Upload
larisa-ioana-zafiu
View
240
Download
5
Embed Size (px)
DESCRIPTION
A Creativity Based Design Process for Innovative Product Design
Citation preview
Ergo
Rebuilding a product design process (PDP) with creativity techniques can effectively assist designers in encouraging
1. Introduction management, research and development, produc-tion, marketing, and decision-making, and re-
ARTICLE IN PRESSNew product development (NPD) is a complexarea of ideas and innovations, involving strategy,
quires linking science and technology/inventionor innovation with the marketplace. The success ofnew products mostly depends on new productdevelopment process and management (Chaturve-di and Rajan, 2000), including effective attainment
Corresponding author. Tel.: +886-6-275-7575; fax: +886-6-274-6088
0169-8141/$ - see front matter r 2004 Elsevier B.V. All rights reserved.doi:10.1016/j.ergodesign creativity, further improving the overall performance of innovative product design. Accordingly, this research
provides a new approach to the design process for industries by taking advantage of a creativity-based design process to
achieve the goals of innovative product design.
r 2004 Elsevier B.V. All rights reserved.
Keywords: Creativity; Design process; Product design; Decision-making model; Systematic approachIn todays highly competitive and uncertain market environment with short product life cycles, product development
must not only satisfy the quality and speed of production, but it must ensure that products themselves have included
innovative values. As creativity plays an important role in new product development (NPD), it can be utilized in search
of novel ideas for innovative product design, and also can be regarded as a helpful tool in advancing NPD output.
In this paper, we developed a creativity method based on the naturally sensuous ability of human beings. We also
proposed a creativity-based design process integrating some systematic design methodologies with a developed
creativity tool. The essence of this proposed design process is not the sole performance of the creativity tool but the
coherent efforts among each involved process technique.
To prove the practicability of this design process, a case study was conducted according to various procedures that
were followed. By applying evolutionary thinking, the sensuous association method, and other systematic approaches
to the design process, a number of concept solutions were produced. An optimal solution was then determined by using
an operative decision-making model based on the weighted generalized mean method.
Relevance to industryAbstractShih-Wen Hsiao, Jyh-Rong ChouDepartment of Industrial Design, National Cheng Kung University, Tainan, Taiwan 70101, ROC
Received 21 February 2003; accepted 10 May 2004
Available online 12 August 2004A creativity-based design process for innovative product designInternational Journal of Industrialn.2004.05.005nomics 34 (2004) 421443
ARTICLE IN PRESS
nal oof knowledge and using it as part of the productdesign process (Poolton et al., 2000). Cooper(1996) indicated that new product development isa vital endeavor for modern industries. Companiesmust learn how to innovate effectively, over-hauling their new product processesfrom ideato launchand incorporating winning ideas forsuccessful new products. However, Grifn (1997)reported that, despite widespread acceptance in theliterature, almost 40% of rms surveyed still useno formalized product development process.Product design is a goal-directed problem-solving
activity that relies heavily on human experience,creative thinking, and related knowledge; it shouldbe done by integrating creativity and innovationtools with axiomatic design methodology for durableproduct development (Goel and Singh, 1998). Villa(1998) indicated that continuing innovation inindustries is dened as: promoting a frequent re-design of offered products as well as of theproduction processes required. Also, he proposedan innovation loop concept and introduced aconceptual model of multi-resolution design process.Majaro (1988) suggested the rst step in theinnovation process is the generation of ideas orcreativity, but an important minority of pioneeringdesign theorists has implied that the most valuablepart of the design process is that which occurs insidethe designers head and partly out of reach of hisconscious control (Jones, 1992, pp. 4647). Unfortu-nately, the black-box view of designing can not beclearly expressed in contemporary design theoriesbased on rational analytics. In fact, design work notonly consists of logical rules dealing with distinctprocedures but also innovative thinking producingintricate creativity.Product design process (PDP) is an essential
factor during the early phase of new productdevelopment, which can be considered a complexset of integrated efforts, including generatingideas, developing concepts, modifying details,and evaluating proper solutions. An inappropriateproduct design process not only affects productlife-cycle phases but also increases the possibilityof failure in new product development. Since the1960s, some design scholars, such as Hall (1968),Archer (1971), French (1971), Pahl and Beitz
S.-W. Hsiao, J.-R. Chou / International Jour422(1984), Hubka (1989), etc. have successivelydeveloped many design processes, taking advan-tage of denite methodology to eliminate theillusion of Designer as black boxes. Theconventional design process can offer designers ahelpful tool to product design and thus increasesNPD output. However, Stevens and Burley (1997)indicated that about 3000 raw ideas are required toproduce one substantially new commercially suc-cessful industrial product. They also found thatNPD requires breakthrough creativity because therst ideas for commercialization are almost nevercommercially viable until they have been substan-tially revised through a thought process involvingbranching (Stevens et al., 1999).Bullinger et al. (2000) once stated: the new
millennium has fostered a look towards the future,accompanied by both hopes and fears. Facingtodays global information society, companies mustfocus on the task of meeting competition andchallenges. A high level of complexity, dynamismand uncertainty characterizes the economic situationfor industry and its production facilities. In todayshighly competitive and uncertain market environ-ment with short product life cycles, product designmust not only satisfy the quality and speed ofproduction, but it must ensure that productsthemselves have included innovative values. Stevenset al. (1999) demonstrated that positive correlationswere found between prots resulting from NPDproject analyses and the degree of creativity of theanalysts evaluating those projects. McAdam andMcClelland (2002) also documented the perfor-mance results using new product ideas and creativitypractices in the UK textile industry. Therefore, thisresearch focuses on the integration of a creativitymethod and systematic design approaches, andproposes a creativity-based design process forinnovative product design. The proposed processcan effectively assist designers in encouraging designcreativity, further improving the overall perfor-mance of innovative product design.
