect is
obtained. In this environment, the users’ needs focus on the
auditory feedbacks of some components and not on their
visual appearances. This is particularly important for com-
ponents such as the knob, the drawer and the door in which
the users’ past experiences with similar products influence
their perceptions and needs.
Form a technical point of view, the VP approach refers to
authors’ prior studies (Bordegoni et al. 2008;Ferrise et al.
2010;Bordegoni et al. 2011), which concern the develop-
ment of the here-presented multimodal Virtual Prototyping
environment and its use in the areas of development and mod-
ification of product shapes and of ‘Customer Evaluation’ of
design solutions. More, the present work builds on the tests
carried out in (Ferrise et al. 2010;Bordegoni et al. 2011)
concerning the level of fidelity, the completeness, and the
flexibility in making modifications to the multimodal envi-
ronment. However, some technical limits, mainly related to
the intrusiveness of the haptic device that affects the quality
and the naturalness of the interaction, as described previ-
ously (Bordegoni et al. 2011), must still be overcome. This
brings about the opportunity to continue research activities
by introducing Rapid Prototyping elements that reproduce
the exact shape of each component that must be manipulated
(the drawer, the door, the knob) and mounting it on the haptic
system as an end effector.
Moreover, the experimental activities could investigate the
implementation of the VP approach from the beginning of
the product development process. This is important not only
for activities concerning the capturing of the VOC in cus-
tomer-centred product development processes for mass cus-
tomisation but in supporting processes of personalisation of
specific products as well. In fact, the personalisation of a
product could be acquired directly by the use of the VP from
a customer, and the stored parameters could be used to man-
ufacture the product for that specific customer.
References
Akao, Y. (1997). QFD: Past, present, and future. In International
symposium on QFD ’97.
Berneburg, A. (2007). Interactive 3D simulations in measuring con-
sumer preferences: Friend or foe to test results? Journal of
Interactive Advertising.
Bordegoni, M., Colombo, G., & Formentini, L. (2006). Haptic tech-
nologies for the conceptual and validation phases of product
design. Computers & Graphics, 30(3), 377–390.
Bordegoni, M., Ferrise, F., Shelley, S., Alonso, M. A., & Hermes D.
(2008). Sound and tangible interface for shape evaluation and
modification. In HAVE 2008—IEEE international workshop on
haptic audio visual environments and their applications, Ottawa,
Canada, 18–19 October 2008.
Bordegoni, M., Polistina, S., & Carulli, M. (2010). Mixed reality
prototyping for handheld products testing, research in interac-
tive design. In Proceedings of IDMME—virtual concept 2010,
Bordeaux, France, 20–22 October, 2010.
Bordegoni, M., Ferrise, F., & Lizaranzu, J. (2011). Use of interactive
virtual prototypes to define product design specifications: A pilot
study on consumer products.InProceedings of IEEE—ISVRI,
Singapore, 2011.
Buxton, B. (2007). Sketching user experiences: Getting the design right
and the right design. San Francisco, CA: Morgan Kaufmann Pub-
lishers Inc.
Clausing, D. P. (1994). Total quality development: A step by step guide
to world-class concurrent engineering. New York: ASME Press.
Cohen, L. (1995). Quality function deployment: How to make QFD
work for you. Reading, MA: Addison-Wesley.
Da Silveira, G., Borensteinb, D., & Fogliatto, F. S. (2001). Mass
customization: Literature review and research directions. Inter-
national Journal of Production Economics, 1(72), 1–13.
de Beer, D. J., Campbell, R. I., Truscott, M., Barnard, L. J., &
Booysen, G. J. (2009). Client-centred design evolution via func-
tional prototyping. International Journal of Product Development,
(8), 1.
123
Author's personal copy
J Intell Manuf
De Fanti, T. A., et al. (2011). The future of the CAVE. Central
European Journal of Engineering, 1(1), 16–37.
Farnell, A. (2010). Designing sound. Cambridge: MIT Press.
Ferrise, F., Bordegoni, M., & Lizaranzu, J. (2010). Product design
review application based on a vision-sound-haptic interface. In
Proceedings of HAID 2010. Lecture Notes in Computer Science,
Vol. 6306 (pp. 169–178).
Gilmore, J. & Pine, J. (1997). The four faces of mass customization.
Harvard Business Review.
Gupta, R., Whitney, D., & Zeltzer, D. (1997). Prototyping and
design for assembly analysis using multimodal virtual environ-
ments. Computer-Aided Design, 29(8), 585–597.
Hainich, R. R., & Bimber, O. (2011). Displays: fundamentals and
applications. Boca Raton: A K Peters/CRC Press.
Hart, C. (1995). Mass customization: Conceptual underpinnings,
opportunities and limits. International Journal of Service Indus-
try Management, 6(2).
