Engineering Emotional Values

in Product Design

-Kansei Engineering in Development

Simon Schütte

Linköping, 2005

Linköping Studies in Science and Technology,

Dissertation 951
Quality Technology and Management
Quality and Human-Systems Engineering
Department of Mechanical Engineering
SE-58183 Linköping
Printed by: UniTryck, Linköping
Linköpings Universitet
Department of Mechanical Engineering

ISBN 91-85299-46-4

ISSN 0345-7524

Cover Design: Rilda Schütte

Figures: Rilda Schütte

© 2005, Simon Schütte

Distributed by:

Linköpings Universitet
Division of Quality Technology and Management
Department of Mechanical Engineering
SE-58183 Linköping, Sweden
Tel: +46-13-28 2448

Kansei Engineering Group:

http://www.ikp.liu.se/kansei
Abstract
Feelings and impressions of a product are important for the decision of purchasing it or
not. Designing attractive products therefore requires knowledge about the feelings and
impressions the products evoke on the customer and the user. Integrating such
affective values in product design requires the introduction of suitable methods into
companies’ product design processes, methods which can capture and convert
subjective and even unconscious feelings about a product into concrete design
parameters. This is sometimes referred to as ‘Affective Engineering’. One
methodology in this context is Kansei Engineering, which has been developed in Japan
in order to design feelings into products. The aim of this thesis was twofold: Firstly, to
improve understanding of the nature of products making emotional impact on the users
and customers. Secondly, to identify and improve methods capable of grasping those
affective values and translating them into concrete product design solutions. This
thesis presents three empirical studies and two methodological papers, relating to
warehouse trucks and laminate flooring. The first study was made on user impact of
warehouse trucks in three different European countries. A second study dealt with
affective values of rocker-switches in work vehicles, such as warehouse trucks. A third
study on this truck type compared the old manoeuvring panel evaluated in the previous
studies with a newly introduced manoeuvring panel in order to validate the impact of
the design improvements made after the first study. Further, a conceptual model on
Kansei Engineering methodology was proposed in a methods paper based on the
experience from the studies performed in order to provide a structure for performing
Kansei Engineering studies. The fifth paper had the purpose of validating and
improving the proposed model using laminate flooring as research object. More
structured ways of identifying design parameters and relevant product properties was
given high priority in the improvement work of the methodology. A model for
spanning the Space of Product Properties was presented and applied. This thesis also
deals with other improvement areas in the methodology and proposed new
developments, including the use of scales, experimental design and validation
methods. In conclusion, Kansei Engineering is a concept and a methodology in strong
development, a framework in which tools and methods are continuously developed,
added and integrated.

i
Acknowledgements
Doing research is always an adventure which can not be mastered alone. On my
journey towards this thesis, I met many people who travelled with me a while and
supported me in both professional and private areas. I would like to express my
deepest gratitude to all of you.
In particular I am grateful to my supervisor Professor Jörgen Eklund and by co-
supervisors Professor Jens Jörn Dahlgaard and Professor Mitsuo Nagamachi,
Hiroshima International University, Japan for supervision, encouragement, enthusiasm
and most important interest in my work.
Moreover I would like to thank Professor Gunnela Westlander, Professor Mats
Lörstad and Dr. Jan Axelsson, Division of Quality and Human Systems Engineering at
Linköping University for their help in projects I carried out and constructive comments
on my papers and thesis.
Warm thanks to Associate Professor Shigekazu Ishihara and Associate Professor
Tatsuo Nishino for welcoming me at Hiroshima International University in Japan and
explaining the secrets of Kansei Engineering. Also many thanks to Toshio Tsuchiya,
from Shimonoseki City University for visiting us and sharing ideas with us.
I also wish to thank Ebru Alikalfa and other masters students who helped to bring
my research forward with their findings.
Many thanks also to my colleagues, former and current, at the Division of Quality
and Human Systems Engineering for providing me a fantastic work atmosphere and
inspiring discussion whether work related or not.
The financial support from BT Industries AB was essential for this project, and
hereby gratefully acknowledged. I wish to thank Sven Wirenhammar, Magnus
Senneryd, Daniel Nåbo, Mats Wingmo and Per Axner for a fruitful co-operation.
In the autumn 2002 I had the opportunity to visit Hiroshima International University
in Japan for a three month stay. In addition to Professor Nagamachi who kindly invited
me, I also want to thank the Sweden-Japan-Foundation for providing me the financial
support.
I also want to thank my family in Germany, my parents, for constantly encouraging
and supporting me, and my brothers for the happiness and joy they have given me.
And last but not least, thanks to the nicest, smartest, and most beautiful women in the
world. Eres mi sol, mariposa!

Linköping June, 2005

Simon Schütte

iii
 List of Contents

1. Introduction ........................................................................................................... 1
1.1. CUSTOMER NEEDS IN FUTURE PRODUCTS........................................................ 1
1.2. LABELLING THE FIELD OF KNOWLEDGE........................................................... 3
1.3. METHODS IN THE AREA OF AFFECTIVE PRODUCT DESIGN............................... 4
1.4. AFFECTIVE QUALITY WORK............................................................................. 5
1.4.1. A Short History on Quality ....................................................................... 5
1.4.2. Evolution of the Concept ‘Quality’........................................................... 6
1.4.3. The Kano Model as a Tool for Attractive Quality Creation..................... 8
1.5. INDUSTRIAL PARTNERSHIP ............................................................................. 10
1.5.1. BT Industries AB..................................................................................... 10
1.5.2. Product Range ........................................................................................ 11
1.5.3. Meeting Customer Demands at BT......................................................... 11
1.5.4. BT REFLEX ............................................................................................ 12
1.5.5. Market Segmentation .............................................................................. 12
1.5.6. Productivity through Ergonomics .......................................................... 13
2. Definitions ............................................................................................................ 15
2.1. AFFECT ........................................................................................................... 15
2.2. AFFECT VS. EMOTION, FEELING AND MOOD .................................................. 16
2.3. DESIGN VS. ENGINEERING .............................................................................. 18
2.4. PRODUCTS ...................................................................................................... 18
2.5. EXPRESSIONS IN SUBJECTIVE ASSESSMENT ................................................... 19
2.6. DATA, INFORMATION, MEANING, KNOWLEDGE ............................................ 20
2.7. SEMANTICS ..................................................................................................... 21
2.8. USER, CUSTOMER, PURCHASER ..................................................................... 22
3. Aims and Delimitations....................................................................................... 23
3.1. AIMS OF THE RESEARCH ................................................................................. 23
3.2. DELIMITATIONS .............................................................................................. 24
4. Research Methodology........................................................................................ 25
4.1. THE AREA OF RESEARCH................................................................................ 25
4.2. RESEARCH ACTIVITIES ................................................................................... 25
4.2.1. Author’s Studies...................................................................................... 26
4.2.2. Student Projects ...................................................................................... 26
4.2.3. Visit in Japan .......................................................................................... 27
4.3. RESEARCH STRUCTURE .................................................................................. 27
4.3.1. Phase I, Licentiate Thesis ....................................................................... 28
4.3.2. Phase II, Doctorate Thesis ..................................................................... 29
v
 4.4. SCIENTIFIC APPROACH .................................................................................... 30
4.4.1. Streams in Scientific Philosophy ............................................................ 30
4.4.2. Quantitative vs. Qualitative Measurements............................................ 31
4.4.3. Building a Bridge between Customers and Feelings ............................. 32
5. What is Kansei? ................................................................................................... 35
5.1. THE NATURE OF KANSEI ................................................................................ 35
5.2. THE ETYMOLOGY OF KANSEI ......................................................................... 37
5.3. DICTIONARY DEFINITION ............................................................................... 37
5.4. PHILOSOPHICAL DEFINITION .......................................................................... 38
5.5. PSYCHOLOGICAL DEFINITION ......................................................................... 39
5.6. DEFINITIONS BY R&D KANSEI RESEARCHERS............................................... 39
5.7. KANSEI VS. CHISEI ......................................................................................... 42
5.8. A HIERARCHY OF THE KANSEI ....................................................................... 43
5.9. MEASURING THE KANSEI................................................................................ 45
5.10. KANSEI AND PRODUCT DEVELOPMENT ...................................................... 47
6. Kansei Engineering ............................................................................................. 49
6.1. HISTORY OF KANSEI ENGINEERING ................................................................ 49
6.2. PERSPECTIVES ON KANSEI ENGINEERING....................................................... 50
6.3. THE PRINCIPLE OF KANSEI ENGINEERING ...................................................... 52
6.4. TYPES OF KANSEI ENGINEERING .................................................................... 53
7. A Proposed Model ............................................................................................... 55
7.1. A PROPOSED STRUCTURE OF KANSEI ENGINEERING ..................................... 55
7.2. CHOOSING THE DOMAIN ................................................................................. 56
7.3. SPANNING THE SEMANTIC SPACE ................................................................... 57
7.3.1. The Procedure of Spanning the Semantic Space .................................... 57
7.3.2. Collection of Kansei Words.................................................................... 58
7.3.3. Tools for Semantic Structure Identification............................................ 58
7.4. SPANNING THE SPACE OF PROPERTIES ........................................................... 59
7.4.1. The Space of Properties as a Counterpart of the Semantic Space ......... 59
7.4.2. Proposing a Model for Spanning the Space of Properties ..................... 60
7.5. SYNTHESIS ...................................................................................................... 62
7.5.1. Relationship Identification ..................................................................... 63
7.6. MODEL BUILDING AND TEST OF VALIDITY .................................................... 64
8. Research Contributions to Improvement of Kansei Engineering .................. 65
8.1. SENSING THE KANSEI ..................................................................................... 65
8.1.1. Proximity of Interaction and Presentation ............................................. 66
8.1.2. Showroom Appeal................................................................................... 68
8.1.3. Indirect Affective Product Design .......................................................... 70
8.1.4. Honest Products...................................................................................... 71
8.1.5. Trend Sensitivity ..................................................................................... 71
vi
 8.2. COLLECTING DATA FOR KANSEI ENGINEERING EVALUATIONS ..................... 72
8.2.1. Scale Types ............................................................................................. 72
8.2.2. Computerised vs. Manual Data Collection ............................................ 75
8.2.3. Constructing Kansei Engineering Questionnaires ................................. 76
8.3. DEGREE OF IMPORTANCE ............................................................................... 78
8.4. DESIGNING ATTRACTIVE QUALITY INTO PRODUCTS ..................................... 81
8.4.1. Connecting the Kano Model to Kansei Engineering.............................. 81
8.4.2. Suggesting a new Mathematical Model.................................................. 82
9. Discussion ............................................................................................................. 85
9.1. KANSEI ENGINEERING–NOT ONLY A METHOD FOR INDUSTRIAL DESIGN ...... 85
9.2. SHORT- CUTS TO KANSEI ENGINEERING? ...................................................... 86
9.3. QUALITATIVE VS. QUANTITATIVE APPROACHES ........................................... 87
9.4. REDUCTIONISM VS. HOLISM ........................................................................... 87
9.5. THE WORDS ARE NOT ENOUGH....................................................................... 88
9.6. ASIAN ADVANTAGES? .................................................................................... 89
9.7. IMPLEMENTING KANSEI ENGINEERING .......................................................... 90
9.7.1. Integrating Kansei Engineering in Product Development Processes .... 90
9.7.2. Integrating Kansei Engineering Data in QFD ....................................... 92
9.8. TANGIBLE AND INTANGIBLE PRODUCTS ........................................................ 94
9.9. INNOVATIVENESS IN KANSEI ENGINEERING? ................................................. 94
9.10. APPLYING KANSEI ENGINEERING IN EUROPE ............................................. 95
10. Conclusions....................................................................................................... 97
11. Future Research............................................................................................... 99

vii
viii
 Appended Papers

Paper A: Comparing the User Impact of Warehouse Truck Design in Europe

SIMON SCHÜTTE AND JÖRGEN EKLUND

Published in ‘Arbete Människa Miljö’ 1/2003, pp. 38-50, and presented in an

earlier version at ‘Conference on Human Affective Design’, Singapore, 2001.

Paper B: Design of Rocker Switches for Work Vehicles

SIMON SCHÜTTE AND JÖRGEN EKLUND

Accepted for publication in ‘Applied Ergonomics’, 2005 and presented in an

earlier version at ‘NES-Conference’, Reykjavik, Island, 2003.

Paper C: Affective Values of Lift Trucks - an Application of Kansei Engineering

SIMON SCHÜTTE, RILDA SCHÜTTE AND JÖRGEN EKLUND

Accepted for publication in the ‘Proceedings of QMOD Conference 2005’,

Palermo, Italy.

Paper D: Concepts, Methods and Tools in Kansei Engineering

SIMON SCHÜTTE, JÖRGEN EKLUND, JAN AXELSSON AND MITSUO
NAGAMACHI

Published in ‘Theoretical Issues in Ergonomics Science’, (2004), Vol.5 (3), p

214-232, 2004

Paper E: Developing the Space of Product Properties Supporting Kansei

Engineering Procedure
SIMON SCHÜTTE

Submitted for international publication

ix
x
 List of Abbreviations

AI Artificial Intelligence
AVI Analogue Visual Scale (compare VAS-scale)
BT BT Industries AB. One of the worlds leading manufacturer of warehouse
trucks.
CA Conjoint Analysis (see Appendix C)
CS Category Score (compare QT1)
EEG Electro-Encephalo-Gram (Measurement of brainwaves)
EMG Electro-Myo-Graphy
HMU Human Machine Unit. Mazda concept (see Appendix E).
KE Kansei Engineering
KES Kansei Engineering System (see Appendix E).
KEW Kansei Engineering Word (see Section 7.3.3)
KW Kansei Word (see Section 7.3.1)
MCC Multi Correlation Coefficient. Similar to R value in linear regression
(compare QT1)
PCC Partial Correlation Coefficient (compare QT1)
PD Product Development
PSI Physical Structure Identification (see Section 7.4 and Appendix B)
PI Proximity of Interaction (see Appendix B)
RI Relationship Identification (see Section 7.5 and Appendix B)
SD (Scale) Semantic Differential (Scale). Invented by Osgood (see Appendix C).
SMB Semantik Miljö Beskrivning, Semanic decriptions of environments.
Method within Affective Engineering (see Appendix C)
SSI Semantic Structure Identification (see Section 7.3.3 and Appendix B).
VAS Visual Analogue Scale (compare AVI-scale) (see Section 8.2.1)
QFD Quality Function Deployment. Method within affective quality work (see
Appendix C)
QT 1 Quantification Theory Type 1 (see Section 7.5 and Appendix B)

xi
xii
 Before Reading

This thesis is one of the first works written in English addressing the field of designing
affective values into products and Kansei Engineering. It is assumed that some readers
wish to make quick reference to certain parts of the thesis. Hence, the thesis is
structured ‘modular’ in order to allow easy access. Some comments regarding the
disposition might therefore be useful in order to facilitate the reading of this thesis. It
consists of two main parts: a compilation and summary of research findings, and
appendixes including published articles.

The purpose of the first part is to present a holistic picture of the research and to relate
the research findings to theory. In chapter 1 and 2, the reader is introduced to the
research field of integrating affective values in products. Methods in the field are
presented briefly, followed by the aims of the thesis in chapter 3, and methodology for
the research in chapter 4. Kansei Engineering methodology is explained more in detail
in chapters 5-6. Chapters 7 and 8 are summarising the research findings from the
authors’ publications including a proposed model on Kansei Engineering methodology
and experiences from applying the methodology in Swedish industry. Chapter 9
discusses the findings and conclusions are drawn in chapter 10.

The second part, the appendixes, include complementary material giving a more
detailed background picture of the area of knowledge. It is divided into five parts:

• Appended papers published by the author

• A guide to tools in Kansei Engineering
• Other methods for measuring affective values in products
• A guide to literature within the field of Kansei Engineering
• An in-detail description of the different Types of Kansei Engineering

Since the area of research is rather new and yet not clearly labelled, many sources are
difficult to find through keywords. Also, many important sources are published in
Japanese or Korean language. Hence, the appendixes are relatively voluminous in
order to guide also readers with a deeper interest.

xiii
 1.
Introduction
はじめに

This chapter introduces and makes the reader familiar with the topic. It explains why it is
necessary to integrate affective aspects into product design and presents existing tools and
methods in short. The field of research is quite new, and a part of this chapter also deals with
today’s definition of it. Finally, affective aspects in the field of quality are shown and the
research partner, BT Industries AB is presented.

1.1. Customer Needs in Future Products

During the recent decades the production capacity of consumer products has increased
in a never before seen scale. New national and international actors arise and world’s
markets are globalizing at high speed. Changes therefore play a more important role
leading to shorter life cycles of products than before, see Brown and Eisenhardt
(1998). Shorter life cycles mean shortened development time of new products and in
combination with hard competition it also entails an increased focus on cost reduction
(IVA, 1999).

At the same time occurring new techniques such as Internet and mobile
communication, flat TV-screens etc., allowed the development of previously
unthinkable products but also forcing manufactures to fast adaptation of product
development and production methods (IVA, 1999). Also, new customer demands arise
due to improved cognitive ergonomics and education for highly advanced products.
Another effect is that new products also become mature more quickly and must be
followed up more frequently starting the Plan-Do-Study-Act (PDSA)- circle, a tool for
constant improvement in the sector of quality over and over again (Deming, 1986).

Together with an increasing demand for higher product quality the situation becomes
even more complicated; products must have higher performance, better function and
they must be developed faster.

Another challenge for companies is even more crucial for companies in these markets.
Strong ‘supertrends’ such as hedonism, spirituality, downsizing and individuality

1
 Introduction

(Jordan, 2001) abandon the traditional focus on functionality and concentrate on

‘softer’ issues such as Hedonic Ergonomics in design and Pleasurable Products and
Interaction (Helander, 2001)

Even if it seems so, today’s market situation is not completely new. Industry has been
challenged many times before and adapted by adopting new ways of working. One
way of improving market shares on markets full of mature and similar products is to
make their product ‘edgy’, i.e. easily recognisable and typical for the brand.

Industry coped with these demands though more active integration of the customers
opinions in the designing phase. In the 1950ies and 1960ies the quality movement was
born. Although functional aspects were in focus in the beginning, usability and
intangible product characteristics soon became more important the early 1980ies
(Childs, 2004).

Today’s development probably goes towards integration of emotional values in

products. Products must appear independent reflecting an individual life style. Also, if
the customer has the choice between products from different manufacturers, which are
equivalent in price and performance a consciously build in ‘good feeling’ can trigger
the final purchase decision.

Consequently, it is the customer’s emotional needs which primarily must be taken care
of. For integrating the voice of the customer different methods exist for industrial use
(Bergman and Klefsjö, 1994), but when it comes to measurements of emotions,
impressions and pleasure, methods are very rare. So research is needed and in the
recent years new research directions such as ‘Emotional Design’, ‘Affective Design’,
‘Affective Ergonomics’, ‘Pleasure with products’ etc. have appeared (ENGAGE,
2005). All these research directions have in common that they try to grasp the
customer’s and users’ personal impression and make it describable or even measurable
in order to evaluate product solution according to their emotional impact.

However, as these research directions are not interconnected, one could say that there
is a great need for methods, but the research is still in its infancy. It is surprising that
there is not more research done on this area since we as human beings apparently have
quite little knowledge about how we interact affectively with products (ENGAGE,
2005).

Helander (2001) identifies the most urgent research needs. Firstly, the measurement
issues and theory formation must be addressed. Secondly, it must be possible to
predict user/customer needs for affect.
2
 Engineering Emotional Values in Product Design

In 2004 a European project was started under the name ENGAGE for just these
purposes. Its goal is to form a knowledge community involving all research directions
in order to define the content of the new research field to be formed.

1.2. Labelling the Field of Knowledge

The emerging research field is currently addressed by many names. The ENGAGE
network lists and define some of the labels which have been used in the area:

• Emotional Engineering
• Affective Design
• Affective Engineering
• Affective Ergonomics
• Design for experience
• Pleasure with products
• Design of metaqualities
• Design for human senses
• Kansei Engineering
• Sensorial Engineering
(ENGAGE, 2005)

The discussion is currently not finished about what the new area will be called but it is
clear that the current lack of agreed terminology must be overcome in order to promote
it properly. When looking on the definitions behind the denominations above it
becomes clear some are more general than others and therefore more suitable. Within
the ENGAGE network also the hierarchy of the different labels will be defined in
order to identify suitable candidates.

The Keyworth Institute in Leeds/GB attempt a definition. ‘Affective engineering is the

study of the relationships between physical and rational products features and their
subjective cognitive or emotional influences on the people interacting with them, and
the use of the knowledge gained to design more satisfying products’
(www.keyworth.leed.ac.uk, 2005). Using the term ‘engineering’ implies a ‘systematic
approach that aims to integrate engineering, scientific and artistic approaches for the
design and delivery of products’ (ENGAGE, 2005)

3
 Introduction

1.3. Methods in the Area of Affective Product Design

The idea of integrating values related to user impression into products is not entirely
new. It developed gradually since the early 1970ies after an almost 20 year long period
of economic growth in Europe which started when the damage from the Second World
War was repaired.

New competitors emerged foremost from outer-European countries. Since markets

were saturated the strategy of producer-controlled supply did not work any more and
the need for new product development strategies became obvious.

Hence, in the late 1970’s companies began integrating customer-oriented techniques

into their product development processes. Companies made their production more
efficient and spent more effort on quality management. As a consequence the variety
of products increased and the quality of the products improved driven mainly by
competitive pressure from Japan (Juran et al., 1974).

Measuring the customers voice in combination with highly flexible production-

systems made it possible from the early 1990’s onwards to satisfy the wants of even
selective customers with individual taste and high demands on quality, adaptability
and personalisation.

Constantly shifting trends in customer demands require new or improved tools

integrating even affective aspects into product development. Such methodologies are
e.g.:

• Semantic Differential Methods (Osgood et al., 1957)

The Semantic Differential Scales SD- Scales are a political instrument for
measuring the affective impact of political streams on the citizens mind. This
tool can also be used in a modified version for product development.

• Conjoint Analysis (Green and Srinivasan, 1978)

Conjoint Analysis is a monetary tool. It is designed to ascertain how much
money the members of a target group are willing to spend on certain product
features.