2. Theoretical background
2.1. The use of creativity techniques
Creativity is still one of the most mysterious
f Industrial Ergonomics 34 (2004) 421443elements in human thinking. Behavior involves
oneare
(2)
(3)
(4)
listening and feelingcatalysis and outburstcourse.
The SAM is used for refreshment of sensuous-ness and association of inspiration, and it can beregarded as a creativity tool for encouragingdesigners potential to produce innovative ideasquickly. The operation of the SAM is describedbelow:
(1) Put together a design team from a group ofcarefully selected people for a product designproject.
(2) Before the SAM is performed, team membersmust be gathered into a discussion room, so
ARTICLE IN PRESS
Fig
nal of Indtwo different levels of creativity, namely personaland social-cultural (Liu, 2000). Heap (1989)explained that creativity is the synthesis of newideas and concepts through the radical restructur-ing and re-association of existing ones, whereasinnovation is the implementation of the results ofcreativity. Although creativity in the designprocess is regarded as a powerful tool, it is oftencharacterized by the occurrence of a creative leap,which is not easy to be described in logical rules ofconventional design theory. Research in creativitytechniques has been done in the past. For example,Geschka (1996) reviewed three creativity techni-ques: brainstorming, visual confrontation, andmorphological techniques, which were developedand have been used in Germany or German-speaking countries since the 1960s. Krohe (1996)presented a view on managing creativity as well asa summary of 22 creativity stimulating techniques,such as bug listing, goal/wish, manipulative verbs,nominal group technique, wildest idea, wishfulthinking, etc. Cross (1997) constructed a proce-dure based on a creative-leap example andgeneric descriptive models of creative design toprovide further insight into the example. Lugt(2000) described four creative problem solvingexperiments involving meetings in which graphicvariations on brainstorming technique were ex-plored. Dorst and Cross (2001) proposed rene-ments to the co-evolution model, and alsosuggested that creativity in the design processcan validly be compared to burst of develop-ment. These techniques are universally employedfor encouraging creativity and assisting people inproducing a wide variety of novel ideas.
2.2. Sensuous association method
Referring to the applicably existing creativitytechniques, we developed a creativity methodbased on the naturally sensuous ability of humanbeings. This method is what we call the SensuousAssociation Method (SAM), whose main aim isto produce creative ideas by both a designersindividual association and surrounding stimula-tion. As shown in Fig. 1, the method contains four
S.-W. Hsiao, J.-R. Chou / International Jourpersonal behaviors of human sensuousness andtive inspiration is increased through teaminteraction and surrounding atmosphere,which are also considered an input source of(5)
creativity output course.Stimulation: furthermore, the designers crea-structuring course.Describing: describe your mental imageformed after extraction and restructuringcourse.Comparing: compare what you look at andwhat you think aboutextraction and re-mation input course.Thinking: think about their origins and evolu-tionary trendsinference and re-association(1)extrinsic inuence of the environment. Theyexpressed as follows:
Looking: look at the involved thingsinfor-
. 1. Diagram of the sensuous association method (SAM).ustrial Ergonomics 34 (2004) 421443 423arranged that a large number of categorized
(4) During the SAM operation, the participant
(6)
Sofand
sphcanqua
3. O
Sprothecrea
nottoothe
lation matrix, interaction matrix/net, SAM in the
ARTICLE IN PRESS
nal o. The essence of this design process isthe sole performance of the creativity
l but the coherence among involved stages ofsomSAMutline of the creativity-based design process
ince creativity plays an important role in newduct development, it has to be integrated intoproduct design process. Hence, we propose ativity-based design process integratinge systematic design methodologies with theused. In such a natural and interactive atmo-ere, not only team members creative abilitiesbe enhanced, but a great deal of practical andlity ideas can be produced as well.ingaropotential hidden in each members mind.
ince the SAM includes four natural behaviorshuman beings: looking, thinking, comparingdescribing, as well as advantageous surround-stimulation, design creativity can be easilywritten down by the recorder.Consequently, members interactions will sti-mulate each others creative inspirations in ahighly conducive environment. The environ-ment is conducive to opening up the creativehas to compare his/her associations withinformation/pictures observation and contem-plation, extracting novel ideas regarding crea-tive concepts.