Hvam, N. H., & Mortensen, J. R. (2007). Product customization.
Berlin: Springer.
Jiao, J., Xu, Q., Du, J., Zhang, Y., Helander, M., Khalid, H. M., Helo,
P., & Ni, C. (2007). Analytical affective design with ambient intel-
ligence for mass customization and personalization. International
Journal of Flexible Manufacturing Systems, 19, 570–595.
Jones, J. C. (1992). Design methods. New York: Van Nostrand Rein-
hold.
Kanno, K., Fernando, N., Bolhassan, A., Narita, S., & Cohen, M.
(2006). “Personal Practically Panoramic” multimodal interfaces.
In IEEE VR 2006 workshop on emerging display technologies.
Kawakita, J. (1991). The original KJ method. Tokyo: Kawakita
Research Institute.
Kim, L., Han, M., Shin, S. K., & Park, S. H. (2008). A haptic
dial system for multimodal prototyping. In 18th international
conference on artificial reality and telexistence (ICAT 2008).
Kotler, P. (1989). From mass marketing to mass customization. Strategy
& Leadership, 17.
Kotler, P., & Keller, K. L. (2006). Marketing management. Englewood
Cliffs, NJ: Prentice Hall.
Kramer, J., Noronha, S., & Vergo, J. (2000). A user-centered design
approach to personalization. Communications of the ACM, (43),
8.
Kroll, E., Condoor, S., Sridhar, S., & Jansson, D. G. (2001). Innova-
tive conceptual design theory and application of parameter anal-
ysis. Cambridge: Cambridge University Press.
Lee, J. Y., Rhee, G. W., & Park, H. (2009). AR/RP-based tan-
gible interactions for collaborative design evaluation of digital
products. International Journal of Advanced Manufacturing Tech-
nology, 45, 649–665.
Loftus, E., & Wells, G. (1984). Eyewitness testimony: Psychological
perspectives. New York: Cambridge Univ. Press.
Luh, Y. P., Wang, J. B., Chang, J. W., Chang, S. Y., & Chu, C. H.
(2012). Augmented reality-based design customization of foot-
wear for children, Journal of Intelligent Manufacturing. doi:10.
1007/s10845-012-0642-9.
Nam, T. J. (2005). Sketch-based rapid prototyping platform for hard-
ware-software integrated interactive products. In CHI 2005—Pro-
ceedings of conference on human factors in computing systems,
Portland, USA.
Norman, D. A., & Draper, S. W. (1986). User-centered system design:
New perspectives on human-computer interaction. London:
Lawrence Erlbaum Associates.
Pahl, G., Beitz, W., Feldhusen, J., & Grote, K. H. (2007). Engineering
design. Berlin: Springer.
Park, H., Son, J., & Lee, K. (2008). Design evaluation of digital con-
sumer products using virtual reality-based functional behaviour
simulation. Journal of Engineering Design, 19(4).
Peck, J., & Childers, T. L. (2003). To have and to hold: The influ-
ence of haptic information on product judgments. Journal of
Marketing, 67, 35–48.
Ren, Y., & Papalambros, P. (2011). Design preference elicitation:
Exploration and learning. In International conference on engi-
neering design, ICED11.
Risdiyono, R., & Koomsap, P. (2011). Design by customer: Concept
and applications, Journal of Intelligent Manufacturing. doi:10.
1007/s10845-011-0587-4.
Rocchesso, D., & Serafin, S. (2009). Sonic interaction design. Inter-
national Journal of Human-Computer Studies, 67(11), 905–906.
Santos, P., et al. (2007). Improve: An innovative application for
collaborative mobile mixed reality design review. International
Journal on Interactive Design and Manufacturing, 1(2), 115–126.
Shaw, M. C. (2001). Engineering problem solving: A classical per-
spective. Norwich: Noyes Pub./William Andrew Pub.
Spence, C., & Gallace, A. (2001). Multisensory design: Reaching out to
touch the consumer. Psychology and Marketing, 28(3), 267–308.
Spool, J. M. (2004). The KJ-technique: A group process for
establishing priorities. Resource document. User Inter-
face Engineering. http://medicina.iztacala.unam.mx/medicina/
Spool%20KJ%20Technique.pdf. Accessed 2 May 2012.
Tseng, M. M., Jiao, J., & Su, C. J. (1997). A framework of virtual
design for product customization. In Emerging technologies and
factory automation proceedings, ETFA ’97.
Ulman, D. G. (2003). The mechanical design process.NewYork:
McGraw-Hill.
Ulwick, A. (2005). What customers want: Using outcome-driven
innovation to create breakthrough products and services.New
York: McGraw-Hill.
Wicker, A. W. (1969). Attitudes versus actions: The relationship of
verbal and overt behavioral responses to attitude objects.Journal
of Social Issues, 25(4), 41–78.
123
Author's personal copy