• Semantic Description of Environments (SMB) (Küller, 1975)

Semantic Description of Environments (in Swedish: Semantisk Miljö
Beskrivning) designed as a method for evaluating architectural structures

4
 Engineering Emotional Values in Product Design

according their aesthetic appearance. This method originally addressed artistic

non-commercial interests.

• Quality Function Deployment (QFD) (Akao, 1990)

QFD is an engineering tool developed in by Japanese Quality technology
experts. It identifies relations between customers’ (functional) needs and
engineering characteristics.

• Kansei Engineering (Nagamachi, 1989)

Kansei Engineering is an engineering tool. It collects the users’ emotional needs
and establishes mathematic prediction models of how the emotional needs are
connected to selected product properties.

In particular the last methodology mentioned, Kansei Engineering is of special

interest, since it is the only tool especially designed for quantifying emotional
customer needs and develop them into products. Kansei Engineering has probably
come furthest in the pursuit of introducing engineering methods into implicit customer
needs.

In contrast to other methodologies Kansei Engineering has the ability to collect and
prioritize the customers’ feelings and distinct customer groups with different ‘tastes’.
It also can collect the product properties which are most important to the user.
However, what distinguishes Kansei Engineering from the method above is its ability
to build mathematical prediction models on how feelings are connected to product
properties.

1.4. Affective Quality Work

1.4.1. A Short History on Quality

The modern history of quality began in the 1920ies in the USA. Sheward and Radford
stand for a systematic research on this area (Axelsson, 1999). They recognised, that
good quality products supports sales. However, it was in Japan that the ideas were
applied in full scale.

Quality thinking was introduced in Japan after the Second World War. Japanese
society had been isolated for many decades before the Second World War and
struggled with social problems caused by the new open society. Japanese industry saw
itself confronted with the competition from the global market. A bad reputation and
bad quality of the products made in Japan made them difficult to sell. However the
Japanese learned quickly to take advantages of the changed situation and were open to

5
 Introduction

new impulses from abroad. They began to manufacture foreign products under license
and tried to improve their quality. Two Americans, W. Edward Deming and Joseph. M
Juran played an important role in quality development.

In the middle of the 1970’s Japanese companies had not only survived the challenges,
but had grown stronger and produced products that were attractive to customers. The
world depression made it possible for Japanese products to enter European and
American markets, since customers ranked quality products at a reasonable price
higher than brand-fidelity. It was mainly cars and cameras, which sold best but soon
Japanese brands took over even big shares of the electronic market. European and
American companies were forced to react, but the Japanese competitiveness could not
be broken, even with increased import taxes on Japanese products. Many companies in
the West became insolvent, the survivors adapted and integrated quality thinking into
their organizations (Bergman and Klefsjö, 1994).

1.4.2. Evolution of the Concept ‘Quality’

The word quality originates from the Latin word ‘qualitas’ and means ‘of what’. In
general usage it describes the property or the nature of things (Cicero, 2004). In an
industrial context the definition of quality shifts depending on the context. Moreover,
the expression ‘quality’ also underwent significant changes during the time of its
existence. Originally, the word was strictly bounded to the area of production control.
Nowadays it has become an everyday word in everybody’s mind indicating everything
from ‘quality vacation’ to ‘quality time’.

In the 1950ies when the quality movement started the expression ‘Quality’ was used as
a means of insuring good quality of goods in industrial production. Famous
researchers in this era were names such as Crosby, Deming, Juran. They saw the task
of quality in the following ways.

‘Conformance to customer requirement’ (Crosby, 1984)

‘Fitness for purpose or use’ (Juran et al., 1974)
‘A predictible degree of uniformity and dependability at low cost suited to market’
(Deming, 1986).
Later, the expression quality was also applied in other areas in industrial activities such
as product development and after-sales. Moreover it was not only restricted to tangible
products but also to intangibles (compare Section 2.4). The original ISO 9000
definition provides a direct connection between quality and the properties of a product:

6
 Engineering Emotional Values in Product Design

Quality is ‘the totality of those properties and characteristics of a product or an

activity that relate to its suitability to fulfil stated requirements.’

In the late 1980ies and 1990ies a new way of thinking was added, apart from the
controlling tasks in industrial production. This was the customer perspective. A new
expression occurred: ‘perceived quality’ (compare Zeithaml (1988)). For the first time
it became possible to measure if the objective quality in production was perceived in
the same way by the customer. With this, even the definitions had to be extended.
Garvin (1988) drew eight dimensions of quality work, of which the first six were
consistent with traditional quality work. New were the last two dimensions: aesthetics
and perceived quality. These were new areas, where almost no tools existed.

1.Performance
2.Features
3.Reliability
4.Conformance
5.Durability
6.Serviceability
7.Aestetics
8.Perceived quality
Garvin summarised the two new dimensions in a single sentence:

‘Quality is in the eye of the beholder!’ (Garvin, 1988)

Others followed and definitions including affective values occurred and revealed
connections to the field of ergonomics:

‘A product/service is of quality when it makes a maximum contribution to the health

and happiness of all people involved in its production, use, destruction, and reuse’
(Aune, 1999)

Tribus defines quality in even more emotional terms: ‘Quality is what makes it
possible for a customer to have a love affair with your product or service. Love is
always fickle. You must be ever on the alert to understand what pleases the customer,
for only customers define what constitutes quality’ (Tribus, 1990).

Focusing on product development processes two aspects of the above presented

definitions are of importance:

• Affective aspects (emotional impact of the new product)

• Design aspects (product properties)
7
 Introduction

It is hypothesised that affective impact on the user is consequently a result of the

composition of the different product properties. Hence, the goal must be to choose the
best combination possible in order to maximize the customer satisfaction. Figure 1
lists some quality dimensions of an article as it is seen by Bergman and Klefsjö (1994).

Figure 1: Some quality dimensions of an article (adapted from Bergman and Klefsjö
(1994)).

1.4.3. The Kano Model as a Tool for Attractive Quality Creation

When Kano, Seraku and Takahashi carried out investigations on customer needs in the
early 1980’s, they discovered that customer needs could be grouped into different
categories on different levels. This discovery led to the introduction of the basics of
the model later called the Kano Model (Kano et al., 1984). Figure 2 shows the Kano
model in its later version. The x-axis displays the degree of achievement and the y-axis
the degree of customers’ satisfaction with the certain type of achievement. Depending
on the product property being considered, different function-characteristics can be
obtained.

8
 Engineering Emotional Values in Product Design

Figure 2: The Kano model (adapted from Kano et al. (1984)).

For example, basic demands like the electrical start of cars or automatic warming
system in houses are considered a must-be quality. Often customers do not express
these demands, since it is self-evident to them. Usually one-dimensional quality is
expressed, i.e. special light metal rims or CD-player in cars, tiles in the bathroom, etc.
The customer is aware that this type of equipment is not standard and expresses a
desire for it. Manufacturers try to surprise the customer by adding a feature, which was
not expected like a sponge for body gel, electrical heated side windows in a car, etc.
This product property might be decisive for the customer buying the product.
Particularly common products manufactured by many different companies can be
made recognisable to the customer as common in this way. This is called an attractive
quality.

Kano points out that almost every product property traverses a certain life cycle. When
the property is newly introduced to the product the customer considers it an attractive
quality. After a certain time the customer gets used to it and expresses it when
purchasing a new product. In the end the feature becomes common and it can be found
in almost every product. Then it has become ‘must-be quality’. Take the remote
control for a TV-set as an example. In 1970 this feature was probably considered very
attractive. Several years later practically all manufacturers could offer a remote
controlled version of their TV-sets and the remote-control became a desirable feature
(one-dimensional quality). Since the number of TV sets sold equipped with remote
control was relatively low, it became a kind of status symbol too. Nowadays remote

9
 Introduction

controls are standard. No TV set is sold without them any longer, since it has become
an expected must-be quality. This is called the life cycle of product features and is
indicated in Figure 2 with the bold arrow. As a consequence manufacturers must find
new attractive features in order to make their products distinctive (Kano et al., 1984).

In the context of integrating feelings into products the Kano model is highly relevant.
Most product properties have an emotional impact on the users mind and the
producer’s goal is to make it positive. Especially interesting in this context is attractive
quality creation meaning that the user should become positively surprised about a
certain feature. This of course is closely connected to the affect.

1.5. Industrial Partnership

In order to conduct the research presented in this thesis a number of studies have been
carried out within the time period between spring 2001 and winter 2004/2005.
However, only the minor part i.e. the most significant studies are presented explicitly
in this thesis. Many minor studies have been conducted in cooperation with companies
such as Electrolux, Hörnell International, Saab Automobile, Scania, , Volvo Tech (see
www.ikp.liu.se/kansei).

However the majority of studies was done in cooperation with BT Industries AB. BT
also financed a considerable part of the authors work.

1.5.1. BT Industries AB
BT Industries AB is located in Mjölby /Sweden. It is one of the leading manufacturers
of industrial handling equipment and the worlds biggest
manufacturer of electrically driven warehouse trucks.

BT was founded in 1946 as a construction and transportation

company. Among others BT also imported industrial handling
equipment and started one year later fabricating hand pallet
trucks. Soon the manufacturing of handling equipment
became the main focus for the new company. From this
moment the company started a period of continuous growth.
After a major reorganization in 1990 BT acquired Raymond, www.bt-industries.com
one of USA’s biggest truck manufacturers. In June 2000 BT was bought by Toyota
Material Handling Company and is now a part of the world’s industrial handling
equipment manufacturer.

10
 Engineering Emotional Values in Product Design

1.5.2. Product Range

BT’s product range stretches from hand pallet trucks to 5 ton diesel fork trucks.
Moreover, it includes applications for horizontal transport, stacking, order picking,
very narrow aisle, and combinations of these items (BT-Industries, 2003).

The product program is complemented with surround services, which include

prophylactic and emergency maintenance in different forms, spare part supply, short-
and long time rental, financing and driver education. These parts account for around
50% of BT’s turn over, which is typical for this branch. (BT-Industries, 2003).

Trucks and services can be composed into individual packages adapted to particular
purpose. Consequently, BT delivers in most cases not only the truck itself, but the
combination of tangible and intangible sub-products.

1.5.3. Meeting Customer Demands at BT

BT Industries AB has experience in designing the customer’s demands into their
products; i.e. trucks and surrounding services. Adapting the services to the customer’s
demands is possible by tailoring the rental and service contracts individually.

Product development methods such as Quality Function Deployment (QFD) make it

possible to collect the voice of the customer regarding a tangible product.
Modularisation contributes to an easy change of components in order to suit the
customer’s demands. In addition, prototypes or parts of the new trucks are tested in
field tests under real conditions, which gives valuable feedback from the actual users.
Also BT works according to Toyotas product development system which at its core
includes the concept of ‘go and see’, i.e. the staff visits places where BT products are
used and can observe the usage and talk to the users (Schill, 2005).

11
 Introduction

1.5.4. BT REFLEX
The BT REFLEX is an electrical driven warehouse reach
truck. Technically the REFLEX is a reach truck which is
characterized by a seating position oblique to the driving
direction in order to reduce the length of the vehicle.
Together with a single wheel in the back this design allows
better manoeuvrability than in conventional designs e.g.
counter balanced trucks. Consequently, a narrow isle width
in the warehouse can be realized. Another contribution to
this is the retractable mast. The driver can move the truck in
front of the stack lift the forks to the intended level then
slide the forks into the pallet by extracting the mast and pick
the load. This leads to an efficient handling and reduces the risk for damage of good or
injuries. The REFLEX is on of BT’s most technically advanced products and is
manufactured in high volumes with good profitability.

1.5.5. Market Segmentation

The market for industrial handling equipment is strongest in industrialized countries.
The total volume of this market was estimated to 600000 units in 2000, of which circa
40% are electrically driven warehouse trucks and around 50% are counter weight
trucks. According to BT, the market for reach trucks can be divided into three main
categories as shown in Table 1.

Table 1: Market segments for Reach Trucks (source: BT Industries).

Group I II III
Frequency of use Low Medium/High High
Ownership Purchase Rental Rental
Lifting Height Low/Medium Medium/High High
Driver Layman/Professional Professional Professional
Qualification
Operating Time <700 h/Year 1000-3000h/Year 2000-3000h/Year
Sales in % 15%-20% 80-85%

Group I is characterised by trucks with low frequency of use, used in small

warehouses with a small number of employees working with various tasks. Material
handling equipment in this sector must be easy to handle by everybody and universally
applicable. Since the majority of these trucks are owned by the company the

12
 Engineering Emotional Values in Product Design

purchasers are relatively price sensitive. Group II and III are rental trucks with
medium to high frequency of use. Those are typically operated in big warehouses
under single or multiple shift conditions. In these cases the operators are professional
truck drivers, requiring an adaptability to their personal working situation.

1.5.6. Productivity through Ergonomics

BT Industries supports joint long-term research in the field around productivity and
ergonomics (PRODERGO). The founders of this project were Eric Berg, BT-
Industries AB and Professor Jörgen Eklund, LiTH. Today this project includes three
forms of cooperation.

The central pillar is cooperation largely based on the work of Ph. D. student, which
also is the basis for this thesis. Often the projects are an integral part in the ongoing
research work. Further, the cooperation includes project work by undergraduate
students and ‘learning laboratories’ where the researchers and personnel from BT have
joint seminars in topics relevant for PRODERGO at regular intervals.

13
Introduction

14
 2.
Definitions
定義

This chapter includes definitions of the most important expressions and concepts used in this
thesis work. Several different sources are quoted and compared to each other, followed by a
motivation of the authors choice.

This work is written in a way that it would not require pre-understanding of technical
specialised expressions. However, the author is aware that parts may be hard to
understand even for the initiated reader. In these cases a special dictionary can be of
assistance.

On the other hand, certain expressions might be common and seem easy to understand,
but in the context of this work they are used in a restricted sense, with a different
perspective or focus than in other contexts. In order to prevent misinterpretations,
misunderstandings or simply to make the reader aware of unusual nuances within
certain expressions, some of the most central and most frequently used terms are
defined here. Moreover, a list of abbreviations is also provided in the beginning of this
book, hopefully facilitating reading.

2.1. Affect
Within the scientific area ‘Affective Engineering’ and ‘Kansei Engineering’
respectively the expressions ‘affect’ and ‘affective’ are used frequently and therefore
hold significant importance. Hence, it is opportune to explain the expression itself and
define its usage for this thesis.

A dictionary definition states that ‘affect is an emotion or subjectively experienced

feeling, or the involvement of such processes…’. Moreover, in abnormal psychology
affect can also refer to ‘emotional expressiveness’ (Encyclopaedia Britannica Online,
(2005).
15
 Definitions

The American psychologist Edward Titchener uses the term in a more specific way
outside the mainstream. He refers to the concept of affect to a pleasantness-
unpleasantness dimension of feeling (Titchener, 1998). This definition comes closer to
the meaning the term is used in this context since also Osgood uses the assumption of
a one-dimensional affect in his Semantic Differential-Scales (Osgood et al., 1957).

In his openscource project ‘Passionate Engines’ The researcher DeLancey, (2002)

illuminates the concept of affect from a pragmatic angle. His concern is to utilize the
findings for building an affective engine using Artificial Intelligence (AI). According
to him, affect is a superior expression embracing all so called states. As states he
defines desire, emotion, pleasure, moods, etc. He also disagrees with Titchener’s
concept of affect as bivalent and mono-dimensional. Instead he suggests an amorphous
structure. DeLancey also makes a distinction between what he calls occurrent affect
and disposition to affects. According to him occurrent affect is a current state, whereas
a disposition to an affect is a description of which occurrent affect is caused in a
certain context. Therefore he concludes that disposition to affect is derived from the
concept of occurrent affect (DeLancey, 2002). This assumption can be strengthened by
the discussion of another AI researcher. Rosalind Picard, says that emotional states can
not be maintained over a long time (Picard, 1997) but can be stored latently in form of
moods.

The expression concept of affect in this thesis is seen as an occurrent state, because the
customer’s immediate reaction on the products in question is of interest. Moreover, the
author chooses to utilise Titcheners view on affect as a mono-dimensional variable in
order to be able to use Osgood’s SD-scales.

2.2. Affect vs. Emotion, Feeling and Mood

As mentioned before the expression affect and affective respectively are used
frequently in this thesis. However, this expression automatically leads to associations
or even overlaps with other expressions which also will be used in this work. A short
discussion is required in order to avoid possible misunderstandings. The way the
expressions are defined is not always the most common definition.

One major expression is the word ‘emotion’. When trying to find a proper definition
the author found many different ones, each of them context defined and more or less
generally expressed. Kleinginna and Kleinginna (1981) did an intensive search on this
topic and identified and categorised about one hundred different definitions. However

16
 Engineering Emotional Values in Product Design

even if it is not possible to present all definitions here, it is possible to define the
differences and joint possessions of the concepts of ‘affect’ and ‘emotion’.

‘Emotion’ is often described as the antithesis to reason (Damaiso, 1996). It was also he
who proved this empirically and showed in his studies the importance of emotion for
decision-making. In the context of affective engineering and AI, the term ‘emotional’
is often used equivalent with the term ‘affective’ (Picard, 1997). However, some
specific definitions see emotions placed on an instinctive basic level, comparable to
the sexual drive (DeLancey, 2002). This is not what is meant in this thesis. In order to
avoid confusions the term emotion is mostly replaced by the term ‘affective’. If the
instinctive basic level of emotions is explicitly addressed the term ‘sentic’ is used. This
term originates from the latin term sentire and emphasizes physical mechanisms of
emotion expression (Clynes, 1977).

Another term appearing rather frequently within this research area is the word
‘feeling’. Feelings can be seen as less subjective phenomena than emotions which
rather evoke emotions than are emotions themselves (Damaiso, 1996). Feelings are on
a lower level of consciousness and are connected to anatomical physical properties.
However there are different types of feelings, depending on in which way an emotion
is triggered. Those types are:

• internal (physical) feelings

• external (social) feelings
(Encyclopaedia Britannica Online, 2005)

Picard, (1997) defines ‘feelings’ strictly as a physical sensory input although she is
aware that they sometimes are used equally with emotional experience. This is not
necessarily the most common definition but in order to avoid confusion the term
feelings is defined in this way for this thesis.

The third expression to be defined in this context is ‘mood’. It is certainly less

common in the research on affective topics, however it plays an important role in
subjective experience. A ‘mood’ is a long term affective state and is triggered by a
combination of emotions (Picard, 1997). The difference to the other concepts
presented above is that people are more aware of their state of mood and therefore can
express it. So if it is attempted to measure the affective values this can preferably be
done indirectly by measuring the mood, which in turn is evoked by external events
(Encyclopaedia Britannica Online, 2005).

17
 Definitions

2.3. Design vs. Engineering

In the field of affective product development two expressions occur frequently naming
the process itself: ‘Design’ and ‘Engineering’. In some cases they even together appear
as ‘Engineering design’. These terms are important, since they are considered as name
for an entirely new field of knowledge.

In literature about affective products the two expressions are sometimes used equally,
sometimes there is a distinction between the expressions. Also the names of the
scientific approaches (Emotional design (Nagasawa, 2002a), Affective Engineering
(Barnes et al., 2004, ENGAGE, 2005), Affective design (Helander, 2003)) show that
Design and Engineering are closely related to each other, because they indicate the
same area of research. Despite that they are not completely identical.

Engineering is according to a dictionary definition ‘the application of science to the

needs of humanity. This is accomplished through knowledge, mathematics and
practical experience applied to the design of useful objects or processes.’
Encyclopaedia Britannica Online, (2005). From this it can be concluded that design is
a tool in the area of engineering.

The definition of ‘design’ supports this thesis: ‘Design is the process of originating
and developing a plan for an aesthetic and functional object…It is used in the areas of
applied arts, engineering, architecture and other such creative endeavours’. Despite the
fact that design both has an aesthetic and functional dimension the expression is
understood by Swedish and German people almost exclusively (aesthetical) industrial
design.

Concluding both design and engineering are creative actions deployed in e.g. product
development processes. The main differences are that ‘design’ tends more to the
artistic side whereas ‘engineering’ is closer related to natural science and mathematics.
This is also the way the expressions are used in this thesis.

2.4. Products
The word ‘product’ is derived from the Latin word ‘productum’, which means result or
gain. Initially it had a strongly limited usage in mathematical science. However, during
the industrial revolution it extended its significance even to the commercial sector and
it became synonymous with industrially manufactured commodities. Traditionally,
these types of goods are tangible products i.e. physical objects.

18
 Engineering Emotional Values in Product Design

With time even non-physical products e.g. services were considered as products, but it
took time until this point of view became accepted. Nowadays the expression product
naturally also includes intangible products (Röstlinger and Goldkuhl, 1999).

Later a third form of product emerged, which were tangible products combined with
intangible features such as delivery, installation and sometimes even maintenance of
i.e. a washing machine. However, the focus hitherto is mostly on the physical part.

With emerging new techniques new trends appear. Quite recently the focus shifted for
many new products from the mentioned third form towards the intangible part i.e.
these products are foremost as services which necessarily requires tangible
components (IVA, 1999). Examples are mobile telephones, rental TV or household
machines, etc.

Figure 3: Product Definition used.

Extrapolating this trend a vision could be that future products whether they are
tangible or intangible, are owned by the company producing them and customers pay
for the access of using it. The driving forces for such change are sometimes external.
For the person transportation sector Toyota predict a model where the company owns
the vehicle and all peripheral services and the customer pays for the transportation
alone.

2.5. Expressions in Subjective Assessment

As many methodologies dealing with qualitative data, also the methods within
Affective Engineering such as Kansei Engineering utilise subjective assessment
methods (compare Section 8.2). These methods require a certain type of jargon.
Although these expressions are internationally recognised Kansei Engineering
researchers sometimes use different words.

Entity/ Concepts
According to Sinclair, (1990), entities are the products to be scaled, for example the
fork-lift trucks in the study for BT Industries (compare Appendix A, Paper A). In
Kansei Engineering also the expression ‘concept’ is used in the same context.

19
 Definitions

Attribute/ (Product) Property

An attribute is the property of the entity which is scaled. This can be an ability of the
product as for the trucks mentioned above, the lifting speed, the colour or the height.
Both terms are common in Kansei Engineering literature.