(5) While creative images appear in ones mindafter the comparison and extraction, they mustbe described in a sensuous phrase (e.g., lovelyand pliant like a little bird, slender waist of abeauty, as sharp-sighted as an eagle, etc.), andinformation/pictures are stuck on the wall-board to make a very creative environment.
(3) When discussion begins, one member shouldbe assigned as a recorder, the other membersshould look quietly at a section of productphotos on the wall-board. The sensuousassociating can be done by the design teammembers to help them think logically aboutthe origins and evolutionary trends of thetarget product.
S.-W. Hsiao, J.-R. Chou / International Jour424process.divergent stage, morphological analysis, analysisof interconnected decision areas (AIDA) in thetransforming stage, and then a decision-makingmodel based on the weighted generalized meanmethod used to evaluate an optimal designsolution in the convergent stage. The frameworkof this creativity-based design process is shown inFig. 2, and the detailed procedural method ofimplementation is illustrated step by step inSection 4.
4. Implementation methods
To illustrate the detailed procedure and to provethe practicability of the proposed design process,methods of implementation were used and a casestudy of appearance design for an electric scooterwas conducted in this section. According to thethree essential stages and framework of thiscreativity-based design process shown in Fig. 2,the procedures of implementation are described inthe following subsections.
4.1. Description of the target product
Electric scooters are considered a new greentechnology for the 21st century. They are alsoThe proposed design process includes threeessential stages: (1) divergence, (2) transformation,and (3) convergence (Jones, 1992, pp. 6369). Thedivergent stage is an analytic process for searchingthe problem space, which can be described asbreaking the design problem into pieces. Thetransformation stage is a synthetic process forgenerating the solution space, characterized asputting the pieces together in new ways. Theconvergent stage is an integration and evaluationprocess for nding applicable sub-solutions andoptimal design solutions, described as testing todiscover the results of putting the new arrange-ment into practice.There are many design methods and techniques
employed in the overall design process: evolution-ary thinking, systematic structure analysis, corre-
f Industrial Ergonomics 34 (2004) 421443regarded as a viable niche market and a potential
ARTICLE IN PRESS
y thin
arts tr
nal oSystematic structure analysis
Data collection & classification
Evolutionary thinking Evolutionar
Partial p
S.-W. Hsiao, J.-R. Chou / International Jourindustry for many countries, especially in Asianareas (Colella, 2000; Tso and Chang, 2003).In order to formulate product development
policy and establish intended goals of designing,related product information (e.g., product speci-cations, design regulations, technical documents,photos, patterns, etc.) should be collected rst, andthen categorized based on its characteristics andattributes.
(Convergent stage)
Evaluation for an optimal solution
Conclusive design
Integration for concept solutions
FFiinnddiinngg aapppplliiccaabbllee ssuubb--ssoolluuttiioonnss
Determining the dimension parameters
Developing fundamental concept solutions
Defining the boundary conditions
Integration
Decision-
SSeeaarrcchhiinngg tthhee pprroobblleemm ssppaaccee
Sensuou
Building
Constru
Fig. 2. Framework of the creati(Divergent stage)Target product
king diagram
ee
f Industrial Ergonomics 34 (2004) 421443 4254.2. Divergent stage of the design process
As product design is considered a problem-solving activity, the problem spacethe locus inwhich problem solving activities take place (Gold-schmidt, 1997), should be dened by expandingthe boundary of design criteria so as to have alarge and fruitful search space in which to look forproblem solutions.
(Transforming stage)
New product
GGeenneerraattiinngg tthhee ssoolluuttiioonn ssppaaccee
Defining dimension parameters
Developing analogous shape parameters
Assembly/Layout analysis
of the sub-solutions among each subsystem
making model based on the weighted generalized means
s association & conceptual description
a correlation matrix
cting an interaction matrix and an interaction net
vity-based design process.
4.2.1. Evolutionary thinking of the target product
Basically, the wheel was a vital invention forhuman civilization that changed the way ofmoving on land and expanded the domain ofhuman activity. To understand the developmentalhistory and evolutionary pathway toward thetarget product, as well as anticipating its possibledevelopment in the future, we collected a large
number of related product photos and arrangedthem in an evolutionary thinking diagram asshown in Fig. 3. By contemplating the diagram,various kinds of two-wheel vehicles were distin-guished, including the bicycle, motorcycle, motorbike, electric bike, scooter, and electric scooter. Wesubsequently assumed the position of electricscooters is different from traditional fuel scooters.
ARTICLE IN PRESS
S.-W. Hsiao, J.-R. Chou / International Journal of Industrial Ergonomics 34 (2004) 421443426Fig. 3. Evolutionary thinking diagram for the target product.
ARTICLE IN PRESS
derl
nal oA. Basic framework
(Un
(Subsystem )
(System )
S.-W. Hsiao, J.-R. Chou / International JourElectrically powered two-wheel scooters howeverwill become a signicant means of transportationfor many urban areas. Moreover, we also identi-ed some design objectives for developing anelegant, harmonious, and humanistically designedelectric scooter.
4.2.2. Systematic structure analysis of the products
components
For reducing complex processes of designactivity and simplifying design objects, the targetproduct should be analyzed based on a systematicstructure. In this appearance design case, anelectric scooter was separated into three subsys-tems: the basic framework, the outer covering, andfunctional accessories, and their involved compo-
Appearance design of an electric scooter B. Outer covering
C. Functional accessories
Fig. 4. Partial parts tree for appearana.Handlebars (p1)b.Front absorber (p2)c.Main frame (p3)
d.Rear absorber (p4)e.Kick stand (p5) f.Back holder (p6) g.Front wheel (p7)h.Rear wheel (p8)
a.Headset (p9) b.Front fender (p10)
ying part of subsystems )
f Industrial Ergonomics 34 (2004) 421443 427nents were grouped into a partial parts tree asshown in Fig. 4. Due to the underlyingparts of subsystems being more subject to appear-ance designing, they should be consideredmajor units in the partial parts tree of this case.The subsystem component set can be expressed asbelow:
P fp1; p2; p3;y; png; 1
where pj is an underlying part of subsystems,j 1; 2; 3;y; n:
4.2.3. Searching the problem space
Following procedures of evolutionary thinkingand systematic structure analysis, critical problemssuch as product evolution, design objectives, and
c.Mudguard (p11)d.Footrest board (p12)e.Body-board (p13)f.Seat (p14)
a.Rear view mirror (p15)b.Instrument panel (p16) c.Headlight (p17)d.Taillight (p18) e.Front turn signal (p19)f.Rear turn signal (p20)
Speedometer
Odometer
Remaining energy indicator
Power/Charge indicator
Turn indicator
Lamps indicator
ce design of an electric scooter.
paronsear
(1)
ARTICLE IN PRESS
Tabl
List
Item I
s1s2
nal of Ind428phrases can be produced. By deleting thesimilar, preposterous, or indeterminatephrases, the selected conceptual descriptionS
wheIn
theboa22 ccreaitem
(2)ts analysis and grouping were identied. Basedinitial identication, the problem space can beched as follows.
Sensuous association and conceptualdescription.Using the SAM described in Section 2.2 toexpand the boundary of design creativity, alarge number of conceptual descriptions5 A bald-headed guy
s6 To obstruct or hinder in the middle
s7 To be deeply attached to each other
s8 To turn upward
s9 Flimsy wings, stacked up
s10 Looking for a needle in a haystack
s11 No need to cover withs3s4e 1
of the conceptual description phrases
no. Conceptual description phrase
Bulging front and raised rear
Round rear end, like a semi-sphere
Incline at 45
A at platereal level
S.-W. Hsiao, J.-R. Chou / International Jourphrases can be arranged into a set as below:
fs1; s2; s3;y; sqg; 2
re si is a single phrase, i 1; 2; 3;y; q:this case study, nearly 100 photos relevant to
scooters were classied and stuck on the wall-rd. Through the use of the SAM, we chose theonceptual description phrases characterized astive ideas of the problem space. They areized as shown in Table 1.
Building a correlation matrix between con-ceptual description phrases and underlyingparts.In order to properly apply the creative ideas tothe development of subsystem components, apracticable correlation between conceptualdescription phrases and each subsystem partshould be identied.Based on Eqs. (1) and (2), the correlationmatrix, R; can be represented as
R ST P s1; s2; s3;y; sqT p1; p2; p3;y; pn
rijqn; 3
where i=1,2,3,y,q; j=1,2,3,y,n.The entries rij are dened by the following entry
rules:Rule 1: If si correlates with pj ; then rij 1; i.e.,
the conceptual description phrase is able to applyto the subsystem part(s).
s13 A drop of tears
s14 To promote step by step
s15 To treat another as ones equal
s16 An unemployed locksmith
s17 To have edges and corners of view
s18 A drooping buttock
s19 A partial lunar eclipse
s20 A guy with big feet
s21 Round as pearls and smooth as jade
s22 Slender waist of a beautys1R
rij theeachU
casewhiandThedescthedesi
(3)tem no. Conceptual description phrase
2 A y with a huge head
ustrial Ergonomics 34 (2004) 421443ule 2: If si does not correlate with pj ; then0; i.e., the conceptual description phrase andsubsystem part(s) are icompatible withother.sing Eq. (3), a correlation matrix for thestudy was constructed as shown in Table 2,
ch helps us to clarify which pairs of partsphrases are compatible with each other.question of how to turn the conceptualription phrases into real design elements forsubsystem parts was the problem space in thisgn case.