Item
The item is a term used mainly by Nagamachi. It means Product attributes/ properties
chosen for Kansei Evaluation. A attribute/ property is called an item when it has
passed the ‘Spanning of the Semantic space’ and is chosen for evaluation with
Quantification Theory Type 1 QT1, a type of linear regression using dummy coding
(compare Appendix B)

Category
Displays the variety of an item. This is also a Kansei Engineering term. An item has a
number of different categories which can vary. For example the colour of a truck can
be orange, red, yellow, etc. These are called categories. The introduction of the
expression category is mainly used for Quantification Theory Type 1 (QT1)

Subject/ Participant
The people used for scaling. In some of the appended papers they are referred to as
participants.

Respondent
The subjects/ participants used in questionnaires. Sinclair (1990) makes a distinction
between subjects and respondents. However since this work deals with data gathered
from questionnaires there is no difference between subjects, participants and
respondents here.

2.6. Data, Information, Meaning, Knowledge

Using scientific methods means that models of the reality are built from a noisy and
fuzzy environment. So from the decision on which raw data to collect in order to
describe or measure a certain phenomenon until the underlying mechanism is properly
understood, the data has to be reduced. Otherwise the human mind cannot cope with it.
In this context three expressions are of interest, indicating the different stages of
‘purification of the raw data’. These phrases are: ‘data’, ‘information’ and
‘knowledge’.

‘Data’ is derived from the Latin term ‘datum’ which means ‘a statement accepted at
face value’ (Encyclopaedia Britannica Online, 2005) i.e. ‘to give’ or ‘given value’. In
20
 Engineering Emotional Values in Product Design

computing and scientific contexts with ‘data’ is meant ‘raw data’ which are numbers,
characters, images or other outputs with no mutal structure.

‘Information’ is a type of non-material input into an organism or device. Information is

often carried by a weak stimuli which requires special equipment for amplification in
order to be visible. In contrast to data information is a result of data reduction by
elimination of noise.

‘Knowledge’ is defined by Plato as ‘justified true belief’. In other words: Knowledge

is the ‘awareness and understanding of facts, truths or information gained in the form
of experience or learning (Encyclopaedia Britannica Online, 2005). Although
‘knowledge’ and ‘information’ consist of true statements it is not equal. Knowledge
exists in the human mind or collective theoretical human mind like the ‘body of
knowledge’. However it can simultaneously exist in other forms as information.
Knowledge is perhaps the human means of storing information, facts, skills,
experiences etc.

Data on its own does not have any meaning. However it can be interpreted and
conditions in data processing systems and become information. Through learning
processes the information is understood and gets a meaning which then results in
knowledge.

2.7. Semantics
In general semantics can be explained as the study of meaning. The expression itself is
derived from the Greek term ‘semantikos’, which means ‘significant meaning’
(Encyclopaedia Britannica Online, 2005). So via meaning semantics is connected to
knowledge.

In the area of linguistics semantics is traditionally the study of meaning of parts of

words, phrases, sentences and texts. A newer application is in the area of mathematics
and computer sciences, semantics stands for the meaning of logic linkages. The
neighbourhood to knowledge here becomes even more visible.

In this thesis the term is used mostly in connection with Semantic Differential Scales
technique (SD-scales) (compare Appendix B and C,). Semantic Differential applies
both on linguistic expressions and product semiotics, i.e. the language of signs and
sign-systems.

21
 Definitions

2.8. User, Customer, Purchaser

New products must fit the market’s demand in order to be economically successful.
Hence, in the area of (affective) product development the customer is (or should be)
the focus of each process ensuring that new products will fit her needs.

By definition a customer is somebody who buys a product in order to achieve a certain

goal with it (compare Figure 4). So there are two aspects in the expression customer;
an economical and a functional. The economical aspect focuses on the purchasing
process whereas the functional describes the usage of the product.

Figure 4: Definition of ‘customer’.

In certain situations it is possible that the purchaser and the user are the same person:
the customer and in other cases not. Products for private consumption are probably
mainly purchased by the future user in person, while the buying decision for products
dedicated for use in a work company environment is often done by a professional
employed purchaser. However, the rules above do not always apply. Sometimes
products for private consumption are bought by another person e.g. as a gift;
sometimes factory-workers may choose between different machines that they are
supposed to use.

Companies often know much about ‘their’ customers, but little about the users behind
them. However for integrating affective values in products exact knowledge about the
user and user behaviour is needed.

In this thesis the expressions ‘user’ ‘consumer’ and ‘customer’ are equally used in the
sense of the person who really interacts with the product in a functional manner. If
explicitly an economical perspective is meant the term ‘purchaser’ or ‘buyer’ is used.

22
 3.
Aims and Delimitations
研究の目的と限界

3.1. Aims of the Research

As pointed out in the previous chapters, the area of research addressed in this thesis is
not entirely new. Affective values in products and attractive products have always
been an issue for customers. However, the questions that have been asked recently are
what makes products desirable on an emotional basis and how this newly gained
knowledge may contribute to the improvement of affective product properties.
In order to answer these questions the aim of this thesis was twofold:

• Firstly, to improve understanding of the characteristics of products making

affective impact on the users and customers
• Secondly, to identify and improve methods capable of grasping affective values
and translating them into concrete product properties

More specifically, the aim was to improve the understanding of the field of knowledge
by giving an overview of Affective Engineering and Kansei Engineering in particular.
This included the application of Kansei Engineering in concrete projects in Swedish
industry.

An important aim was also to analyse and improve Kansei Engineering methodology,
to summarise the findings, and on the basis of this to propose a general model on
Kansei Engineering methodology.

23
 Aim and Delimitaions

3.2. Delimitations
Kansei Engineering is in general possible to apply for both artefacts and services.
However, the studies in this thesis are made on physical products only. Hence, the
conclusions are mainly valid for this type of product.

Kansei Engineering methodology deploys different methods found in other fields of

research. In this thesis, these methods did not receive any further development. They
were, however, adapted with the purpose of being applied in Kansei Engineering.

Also, Kansei Engineering includes a number of statistical evaluation tools which

originate from other areas or knowledge fields. These were not developed further,
except necessary changes for adaptation.

The focus of this thesis has not been to suggest or improve specific product
development processes, but to use Kansei Engineering in product design for improving
existing products and product concepts

24
 4.
Research Methodology
研究方法

This chapter gives a brief overview on the segmentation of the research area. Also some of
the main research activities carried out for this thesis are presented. Finally, a structure of
the research is given and its phases are explained in detail.

4.1. The Area of Research

Integrating affective values in products is essential to many different areas of research.
This includes consumer research, ergonomics and quality research for identifying
customer needs via the product design and engineering marketing and surround
services.

Examples of research in Europe in the areas of product design and (affective)

engineering topics are found at e.g. the University of Leeds (Barnes et al., 2004,
www.keyworth.leed.ac.uk, 2005). Research on affective design is carried out at Delft
University of Technology (Hekkert, 1999). Customer research is performed in Sweden
at Chalmers University of Technology in Göteborg (Rosenblad, 2000). Research on
affective impact of buildings is conducted at Lund University (Küller, 1991). Kansei
Engineering research at Linköping University covers the areas of ergonomics,
industrial design, mechanical engineering, psychology and quality.

4.2. Research Activities

Different activities were part of the 4 years of research finalised in this thesis. Beside
his own studies the author also supervised a number of projects carried out by
undergraduate students. A highlight was also the 3 month visit as a guest researcher at
Hiroshima International University.

25
 Research Methodology

4.2.1. Author’s Studies

The first steps were to understand the ideas and the basics of Kansei Engineering
properly. Based on this knowledge, several small case studies could be evaluated
including few Kansei Words and a limited number of product properties. Kansei
Engineering literature mentions a number of different evaluation tools which were
tested in this context. In parallel, the author conducted a study on BT fork lift trucks in
order to find out how the products are perceived by the actual users in different
European countries. Kansei Engineering could be developed in a way that met BT’s
demands on efficiency and time.

The next step was to find support for the gathered result in the BT’s organisational
structure and its development process. In Japanese companies, Kansei Engineering is
often run separately from other development procedures and is able to deliver the
result independently from the actual company philosophy. However, the author chose
to introduce Kansei Engineering as an integrated part in product development. Hence,
the approach here was to inspect the different methods used in product development
processes and find entry points for Kansei Engineering data. This task was
accomplished by an in depth scrutiny of common product development processes and
related methods for gathering customer information. Once again BT offered the
opportunity to conduct studies and the author joined the pre-planning phase of a new
forklift-truck model in order to provide and condition the data gathered from the
previous study to the on-going project.

4.2.2. Student Projects

A number of projects were carried out by students in co-operation with industrial
companies. The author acted as supervisor and coordinator for these projects. Together
with BT, Saab, and Scania the manipulation feeling with switch keys was evaluated
(Rydman and Sandin, 2000). Another study dealt with the driving feeling of BT
warehouse trucks (Elsmark, 2000). In cooperation with Electrolux AB a study on
affective impact of vacuum-cleaners was done (Skogman, 2002). In the same period
also the opinion of welders on auto-darkening welding visors were tested for Hörnell
International (Burkhardt, 2001, Arnold, 2002). At Volvo Tech a group of students
evaluated the presentation of textile surfaces in vehicle environment using VR-
technique (Frisk and Järleskog, 2003). Also, a project was done on laminate flooring
together with Pergo AB (Lindberg, 2004).

26
 Engineering Emotional Values in Product Design

4.2.3. Visit in Japan

After finishing the Licentiate thesis the author visited the Institution of Kansei
Ergonomics at Hiroshima International University as a guest researcher for a three-
month period in the autumn of 2002. The purpose of the visit was to intensify the
exchange of experience on the area of Kansei Engineering and learn about tools and
their application in the area. Moreover, BT Industries had an interest of extended
exchange with its parent company; the Toyota Material Handling Company in Aichi
prefecture in central Japan.

Initially orienting conversations with the researchers at Hiroshima International

University took place. The author presented problems, which occurred in the European
Kansei Engineering studies done at Linköping so far. New approaches were prepared
to be tested back in Sweden. This was also the start of a series of regular seminars
throughout the period. Before the author had left Sweden, data was collected for a joint
study on cultural differences in affective impression of the form ration (ratio of high to
width) of kitchen furniture elements e.g. refrigerators. Together with data collected at
the same time in Japan, an analysis was carried out where the newly learned tools were
applied. The result was presented in the form of a report (Garcia and Schütte, 2002). In
the final stage the author visited Toyota’s Takahama plant in Aichi prefecture several
times and learned about their production system as well as the tools used for affective
product development. These results benefited the PRODERGO project (see Section
1.5.6).

4.3. Research Structure

In the following the research plan used for this thesis work is presented. It follows the
PDSA circle used in quality management (Deming, 1986). Knowledge about Kansei
Engineering is developed gradually by literature research, followed by case studies,
and finally deployed in real products and retested. The planning of the following
studies was based on the results of previous research sets using an evolutionary
process where more knowledge about the Kansei Engineering methodology was
gained. This approach is corresponds to the concept of ‘Kaizen’, a Japanese method
insuring that results learned are sustained (Imai, 1991). Figure 5 presents the resulting
research structure for this thesis work.

27
 Research Methodology

Figure 5: The research process of this thesis.

In the Swedish academic world the doctoral studies are divided into two parts. The
first part includes basic studies about the research topic and is usually completed with
the Licentiate thesis. In the second part more depth is gained and finally the Doctoral
thesis sums the findings up.

4.3.1. Phase I, Licentiate Thesis

The aim of the Licentiate thesis which was roughly identical to the first phase was ‘to
understand and apply a methodology, measuring and translating the psychological
feeling that customers have about a certain product’ (Schütte, 2002). Hence it seemed
to be suitably subdivided it into thee parts:

• Understanding Kansei Engineering

• Making Kansei Engineering work
• Application in full scale on Kansei Engineering

The first stage was characterised by an intensive literature study on areas related to
affective methods in product development. In particular the studies of Elsmark (2000)
and Nolimo-Solman, (2001) were important bases. The purpose was mostly the

28
 Engineering Emotional Values in Product Design

understanding and testing of the method. Research objects were chosen due to their
simplicity e.g. soaps and wristwatches.

Making Kansei Engineering work was a challenge, which was not expected from the
beginning. Some of the tools used in Japanese Kansei Engineering studies were not
applicable due to different reasons and had to be replaced by other alternatives. This
was done in several studies made by masters students closely followed and supervised
by the author. Examples here were the studies on vacuum-cleaners, welding visors and
on textile surfaces in vehicle environments as earlier mentioned.

Bringing together all the experiences collected in the first year, a Full scale study on
reach trucks was carried out in three different European countries (see Paper A in
Appendix A). This was the first study using Kansei Engineering alone.

The Licentiate thesis was then finished by proposing a general model for conducting
Kansei Engineering Studies. At that time the model was only built on the very limited
experiences from the European studies. It was up to further research to validate and
refine it.

4.3.2. Phase II, Doctorate Thesis

The second phase of the research presented in this thesis is based on previous results
from both phase I and the Japan visit. The proposed model of Kansei Engineering
procedure as developed in the Licentiate thesis, and the new and refined tools from
Hiroshima International University played a significant role. Like the first phase the
second phase is subdivided into stages:

• Analysing Kansei Engineering Methodology

• Improving Kansei Engineering Methodology

In the first stage of the second phase, the previously gathered information is analysed
and partly revaluated. New studies were made on e.g. rocker switches (Appendix A,
Paper B) among other things in order to verify deductively derived findings. Other
studies were carried out by undergraduate students supervised by the author. Beside
the results the studies also tended to confirm the model suggested in the Licentiate
thesis. The new data significantly underpinned the assumption in the paper on tools
and methods in Kansei Engineering (see Paper D in Appendix A). In parallel a small
book was written summarising the ideas of Kansei Engineering in the first book in
English language (Schütte and Eklund, 2003).

29
 Research Methodology

In the last stage, conclusions about the knowledge gained were drawn and
Improvements of Kansei Engineering methodology were done. Firstly, a theoretical
framework was proposed which later was confirmed by a study in co-operation with
PERGO AB (Appendix A, Paper E). Another study on manoeuvring panels in fork-lift
trucks was done with the same purpose. This study was at the same time a follow up of
the first full-scale Kansei Engineering study made on the previous model of fork-lifter.
Its purpose was to determine whether the new model really was improved on an
affective level. Two papers summarise the findings (compare Appendix A, Paper A
and C).

4.4. Scientific approach

4.4.1. Streams in Scientific Philosophy

The history of science shows mainly two streams. On one side the rational-empirical
approach which refers to the early Greek scientists like Aristotle, on the other side the
intuitive-theological approach, which refers to the early forms of biblical studies.
During the following centuries the dominance of the two forms shifted, depending on
the regime in power and spirit of the age, but both forms existed in parallel and
influenced each other. Today these mainstreams are called positivism and
hermeneutics.

Positivism
The positivism grew from Aristotle’s logic which was based on empiricism. Around
1930 Comte (1979) recognised that different scientific methodologies were used in
different scientific fields. He pleaded for a homogenous methodology, ‘which could
guarantee positive knowledge’. This was the starting point for positivism.

One of positivism’s most essential assumptions is that existence of a true reality- a

reality, which is independent of the researcher’s beliefs and convictions. Knowledge
can only be gathered from observations and has to be documented in a logical,
analytical or mathematical way in order to give the statements a universal character
and make the observations repeatable.

Scientific knowledge is according to Comte, an instrument for predicting future

events. These predictions can be based on probability terms, which in turn have to be
based on sensory data. Useful scientific data must be measurable and express
observable laws. This make the positivistic approach suitable for natural sciences.

30
 Engineering Emotional Values in Product Design

Hermeneutics
Hermeneutics is a Greek expression and refers to Hermes, the messenger of the gods.
During the Middle Ages, Protestant priests were called Hermeneutics, since they could
interpret the Bible. These priests were representatives of the above mentioned
intuitive-theological approach.

A Hermeneutic perspective assumes that human expressions follow a universal

pattern, which is expressed by actions, language, gestures, etc. Since humans live in
the context of the present their actions will be based on history and controlled by
previous experience. The Hermeneutics state that no fact is static, but everything can
(and has to be) interpreted. This is valid even for material things. Reality is what
humans consider to be real and consequently subjective. The basis of human thinking
is language, which forms the way of thinking, clarifying standpoints and evaluating as
well as understanding and having a perspective on life.

Hermeneutics strives after an understanding and an interpretation of the human context

rather than ‘hard’ measurements and establishing of laws. Moreover, according to the
Hermeneutic perspective it focuses on the overall impression of things. Whereas
positivism examines a phenomenon in its parts, hermeneutics considers that the sum of
the parts results in a greater meaning. Hence, hermeneutics is especially suitable for
human science approaches.

4.4.2. Quantitative vs. Qualitative Measurements

In recent years much has been written manifesting the distinction between qualitative
and quantitative research methods. It appears as if one rules out the other in a research
approach. In practice this is not the case. Both methods can be used within the same
research problem (Patel and Tebelius, 1987). The choice of which method is most
suitable depends on what kind of information is required.

If the purpose of a study is to reduce the facts gathered to mathematical figures, and
formulas to provide a measurability of the investigated phenomenon: a quantitative
approach should be chosen. Among practitioners, quantitative measurements are often
considered to be more objective (Black, 1999). On the other hand qualitative methods
suffer from a lack of detailed description, especially in poorly developed research
areas (Strauss and Corbin, 1990). Since these are the demands the positivistic view
makes on its investigation methods, hermeneutics often rejects the use of this method.

Qualitative measurements are mostly used in hermeneutics. The research question

aims for a deeper understanding and interpretation of a phenomenon. With a holistic
31
 Research Methodology

view upon the settings, and the humans involved, the context becomes important
(Taylor and Bogdan, 1984). The strength of qualitative methods lays in their ability to
support understanding of details and correlate them to the context. Overlying patterns
becomes detectable (Patel and Tebelius, 1987). Findings on the other hand are not
arrived at by means of statistical procedures or other means of quantification (Strauss
and Corbin, 1990).

4.4.3. Building a Bridge between Customers and Feelings

Starting from an easy experiment, Fibonacci developed a series of numbers, nowadays
known as Fibonacci-numbers. This series is commonly used in art i.e. for determining
the ‘golden section’ and can be seen as a link between mathematics and art, and
qualitative and quantitative science (Liedman, 2001).

Market research utilises social scientific theories in order to reveal customers’

demands and attitudes. Whereas product design is primarily based on natural science’s
mathematical terms. Integrating the findings from market research into product design
requires a bridging of the different areas. The main task is to translate feelings,
perceptions, individual experiences and understandings into ‘hard’ measurements and
mathematical models which in turn have to be falsifiable using both qualitative and
quantitative methods. Figure 6 shows this graphically.

Figure 6: Scientific Context (adapted from Patel and Teblius (1987) and Starin et al.
(1991)).
Starting at a hermeneutics-context data about the emotional impact of products is
gathered. Using induction hypotheses theories are built describing the relations either
quantitative or qualitative. The models which had been gathered in that way are a
linkage between a certain impression and e.g. a product property which has to be

32
 Engineering Emotional Values in Product Design

verified using deduction methods. This means that the bridge between customer
feelings and product properties has to be double tracked.

33
Research Methodology

34
 5.
What is Kansei?
感性とはなにか?

The term Kansei is a Japanese expression which with no direct corresponding term in
English. This chapter makes an attempt to explain the concept to the reader. After a brief
illustration the etymology of the Japanese term is presented as well as several eminent
definitions. This is followed by a more detailed explanation of the concept. Finally
measurement methods are presented, among them the Semantic Differential Scales Method.

5.1. The Nature of Kansei

The Japanese word Kansei is a multi-facetted expression which is not readily
translated to other languages. This is a result of the fact that it is so closely connected
to the Japanese culture.

However, the use of the expression Kansei requires a proper understanding of the
underlying concept in order interpret and translate crucial aspects. This chapter will
give definitions and explanations for the phenomena starting with an example. A few
moments of concentration on the painting below will give a Kansei.

Looking at the painting on the following page may make some people feel the warm
sun on the skin and a soft breeze in the hair. Perhaps there is a butterfly dancing in the
sunlight coming through the tight leaves of the mighty trees? And perhaps there is the
smell of summer and a taste of fresh strawberries.

That is what Kansei is about!

35
 What is the Kansei?

Figure 7: ‘Frukost’ by Carl Larsson.

‘Kansei is an individual’s subjective impression from a certain artefact, environment

or situation using all the senses of sight, hearing, feeling, smell, taste [and the sense of
balance (annotation by the author)] as well as recognition’ (Nagamachi, 2001). Of
course, this might seem irrational but in most cases even your personal Kansei may not
be so unique and it can be grouped, categorised and measured in order to use it for e.g.
product design.

In the following the expression Kansei and its background is illuminated from
different angles finishing with a concluding definition used by the author in this work.

36
 Engineering Emotional Values in Product Design

5.2. The Etymology of Kansei

Japanese language possesses two different types of alphabets (Hiragana, Katakana)
and Kanji characters. The Kanji has its roots in the Chinese culture and each character
expresses a whole meaning. However, a Kanji character usually has multiple meanings
and only the combination and arrangement with other signs creates a definitive
meaning.

The Japanese term ‘Kansei’ consists typically of two different Kanji- signs ‘Kan’ and
‘Sei’, which in combination means sensitivity or sensibility (Lee et al., 2002,
Nagasawa, 2002a). But first in a technical or psychological context, the expression
receives its final meaning which is used here. Etymologically the term Kansei splits
into two signs of which the second sign derivates from two other signs (compare
Figure 8 below)

Figure 8: Etymology of Kansei (Lee et al., 2002).

5.3. Dictionary Definition

According to the ‘Dainihon Japanse Dictonary’, Kansei apparently is the abbreviation
of the word ‘kanjusei’ which means translated to English ‘sensitivity’. Moreover the
dictionary also provides a more accurate definition:

Kansei: Sensitivity of a sensory organ where sensation or perception takes place in

answer to stimuli from the external world.

In this context Nagasawa (2002a) states that according to his understanding there is an
enhanced meaning beyond the pure word of sensitivity; Kansei also includes a part of
sentiment in the meaning as in classical literature of the old days. He quotes an old

37
 What is the Kansei?

Japanese source from Ukiyozoshi entitled ‘Ten-Inch-Diameter Mirror of

Homoerotism’, where Kansei adopts the meaning of ‘feeling the core’.