Constructing an interaction matrix and aninteraction net.An interaction matrix is similar to that of acorrelation matrix, which helps us toanalyze the interaction between any two partswithin the underlying parts. The interaction
ARTIC
LEIN
PRESS
Table 2
The correlation matrix between conceptual description phrases and underlying parts for the design
Description
phrases
Basic frameworks Outer covering Functional accessories Others
Handle-
bars
Front
absorber
Main
frame
Rear
absorber
Kick
stand
Back
holder
Front
wheel
Rear
wheel
Headset Front
fender
Mudguard Footrest
board
Body-
board
Seat Rear
view
mirror
Instrument
panel
Headlight Taillight Front
turn
signal
Rear
turn
signal
p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 p11 p12 p13 p14 p15 p16 p17 p18 p19 p20
s1 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0
s2 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0
s3 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0
s4 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0
s5 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0
s6 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0
s7 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
s8 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
s9 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
s10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
s11 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0
s12 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 1 0 0 0
s13 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1
s14 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
s15 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0
s16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
s17 0 0 0 0 0 0 0 0 0 1 1 0 1 0 0 0 0 0 0 0
s18 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
s19 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0
s20 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0
s21 0 0 0 0 0 0 0 0 1 1 0 0 1 0 0 0 1 1 0 1
s22 0 0 0 0 0 0 0 0 0 1 1 0 1 0 0 0 0 0 0 0
S.-W
.H
siao
,J
.-R.
Ch
ou
/In
terna
tion
al
Jo
urn
al
of
Ind
ustria
lE
rgo
no
mics
34
(2
00
4)
42
1
44
3429
B
wheT
beloR
1:R
bij R
bij R
comU
scooF
searactishointethepair
ARTICLE IN PRESS
ying
Bb
1
1
0
0
0
0
1
0
1
nal omatrix, B; can also be presented asTable 3
The interaction matrix between any two parts within the underl
Aa Ab Ac Ad Ae Af Ag Ah Ba
Aa 1 1 0 0 0 1 0 1
Ab 1 1 0 0 1 0 0Ac 1 1 1 1 1 0Ad 0 0 1 1 0Ae 0 1 1 0
Af 0 0 0
Ag 1 0
Ah 0
Ba
Bb
Bc
Bd
Be
Bf
Ca
Cb
Cc
Cd
Ce
Cf
S.-W. Hsiao, J.-R. Chou / International Jour430PT P p1; p2; p3;y; pnT p1; p2; p3;y; pn
bijnn 4
re i 1; 2; 3;y; n; j 1; 2; 3;y; nhe entry rules of entries bij are dened asw:ule 1: If pi directly interacts with pj ; then bij
ule 2: If pi indirectly interacts with pj ; then1:
ule 3: If pi has no interaction with pj ; then0:
ule 4: pi and pi itself (i.e. i j) no need to bepared.sing Eq. (4), an interaction matrix for theter design was carried out as shown in Table 3.or extending the use of interaction matrix andching the problem space concerning the inter-ons of underlying parts, an interaction netuld be constructed based on the result of theraction matrix operation. As shown in Fig. 5,constructed network shows not only whichs of parts are connected, but also distinctlyillustrates the relative importance or interdepen-
parts
Bc Bd Be Bf Ca Cb Cc Cd Ce Cf
0 0 0 0 1 0 0 0 0 01 0 0 0 0 0 0 0 0 00 1 1 1 0 0 0 0 0 00 0 1 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 1 1 0 0 0 0 0 01 0 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 0
0 0 0 0 1 1 1 0 1 01 1 1 0 0 0 1 0 1 0
0 0 0 0 0 0 0 0 0
1 1 0 0 0 0 0 01 0 0 0 1 0 1
0 0 0 0 0 0
0 0 0 0 0
1 0 0 0
0 1 0
0 1
1
f Industrial Ergonomics 34 (2004) 421443dence among the underlying parts.Analyzing Fig. 5, we found that interaction was
closer between the basic framework and the outercovering, whereas the functional accessories weremore independent than the other subsystems.Considering all parts, the main frame of thescooter (Ac(6+4)) interacted with 10 parts, includ-ing six direct interactions and four indirectinteractions. The body-board (Be(6+3)) interactedwith nine parts, including six direct interactionsand three indirect interactions. The front fender(Bb(4+5)) interacted with nine parts, including fourdirect interactions and ve indirect interactions.The analysis demonstrates the appearance designof an electric scooter should focus on the mainframe for the basic framework design, and thebody-board and the front fender for the outercover design. Most of the functional accessoriesdirectly interacted with the headset (Ba(5+1)),including the rear view mirror (Ca(2+0)), theinstrument panel (Cb(1+1)), the headlight(Cc(2+2)), and the front turn signal (Ce(2+2));hence, they could be considered for collocationwith the headset in the design.
ARTICLE IN PRESS
6+3)
(5+1)
(2+3)d
among the
nal of Ind(6+4)
(1+3) ((3+1)
(1+4)
(1+7)
(2+1)
(1+2) Ac
AAe
Ag
Bd
AfAh
Bf Be
Fig. 5. Interaction net4.3.
Mprodesisolubetakimak
4.3.
Fticamussub
(1)(1+2)
(2+2)
(1+1)
(4+5)
Cd
BbCf
Ce
S.-W. Hsiao, J.-R. Chou / International JourTransforming stage of the design process
any design concepts were created from theblem space through the divergent process. Thegn concepts regarded as radical problemtions are not concrete enough. They shouldtransformed into a denite solution space,ng advantage of positive design elements toe the concepts a reality.
1. Generating the solution space
or converting the design concepts into prac-l design elements, the denite solution spacet be generated corresponding to the threesystems we designated before.
Dening dimension parameters for the basicframework design.According to the analysis of Fig. 5, the basicframework design should be based on themain frame, whose key points are not only toensure the structural strength but also to meetthe dimensional parameters, such as center-of-gravity position (CG), wheelbase, overall
(2): Direct relation between parts : Indirect relation between parts
(4+2)Ab
underlying parts.(1+1)(2+2)
(2+0)
(2+1)(3+3)
Ba Ca
Cc Cb
BcAa
ustrial Ergonomics 34 (2004) 421443 431length, overall width, overall height, etc. Sincethe CG position of an electric scooter isdifferent from that of a conventional fuelscooter (Lai et al., 2003), it must be preciselydetermined. In this design, over-sized and no-inner-tube tires with front-disc and rear-drumtype of brakes were used. The signicantdimension parameters of the basic frameworkdesign are shown in Fig. 6, and they can beconsidered a solution space in teams of thebasic framework subsystem.Developing analogous shape parameters forthe outer cover design.Since outer covering is the most importantfactor in appearance design of an electricscooter, it must be considerably well-thought-out. Based on the result of Table 2,if a conceptual description phrase correlateswith any underlying part (i.e. rij 1), itshould be turned into an analogous shapeparameter by image outlining of the phraseand parts feature, and the image outlines alsoshould be represented in a morphologicalchart.
ARTICLE IN PRESS
nal o432Aconideachaof tlogiparfendtwobodThedesipermpar
(3)S.-W. Hsiao, J.-R. Chou / International Jours shown in Table 4, the morphological charttains some concrete design elements of creatives originating from the SAM operation. Thert can be considered a solution space in teamshe outer covering subsystem. In the morpho-cal chart, we have four pieces of shapeameters for the headset (Ba), six for the fronter (Bb), four for the mudguard (Bc),for the footrest board (Bd), six for the
y-board (Be), and three for the seat (Bf).oretically, there are 3456 sets of conceptualgn possibilities in the solution space byuting and composing these pieces of shape
ameters.
Assembly/layout analysis of the functionalaccessories.
Acleasetstion
Fig. 6. The signicant dimension parameters of Industrial Ergonomics 34 (2004) 421443As we found in Fig. 5, the functionalaccessories are more independent of butnecessary for driving safety because they mustbe appropriately collocated with the othercomponents. The analysis of interconnecteddecision areas (AIDA), an analyticalmethod for dealing with the determinationsof related decision-making concerning designproblems (Luckman, 1967), can be applied tofacilitate layout design of the functionalaccessories.
s shown in Fig. 7, the AIDA method canrly help us to understand all the compatibleregarding the layout conditions of the func-al accessories. There are 15 total pairs of
f the basic framework design.
ARTICLE IN PRESS
r des
nal oTable 4
Morphological chart of the shape parameters for the outer cove
B 1 2 3
S.-W. Hsiao, J.-R. Chou / International Jourcompatible solutions shown in Table 5, which wereidentied by the analysis of the decision areas.These solutions also can be considered a solutionspace in terms of the functional accessory sub-system.
a. Headset
s5 s11 s12
b. Front fender
s21 s3 s1
c. Mudguard
s17 s19 s11
d. Footrest board
s15 s 6
e. Body-board
s1 s13 s18 s19 s3 s22` ` `
f. Seat
s 7 s15 s14ign
4 5 6
f Industrial Ergonomics 34 (2004) 421443 4334.4. Convergent stage of the design process
Because of a large number of problem solutionswithin the solution space, they must be convergedand integrated into formal solutions, not only for
s 21
s 4 s 6 s 11 s17 s22
s 22
s 2 s 18 s 21 s17 s 9` ` `
s 21
` `
ARTICLE IN PRESS
Ceurn signal
Bbront fender
tiona
nal of Ind434Cb Instrument panel
Cc Headlight
Fig. 7. AIDA for the funcndan o
4.4.