5.4. Philosophical Definition

According to Lee et al. (2002) the problem field of ‘Kansei’ was originally addressed
by the German philosopher Baumgarten. In 1742 Baumgarten gave an initial lecture
on aesthetics which was followed up eight years later when he published his
unfinished Latin book ‘Aesthetica’ in 1750, in which he introduced the philosophical
theoretical framework and expression of Aesthetics (Schweizer, 1973). Baumgarten
did not only want to give a theory of fine arts, but also intended to enhance philosophy
with a ‘Wissenschaft der sinnlichen Erkenntis (cognitio sensitiva)’ (Science of sensual
awareness) (Baumgarten, 1961). He argued in favour of two aspects in Aesthetics: the
sensitive-theoretical and the more common artistic-rhetorical, i.e. the receptory and the
productive meaning of Aesthetics.

Even another famous German philosopher, Immanuel Kant, picked up this topic about
20 years later when he wrote his main work ‘Die Kritik der reinen Vernunft’ (Kant,
2004). When the manuscript was translated of Teiyu Amano in 1921, the word Kansei
was used as translation for the German expression ‘Sinnlichkeit (cognitio sensitiva)’
(Nagasawa, 2002a). As a German, the author immediately understands that the word
Sinnlichkeit, as the expression Kansei, does not have a direct translation in English.
What Kant meant is ‘the perception based on senses in contrast to the abstract
perception of the mind without sensory input’ (Brockhaus Universallexikon, 2000). In
other words ‘Sinnlichkeit’ is what enables a human being to receive an image of an
object (or situation), which can not be grasped by reason. E.g. a sunny warm spring
day can trigger impressions (e.g. happiness) which can not be processed by reason
alone. According to Kant the Kansei/’Sinnlichkeit’ plays an important role when a
human subject comes in contact with the real world (Nagasawa, 2002a). The Kansei
therefore is a basic part of a complex combination and reciprocal effect of cognition
and emotion.

Originally Kansei is a term from the Japanese epistemology, which is a branch of

anglo-saxon philosophy. It is the science of knowledge that tries to find principles and
rules about how knowledge it gained (Brockhaus Universallexikon, 2000). Kansei is
considered to be a ‘passive mental process’ activated by the external world. According
to Nagasawa (2002a) the epistemological approach distinguishes between the concepts
of sensitivity (Kansei), understanding and reason. Human beings receive the outer
world by Kansei; carry out analysis, integration and extraction activities on it by
understanding and systemize and unify it by reason.
38
 Engineering Emotional Values in Product Design

Kansei is seen as sensual intuition providing the material for understanding, and
through this the mental ability of experiencing affective values like feeling emotion
and desire. Nagamura (1991) brings this into a neuro-physiological context. He states
that the Kansei is processed by the right half of the brain, processing analogue and
fuzzy data whereas the understanding possessing logical speculative nature is (reason)
treated by the left brain side, whose specialisation is digital data processing.

5.5. Psychological Definition

The expression Kansei itself did not exist in Japanese Psychology at all until recent
years. It is connected to concepts such as sensation, perception and cognition which
are accepted psychological terms. They belong to a sequential process called the
cognitive continuum. In this context, sensation, perception and cognition are processes
which generally are independent but can be interconnected. Nagamura, (1991)
illustrates this in an example. He says that a light stimulus can be sensed and perceived
without cognition. No cognition is required if somebody is asked whether she sees the
light or not. If one is asked how it looks like even cognition is required and the three
processes work together in the cognitive continuum, previously mentioned.

5.6. Definitions by R&D Kansei Researchers

When dealing with Kansei for practical use the theoretical definitions can be useful in
order to put the basics of the work on solid ground. However, to some extent they
contradict each other and may be imprecise regarding certain practical details, making
Kansei difficult to handle. On the other hand researchers using the concept of Kansei
for different reasons cannot assume that the expression is commonly known.
Consequently, some of them documented their own view on Kansei according to their
purpose and context.

According to Lee et al. (2002) Kansei incorporates the meaning of the words:
sensitivity, sense, sensibility, feeling, aesthetics, emotion, affection and intuition (see
Figure 9 below).

39
 What is the Kansei?

Figure 9: The meaning of Kansei inspired by Lee et al. (2002).

Shimizu et al. (2004) sees Kansei closely related to ‘sophisticated human abilities such
as sensibility, recognition, identification, relationship making and creative action…’.
This definition does not only include the sum of the meanings of words translating
Kansei to English but they see Kansei also as a part of a process binding together these
concepts. Kansei is defined as an internal concept with three basic pillars of taste/
sentiment, Feeling and Emotion. These basics continuously interact with each other by
triggering certain actions. His model is presented in Figure 10.

Figure 10: A model of the Kansei according to Shimizu et al. (2004).

According to the Japanese Society of Kansei Engineering (JSKE) ‘Kansei is the
integrated function of the mind, and various functions exist during receiving and
sending [signals]. Filtering, acquiring information, estimating, recognizing, modelling,
40
 Engineering Emotional Values in Product Design

making relationship, producing, giving information, presenting etc. are the contents of
Kansei’ (JSKE, 2004). In contrast to Shimizu et al. (2004) Kansei is not only an
internal process but a process in constant contact with the outer world receiving
external information, processing it and reflecting it back to the outer world.

Figure 11: The Kansei according to Japanese Society of Kansei Engineering (JSKE,
2004).
According to Nagamachi (2001) Kansei is ‘the impression somebody gets from a
certain artefact, environment or situation using all her senses of sight, hearing, feeling,
smell, taste [and sense of balance (annotation by the author)] as well as their
recognition’. Due to his psychological background he sees sensation, perception and
cognition as separate processes, but Kansei does only exist if they are cooperating.

Nagasawa (2002a) has a similar point of view. Like Nagamachi, he prefers the
psychological view, and combines it with a part interacting with the outside world. He
emphasises the connection between reason and Kansei by making a parallel process
model. In his model the incoming stimuli is simultaneously processed by Kansei and
reason which results in an occurring sentiment and an outside reaction. He also gives
an example which is presented in the following:

1. Light of 650nm wavelength from a red flower reaches the eye

2. An image is built up on the epiploon and photo-receptor cells are stimulated
3. Physiological information is transferred to the brain
4. The perceived information is compared with knowledge data from past
experiences and recognised as a red flower
5. At the same time a sentiment or emotion like ‘beautiful’ or ‘passionate’ are
activated by the image.

41
 What is the Kansei?

6. The sentiment occurring in the mind is expressed by words, patterns,

behaviours, facial expression etc.
(Nagasawa, 2002a)

In contrast to other models presented this model is the most comprehensive, since
Nagasawa considers the philosophical approaches as ‘rather difficult to understand’.
Furthermore, he states that engineering staff tend to find the concept of Kansei
confusing and an opposite to reason and intellect. Hence, he offers them a model
fulfilling their expectations by using engineering terms for explaining a psychological
phenomena.

5.7. Kansei vs. Chisei

Emotion and feeling have traditionally found its counterpart in reason. When
Baumgarten wrote his ‘Aesthetica’ (Baumgarten, 1961), one of his aims was to create
an opposite pole to the field of logic and ratio described in Aristoteles’ ‘Organon’
(Schweizer, 1973). Also Kant realised the short comings of ‘pure reason’ and the
necessity of creating a counter part (Kant, 2004). In the field of physiology Damasio
(1996) could prove that emotion and reason in fact are hard wired in the human brain.

In practice Picard (1997) built interaction models on reason and emotion. On another
area, in marketing experts distinguish between ‘feeling appeal’, i.e. commercial
applying to ‘emotional, subjective impressions of product features’ and ‘thinking
appeal’, i.e. commercial applying to ‘logical, objectively verifiable product features
(Liu and Stout, 1987).

Lee et al. (2002) refers to this area as ‘Kansei’ and its counterpart ‘Chisei’. Whereas
the concept of Kansei is closely connected to affective, emotional values of human
beings, Chisei ‘works to increase the knowledge or understanding which is matured by
verbal descriptions of logical facts’ (Lee et al., 2002). Both have in common that they
are triggered by a sensory input, which is mapped from both perspectives. The Kansei
then builds affection, feelings and emotions, which in turn lead to creativity; the Chisei
or reasoning builds logics, recognition and understanding which then become
knowledge. Figure 12 displays this.

42
 Engineering Emotional Values in Product Design

Figure 12: Model of the Kansei/Chisei.

In order to be able to cope with the term Kansei the expression has to be defined
according to the author’s point of view, based on existing models presented before.
Summarising these definitions, the following model can be built as seen in Figure 13.
A certain sensory input from at least one of the senses is mapped and build the Kansei.
Outgoing from this concept, subjective values like affection, feeling, emotion and
intuition are built up which in turn trigger human creativity.

Figure 13: Model of the Kansei as it is used in this thesis.

5.8. A Hierarchy of the Kansei

As explained in the previous sections a specific Kansei arises when a human being is
subjected to an artefact in a certain environmental context (Osborn, 1953). A walk
through the forest lying in the first warm spring sun can e.g. evoke Kanseis such as
beautiful, nice, astonishing, exiting, etc. However these Kanseis are just a small part of
the overall experience. One might feel glad about that the winter is over and that the
trees are in leaves once more. One might also feel enjoyment about the situation and
look forward to the future. This all might end up in a big relief and an unspecified
feeling of happiness.

This ‘situational complexity’ must be handled and the strategy chosen is to apply a
hierarchic thinking. Certain simple Kanseis create higher Kanseis which in turn build
up a general Kansei (Figure 14). The first degree Kansei is a collection of many

43
 What is the Kansei?

transitory Kanseis, which appear spontaneously and build a higher degree Kansei.
Many of these higher degree Kanseis can then be summarised in only one (or a few)
general Kansei. In contrast to the higher degree Kansei the lower degree Kansei arises
instantly where as higher degree Kansei take some time to be build up.

Figure 14: The hierarchy of the Kansei.

Mori (1998) became aware of this fact. He states that the lower degree Kanseis often
are more individual than higher degree Kanseis. This means that e.g. many people
have a feeling of happiness (general Kansei) when walking through the forest on a
warm spring day, but not all do necessary experience it as beautiful (low degree
Kansei). Mori (1998) therefore says that the hierarchical structure of the Kansei must
be mapped clearly in order to understand the unconscious processes (Mori, 1998).

A very common general Kansei for Japanese people is the Kansei ‘high grade’. It
stands for both quality aspects and aspects regarding the social status of the owner. In
Europe research has not been conducted long enough in order to give a final solution
about European general Kansei. However, an indication is that in many studies the
expression ‘Quality’ in the sense of durable and value for money often occurs as a
higher-level Kansei.

44
 Engineering Emotional Values in Product Design

5.9. Measuring the Kansei

The Kansei is an individual mind structure. Understanding others’ Kansei is not
always easy and involves a large amount of empathy and experience. Hence, it might
be advantageous to find rules and methods for quantifying the Kansei structure in
order to make individuals’ Kansei reasonably comparable.

However, the question arising is how the individual Kansei can be grasped and
converted into information useful for product development. Kansei is an internal
sensation, but at present it can only be measured using methods based on
externalisation. Therefore, a series of standard measurement methods has been
developed, interpreting:

• People’s behaviours and actions

• Words (spoken)
• Facial and body expressions
• Physiological responses (e. g. Heart Rate, EMG, EEG)
(Nagamachi, 2001)

Nagasawa agrees with Nagamachi regarding this categorisation. He structures

Nagamachi’s model and distinguishes between two principally different measures.

1. Physiological measures, which are measures of physiological responses,

behaviours, and body expressions (compare Nagamachi (2001)) generated by
‘external stimulation’
2. Psychological measures, where he mainly refers to the Semantic Differential
Scales Method (SD method) (compare Appendix B)
(Nagasawa, 2002a)

All of the previously mentioned methods have been applied successfully on specified
areas of the Kansei. However, feelings and impressions are complex structures, which
require sensitive measuring instruments. Unfortunately, even the most powerful of
those measurement methods is not able to reveal the complete Kansei of someone, but
only minor parts of it. This makes it necessary to carefully specify the interesting part
of the Kansei and choose an appropriate tool for conducting the mapping.

45
 What is the Kansei?

Figure 15: The excerpt from Kansei measured by different methods.

Because of the delicate structure of Kansei, which often includes small nuances of
emotional impressions, mapping methods must be sensitive enough to show these
shadings. Methods meeting these demands are preferably based on semantic
descriptions, whereas methods measuring physiological responses (EEG, EMG, Heart
Rate) are often regarded as too inexact, and methods interpreting facial expressions
can often only be used in combination with semantic description methods. Nagasawa
(2002a) also warns for over-interpreting the result from physiological methods of
assessment. He states that ‘autonomic nerve reflections are not the Kansei itself, but
only correspond to the Kansei’. This makes physiological measuring methods to an
indirect measuring method (Nagasawa, 2002a).

Measurements of semantic exploration and description of the Kansei are foremost

interview techniques as Means End Analysis (Reynolds and Olson, 2001) (see
Appendix C), etc. Another approach is Semantic Differential Scales Method by
Osgood (1957). It is based on subjective estimations of concepts which are ranked
against each other (Osgood et al., 1957). Nagasawa (2002a) admits that this method
cannot be objective due to the arbitrariness of the responses. However, for the
purposes of positioning and identifying a rough picture of the Kansei, it is more
suitable than evaluating data gathered by physiological measurement methods.

46
 Engineering Emotional Values in Product Design

5.10. Kansei and Product Development

Design and development of new products and product concepts has always been a
challenge. Internationalisation, technological development and economical pressure
contribute to an increasing competition in practice on all international markets. An
increased number of products available, sometimes in combination with decreased
purchasing power of the customers force companies to re-consider their product
development strategies (Shimizu et al., 2004).

Products, which have been newcomers not long ago (e.g. mobile phones, handheld
computers etc.) are now becoming mature products and sales are not increasing as
before. Quick model changes, technical updates or price reductions in order to improve
the turnover can not be sufficient solutions.

Due to the new situation, customers’ demands and expectations change. An increasing
number of people want to express their individuality. Even mass-produced-products
have to be adaptable to individual demands regarding form, design and function
(Shimizu et al., 2004).

A long period of ergonomic development is followed by the pursuit of other sectors

like cognitive ergonomic design in order to support customers’ ability to understand
technical gadgets in an easier way. E.g. new electronic products are preferred to be
small, with many functions and easy intuitive access to them. Car stereo and
navigation systems require operation with as little visual attention as possible and
truck manufacturers seek for ways to increase the efficiency of their vehicles by
designing more ‘feeling’ into the controls.

In general products are expected to be of high quality. However, they are also
supposed to express high quality by their design and communicate this to the
customers.

Eventually many customers make their final decision unconsciously based on rather
subjective factors. They purchase the product, which ‘feels’ better, and are often
unable to explain why. Taking this ‘feeling’ into account already in the design process
can give a substantial selling advantage (Söderlund, 2003).

It is not easy to decide what property of the product evokes a certain Kansei and how
the Kansei is influenced when this property is changed. Moreover, the Kansei depends
in many cases not only on one product property, but on the composition and balance
between them. The decision process, whether a bit of chocolate is good or not or if the
47
 What is the Kansei?

new car has good quality is made unconsciously. Only the result of such a thinking
process can be expressed in words.

Concluding, it would be advantageous if the Kansei could be used for analysing

unexpressed and unconscious needs of customers and to develop such needs into an
‘affective’ specification list. Such a list would contain valuable information about how
product properties or combinations of properties are perceived by the customer group
examined, and the information will be used for drawing conclusions about trends in
future products.

48
 6.
Kansei Engineering
感性工学

This chapter gives an overview of the history of Kansei Engineering as a product design
methodology. Also perspectives by different researchers are given. Finally the theoretical
principles and different types of Kansei Engineering are presented.

6.1. History of Kansei Engineering

Mazda Motor Company manager K. Yamamoto used the term Kansei Engineering for
the first time when he delivered a speech at Michigan University in 1986 (Yamamoto,
1992). Since then the term has been used by many researchers within the area.
However, the research field is much older. Before Kansei Engineering was coined the
term Sensory or Sensitivity Engineering was used (JSKE, 2004). Already in the
1970ies Nagamachi presented an affective product development method which he
called ‘Emotional Engineering’..

Today Kansei Engineering is a inter-disciplinary product design methodology that

extends over the humanities, social sciences and natural sciences. It integrates
affective elements in products already in the development phase.

Nagamachi was a researcher pioneering the development of Kansei Engineering in an

academic context. His approach was to develop Kansei Engineering as an ‘ergonomic
consumer-oriented technology for new product development aiming at implementation
of a consumer’s demand in the product. He defined Kansei Engineering as a ‘…
technique for translating the human Kansei into product design elements’ (Nagamachi,
1989). The term Kansei used in this context is narrower than the original meaning.
Nagamachi states: ‘Kansei is a Japanese word which implies a customer’s
psychological feeling and image regarding a new product (Nagamachi, 1997a).

49
 Kansei Engineering

6.2. Perspectives on Kansei Engineering

Kansei Engineering is first and foremost a product development methodology, which
translates customer’s impressions, feelings and demands on existing products or
concepts to design solutions and concrete design parameters (Nagamachi, 1989,
Nagamachi, 1994a). Secondly, it shows how Kansei is translated into design (Schütte,
2002).

Figure 16 presents the basic idea: The Kansei is measured and inserted into a system
which gives recommendations for design solutions. The system is here shown as a
computer system, but in simpler terms it can also be presented as a mathematical
regression model.

Figure 16: Principle of a Kansei Engineering System adapted from Nagamachi

(1989).
According to Nagamachi (2001) there are three focal points of Kansei Engineering:
• How to accurately understand customer Kansei
• How to reflect and translate Kansei understanding into product design
• How to create a system and organisation for Kansei orientated design

The Japanese Society of Kansei Engineering sees in Kansei Engineering a web, which
‘spins and weaves every thing (from an atom to spirit)’. They see in Kansei
Engineering a link between society and its people with varying cultural backgrounds,
and from this resulting demands on products. Figure 17 portrays JSKE’s visualisation
of Kansei Engineering.

50
 Engineering Emotional Values in Product Design

Figure 17: The structure of Kansei Engineering is a network according to JSKE

(2004).
Shimizu et al. (2004) agrees with the JSKE, regarding the fact that Kansei Engineering
is a network embracing all areas of society. Moreover, the authors, imply that Kansei
Engineering is not only applicable for product development, but for all areas of the
human existence. In order to illustrate this they roughly divide the field of Kansei
Engineering into three parts: (1) Sensory Engineering which is a product design
support evaluating ‘short term emotions’, (2) Kansei product technology, focusing on
‘medium term reactions’ and (3) Kansei culture, society and philosophy focusing on
‘taste and sentiment over the long term’. For increased understanding Shimizu et al.
(2004) developed a model resembling Maslow’s ladder. The degree of involvement of
Kansei thinking is indicated by a new step in this model (see Figure 18).

Figure 18: The region of Kansei Engineering (Shimizu et al., 2004).

51
 Kansei Engineering

6.3. The Principle of Kansei Engineering

Different functional models have been drawn on Kansei Engineering. The most
common type of model focuses on practical process mapping of grasping and
translating the Kansei into products properties, as described before. Nothing has been
written about how Kansei works ‘inside’ the human being. Outgoing from the model
in Figure 12 the author tries to give a explanation about what happen inside the human
brain, what the outcome is from an psychological point of view and how reliable the
methodology is holistically. As described before, a sensory input from one or several
of the senses of hearing, sight, smell, taste and touch leads to the building of Kansei.
Humans also have a sense of balance and proprioception. At the same time a Chisei is
built up by the same input, building knowledge through learning processes.

Kansei Engineering utilises certain stimuli (usually product samples and describing
words) which are fed into the system. The output from the system is recorded usually
in a questionnaire form. This data constitute in fact a representation of the people’s
Kansei on the product.

Figure 19: The working principle of Kansei Engineering.

Consequently, using Kansei Engineering techniques does not necessarily mean that the
outcome of the study reflects the true Kansei somebody has about a certain product. It
is just a picture taken at a certain point of time under certain circumstances. As with
every picture the mapping of the persons Kansei is flatter, less contrasted and more
static than the original. However, it could be seen that the greater the number of
respondents is, the better is the representation of the Kansei. Anyway, a certain caution
is necessary dealing with Kansei Engineering data.
52
 Engineering Emotional Values in Product Design

6.4. Types of Kansei Engineering

As seen previously, today’s Kansei Engineering is widely spread into many areas of
society. Many researchers have worked in this area and made their contribution to the
development of Kansei Engineering. Moreover, Kansei Engineering is expanding in
many new areas including new innovative tools which are added to the original
methodology. Kansei Engineering as a unitary methodology has grown more complex
due to the plurality of its applications.

Nagamachi (1997a) collected all these applications on Kansei Engineering and

grouped them according to the tools included and task areas. From these groups he
identified so called types of Kansei Engineering. At present six different types of
Kansei Engineering exist:

• Kansei Engineering Type I- Category Classification

In Kansei Engineering Type I, a product strategy and a market segment is
identified and developed into a tree structure identifying the customer’s
affective needs. These affective needs or Kanseis are then connected manually
to product properties.
• Kansei Engineering Type II- Kansei Engineering System
Kansei Engineering Type II is often a computer aided system using interference
engines and Kansei databases (Nagamachi, 2001). The connections between
Kansei and product properties are made using mathematical statistical tools.
• Kansei Engineering Type III- Hybrid Kansei Engineering System
Kansei Engineering Type III is also a computer database system similar to the
second type. However, it can not only suggest suitable product properties from
an intended Kansei, but also predict the Kansei that product properties elicit,
e.g. a using prototype or mock-up.
• Kansei Engineering Type IV- Kansei Engineering Modelling
The forth type of Kansei Engineering focuses on building mathematical
prediction models. These models are more strongly validated as in the Types II
and III.
• Kansei Engineering Type V- Virtual Kansei Engineering
Kansei Engineering Type V integrates Virtual Reality (VR) techniques with
standard data collection systems. This type replaces the presentation of real
products with VR representations.

53
 Kansei Engineering

• Kansei Engineering Type VI- Collaborative Kansei Engineering Designing

In Kansei Engineering Type VI, the Kansei database is accessible via Internet.
Such design supports group work and concurrent engineering.
(Nagamachi, 1997a, Nagamachi, 2001)

These types of Kansei Engineering are presented in more detail in Appendix E.

54
 7.
A Proposed Model
提案するモデル

In this chapter a general model on Kansei Engineering methodology is proposed. The content
of a chosen product domain is mapped from both a semantic and a physical perspective. Each
Kansei can then be linked to corresponding physical properties. After a validation, prediction
models can be built. Suitable tools are also given in this chapter.

7.1. A Proposed Structure of Kansei Engineering

Even if the procedure of Kansei Engineering at a first glance seems to be strongly
dependent on the individual research context, there are in fact similarities in the
procedures and the tools used for evaluation. Based on a literature review, the author
proposed a framework on Kansei Engineering methodology. Figure 20 portrays this
framework.

Based on an earlier chosen domain the idea behind the product can be described from
two different perspectives: The semantic description; and the description of product
properties.

These two descriptions each span a kind of vector space. Subsequently these spaces
are analysed in relation to each other in the synthesis phase indicating which of the
product properties evokes which semantic impact. After these steps have been carried
out, is it possible to conduct a validity test, including several types of post-hoc
analyses. As a result of this step, the two vector spaces are updated and the synthesis
step is run again. When the results from this iteration process appear satisfactory, a
model can be built describing how the Semantic Space and the Space of Properties are
associated.

55
 A Proposed Model on Kansei Engineering

Figure 20: A proposed Model on Kansei Engineering.

7.2. Choosing the Domain

Choosing the domain includes the selection of a target group, market-niche, and
specification of the new product. Based on this information, product samples are
collected, representing the domain. The Kansei Domain can be understood as the ideal
concept behind a certain product.

Despite the fact that a circle can never be drawn perfectly round, everybody knows
what the perfect idea of a circle is. The Kansei domain is dealt with in the same way. It
is an abstract super-ordinate mind structure while the representative products are either
tangibles or intangible samples from this domain. As a result, a domain includes both
existing products, concepts and even still unknown design solutions.

The task in this first step is to define the domain and find representatives (products,
drawings, samples, etc) covering an as big as possible part of the domain.

56
 Engineering Emotional Values in Product Design

7.3. Spanning the Semantic Space

The Kansei is hierarchic. This means that one higher-level Kansei join together several
lower-level Kansei and facilitates in this way the representation of the customers’
affective values (compare Section 5.8). For example the low-level Kanseis of ‘slow’,
‘fast’, ‘indolent’, ‘agile’, ‘quick’ and ‘speedy’ can be summed up to a single higher-
level Kansei ‘kinetic’.

In Kansei Engineering only higher level Kanseis are connected to product properties in
the synthesis phase in order to achieve a better generalisation of the results. Spanning
the Semantic Space identifies these higher level Kanseis from a great number of
semantic expressions. Although the expression ‘Semantic Space’ origins from Osgood
(1957), more methods than his Semantic Differential Scales are available nowadays.

7.3.1. The Procedure of Spanning the Semantic Space

For practical reasons spanning the Semantic Space has been subdivided into three
steps as presented in Figure 21. Using the desired domain as a starting point, low level
Kanseis also called ‘Kansei Words’ are collected, describing the considered product
semantically (compare Section 7.3.2). In ‘Kansei Structure Identification’ higher-level
Kanseis are then identified from this set using a number of possible tools (compare
Section 7.3.3). In a Kansei Engineering literature these higher-level Kansei sometimes
are also referred to as ‘Kansei Engineering Words’. Finally, the data is compiled in a
standardised way in order to facilitate the following synthesis phase. If important
Kansei Words are missed in this step, the result may have significantly limited
validity. Hence, it is better to select a few more words than necessary.

Figure 21: Procedure of Spanning the Semantic Space (see Paper D, Appendix A).

57
 A Proposed Model on Kansei Engineering

7.3.2. Collection of Kansei Words

A Kansei Word is a word describing the product domain. Often these words are
adjectives but other grammatical forms are possible. For example when describing the
domain ‘fork lift truck’, adjectives like effective, robust, quick, etc. but also verbs and
nouns such as ‘accelerate’/ ‘acceleration’ can occur (Schütte and Eklund, 2001). In
order to get a complete selection of words all available sources have to be used, even if
the words emerging seem to be similar or the same. Suitable sources can be:

• Magazines
• Pertinent Literature
• Manuals
• Experts
• Experienced Users
• Relating Kansei Studies
• Ideas, visions

An important point is to translate ideas and visions into Kansei Words because non-
existing solutions should also be considered. In this way Kansei Engineering can be
used as a creative product development tool, which generates innovative solutions.
The task is to describe the domain, not the existing products. Depending on the domain
considered, the number of existing Kansei Word generally varies between 50 and 600
words (Nagamachi, 1997a). Since it is of great importance to cover the whole Kansei,
the word collection is continued until no new words occur. The data gathered will
critically influence the validity of the results if important words are missing.

7.3.3. Tools for Semantic Structure Identification

For the identification of the Kansei structure, different methods are developed, tested
and made available for use. These methods can roughly be divided into two different
groups:

1. Manual Expert Methods

2. Statistical Methods

Manual Expert Methods

Expert groups often prefer manual methods. The Kanseis are grouped and summarised
according to the participant groups preferences and needs. Tools supporting this are:

58
 Engineering Emotional Values in Product Design

• Affinity Diagram (Bergman and Klefsjö, 1994) (see Appendix B)

• Designers choice
• Interview techniques

Statistical Methods
The major disadvantage of manual methods is that experts can fail. An alternative is to
ask the users of the products about their Kansei and what they consider to be the
important Kansei. Typically this is done by a questionnaire given to a customer group.
Using statistical methods to evaluate of the gathered material quantifies the affinity
between the different Kanseis. The following statistical methods are available today:

• Principal Component Analysis (Osgood et al., 1969) (see Appendix B)

• Factor Analysis (Osgood and Suci, 1969) (see Appendix B, C)
• Cluster Analysis (Hair et al., 1995) (see Appendix B)
• Quantification Theory Types II, III, IV (Tsuchiya, 2004)
• Neural Networks (Ishihara et al., 1996)

7.4. Spanning the Space of Properties

7.4.1. The Space of Properties as a Counterpart of the Semantic

Space
As shown in Figure 20 the product domain is described both from a semantic
perspective and a physical perspective. Both perspectives are presented in the form of
vector spaces.

However there are significant differences in the theoretical background of the two
spaces. Whereas the semantic descriptions possess a well researched theoretical
background based on e.g. Osgood’s Semantic Differential Technique (Osgood and
Suci, 1969), there is no similar theory for the Space of Product Properties. Hence,
there is no consistent way of developing the Space of Properties. At the same time, the
selection of properties is essential (Nagasawa, 2002b). However, few studies really
evaluate the affective impact and the importance of the product properties on the user.
Often they are assumed to be relevant, given by the client company, or even chosen
randomly. In the majority of cases however, the product properties are chosen due to
the feasibility of producing product examples to present in the study (Kanda, 2002).

How can it be ensured that the properties chosen are really relevant to the user/
customer in the examined context? What happens if a trait chosen for selection is not
important to the user? To illustrate, in a study the participants are asked to make a
59
 A Proposed Model on Kansei Engineering

statement on the quality impression of a postal service. The samples differ in many of
their properties and especially the delivery time and ability to track the batch may be
of importance. If these properties are not chosen for evaluation the final result will not
give a correct answer. Worse, it will not be possible to determine that there is a
property missing. Consequently, it is necessary to rate the importance of the different
product properties and make this a criterion for the selection.

There are certainly methods capable of making an adequate selection of product

properties for Kansei Engineering but the problem is that they are not structured, nor
generally tested for this task.

This section will make an attempt to assemble methods for different studies, but also
use methods from other areas used for similar purposes. Probably the most important
demand is to provide a structured approach for constructing the Space of Properties.

7.4.2. Proposing a Model for Spanning the Space of Properties

The systematic collection of Kansei Engineering Properties i.e. properties usable for a
Kansei Engineering study, follows the model of the collection and selection of Kansei
Engineering Words. It roughly can be subdivided into three steps as shown in Figure
22. In the collection step inspirational material regarding a product domain is collected
from a variety of sources and potential properties are identified. In a second step they
are sorted according to certain rules. The number of properties is narrowed by
selecting the most important properties. Properties passing on to the following step
possess only the highest affective impact for further evaluation.

Finally, example products are found possessing those properties chosen and
representing in this way the space of properties. Depending on the method used for
relationship identification the assembly of products can vary. In conformity with the
building of the Semantic Space, the raw data is collected from different sources.

60
 Engineering Emotional Values in Product Design

Figure 22: Spanning the Space of Properties (see Paper E in Appendix A).
Typically, existing products provide a wide variety of potential properties, which can
be integrated into new products. Getting inspiration from already existing products is
one of the most common ways of identifying relevant properties. As seen in the left-
hand column of Figure 22, sources for collection are usually found in literature,
technical datasheets, magazines etc. For identification often an assembly list of
properties is sufficient. The determination of importance and selection of properties
with the highest importance and affective value is preferably done by customer
representatives. To facilitate the work of gathering raw data tools such as focus groups
or one-to-one interviews can be used. For determining the importance, e.g. pareto
diagrams (Bergman and Klefsjö, 1994) can be useful.

In almost all Kansei Engineering studies carried out within industrial product
development projects, a central specification must also fit the company image.
Companies therefore tend to integrate characteristic features in their products. The
right column in Figure 22 identifies that coming from the companies existing products
are the product properties which are unique to the company are identified. Together
with company marketing experts, the relative importance of these properties is
determined. Usually the number of image properties is so small that no special tool
needs to be deployed.

61
 A Proposed Model on Kansei Engineering

The central column in Figure 22, however, is the integration of new product concepts.
Kansei Engineering has been criticised for not being innovative. This part displays
how creative thinking and new ideas can be integrated into Kansei Engineering as a
method. As a main source the designers mind is used. A designer can make a sketch, a
mock-up or a prototype of the whole product or parts of it. Thereby s/he creates
potentially new properties, which are apprised and selected by an expert group.

However, Figure 22 also displays that these processes do not necessarily take place
separately and in isolation. In contrary, they influence each other as indicated by the
arrows. The designer might get inspiration from both existing products and company’s
image, which is then developed into a new solution. This new concept in turn might
influence the companies decision making about which product properties to select as
relevant image. Also, new trends identified by the designers may influence the choice
of product properties from existing products.

Finally, all selected properties are brought together to one set of product properties
from which representing products are determined or mocked up to be used in the
following synthesis step. A practical example of the proposed model in Figure 22 is
presented in the appended Paper E in Appendix A.

7.5. Synthesis
In the synthesis step the Semantic Space and the Space of Properties are linked
together as displayed in Figure 23. For every Kansei Word a number of product
properties are found, affecting the Kansei Word. Ishihara et al. (1998) conducted a
study on beer can design. His results showed e.g. that the score of the Kansei Word
‘bitter’ is most affected by the colour of the can and the shape of the logo. In fact a
black colour in combination with a non-oval logo evoked a strong bitter Kansei,
whereas a white can with an oval logo involved the opposite Kansei.

62
 Engineering Emotional Values in Product Design

Figure 23: Synthesis Phase.

The research into establishing these links has been one of the core parts of
Nagamachi’s work with Kansei Engineering in the last few years. At present a number
of different qualitative and quantitative tools are available. Since the incoming data is
stacked in a standardised way every tool can be used, and it is even possible to use
different tools and compare the results afterwards in order to reveal the most suitable
tool.

7.5.1. Relationship Identification

This relationship identification conducted in the synthesis phase is the core of Kansei
Engineering technology. Whereas the identification of the semantic structure is carried
out in different forms even in other contexts as SMB (Küller, 1975) or Semantic
Differential Method (Osgood et al., 1957), the translation of the determined Kansei is
exclusively performed in Kansei Engineering. Due to this fact this part of Kansei
Engineering has been in focus for research since the beginning of its existence.

A number of tools has been developed and are used for this part. Even here the same
categorisation can be made into the three different areas:

1. Manual Methods
2. Statistical Methods
3. Other Methods

Manual Methods
Manual methods for connecting the Kansei and the different product properties are
easy to perform and require comparatively small resources. These are the oldest tools
are preferred by practitioners. One tool is:

• Category identification (KE Type I), (Nagamachi, 1997a, Nagamachi, 2001)

63
 A Proposed Model on Kansei Engineering

Statistical Methods
As in semantic structure identification, statistical methods are used for treating great
amounts of data from questionnaires. The tools used here have to be modified in order
to fit the requirements of Kansei Engineering. Some possible tools for statistical
treatment are:

• Regression Analysis (Ishihara, 2001)

• General Linear Model (Arnold, 2002)
• Quantification Theory Type I (Komazawa and Hayashi, 1976)

Ranking/ Rating Methods

• Generic Algorithm (Nishino et al., 1999)
• Fuzzy Set Theory (Shimizu and Jindo, 1995)
• Rough Set Theory (Nishino et al., 2001, Mori, 2002)

7.6. Model Building and Test of Validity

Finally a mathematical or non-mathematical model is built depending on the synthesis
method chosen. However, before using the model as a prediction model for future
products it has to be validated. At present, not only are validation methods for the
Semantic Differentials available, but there is a need for a more integrative validation
concept.

64
 8.
Research Contributions to Improvement
of Kansei Engineering
感性工学の発展への寄与

This chapter collects the research contribution on the development of the methodology. The
first part deals with topics of how sensory organs support and limit the building of the Kansei.
This affects the way Kansei questionnaires are constructed. Finally, a model on determining
the degree of attractiveness of a product is proposed.

8.1. Sensing the Kansei

As shown in the introduction of this work the Kansei as an internal sensation is closely
related to the external senses. The external senses deliver the input which is needed to
build up a Kansei and react in an appropriate way.

There are differences in the importance of the sensorial input. Sight is considered as
the most important sense. The other senses of hearing, smell, taste and touch are
usually used in a complementary manner. There are models ranking and arranging
them in accordance to the frequency of usage and importance (Penfield and
Rasmussen, 1950).

However, in Kansei Engineering context, these models are not very helpful. In fact, in
many cases they are irrelevant, because the significance of sensory organs can vary
between individuals and between products. For example for determining the quality of
a cup of coffee the taste and smell are most suitable, the haptics of a textile surface is
evaluated by the sense of touch and the sound of a car engine requires hearing as
sensory organ. However, in most studies on Kansei products, all senses are needed; the
degree or importance however depends on the product in question (Nagamachi, 2001).

65
 Research Contributions

8.1.1. Proximity of Interaction and Presentation

As seen, sensory organs play important roles in how the Kansei is achieved. But as
mentioned earlier, the Kansei is individual. So besides the senses involved in this
process; also the personal interest and experience contribute to build a personal
Kansei. A first examination of the object is usually done visually. Depending on the
individual Kansei and interest, people can then determine whether or not to intensify
the examination by adding more sensory input. In other words, if a person spots
something interesting, he might want to not only look on it but also touch and smell
even taste the object.

Concluding, the degree of involvement of sensory organs has a strong effect on the
Kansei. Picard (1997) calls this the ‘affective channel width’. She concludes that
computers affectively interacting with human beings have to provide a couple of
signals, which can be sensed by human beings. In general it can be said that the more
affective signals are given, the more clearly the message can be understood. For
Kansei Engineering this means that, the more senses involved the better the mental
picture (more intensive Kansei).

Other relevant aspects are physical factors such as the time period, the way, the
intensity of interaction, previous knowledge about the product and its features, etc.
Wikström (2002) shows in her study on stoves that the rating on semantic scales
changes significantly when the participants use the stove instead of looking at it.

Further humans perform activities quicker the more often they have been repeated.
People confronted with a new product will take some time to learn about its features.
In a later state they perform faster and start to demand new properties. Cooper and
Reimann (2003) state that the more familiar a user becomes to a computer software,
the more they want to adapt it. The aspects presented above have crucial consequences
for Kansei Engineering studies but in spite of that seldom addressed in English Kansei
Engineering literature.

The Affective Flow

Time and economic resources are crucial in the design of studies. For Kansei
Engineering studies this is especially true when it comes to the time resource, since
product trends can be rather short-lived (compare Section 8.1.5). Consequently, studies
should be designed as quick as possible and as economically as possible, but at the
same time deliver correct results. The goal is to allow the participants an as complete
Kansei as possible, but not to provide a wider affective channel than necessary. For

66
 Engineering Emotional Values in Product Design

example evaluating the sound of closing car door does not make it necessary to
provide the whole car or even a picture of it.

Figure 24: The affective flow insprired by Picard (1997) and Nagamachi (1994b).
Figure 24 illustrates the way the information travels when a Kansei is built. This may
be referred to as the affective flow. A certain product property possesses attributes
which are transferred by different ‘wires’ in the affective channel. This information is
then received by the users senses and transformed into a Kansei. However, this is an
ideal view. In praxis there are obstacles on the way, limiting or even cutting of the
semantic flow. In Figure 24 these obstacles are referred to as Proximity of Presentation
and the Proximity of Interaction, and act as ‘affective windows’ which limit the
affective flow. In Kansei Engineering the ‘affective windows’ must be set in a way
that information necessary for building up a sufficiently complete Kansei passes, but
unnecessary information is blocked.

Proximity of Presentation
Product properties need certain senses to be transferred into a personal Kansei. In
order to sense the Kansei fully, certain affective channels (Picard, 1997) must be used
as shown in Figure 24. This means that the way a product is presented plays an
important role. For example, the Kansei of a piece of chocolate is not satisfactory
transferred by a picture, since the smell and taste are important stimuli which are
suspended in this case and the user can not interact with the product in the way s/he
normally would do. Hence, the ‘affective window’ must allow olfactorial, visual and
flavoural information pass.

In general it can be said that the following two points have to be considered for each
Kansei Engineering study.
• Definition of channels needed in order to give a full Kansei
• Definition of necessary degree of interaction
67
 Research Contributions

These definitions automatically devise the ‘(minimal) Proximity of Presentation’.

Proximity of Interaction
When performing a Kansei Engineering study, the goal must be that all participants
experience the Kansei of the product in question. Otherwise, the impact of the
affective values cannot be measured. The ‘affective window’ must be set in a way that
the information sent by the product is really sensed by the user. Three points can be
identified which have major influence on how well the Kansei is transferred.

• prior experience of the products,

• interest in the products
• degree of interaction

A study carried out on office chairs comes to similar results. In this paper the
expression ‘Proximity of Interaction’ is suggested for describing this phenomenon
(Eklund and Kiviloog, 2003). Experience by the author shows that a high Proximity of
Interaction i.e. participants with good prior experience, high interest and a high degree
of interaction usually deliver better results than other groups.

Paper B in Appendix A presents a study of rocker switches in work vehicles. Different

samples are presented to the respondents separately (instead of fitted in a dashboard)
in order to be evaluated on Kansei Engineering questionnaires. Such limited Proximity
of Presentation narrows the affective channel and the limited opportunity to interact
leads to an incomplete Kansei. However, the achieved Kansei was considered to be
sufficient in order to draw conclusions about the affective impact of rocker switches.

8.1.2. Showroom Appeal

Users interact with products in different ways depending on the experience they have
with it. First the users learn about the product features and how to use them. Finally,
they become experienced users and they might wish to improve or adapt the product,
additionally, depending on the user’s stage, the Kansei changes. Cooper and Reimann
(2003) indicates that novice users, normal users and expert users have fundamentally
different demands on computer software due to their varying experience.
Consequently their impression changes when they advance more. They therefore
suggest that software should be designed for the medium user, and features should be
added such as online help for novices and shortcuts for experts. Another example
comes from Asatekin (1975) who points out that soft seats in furniture or in
automobile use, are experienced as comfortable in short term use, but uncomfortable in
long term use. His conclusion is that seats aimed for long term comfort must not be too
68
 Engineering Emotional Values in Product Design

soft in order to avoid discomfort. Eklund and Kiviloog (2003) call the short term
experience ‘showroom appeal’.

Eklund and Kiviloog (2003) further discuss that manufactures could purposely create
products which appear attractive in the showroom in order to increase sales, but loose
attraction as soon they are used under real long term conditions. Seen from an
economic perspective such behaviour does not create sustainable growth for a
company. It is therefore not a good strategy. If a company wants to satisfy customers
and make them return, they must produce products which keep up with the
expectations the users have on the company and get from the product in the showroom
(Juran et al., 1974).

Performing a Kansei Engineering study necessitates a decision in advance of whether

the solution should support the long or short-term appeal This will influence the
experimental design for the study. Following the ideas of Cooper and Reimann (2003)
the product should generally be designed for long-term use. In order to improve the
showroom appeal the relevant product properties can be designed in a way that reflects
the intended short-term experience using Kansei Engineering.

In Paper C in Appendix A the comparison of affective values of manoeuvring panels

of fork lift trucks are examined. One aspect of the study is the quantification of
affective impressions of the different surfaces. A corresponding Kansei is the Kansei
Word ‘nice to touch’.

Figure 25: Manoeuvring panel for a BT REFLEX; left: old model, right: new
model.
From the study it can be seen that the Kansei ‘nice to tough’ has been improved with
the new model. This is a result of the re-engineered surface structure and improved
integration of softer parts around the steering wheel and at the wrist support. However,
the evaluations are done on brand-new models in a laboratory environment. These

69
 Research Contributions

results will probably change over time since wear and tear have an impact on the
subjective perception.

8.1.3. Indirect Affective Product Design

In general, it is preferable to let potential customers interact intensively with the
product, before making a buying decision. Doing so enables the potential customers to
learn about the product and to get a more complete impression, especially about
product properties which normally are hidden and require some interaction to find out.
Also certain property’s quality does not appear in the first place but only after some
time of interaction, (compare Sections 8.1.1 and 8.1.2).

In reality it is not possible to communicate the Kansei completely. A certain

showroom feeling will remain. In cases where the affective channel width is limited
due to the choice of communication medium, such as Internet or catalogues, it is
particularly important to make efforts to strengthen the Kansei. In all cases the
customers will make their buying decision on incomplete technical and affective data.
First after some time of usage the user will be able to determine whether the product
was a good buy or not.