Capp
(1)
Tabl
The
Func
Rear
Instr
Head
Taill
Fron
RearCa (Rear view mirror ) Front t
Ba (Headset) (F
Aa(Handlebar)
S.-W. Hsiao, J.-R. Chou / International Jouring applicable sub-solutions but for evaluatingptimal design solution as well.
1. Finding applicable sub-solutions
orresponding to the transforming process, thelicable sub-solutions can be found as follows.
Determining the dimension parameters for thebasic framework design. One vital purpose ofproduct design is production. Thus, theinvolved dimension parameters must be deter-mined. According to the dened dimensionparameters shown in Fig. 6, the signicantdimensions of the basic framework designwere determined as follows: L1=1200mm,L2=1240mm, H1=235mm, H2=520mm,H3=645mm, H4=610mm, H5=745mm,H6=870mm, H7=1020mm, W1=120mm
(2)
(3)
e 5
compatible solutions identied by the analysis of the decision areas
tional accessory Pair(s)
view mirror 3
ument panel 2
light 3
ight 1
t turn signal 4
turn signal 2l acce Taillight
Be(Body-board)
)
ssory layout design.Cd
Cf(Rea r turn signal)
ustrial Ergonomics 34 (2004) 421443(rear wheel width: 130mm), W2=555mm,W3=320mm, u=28, and v=38.Developing fundamental concept solutions forthe outer cover design.Based on morphological analysis shown inTable 4, we can theoretically produce 3,456sets of conceptual possibilities; however, sub-sequent to concept assessment, there were justsix sets of fundamental solutions chosen byrudimentary evaluation of 102 idea sketches.The six sets of fundamental concept solutionsare shown in Fig. 8.Dening the boundary conditions for thefunctional accessory layout design.For rational and practical options, boundaryconditions of the compatible solutionsshould be dened. As shown in Table 6, thedenition of boundary conditions can help us
Compatible sets
CaAa, CaBa, CaBb
CbAa, CbBa
CcAa, CcBa, CcBb
CdBe
CeAa, CeBa, CeBb, CeCa
CfBe, CfCd
4.4.
con
Bcon
ARTICLE IN PRESS
nal oS.-W. Hsiao, J.-R. Chou / International Jourto clarify the layout condition of each compa-tible solution, as well as to make an appro-priate selection within the compatible sets.
2. Integration of sub-solutions for developing
cept solutions
ased on the six sets of fundamentalcept solutions shown in Fig. 8, the concept
Fig. 8. Six sets of fundamental concepf Industrial Ergonomics 34 (2004) 421443 435design for the electric scooter must befurther developed by integrating the involvedsub-solutions of the basic framework and thefunctional accessory subsystems. Throughcomprehensive evaluation and integration,we developed three sets of concept solutionsfor the electric scooter design as shown inFig. 9.
t solutions in outer cover design.
ARTICLE IN PRESS
Table 6
Denition of boundary conditions for the functional accessory layout design
C Compatible solution Compatible
set
Description of the condition
Rear view mirror (Ca) To set on the Handlebar (CaAa) If the shape parameter of Headset is Ba1
or Ba2, the Rear view mirror can be
directly set on the Handlebar.
To set on the Headset (CaBa) If the shape parameter of Headset is Ba3
or Ba4, the Rear view mirror can be
easily set on the Headset.
To set on the Front fender (CaBb) It seems untting for our target product
due to ergonomic considerations, but it
can be considered to be used in a larger
type of electric scooter design.
Instrument panel (Cb) To set alone on the Handlebar (CbAa) If the shape parameter of Headset is
Ba2, the Instrument panel can be
independently attached to the
Handlebar.
To integrate with the Headset (CbBa) If the shape parameter of Headset is
Ba1, Ba3 or Ba4, the Instrument panel
can be integrated into the Headset.
Headlight (Cc) To set alone on the Handlebar (CcAa) If the shape parameter of Headset is Ba1
or Ba2, the Headlight can be set alone
on the Handlebar.
To integrate into the Headset (CcBa) If the shape parameter of Headset is Ba3
or Ba4, the Headlight can be integrated
into the Headset.
To combine with the Front fender (CcBb) If the shape parameter of Headset is
Ba2, Ba4 or the Front fender shape is
Bb2, Bb3 or Bb5, the Headlight can be
combined with the Front fender.
Taillight (Cd) To set alone on the Body-board (CdBe) This is a common use in current design.
Besides, it also can be set on the Seat
(CdBf) or the Back holder (CdAf),
but both of the locations are not
considered feasible because of their
impracticality for mass production.
Front turn signal (Ce) To x on both sides of Handlebar (CeAa) This position provides for good signal
visibility because both sides of the
Handlebar represent the maximum
width of the electric scooter frame.
To attach to the Headset/combine with
the Front fender
(CeBa) /
(CeBb)
This is a common design for todays
fuel-powered scooters.