Kansei Engineering can help to transmit the product qualities the user cannot sense
immediately, using alternative channels. It is possible to bring out certain key-qualities
using more than one channel. The author refers to this phenomenon as indirect
affective design i.e. properties which are not directly or instantly detectable are
presented using other affective channels.

An example can be found in the previously cited study on lift truck manoeuvring
panels (Paper C, Appendix A). The intention of the designers was to design a ‘durable’
truck. This attribute does often not show until the truck has been used for a long time.
However, there are ways of designing parts of the truck in a way that they appear
‘durable’ already in the showroom in order to reflect the ‘hidden’ qualities. The study
showed that the new panel (compare Figure 25) had a significantly lower impression
of ‘robust’ than the old one. This is probably a result of the optical separation of the
steering wheel from the rest of the panel. However, this was intended and at the same
time the Kanseis of ‘ergonomic’, ‘nice-to-touch’ and ‘comfortable’ could be
strengthened.

70
 Engineering Emotional Values in Product Design

8.1.4. Honest Products

As seen, product design can be used in order to express desired Kanseis and suppress
undesired Kansei at the same time. Kansei Engineering makes it is possible to optimise
the appearance of a product in a desired direction, almost independent of its real
qualities. It is therefore possible to give the product the face of properties it does not
possess in reality. It can also be speculated that manufactures could purposely create
products which appear attractive in the showroom, but loose attraction as soon they are
used under real condition. This in order to increase sales (Eklund and Kiviloog, 2003).
This would, however create incongruent impressions of the product. If a company
wants to satisfy customers and make them return, they must produce ‘honest’ products,
i.e. products which keep up with the expectations the customers and users have on the
company and their expectations of the product initially.

An example of products, which sometimes are perceived incongruent to their

appearance, is laminate flooring. Paper E in Appendix A presents the results of a study
on this product type. It could be seen that people partly perceive laminate flooring as a
dishonest product, since it imitates natural structures such as grains in wooden parquet
or structure of ceramic flooring instead of using original materials. This has a
significant negative impact on e.g. the Kansei of ‘natural’ (compare also Lindberg
(2004)).

8.1.5. Trend Sensitivity

Trends and fashion have driven the development of new products since the beginning
of industrialisation. As mentioned there are trend research institutes such as the
Contemporary Trends Institute in London focusing on trends and predicting the
direction in which products evolve. New products and trends change the perspective
on the older products. Changes in lifestyle brought forward by new products will
therefore also change the affective impact products have (Jordan, 2001). In Kansei
Engineering research this has recognised a long time ago. Especially that the validity
of the semantic descriptions is perishable. Also, Kano (1984) poses that the way
anciently attractive product features are perceived changes quickly.

As a consequence, each time a Kansei Engineering study is made, new Kansei Words
and product properties are collected in order to consider this effect. The strength of the
trend sensitivity is also dependent on the product in question. Products with high trend
sensitivity are generally products with short product life cycle time such as fashion
products, whereas products with low trend sensitivity tend to have longer life cycles.

Yun et al. (2001) states that mobile phones are highly trend sensitive. Küller (1991)
draw the conclusion that semantic descriptions for houses have been practically stable
71
 Research Contributions

for more than 20 years (compare Appendix C). In Paper B, Appendix A the data on
Rocker switches were collected in two rounds with 15 months in-between. No
significant difference in the data could be seen, leading to the conclusion that rocker
switches are relatively trend stable. The results of Paper A and Paper C in Appendix
A on the affective impact of warehouse trucks, did not show significant differences
either. Consequently, warehouse trucks are also relatively trend stable products.

8.2. Collecting Data for Kansei Engineering Evaluations

Most Kansei Engineering evaluations involve customer surveys. The main components
of these surveys are questionnaires including different kinds of rating scales. In order
to minimise the loss of information a good explanation, a clear structure and quick
answering opportunities should be provided in order to increase the
comprehensiveness. The quality is dependent on the presentation of the questions.

8.2.1. Scale Types

Very commonly used methods in subjective assessment are rating methods. They are
widely known and provide relatively rich information if compared to other methods
such as ranking methods. Moreover, data from a great number of participants may be
collected with a minimum of resources.

Several different rating methods can be mentioned. Thurstone’s Paired comparison

technique, Thurstone’s ‘Equal-Appearing Intervals’ or Likert’s ‘Summated Rating’
method also known as ‘Likert scale’, are commonly used However these scaling
methods are suitable if one wants to evaluate a number of different entities. It is more
rare that one wishes to assess a group of entities just for one attribute alone. Doing so
the Semantic Differential Scales (SD-scales) are more useful (Guilford, 1971).

Naming the anchors

In order to understand what the subject is supposed to do, the scales are named at the
extremes. These are called the anchors. Choosing the labels can have a crucial impact
on the results. The labels must be easy to understand for the subject and have to refer
to the object of the study.

In Kansei Engineering every Kansei Word is attached to an individual scale. The way
the extremes are handled is done differently by different researchers. This is due to
cultural differences, deviating experiences or for practical reasons.

72
 Engineering Emotional Values in Product Design

Osgood (1957) uses synonym and antonym for spanning the range of rating (compare
Figure 26 (a)). This allows reducing the number of ratings to a minimum since both
words are rated simultaneously. On the other hand, it sometimes is difficult to find
words having exactly the opposite meaning. As an example the word comfort can be
mentioned. It is shown that discomfort has a different content and can therefore not be
used as an opposite on such type of scales (Zhang et al., 1996). Moreover, this type of
data is difficult to handle by Quantification Theory Type I (compare Appendix B).

Nagamachi and many other Japanese researchers use the Kansei Word as an extreme
on the left side of the scale, whereas he adds a ‘not at all’ on the right side
(Nagamachi, 2001) (compare Figure 26 (b). Doing so, liberates from the force to find
opposite meaning and makes the scale easy to understand and quick to complete for
the participant. Disadvantages are then again a skewed distribution. People experience
the scale as un-balanced and see the neutral value more to the left side of the scale.

Hence, the author chose to combine the advantages of both scales at the same time
excluding most of the problems. The Kansei Word is placed on top of the middle of
the scale, while the anchors are labelled as ‘not at all’ and ‘very much’ (compare
Figure 26(c)). This constellation delivers good data distribution. Together with a 7-
point or VAS scale it is one of the most comprehensive solutions for the subjects.
Küller (1975) uses such scales in the Semantic Description of Environment (SMB)
method (compare: Appendix C). Nevertheless, some disadvantages still remain. The
extremes are in many cases considered to be indefinite which in turn means that the
distances are not considered to be completely equal. Even if this effect is slighter than
in the other cases, even this scale must be called an ordinal scale. Another problem is
that subjects which have no opinion or understanding regarding a certain Kansei
Word, feel forced to check ‘somewhere in the middle’ biasing the result. Therefore if
the mean value of a distribution is around the middle value of the scale, it could be
because the word either is meaningless for the object evaluated or subjects did not
understand the word properly.

73
 Research Contributions

Figure 26: Typical scales used for semantic evaluations. Scale (a) is Osgood’s
original SD scale (Osgood et al., 1957), (b) is a scale used by many Japanese Kansei
Engineering researchers (Ishihara, 2001) and (c) is a modified SD scale by (Küller,
1975).

How many Scale Steps?

Osgood (1957) uses 7 point Semantic Differential Scales gathering for evaluation.
Nagamachi and many Japanese researchers use this type of scale (Nagamachi, 2001,
Nishino, 2001). However they use a 5 point scale. Also Sinclair (1990) provides a 5
point SD scale. The advantages of SD-scales are that they are recognisable by the
participants since many of them have prior experience of working with them
(Guilford, 1971).

One problem especially with 5 point-scales is that the type of distribution of the data is
difficult to determine. Moreover, the 5 point scale sometimes is experienced as to
narrow (see Paper A in Appendix A), in particular when a neutral point is located in
the middle. Participants are experiencing the extremes 1 and 5 as overly extreme
statements and the remaining three points are not sufficient for making a proper
estimation (Schütte, 2002).

Therefore it might be better to choose a 7-point scale in those cases. (Küller, 1975)
uses 7-point scales in SMB for similar tasks as in Kansei Engineering. A 7-point scale
allows more sensitive ratings, while it is as comprehensive and quick to use as a 5-
point scale. However, the problems with determination of the data distribution due to
the low number of discrete steps remain.

In medical science, another type of scale is used, the so called Visual-Analogue Scale
(VAS), sometimes also called ‘Quality of Life Scale’ according to its application. It is
basically a 100mm horizontal strip, with extreme statements on both ends (Figure 27
(a)). The participants mark their estimation with a cross on it. Despite the fact that this
scale possesses discrete steps, the sheer number of them (100) makes it appear as
continuous for the participants. It is therefore very sensitive and has no technical
details such as numbers etc. that can confuse the subject. The disadvantage is that it is
74
 Engineering Emotional Values in Product Design

not commonly known and therefore not easily understood by all participants.
Moreover, it is not completely linear (even if this effect is much smaller than in the
other scales presented). This type of scale have been used several times before
(compare Paper A, B and C in Appendix A).

Figure 27: Rating scales used in the experiment in this work. Above a 100mm VAS
scale, below a 7 degree modified SD-scale.

Statistical Treatment
Another aspect in this context is that most statistical treatment methods and in
particular the methods used in Kansei Engineering require data from continuous
scales, i.e. an interval or ratio scale must be diploid (compare Guilford (1971)).
However, the SD-scales used here do in fact deploy an ordinal scale. Consequently,
the SD-data could not be used e.g. for factor analysis as Osgood (1969) does. Also the
data must have a bivariate normal distribution, which e.g. is not the case in the 5 point
SD scale used by many Japanese researchers (Figure 26(b)). Nevertheless, experience
shows that even these ‘dirty’ ordinal scales lead to similar conclusions as data from
interval scales, and Visual Analogue Scales. Paper B in Appendix A also supports that.

8.2.2. Computerised vs. Manual Data Collection

Mass data collection dates back to the period 1930 to 1950 when the increased demand
for data collection in sufficient volume occurred in order to provide statistical
significance (Drury, 1990). Traditionally, questionnaire data collection is performed
on a sheet of paper, which is given to the participants. This type of collection method
is still up to date and because of its simplicity and not requiring much knowledge, is
still the most common method. Nevertheless, with the arrival of computers, new ways
of collecting data occurred. Data can now be collected using special software on
personal computers. Internet questionnaires are a recent advancement of this
technology.

In this work both conventional and computerised data collections were used. The
computerised data collection was performed both using mobile units and Internet

75
 Research Contributions

surveys (compare Papers A in Appendix A). Doing so has many advantages.

Compared to manual data collection, the data can be collected and stored
electronically. An additional data transfer from paper to computer is omitted and
thereby the risk of errors in data transfer is drastically reduced. Moreover, it becomes
easier to transfer the data into other evaluation programs. This reduces the collection
time, allows the collection of more data and thereby improves the statistical strength.
Common commercial software enables the researcher to perform rapid statistical
operations and prepare the data already during the data collection for further
evaluation. Optional functions such as automatic pre-evaluation of the data, makes it
possible to check the data quality in an early phase and result in improved control of
tendencies and suitability.

Other practical advantages are that studies using computerised data collection can be
performed in other places without the physical presence of a researcher. Using the
same software layout for all participants, every participant experiences an almost
identical test environment, which reduces nuisance and increases statistical strength.
The disadvantage is that an electronical device is needed and participants with no or
low computer experience are distracted. Internet surveys might be opened incorrectly
by different web browsers or the access might be prohibited by a firewall as used in
many big companies.

However, in order to overcome these obstacles, Drury (1990) foresees improved

computers, which have better input devices and displays. Together with people
becoming more used to common computer software, he assumes the problems stated
above will soon disappear. But this is not the main advantage in using computers. The
biggest opportunity lays in the future development of software and hardware. Future
data collection will be carried out by intelligent software, which can adapt to the
individual respondent and ask question more naturally (Drury, 1990). Intelligent tools
can evaluate data starting with the first input and interpret the respondents’ behaviour;
doing so the participant can be supported optimally. This means that the gap between
questionnaire and one to one interview can possibly be closed.

8.2.3. Constructing Kansei Engineering Questionnaires

The future visions drawn in the previous chapter are however, not yet existing.
Questionnaires are still relatively inflexible and made in advance. Therefore it is
important to design them in a way that is less disturbing for data quality and allows the
respondent to fill in as much of his/her own opinion as possible.

76
 Engineering Emotional Values in Product Design

The structure of the questionnaire

The structure of the questionnaire depends on different aspects such as purpose,
respondent group, available resources, aims and also the researchers personal style.
However two main types can be differentiated as shown in Figure 28.

Figure 28: Two types of Questionnaires. Left: Semantic Differential style; right:
Likert style.

On the left hand side the traditional Semantic Differential Scales are presented. One
entity is placed above and rated according to a number of Kansei Engineering Words.
This type of scale is the most common scale for both Semantic Differential studies and
Kansei Engineering studies. It allows rating the entities (products) separately
according to the semantic dimensions spanned by the Kansei Engineering Words
(KEW). The example on the right hand side presents a modified version. It evaluates
the same facts, but is built resembling a Likert scale. The KEW is on top and the
products are rated below. This design allows to make a rating at the same time as the
products are ranked.

Both types and several hybrids have been used for this work. Despite from the fact that
no special examination on the suitability of the two structure types has been done, no
problem with the resulting data which could be related to the structure design, was
detected. However, careful consideration and adaptation to the purpose was always
carried out before.

77
 Research Contributions

Randomising Questionnaires
The Questionnaire design will almost automatically influence the way the respondent
thinks. It is certainly positive to guide the respondents’ thoughts in a certain way but
too obvious guidance leads the customer in the wrong direction. For example rating
the Kansei Engineering word ‘easy’ before the word ‘precision’ could result in a better
rating for the second word than as if it was given separately or in context with other
words.

Guilford (1971) suggests randomising the order of appearance of the questions in order
to avoid such problems. Randomising the Kansei Engineering Words helps to spread
out such effects and treat the fluctuations using statistics. For most of the studies in
this work, the order of appearance of both concepts and Kansei Engineering Words
was randomised. Moreover, computerised data collection made it possible to
randomise the data and present the rating scales separately to the participant. Although
the effect of randomisation on the result was not explicitly measured, it is obvious that
randomized questionnaires produce less biased data than non-randomised ones.

Language in Questionnaires
Some studies included in this work were carried out either in different countries or
with foreign students. The problem occurring was that the terms and expressions had
to translated to be the respective language.

Translating the Kansei Engineering Words is a special problem, since translations in

many cases are not completely accurate. For example, the word design is used in both
English speaking countries, Germany and Sweden. However, a native English speaker
links the expression both to engineering design and artistic design. In Sweden and
Germany the word is almost exclusively used for artistic design. So if the word is used
in both contexts the result will be biased due to the different understandings. This
example is quite obvious and easily fixed by adding a short term of explanation, but
native German or Swedish words are not that easily translated. Unfortunately the
author could not find a final satisfying solution for this problem. Even many Japanese
researchers observed this problem without solving it (compare Lee et al. (2002)).

8.3. Degree of Importance

The problem all affective engineering methods have in common is that not all aspects
of a products affective value can be accounted for. For Kansei Engineering this means
that not all semantic descriptions can be evaluated as Kansei Words and not all product
properties can be examined regarding their affective impact on the user. The reason for
this lays in the sheer quantity of possible aspects and the hard limited processing
78
 Engineering Emotional Values in Product Design

ability of the human mind. Only if the data is prepared into useful information for the
human mind, the meaning of it can be understood and transferred into real knowledge
(compare Section 2.6). In research, facilitated models are built in order to make the
problem more comprehensive.

In the case of Kansei Engineering, building a model means that the semantic
descriptions becoming Kansei Words as well as the product properties to be evaluated
must be prioritised and selected. For this certain factors can play a role:

• The strongest affective impact

The semantic or physical description making the strongest impression on the
user are preferably those to be selected. Which description makes the strongest
affective impact is totally dependent on the context, the research question and
the product type.

• Behaviour
The behaviour of how people interact with the product indicates what properties
receive their intention. Verbal expressions they use while interacting with the
product can be included in the Semantic Space. This type of data can be
gathered using observational methods (compare Section 8.2).

• Experimental design
Another factor is the intention of the experimenter, who sets the rules of how
the participants are supposed to interact.

• Interest and background

The users interest and background in the product plays a major role. In general
it can be said that the more experienced the participants are with this type of
product the more they will focus on the ‘Achilles heel’ of the products.
Moreover they will be able to express themselves in more appropriate terms.

Spanning the Semantic Space according to Osgood et al. (1957), includes the
prioritisation for the Semantic descriptions. This is done by either factor analysis or
manual system such as card systems (Ishihara, 2001) or affinity diagrams (Bergman
and Klefsjö, 1994). The previously mentioned techniques collect possible semantic
expressions, assemble them into groups with words possessing similar meaning and
finally select words covering the most important part of the Kansei (see Figure 15).

79
 Research Contributions

As pointed out previously the Space of Properties is not well researched. Determining
the degree of importance for a property is not specifically documented in English
literature. However, the literature research within this work gave at least some hints
about how the prioritisation could be performed. Together with the experience the
author gathered from practical work in Kansei Engineering studies, three different
ways could be found:

• Interview studies
Typically these studies are carried out in Focus groups, but also One-to-One
interviews occur. One example is a study carried out by the women’s
underwear manufacturer Wacol, where 3000 women were interviewed in order
to find their preferences (Nagamachi, 1997b).

• Observational studies
Observational studies can aid the finding of the most relevant product
properties by observing how the user interacts with the product in question.
Properties used frequently, are likely to be the most important. However, those
studies are often carried out in association with interview studies. Examples
here are studies on vehicle interior (Ishihara et al., 2001) and an overall car
design (Mazda Miata) (Nagamachi, 2001) where drivers behaviour was
observed. Another study was done on hair-treatment products using data from
audio-visual observation of buyers at the shop shelf (Nagamachi, 2001).

• Expert knowledge
This is the most common type of determination. Experts involved in the product
development process and possessing knowledge about marketing strategies are
useful for this task. As method e.g. affinity diagrams can be used. This has been
done in most of the studies carried out for this work; in several cases in
combination with interview studies. Examples are studies made in close
cooperation with company development departments.

80
 Engineering Emotional Values in Product Design

8.4. Designing Attractive Quality into Products

8.4.1. Connecting the Kano Model to Kansei Engineering

The Kano model (Kano et al., 1984) is used in order to create attractive products. As
seen in Section 1.4.3 different types of quality can be recognized. Successful product
development means that the development team must meet the basic expectations (the
must-be quality) as well as the one-dimensional quality, which correspondents to the
expressed expectations a customer has on the products. However, this is everyday
work for product designers. What is considerably more complex but necessary is the
attractive quality creation in new products. Once an attractive new property is
integrated into the product, the customer is surprised and delighted and is tempted to
purchase the product.

One way of making the degree of attractiveness visible on the base of Kansei
Engineering data was proposed by Schütte (2002) who claims that the overall
impression of a product can be estimated in relation to competing products (see
Equation 1). The Kansei Score for each Kansei Word is added and related to rating of
ideal values. Multiplying by 100 gives the percentage certain product rates against the
imaginary ideal product. The application of this procedure to a number of different
products allows a comparison of the overall impression.

∑ KS Pr oduct i
Equation 1: i =0
n
⋅ 100 = SK [% ]
∑ KS Ideal i
i =0

With:
SK .................................Degree of attractiveness compared to the ideal value
KSProduct i ........................Kansei Score for a certain product regarding a Kansei Word
KSIdeal i ...........................Kansei Score for the ideal product
i......................................Consecutive number for Kansei Words
n.....................................Total number of Kansei Words
(Schütte, 2002)

However, this formula is linear and assumes that not only the absolute affective value
for a semantic impression is measured, but also an ideal value, i.e. a value indication
how the product should be designed.

81
 Research Contributions

Axelsson et al. (2002) used a similar approach in order to tie Kansei Engineering
results to the Kano model. In contrast to Schütte (2002) they decided to approximate
the affective impact with a quadratic function suggested by the Kano model.

Figure 29: Quadratic model of user perception from Axelsson et al. (2002).

As seen in Figure 29, three different products, in this case incontinence protection for
the elderly are rated on a 7-point SD-scales. The result is plotted in a radar diagram
and the area included by the particular products is calculated as a percentage from the
theoretical maximal area (all Kansei Words are rated ‘7’). Since the area function is a
quadratic function the results fit the Kano model. Plotting the results from an
ergonomic rating of the time the caregiver stands flexed, gives the diagram presented
in Figure 29. Finally, the ranges of ‘must-be’-quality, one dimensional quality and
attractive quality are set and related to the feeling, in Axelsson’s model referred as
negative, neutral and positive feeling. Calculating the area included gives a quadratic
function.

8.4.2. Suggesting a new Mathematical Model

Axelsson (2002) and Schütte (2002) fit their mathematical models to the respective
product leading to different results. In Schütte’s case, the model is linear whereas
Axelsson suggests a quadratic model. In reality feelings are certainly neither linear nor
quadratic; the equations used can therefore only be approximations.

Equation 2 suggests a mathematical model which can be seen as an improvement from

both previous models. The Kansei scores are made dimensionless by dividing by the
number of points on the rating scale used and summed up. Later the result is divided
by the number of Kansei Words used and the result is given in percentage. Depending
on which dignity is used different characteristic can be achieved. Since only d=1 and

82
 Engineering Emotional Values in Product Design

d=2 have been tested before, other product types might require different
characteristics. Consequently the power ‘d’ can be chosen freely in order to match the
product type in question.

d
1 n  KS Pr odukt i 
 ∗ 100 = SK [% ]
Equation 2: ∑
n i =0 
I KW
a 


With:
SK .................................Relative degree of attractiveness
KSProduct i ........................Kansei Score for a certain product regarding a Kansei Word
i......................................Consecutive number for Kansei Words
n.....................................Total Number of Kansei Words
d.....................................Dignity
a.....................................Number of points in the rating scale
IKW .................................Degree of Importance for the respective Kansei Word

The model can be used in order to compare different products which have been rated
on e.g. SD-Scales. Products receiving higher SK values are experienced as more
attractive than others. Moreover it is possible to define boundaries for negative, neutral
and attractive feeling according to Kano (1984).

This measurement is quick and comprehensive, however it still might fail if a certain
Kansei Word has extremely low ratings but high affective impact on the user. In such
cases the total rating can be high, but the customer do no really like the product due to
an important affective aspect missing. Hence the model includes the value IKW, which
expresses for the relative importance of the Kansei Word used.

In order to clarify the usage of the proposed mathematical model an example is

presented in Table 2. The data from three rocker switches were selected from the same
data used for Paper B, Appendix A. Moreover, the Kansei Words distinct, durable and
quality and the Importance factors IKW were chosen based on the final results from the
same paper. Frequency analysis with pareto diagrams was made in this case. Finally,
Equation 2 was deployed for both d=1 and d=2.

83
 Research Contributions

Table 2: Data collected for rocker switches.

Avarage Kansei Score
Importance
Kansei Word factor
IKW

Distinct 0,3 2,6 5,5 3,9

Durable 0,7 3,7 4,1 4,9
Quality 0,9 3,7 3,0 4,8
SK Linear (d=1) 31% 34% 42%
SK Quadratic (d=2) 12% 12% 25%

It can be seen that the three switches differ for the three Kansei Words. However, the
first two switches are more similar than the third, which is also reflected in the results.
The order of the switches is the same for both assumed linear and quadratic
correlation. The quadratic assumption distinguishes the switches more than a linear
assumption, i.e. only switches with very high ratings are considered to be attractive.
So, the quadratic approach probably reflects the reality better then the linear approach.

84
 9.
Discussion
議論

9.1. Kansei Engineering–not only a Method for Industrial Design

Many Kansei Engineering studies are made presenting pictures or computer graphics
to the participants. This seems to be a natural and convenient way of carrying out
Kansei Engineering studies. Kansei evaluations that don’t provide real products for
participant evaluation save time and money.

On the other hand doing so strongly restricts the way people can interact with the
product. The number of senses used for experiencing the Kansei is limited to the visual
sense. Consequently the Kansei is incomplete or even misleading (compare Section
8.1.1). In fact, by presenting a picture or a computer model, only the exterior shape
and industrial design is evaluated. This is not always sufficient, since understanding
design in many cases also involves other senses such as the tactile sense. One could
argue that using visual presentations is only a sufficient method for indirect affective
design (compare Section 8.1.3), i.e. certain ‘concealed’ qualities are presented
indirectly by other factors e.g. shape and design. For example the Kanseis of ‘Quality’
and ‘Robustness’ can be expressed by suitable industrial design. This is certainly
correct, but the product must show the intended quality in all its details, which might
require a broader affective channel (see Section 8.1.4).

Evaluating product properties that can be sensed solely from pictures of the products
leads to ‘shallow’ results. Kansei Engineering may also evaluate ‘concealed’ product
properties which are deeply merged with the product itself and give recommendations

85
 Discussion

of how to improve them. An example of this is the driver feeling in vehicles. It is

undeniable that certain cars ‘feel’ better than others on e.g. curvy roads. This depends
on how the engine, the power train, the steering and the wheel suspension are to
designed harmonise. In this process a large number of variables are involved. Kansei
Engineering may identify those with the highest affective impact and give concrete
technical values of how they should be tuned in order to match a certain Kansei, e.g.
‘sporty’, ‘control’, ‘safe’ etc.

9.2. Short- Cuts to Kansei Engineering?

One reason for presenting pictures to the participants and thereby restricting the
affective channel width might depend on limited time and economic resources. Kansei
Engineering studies require time, financial resources and expert knowledge. The
authors experience in different companies was that they found the method of Kansei
Engineering highly interesting, but requested a ‘lighter’ version. One opportunity to
less time demanding Kansei Engineering studies lies in careful preparation. This
includes determination of which properties that have the highest affective impact. The
proposed model of Kansei Engineering methodology in Section 7.1) makes a
contribution to this. From this it can be determined which senses are required and limit
the affective channel by setting the ‘affective windows’ in the proximity of interaction
and the proximity of presentation respectively (compare Section 8.1.1).

This certainly makes the process more effective and saves time and financial
recourses. However, the whole process of collecting words and properties, and
prioritising them before the evaluation on SD-questionnaires and synthesis can be
carried out, must be performed each time. So, expert knowledge is still required. The
question arising is whether it is possible to simplify some work intensive steps such as
the erection of the two spaces of semantics and product properties from e.g. previous
studies. Küller (1975) shows in his SMB method how he collected descriptive words.
Using factor-analysis of data gathered from thousands of participants, he finally found
8 factors relevant for architectural structures. Also for Kansei Engineering similar
attempts has been made erecting databases including Kansei Words. (Nagamachi,
2001). It is also imaginable that similar developments can be done for the Space of
Properties.

Regarding the synthesis phase and the different mathematical statistical tools,
simplifications can surely be done. In fact there are already computer programs on the
market, performing e.g. QT1 analysis according to the ‘black-box’-principle.

86
 Engineering Emotional Values in Product Design

9.3. Qualitative vs. Quantitative Approaches

Scientific theories must be falsifiable i.e. they must be verifiable and disprovable. This
means in consequence that findings must be documented in a suitable way. In the area
of technology and engineering, quantitative measurements is the preferred way of
documenting findings. Mathematical and statistical tools can be applied. In other areas
such as in humanistic science, qualitative measures are common. Since Kansei
Engineering is located in between different research areas using different types of
measurements, confusion has occurred. Traditionally, feelings and emotions, in brief
the Kansei - have been treated by behavioural science, psychology, and sociology etc.
using qualitative measurements. Although engineers realised early that the Kansei is
an important factor in product design they have been unable (and sometimes even
unwilling) to treat these topics successfully in many cases due to the uncommon
measurement type.

The forerunners of Kansei Engineering and other methodologies in affective product

development realised this problem and designed Kansei Engineering as a methodology
bridging the gap between techniques and human science. As a result Kansei
Engineering can not only quantify the formerly ‘soft’ topics of Kansei, but also build a
mathematical statistical correct prediction model of customer behaviour. If wanted
every form of qualitative measurement can be suspended. At a fist glance this seems to
be the ultimate solution. Presenting Kansei in figures in fact increased technician’s
interest in affective methods and promoted the dissemination of Kansei Engineering as
a method in product development. For the first time many engineers had the
impression that they could ‘understand’ feelings.

However, using Kansei prediction models without knowing the limitation may lead to
seriously misleading predictions. As in solving an equation one should have a certain
expectation about the result. When engineers are equipped with a tool such as Kansei
Engineering, it takes time to develop the experience required. It is therefore indicated
to educate engineers better in the background and assumptions of affective product
design. Failures of new product’s affective value prediction may be avoided that way.

9.4. Reductionism vs. Holism

It is characteristic for reductionistic approaches that the reality is reduced to a
(mathematical) model. In most natural sciences this is the preferred approach to
explain phenomena isolated from its context in order to simplify and purify. Among
others engineering science is a strong representative for this view. Holistic approaches
on the other hand can often be found in humanistic science. This philosophy arose as a

87
 Discussion

reaction on reductionism in natural science. It assumes that a system is more than the
sum of its parts. Alternative medicine methods such as practiced in traditional Chinese
medicine are representatives for this philosophy. The symptom is not treated isolated,
but as a consequence from many different causes in the environment. However,
recently holistic approaches have also been discussed in natural science. They are
referred to as anti-reductionistic tendencies (Helm et al., 2004).

Kansei Engineering in its original meaning supports the reductionistic perspective in

natural sciences. Consequently, Kansei Engineering performs a prioritisation of both
semantic descriptions and physical descriptions and builds a model which is only
based on the most important factors for the products in question ignoring factors with
minor impact on the total Kansei. This is probably due to the prevailing paradigm in
the application areas of Kansei Engineering, i.e. product development departments.
One reason that Kansei Engineering appears attractive to engineers might be the fact
that it uses a similar approach in problem solving that most staff in product
development use. One advantage of a reductionistic approach is that Kansei
Engineering can be spread more easily to practitioners in product development
departments. Reducing the number of influencing factors to a minimum, i.e. finding
the most essential semantical and physical descriptions of the product domain,
facilitates the understanding, enhances the knowledge of the product and makes the
affective needs of the customers clearer to the designers (Helm et al., 2004).

Applying Kansei Engineering without proper knowledge of the background and the
prediction models limitations can lead to mistakes which could be avoided using a
anti-reductionistic perspective. Omitting factors which presumably does not have a
measurable effect on the customer Kansei can have its risks. Even small effects can
sum up to a big impact if they all point in the wrong direction. Additionally if they are
not in the prediction model, somebody with a reductionistic perspective in mind might
not find the reason for the deviation. Certain details of the Kansei cannot be caught
using Kansei Engineering. This is a direct consequence of its reductionstic view.
However in humanistic science there might be holistic approaches, which can deal
with the missing parts of the Kansei; even not quantified. Although all studies in this
thesis are designed from a reductionistic perspective the respondents were also
supposed to give an opinion on the total impression of the product in question
(compare Papers A, B, C and E in Appendix A).

9.5. The Words are not enough

The structure of the Kansei is explained in detail in chapter 5. Roughly it can be said
that the Kansei is a subordinated mind structure, which connects exterior stimuli to
88
 Engineering Emotional Values in Product Design

emotions. Most emotions are generated in the brain in an area which is called the
limbic system. From an evolutionary perspective the limbic system is much older than
the rest of the human brain. It includes basic function such as heart and breath
regulation, blood pressure, etc. and controls the way we emotionally react on external
stimuli (Damaiso, 1996). In this way the limbic system protects us e.g. by generating
an emotion of fear in dangerous situations. However, all these processes take place
unconsciously since the consciousness is placed in the ‘newer’ parts of the brain, e.g.
the frontal lobes in the cerebrum. Physiologically the limbic system is ‘hard-wired’ to
the cerebrum but the connections are few. Moreover a larger number of connections
lead from the limbic system to the cerebrum then vice versa. This means that human
beings can control their emotions to a certain degree consciously, but not completely
(Damaiso, 1996). For the Kansei this means that it exists in the form of emotions in the
limbic system. When the Kansei is measured by external methods, such as words, only
the impressions and emotions can be described which reaches the consciousness
(compare Section 6.3). Consequently, the description of the Kansei is incomplete.

According to researchers in the field of Kansei Engineering, there are other tools in
order to measure the Kansei. Nagamachi mentions measurements such as EMG, EEG,
eye movement, heart rate and galvanic skin sensitivity (Nagamachi, 2001) (see also
Section 5.9). As mentioned above these reactions are directly controlled by the limbic
system and reach the measuring apparatus unfiltered by any conscious process.
Unfortunately, the sensitivity of these measures is rather low, which makes it difficult
to use in this context. However, as a complementing additional measure to semantic
descriptions these methods have been proven to be useful (compare Picard (1997)).

So, the dilemma remains. At present it is not possible to measure the Kansei
completely. The most effective way is therefore to give a stimulus to the participants
and measure the change in the measurements we have today as explained in Section
5.9.

9.6. Asian Advantages?

During the research for this thesis the author has viewed many sources of information
on the topic of integrating affective values in products. As mentioned before most
publications on Kansei Engineering originate from Japan and Korea and are published
in the respective languages. This might not be surprising since the method itself was
founded there. However, the research on such methods seems to be largely limited to
these regions. Western researchers noticed the area quite recently around the beginning
of the new millennium, and they are just a few. It can also be seen that companies
from the Far East are most effective to integrate affective techniques into their internal
89
 Discussion

processes as a product philosophy. These are not small companies, but big
conglomerates such as Mazda, Toyota, Sharp, Sony, etc. Examples of Western
companies are fewer and those who work with affective methods, often co-operate
with one of the bigger Japanese conglomerates. Maybe the need for designing feelings
into the products is exaggerated? Maybe there no need for such methods in Western
industry? Maybe Western customer products can still be sold even with the
competition from Asian companies.

The authors experience tells another story. Even if just few Western companies
explicitly think in similar terms, the need is there. In particular the market situation of
many similar products makes it necessary to produce products which ‘feel’ better.
Many decision makers know about this, but they struggle with problems. In particular
translating the emotions into product solutions is vague and often dependent on trial
and error. Many companies therefore rely completely on experienced designers’
intuition. The backup from universities is limited and consultants can seldom offer
tools powerful enough. In Europe just a handful of universities work on these topics
and are still in the very early stages. This becomes even more severe considering that
not until 2004 a project financed by the European Union started in order map the area
of affective product development work (ENGAGE, 2005). So the need of research and
implementation is eminent.

9.7. Implementing Kansei Engineering

In Japan, Kansei Engineering is often considered as an independent product
development philosophy, which typically is carried out in concurrent engineering
processes. However, since the methodology is little known in European industry, it
turned out to be helpful to pinpoint possible entry points for Kansei Engineering
methodology in the company’s individual product development processes. In practice
this means to identify product development methods such as QFD etc., which are
suitable for transferring Kansei Engineering results into conventional product
development.

9.7.1. Integrating Kansei Engineering in Product Development

Processes
The major field of application for Kansei Engineering today lays in industry. As
shown earlier, Kansei Engineering is mostly deployed as a tool for product
development, improvement and innovation creation. Kammerlind and Schütte (2001)
and Antoni and Schütte (2002) therefore focus on existing product development
processes in companies and the integration of Kansei Engineering in them.
90
 Engineering Emotional Values in Product Design

As an practical example Antoni and Schütte (2002) mention BT Industries AB

representing an industrial company in the vehicle sector. Like many other companies
BT Industries AB uses a Stage-Gate model. As a standard product development Stage-
Gate model (Cooper, 1998) can be quoted with his model portrayed in
Figure 30. Within each stage an exactly defined part of the product development
process is carried out and subsequently reviewed by the project steering committee. If
the result is sufficient, the project may pass the gate and continue, otherwise it is sent
back for revision or rework. The Stage-Gate process model allows an identification of
when Kansei Engineering data has been found to be most useful for the product
development process.

The Kansei Engineering methodology has been applied on the product at different
levels and at different stages in the product development process. A macro-level
investigation was used on whole product concepts, whereas micro-level studies are
used for detailed studies on product parts after the concept is specified and follow-up
investigations which give feedback to the earlier stages of the process. These three
types can be recognised in the Stage-Gate model in Figure 30.

Figure 30: A Stage-Gate process for product development, adapted from Cooper
(1998).

During the Preliminary investigation stage a quick investigation can be conducted,

which will result in a large number of potential Kansei Engineering projects. This is
based largely on desk research and therefore inexpensive. Kansei Engineering data
from rough macro-level studies on different competing products can be a valuable
information source for the pre-selection of product concepts.

91
 Discussion

In the second stage a more detailed investigation is carried out. Typically, market
studies are included and Kansei Engineering can support the forthcoming decisions by
focusing more carefully on selected product parts (micro-level investigations). After
this point the actual development process is started. Kansei Engineering data can even
support this process by making the designers aware of the Kansei their work may
evoke.

In the following testing and investigation stage (follow-up investigations) Kansei

Engineering is able to reveal whether the new product will fulfil the requirements
regarding emotional impact or not. At this point small changes such as colour setting,
tuning parameters or changing minor modules can still be made.

9.7.2. Integrating Kansei Engineering Data in QFD

Many companies use standardised methods for the translation of customer wants and
needs into product and process properties (compare e.g. Gustafsson (1998)). One of
the most common methods is QFD. Since BT Industries uses QFD, especially in the
second stage of their development processes (compare Cooper (1998)) a recurring
question was if and how Kansei Engineering can contribute or be combined with QFD.
A combination of deductive work and reanalysis of empirical studies could spot five
potential approaches. Kansei Engineering data can be used in order to identify
customer needs and determine their importance, facilitating the setting of target values
for technical data and perform benchmarking between different products and brands
and quantify the relationships in the relationship matrix in a more exact way (Figure
31).

Figure 31: Entry points for Kansei Engineering data in QFD.

92
 Engineering Emotional Values in Product Design

In the relationship matrix the customer’s needs are linked to the technical responses.
The strength of the relationships is determined by more or less qualified guesses of
experts distinguishing between strong, medium and weak relationships. Further
refinement of the scale complicates the decision process and as a consequence
increases the evaluation time. However, in some cases more detailed information is
desirable. Kansei Engineering can quantify those relationships by building
mathematical models of the customer’s Kansei. Moreover, statistics allow examination
of whether different customer group have different opinions which in turn can lead to
the development of alternative product concepts.

In QFD the customer needs are collected using focus group interviews, activity
analysis etc. The data gathered is evaluated and interpreted by i.e. Voice of customer
tables, etc. Those tools try to grasp even implicit needs and wants and link them to
product parameters. However, interpreting the customers’ statements is very difficult
and requires much experience. Kansei Engineering is designed for evaluating the
unconscious wants and needs, and delivers statistical mathematical values for the
connection between a need and its technical response. Regarding furniture the
customer may describe her home as ‘elegant’. One way is to treat this statement with
VOCT, and feed it into the ‘House of Quality’. An alternative way is to make a Kansei
Engineering evaluation and get an exact translation into product properties with only
minor influence from the researcher. In the following step within the QFD procedure,
customer needs are ranked after their importance. QFD offers a number of different
tools, based on (subjective) expert knowledge. Kansei Engineering offers the
opportunity to collect the customer’s attitude about the particular importance of the
different product properties. Paper B in Appendix A describes how the type of
importance can be valued and ranked by customers (see also Section 8.3).

A third way Kansei Engineering can be used in the QFD process is the setting of the
target values for the product properties. For every customer need, a number of product
properties can be identified. Collecting additional information about what the ideal
product should feel like, gives a clear indication about in which direction the Kansei
Score of the certain Kansei Word has to be adjusted to suit the customers’ needs in a
better way. Since it is known how and in which way the product properties are
connected to the certain word, it becomes clear how the Kansei can be adjusted. As a
result Kansei Engineering can help to set target values for the technical specifications.

Success in international markets does not only require a customer-focused design, but
also knowledge about the competition situation in the specific market segment. QFD
provides a product comparison where different products within the same segment are
compared regarding the degree of fulfilment of a specific customer need. By
93
 Discussion

comparing the results of Kansei Engineering studies made for different competing
products, a benchmarking profile can be developed and integrated into the ‘House of
Quality’. Such a combination of the ‘House of Quality’ and Kansei Engineering allows
a special profiling of the products expressed Kansei.

9.8. Tangible and Intangible Products

Products may be physical artefacts or services. A more general definition is that
products are the result of all kinds of man-made allowances towards an intended
purpose. In order to discriminate between services and artefacts the terms tangible and
intangible products was coined (Röstlinger and Goldkuhl, 1999) (compare also Section
2.4.).

Kansei Engineering traditionally treats only tangible products. Some few intangible
products such as web pages etc. have been examined, but there is now a strong focus
on integrating these products types into the methodology. However, as shown before,
the market of intangible products is growing and new types of products combining
both tangible and intangible parts occur. An example of these types of products are
mobile communication devices. An artefact, the phone itself is needed, but this is just
a smaller part. The service behind it, is more cost intensive and more relevant to the
customer. In principle, Kansei Engineering can cope with these products too. However
minor technical modifications have to be made to the methodology. For example, the
collection of the semantic and physical description must be enhanced towards the
intangible properties. New tools might be found. Also the name ‘physical description’
does not longer apply. There is no general difference in how the incoming data is
treated. This means that the synthesis phase and the model building in Figure 20 will
be applicable.

The main problem is the lack of experience with this type of products. Defining the
domain might be more complex. Whereas tangible products have clear physical
limitation, the decision of what property is part of combined tangible/intangible
products is much more difficult.

9.9. Innovativeness in Kansei Engineering?

Critics of the method have claimed that Kansei Engineering has a lack of
innovativeness. However, Kansei Engineering has been involved in the development
of many successful products such as the Mazda Miata or the Camcorder from Sharp
(Nagamachi, 1999), but the argument is that Kansei Engineering itself can not
contribute to the new features.

94
 Engineering Emotional Values in Product Design

Kansei Engineering methodology is often used in order to evaluate already existing

products. It can determine which product features are important for a certain
impression or Kansei and give recommendation about how to combine them with each
other in order to achieve the intended Kansei or even combinations of different
Kanseis. The reason for this is that the customers/ users of a product must have
sufficient experience in order to determine the its Kansei. This is not the case if the
product type is completely new. It is difficult for people to have opinions of products
the have not interacted with. Hence, Kansei Engineering is more useful for evaluating
mature products. People are used to them and have clear opinions about them. They
can also evaluate new features in old products. Even if Kansei Engineering can not
directly take part in the generation of new ideas and concepts, it presents suggestions
and delivers a ground of discussion which triggers creative processes in product
development staff. This is the way it has been used in the cases of Mazda and Sharp.

In most cases product development is not about revolutionising the product type. It is
more about attractive quality creation (Kano et al., 1984), i.e. a gradual improvement
of the products affective values by adding a few attractive qualities and at the same
time ensuring one-dimensional and must-be-quality (compare also Section 1.4.3).
Kansei is designed to accomplish this. However, there are development opportunities
in Kansei Engineering which also have been addressed previously. One is the
integration of creativity tools. These are techniques collecting and ‘harvesting’ new
ideas. Some of them have already been suggested for usage in Kansei Engineering
such as brainstorming, Pareto-diagrams, Card-systems and Affinity diagrams. So there
is no reason why no more methods can be included into Kansei Engineering.

A final point is the product image and the brand image. Some brands have the
reputation of being progressive and include more new solutions than other more
conservative brands. The need and the degree of developing innovative products also
vary with the target group (Naoki, 2002).

9.10. Applying Kansei Engineering in Europe

Kansei Engineering has been used in the Far East by many researchers and companies
since the 1970ies. Its application technique is well developed. However starting to use
Kansei Engineering in Europe resulted in unexpected difficulties. Partly, these
difficulties depended on missing knowledge about the method, but other problems
seemed to depend on cultural differences between Japan and Sweden/Europe. One of
the difficulties was that participants asked what the purpose of the study was. Some
suspected a psychological test, and refused to attend. Suitable explanations took much
time, required the presence of the researcher and destroyed some of the intuitiveness
95
 Discussion

needed since the participants tried to understand instead of giving their first
impression. Another hitch, which was probably related to the above was that
participants were not used to the SD-scales used. It could be seen that people ranked
erroneously or did not understand the task. This was more severe the more ‘non-
academics’ there were among the participants. In this context also the number of rating
points resulted in confusion. Japanese researcher often use 5-point scales. These types
of scales worked well in Germany and France, but in Sweden the results did not show
as good a distribution.

However one of the most severe difficulties which took many experiments to solve
was the fact that participants in Europe could not make as many ratings as in Japanese
studies. This showed itself by incomplete data sets, obviously wrongly checked data
sheets and bad data quality due to fatigue (compare Schütte (2002)). According to
Japanese researcher the rating of up to 20 products with 100 Kansei Words each was
no problem in Japan, but in Europe is was. The consequences were lower statistical
power due to the low number of voluntary participants and insufficient data quality.
The solution was to reduce the number of ratings. As a rule of the thumb 15 minutes is
the maximum including introduction to the products in question. So the maximum
number of ratings lays around 200 which equals ca. 20 Kansei Words for 10 products.
The reasons for the difficulties described include the fact that both researchers and
participants are untrained in the application of the method in Europe. However, some
reasons can also be found in cultural differences.

96
 10.
Conclusions
結論

Increasing customer demands on product design have resulted in that not only
functionality and usability of the products have to be fulfilled. Today, many companies
meet challenges that force them to put strong emphasis on also integrating affective
aspects in their product design, something that now has become distinguishing
qualities of products. Different methodologies have been developed and integrated in
product design processes in order to measure the affective impact of different products
on customers and users. Kansei Engineering is one of these methods, aimed at
determining relationships between on the one hand the feelings and impressions of the
humans and on the other hand product properties. The special characteristic of this
methodology is that it identifies concrete product design solutions, based on the
feelings and impressions of the users and customers. For this purpose (mathematical)
prediction models are built. Even if quantitative measurements in most cases are used
as the basis for these relationships, there are also studies and elements that require
qualitative approaches. The studies in this thesis have demonstrated the ability of the
Kansei Engineering methodology to evaluate, improve and validate product design
changes towards a desired impact. They have also identified improvement areas in the
methodology and proposed new developments, including the use of scales,
experimental design and validation methods. Further, a conceptual model of Kansei
Engineering was proposed in order to provide a structure for performing Kansei
Engineering studies and studies within the field of affective design. This conceptual
model has become a useful tool for explaining and applying the concept. More
structured ways of identifying design parameters and relevant product properties have

97
 Conclusions

been given high priority in the improvement work of the methodology. A model for
spanning the Space of Product Properties is presented and applied. It was found to
support study performance, make it easier and more structured. Advantages of using
Kansei Engineering are that abstract feelings are visualised and made comprehensible.
Thus, it may provide a structured support for integrating affective values into product
design, especially in early and late stages of the product development process.
However, performing full Kansei Engineering studies takes time, resources and special
competence of the facilitator. Moreover, it is difficult to learn the methodology from
publications available in English. In conclusion, Kansei Engineering is a concept and a
methodology in strong development, a framework in which tools and methods are
continuously developed, added and integrated.

98
 11.
Future Research

Kansei Engineering and the area of affective values in products is developing quickly
and the experience of application of different methodologies is limited. Most of them
need more development both regarding general parts and adaptation to application
areas.

According to the proposed model on Kansei Engineering there is much further

development potential. One of the most important points is the spanning of the Space
of Properties, where there is a need for further research. Methods for the synthesis
phase exist, but there is a need for better adaptation to the context. This is closely
connected to the development of improved validation tools.

Kansei Engineering as used in this thesis was limited to evaluations of physical

products. However, services are a growing sector and an important type of product.
Also, combination of artefacts and services occur increasingly. The experience with
this product types is partial since not many studies has been performed in these areas.
One challenge for the future is to extend Kansei Engineering into this direction.

Kansei Engineering is has the potential to evaluate the affective impact of product
parts the user is not even aware of. Unfortunately, many product designers do not
exploit this ability and many studies often focus on apparent features such as exterior
design.

A weakness of Kansei Engineering is its consumption of time and resources and the
need of highly competent facilitators. There are good opportunities for automation of
99
 Future Research

data collection and evaluations. Also databanks can be built storing information of e.g.
exemplary sets of Kansei Engineering Words and sets of product properties for certain
product domains. Also, methods that are easier to apply, based on qualitative data need
to be developed. This could finally lead to further development of qualitative methods
based on interviews with quick and easy evaluation as in e.g. means-end-analysis.
There is also a great need of knowledge summary reports and state of the art
descriptions in order to further the developments in this field.

100
References

Akao, Y. (1990). History of Quality Function Deployment in Japan. Hansa Publisher,

Antoni, M. and Schütte, S. (2002). Learning from Product Development Projects with Kansei
Engineering. Proceedings of QMOD, Pusan Korea, p.
Arnold, K. (2002). Towards increased Customer Satisfaction. IKP/KMT, Linköping,
Linköping University.
Asatekin, M. (1975). Postural and physiological criteria for seating. A review. METU Journal
of the Faculty of Architecture, vol. 1, p. 55-83.
Aune, A. (1999). Leadership and Management for Quality- and a Discipline called Quality
Technology. The Nordic School of Quality Management, Edvardsson, B. and Gustafsson, A.
(ed.), Studentlitteratur, Lund, pp. 62-77.
Axelsson, J. R. C. (1999). The Dynamics of Quality and Work Life Improvement. The Nordic
School of Quality Management, Edvardsson, B. and Gustafsson, A. (ed.), Studentlitteratur,
Lund, pp. 78-136.
Axelsson, J. R. C., Cederqvist, T. and Magnusson, B. (2002), Ergonomics in Hygiene
Products. Report Institute of Technology Linköping University, Linköping.
Barnes, C. J., Childs, T. H. C., Henson, B. and Southee, C. H. (2004). Surface finish and
touch- a case study in a new human factors tribology. WEAR, vol. 257, p. 740-750.
Baumgarten, A. G. (1961). Aesthetica. Georg Olms Verlagsbuchhandlung, Hildesheim.
Bergman, B. and Klefsjö, B. (1994). Quality. Studentlitteratur, Lund.
Black, T. R. (1999). Doing quantitative research in social sciences: an integrated approach
to research design, measurement and statistics. Sage, London.
Brockhaus Universallexikon. (2000). F.A. Brockhaus, Wiesbaden.
Brown, S. and Eisenhardt, K. (1998). Competing on the Edge. Harvard Business School Press,
Boston.
BT-Industries (2003), Annual Report. Report Mjölby/ Sweden.
Burkhardt, D. (2001). Designing The Customer's perception into a welding helmet.
IKP/KMT, Linköping, Linköping University.
Childs, T. H. C. (2004). Personal communication.
Cicero, M. T. (2004). Selected works. Penguin, London.
Clynes, D. M. (1977). Sentics: The Touch of Emotions. Anchor Press/ Doubleday,
Comte, A. (1979). Om positivism. Korpen, Göteborg.
Cooper, A. and Reimann, R. (2003). About Face 2.0: The Essentials of Interaction Design.
Wiley Publishing, Inc, Indianapolis.
Cooper, R. G. (1998). Product Leadership- Creating the Process from Idea to Launch.
Perseus Books, Reading, Massachusetts.
Crosby, P. B. (1984). Quality without tears: The art of hassle-free management. McGraw-
Hill, New York.

101
 References

Damaiso, A. R. (1996). Descartes' error: emotion, reason and the human brain. Papermac,
London.
DeLancey, C. (2002). Passionate Engines. Oxford University Press, Oxford.
Deming (1986). Out of the Crisis. University of Cambridge, Press Syndicate, Cambridge.
Drury, C. G. (1990). Computerized data collection in ergonomics. Evaluation of Human
Work- A practical ergonomics methodology, Wilson, J. R. and Corlett, E. N. (ed.), Taylor and
Francis, London, pp.
Eklund, J. and Kiviloog, L. (2003). Kansei ratings and time dependencies. Proceedings of
Proceedings of the XVth Triennial Congress of the International Ergonomics Association,
Seoul, Korea, p.
Elsmark, L. (2000). Driving Kansei in Theory and Practice. Division of Industrial
Ergonomics, Dept of Mechanical Engineering, Linköping, Linköping Institute of Technology.
Encyclopaedia Britannica Online (2005). http://search.eb.com/.
ENGAGE (2005), European Project on Engineering Emotional Design Report of the State of
the Art- Round 1. Report Valencia.
Frisk, M. and Järleskog, H. (2003). User Impressions of Virtual Surfaces. Department of
Technology, Linköping, Linköping University.
Garcia, R. and Schütte, S. (2002), Comparing Aesthetical Experience from the Form Ration of
Refrigerator Design in Japan and Sweden- An Application of Kansei Engineering. Report
Hiroshima International University, Department of Kansei Ergonomics, Hiroshima.
Garvin, D. A. (1988). Managing Quality. The Free Press, New York.
Green, E. P. and Srinivasan, V. (1978). Conjoint Analysis in consumer research. Journal of
consumer research 5, September, vol. p.
Guilford, J. P. (1971). Psychology Methods. McGraw-Hill Publishing Company, New York.
Gustafsson, A. (1998). QFD Vägen till nöjdare kunder i teori och praktik. Studentlitteratur,
Hair, J. F., Anderson, R. E., Tatham, R. L. and Black, W. C. (1995). Multivariate Data
Analysis with Readings. Prentice-Hall, London.
Hekkert, P. (1999). Editorial. Proceedings of International Conference of Design & Emotion,
Department of Industrial Design, Delft University of Technology, Delft, The Netherlands, p.
Helander, M. G. (2001). Preface. Proceedings of International Conference on Affective
Human Factors Design, Asean Academic Press, Singapore, p.
Helander, M. G. (2003). Hedonomics- affective human factors design. Ergonomics, vol. 46, p.
1269-1272.
Helm, R., Scholl, A., Manthey, L. and Steiner, M. (2004). Measuring customer preferences in
new product development; comparing compositional and decompositional methods.
International Journal of Product Developmetn, vol. 1, p. 12-29.
Imai, M. (1991). Kaizen: att med kontinuerliga, stegvisa förbättringar höja productiviteten
och öka konkurrenskraften. Konsutlförlaget i samarbete med Kaizen Institute of Europe,
Uppsala.
Ishihara, S., Ishihara, K., Nagamachi, M. and Matsubara, Y. (1996). Neural network approach
for Kansei analysis on milk carton design. Proceedings of Human Factors in Organizational
Design and Management - V, Elsevier Science B.V., Breckenridge, USA, p. 7-12.

102
Ishihara, S., Ishihara, K. and Nagamachi, M. (1998). Hierarchical Kansei analysis of beer can
using neural network. Proceedings of Human Factors in Organizational Design and
Management - VI, Elsevier, Amsterdam, p. 421-425.
Ishihara, S. (2001). Kansei Engineering Procedure and Statistical Analysis. Workshop at
International Conference on Affective Human Factors Design, Singapore.
Ishihara, S., Ishihara, K. and Nagamachi, M. (2001). Kansei Engineering Analysis on Car
Instrument Panel. Proceedings of International Conference on Affective Human Factors
Design, Asean Academic Press, Singapore, p.
IVA (1999), Styrkor, svagheter, möjligheter och hot. Report 91-7082-657-9,
Jordan, P. W. (2001). New Century Supertrends: Designing a Pleasurable Future. Proceedings
of International Conference on Affective Human Factors Design, Asean Academic Press,
Singapore, p. 3-8.
(2004). www.jske.org.
Juran, J., Gryna, F. M. and R.S., B. (1974). Quality Control Handbook. McGraw-Hill, New
York.
Kammerlind, P. and Schütte, S. (2001). Integrating Kansei Engineering and QFD in Product
Development. Proceedings of Q-MOD Conference, Linköping/Sweden, p.
Kanda, T. (2002). Classification of Menus on Home Dining Tables Based upon Human Meal
Kansei. Kansei Engineering International- International Journal of Kansei Engineering, vol.
3, p. 9-14.
Kano, N., Seraku, N. and Takahashi, F. (1984). Attractive quality and must be Quality.
Quality, vol. 14 No.2, p. 39-44.
Kant, I. (2004). Kritik av det rena förnuftet. Thales, Stockholm.
Kleinginna, P. R. and Kleinginna, A. M. (1981). A categorized list of emotion definitions,
with suggestions for consensual definition. Motivation and Emotion, vol. 5, p. 345-379.
Komazawa, T. and Hayashi, C. (1976). A Statistical Method for Quantification of Categorical
Data and its Applications to Medical Science. de Dombal, F. T. and Gremy, F. (ed.), North-
Holland Publishing Company, pp.
Küller, R. (1975). Semantisk Miljö Beskrivning (SMB). Psykologiförlaget AB Liber Tryck
Stockholm, Stockholm.
Küller, R. (1991). Environmental assessments from neuropsychological perspective.
Environment Cognition and Action: An Integrated Approach, Gärling, T. and Evans, G. W.
(ed.), Oxford University Press, New York, pp.
Lee, S., Harada, A. and Stappers, P. J. (2002). Pleasure with Products: Design based Kansei.
Pleasure with Products: Beyond usability, Green, W. and Jordan, P. (ed.), Taylor & Francis,
London, pp. 219-229.
Liedman, S.-E. (2001). Ett oändligt äventyr. Albert Bonniers Förlag,
Lindberg, A. (2004). First Impressions Last- A Kansei Engineering Study on Laminate
Flooring at Pergo. Department of Technology, Linköping, Linköping University.
Liu, S. S. and Stout, P. A. (1987). Effects of Message Modality and Appeal on Advertising
Acceptance. Psychology & Marketing, vol. 4, p. 167-187.
Mori, N. (1998), Chain-like Structure which appear in Causal Reasoning concerned with
Human "Kansei" words- Study on Basic Framework of Design Assist System using "Kansei"

103
 References

Engineering. Report Tokyo Institute of polytechnics, Department of Design, Faculty of arts,

Tokyo.
Mori, N. (2002). Rough Set Approach to product Design Solution for the Purposed "Kansei".
The Science of Design Bulletin of the Japanese Society of Kansei Engineering, vol. 48, p.
Nagamachi, M. (1989). Kansei Engineering. Kaibundo, Tokyo.
Nagamachi, M. (1994a). Kansei Engineering: An ergonomic technology for a product
development. Proceedings of IEA '94, p. 1220-122.
Nagamachi, M. (1994b). Kansei Engineering: A consumer-oriented technology. Proceedings
of International Conference of ODAM, p.
Nagamachi, M. (1997a). Kansei Engineering: The Framework and Methods. Kansei
Engineering 1, Nagamachi, M. (ed.), Kaibundo Publishing Co. Ltd., Kure, pp. 1-9.
Nagamachi, M. (1997b). Requirement identification of consumer's needs in product design.
Proceedings of IEA '97, Finnish Institute of Occupational Heath, Tampere, Finland, p. 231-
233.
Nagamachi, M. (1999). Kansei Engineering: A New Consumer-Oriented Technology for
Product Development. The Occupational Ergonomics Handbook, Karwowski, W. and Marras,
W. S. (ed.), CRC Press LLC, pp. 1835-1848.
Nagamachi, M. (2001). Workshop 2 on Kansei Engineering. Proceedings of International
Conference on Affective Human Factors Design, Singapore, 2001, p.
Nagamura, N. (1991). Measurement of Kansei. Handbook of Advanced Technology of Kansei
Measurement, Kuriyama, H. (ed.), Science Forum, pp.
Nagasawa, S. y. (2002a). Kansei and Business. Kansei Engineering International-
International Journal of Kansei Engineering, vol. 3, p. 2-12.
Nagasawa, S. y. (2002b). Revision and Verification of "Seven Tools for new Product
Planning. Kansei Engineering International- International Journal of Kansei Engineering,
vol. 3, p. 3-8.
Naoki, S. (2002). Emotional Program Bible. Water Studio Co., Tokyo.
Nishino, T., Nagamachi, M., Ishihara, K., Ishihara, S., Ichitsubo, M. and Komatsu, K. (1999).
Internet Kansei Engineering System with Basic Kansei Database and Genetic Algorithm.
Proceedings of TQM and Human Factors, Centre for Studies of Humans, Technology and
Organization, Linköping, Sweden, p. 367-372.
Nishino, T. (2001). Exercises on Kansei Engineering. Hiroshima International University.
Nishino, T., Nagamachi, M. and Ishihara, S. (2001). Rough Set Analysis on Kansei
Evaluation of Color. Proceedings of The International Conference on Affective Human
Factors Design, Asean Academic Press, Singapore, p.
Nolimo-Solman, K. (2001). Analysis of interaction quality in human-machine systems.
Institute of Technology, Linköping, Linköping University.
Osborn, A. F. (1953). Applied Imagination. Creative Education Foundation Press, New York.
Osgood, C. E., Suci, G. J. and Tannenbaum, P. H. (1957). The Measurement of Meaning.
University of Illinois Press, Illinois.
Osgood, C. E. and Suci, G. J. (1969). Factor Analysis of Meaning. Semantic Differential
Technique - a Source Book, Osgood, C. E. and Snider, J. G. (ed.), Aldine Publishing
Company, Chicago, pp. 42-55.

104
Osgood, C. E., Suci, G. J. and Tannenbaum, P. H. (1969). Measurement of Meaning.
Semantic Differential Technique - a Source Book, Osgood, C. E. and Snider, J. G. (ed.),
Aldine publishing company, Chicago, pp. 56-82.
Patel, R. and Tebelius, U. (ed.) (1987). Grundbok i forskningsmetodik, Studentlitteratur,
Lund.
Penfield, W. and Rasmussen, T. (1950). The cerebral cortex of man: a clinical study of
localization of function. Macmillan, New York.
Picard, R. (1997). Affective Computing. Massachusetts Institute of Technology,
Reynolds, T. J. and Olson, J. C. (2001). Understanding Consumer Decision making-The
Means-End Approach to Marketing and Advertising Strategy. Lawrence Erlbaum Associates,
Publishers, New Jersey.
Rosenblad, E. (2000). Konsumentteknik: En kunskapsöversikt. Ord &Form, Uppsala.
Rydman, K. and Sandin, J. (2000). Kansei Engineering-methodbeskrivning och tillämpning
på tryckkänsla hos tangenter. Linköping/Sweden, Institute of Technology, Department of
Mechanical Engineering.
Röstlinger, A. and Goldkuhl, G. (1999), Produktbegreppet- En praktikteoretisk
innebördesbestämmning. Report 1999:07, Centre for Studies of Humans, Technology and
Organisation, Linköping.
Schill, H. (2005). Praktiska erfarenheter och resultat med Lean Production (Toyota
Production System). Proceedings of Qualitetsdag, IKP/ KMT, Linköping, p.
Schweizer, H. R. (1973). Ästhetik als Philosophie der sinnlichen Erkenntnis. Schwabe& Co
Verlag, Basel/ Stuttgardt.
Schütte, S. and Eklund, J. (2001). An Approach to Kansei Engineering- Methods and Case
Study on Design Identity. Proceedings of Conference on Human Affective Design, Asean
Academic Press, London, Singapore, p.
Schütte, S. and Eklund, J. (2003). Product Development for Heart and Soul. Linköping
University, Department for Quality and Human Systems Engineering, Linköping.
Schütte, S. T. W. (2002). Designing Feeling into Products-Integrating Kansei Engineering
Methodology in Product Development. Institute of Technology, Linköping, Linköpings
university.
Shimizu, Y. and Jindo, T. (1995). A fuzzy logic analysis method for evaluating human
sensitivities. International Journal of Industrial Ergonomics, vol. 15, p. 39-47.
Shimizu, Y., Sadoyama, T., Kamijo, M., Hosoya, S., Hashimoto, M., Otani, T., Yokoi, K.,
Horiba, Y., Takatera, M., Honywood, M. and Inui, S. (2004). On-demand production system
of apparel on basis of Kansei engineering. International Journal of Clothing Science and
Technology, vol. 16, p. 32-42.
Sinclair, M. A. (1990). Subjective assessment. Evaluation of Human Work- A practical
ergonomics methodology, R., W. J. and Corlett, E. N. (ed.), Taylor and Francis, London, pp.
58-88.
Skogman, A. (2002). Measurements of user feelings regarding existing vacuum cleaners with
Kansei Engineering- An application of Kansei Engineering and other methods. Luleå, Luleå
University.
Strauss, A. and Corbin, J. (1990). Basics of Qualitative Research: Grounded Theory
Procedures and Techniques. SAGE Publications Ltd., Newbury Park CA.

105
 References

Söderlund, M. (2003). Emotionsladdad marknadsföring. Liber AB, Malmö.

Taylor, S. J. and Bogdan, R. (1984). Introduction to Qualitative Research Methods. John
Wiley& Sons, Inc.,
Titchener, E. B. (1998). An outline of psychology. Thoemmes, Bristol.
Tribus (1990). Newsletter ASQC Statistics Division, vol. p. 3.
Tsuchiya, T. (2004). Personal communication.
Wikström, L. (2002). Produktens budskap- Methoder för värdering av produkteres
semantiska funktioner ur ett användarperspektiv. Institutionen för produkt- och
produktionsutveckling, Göteborg, Chalmers Tekniska Högskola.
(2005). www.keyworth.leed.ac.uk.
Yamamoto, K. i. (ed.) (1992). Japans Automotive Industry: It's Strength, Special Lectures by
Top Management, Faculty of Business Administration, Asia University, Tokyo.
Yun, M. H., Han, H. S., Kim, K. J., Kwahk, J., Sang, H. W. and Kim, J. S. (2001).
Incorporating User Satisfaction into the Look and Feel of Wireless Phones. Proceedings of
International Conference on Affective Human Factors Design, ASEAN Academic Press,
London, Singapore, p.
Zeithaml, V. A. (1988). Consumer Perceptions of Price, Quality and Value: A Means-End
Model and Synthesis of Evidence. Journal of Marketing, vol. p.
Zhang, L., Helander, M. G. and Drury, C. G. (1996). Identifying Factors of Comfort and
Discomfort in Sitting. Human Factors, vol. 38, p. 377-389.

106