To integrate into the Rear view mirror (CeCa) The pair of CeCa provides more
signal visibility than CeAa, but the
design may be expensive because of the
need for standardization of parts.
Rear turn signal (Cf) To integrate in the Body-board (CfBe) This design gives a smooth and aesthetic
look.
To be combined with the Taillight (CfCd) This is an easy assembly but provides
poor functional performance.
S.-W. Hsiao, J.-R. Chou / International Journal of Industrial Ergonomics 34 (2004) 421443436
ARTICLE IN PRESS
nal oS.-W. Hsiao, J.-R. Chou / International Jour4.4.3. Evaluation of concept solutions for an
optimal solution
For further conclusive design, an optimalsolution must be determined among the three setsof concept solutions. In this study, we developedan operative decision-making model based on theweighted generalized mean method. The formulaof the weighted generalized means is dened asbelow (Klir and Folger, 1988, pp. 5861):
haa; w Xni1
wi aai
!1=a; 5
Fig. 9. Three sets of concept solutionf Industrial Ergonomics 34 (2004) 421443 437where wi is the parameter weight; wiX0;Pni1 wi 1; ai the relative value of evaluation
factors, the a the variable parameter, NoaoN:The major difference between the weighted
generalized means and the conventional weightedmeans is the involvement of a parameter. By varyingthe a parameter within the interval of N;N; wecan derive a homologous ha value based on a multi-factor analysis, and then draw the a ha curves.The a ha curves can be used to rank thealternatives harmoniously so as to accuratelyidentify an optimal solution among them.
s for the electric scooter design.
ARTICLE IN PRESS
A
6
9
7
6
7
9
7
4
7
9
6
8
Inn
nal oTable 7
Statistical results of the quantitative judgments
Alternative Evaluation factor Good (0.5)
Solution 1 Humanistic 51
Innovative 80
Stylish 62
Aesthetic 58
Solution 2 Humanistic 69
Innovative 87
Stylish 66
Aesthetic 50
Solution 3 Humanistic 66
Innovative 84
Stylish 56
Aesthetic 70
Table 8
The weighted parameters of the evaluation factors
Score Factor
Humanistic
S.-W. Hsiao, J.-R. Chou / International Jour438The decision-making model for evaluating anoptimal concept solution was developed andillustrated in the following steps:
Step 1: Establishing evaluation parametersIn accordance with product design objectives,
we identied four evaluation factors: humanistic,innovative, stylish, and aesthetic, and threeevaluation grades: good (score: 0.5), acceptable(score: 0.3), and ordinary (score: 0.2). Based onquantitative judgments by design team members,the statistical results are shown in Table 7.The relative values for each evaluation factor
involved in the design solutions are summarized ina matrix below:
A aji
340:52 0:83 0:64 0:59
0:68 0:88 0:67 0:50
0:67 0:85 0:58 0:71
264
375Solution 1
Solution 2
Solution 3
:
6
Step 2: Setting up the weighted function
Sum 174 168
Average 4.35 4.2
Naturalization 0.29 0.2cceptable (0.3) Ordinary (0.2) Relative value
2 40 0.52
2 77 0.83
5 53 0.64
3 55 0.59
6 54 0.68
5 80 0.88
0 65 0.67
8 53 0.50
5 58 0.67
3 75 0.85
2 57 0.58
1 58 0.71
ovative Stylish Aesthetic
f Industrial Ergonomics 34 (2004) 421443In order to distinguish the relative importanceof the evaluation factors, we dened the weightedfunction as follows:
wi I1=humanistic I2=innovative I3=stylish
I4=aesthetic; 7
where Ii is the weight for each evaluation factor;P4i1 Ii 1:Further, we used an interaction matrix to
analyze the relatively important degree amongeach pair of evaluation factors, and arranged theminto a weighted parameter list shown in Table 8.With the results, we constructed the weighted
function and converted it into a matrix as below:
wi 0:29=humanistic 0:28=innovative
0:26=stylish 0:17=aesthetic; 8
wi 0:29 0:28 0:26 0:17 : 9
Step 3: The operation of weighted generalizedmeans
156 102
0 3.90 2.55
8 0.26 0.17
ARTICLE IN PRESS
0.5
1.0
nal oh
S.-W. Hsiao, J.-R. Chou / International JourLet i 4; substituting the data of matrixes (6)and (9) into formula (5),i.e.
hja w1aaj1 w2aaj2 w3a
aj3 w4a
aj4
1=a;
NoaoN; j 1; 2; 3; 10
where j represents the jth solution.After calculation, we obtained an a ha relation
diagram as shown in Fig. 10.Step 4: Analysis of the results
-50-100 - 0
SSS
Fig. 10. The a ha relation diagram of the
Y
f(c)
X1 X2X0 a X
0
Fig. 11. Using Simpsons rule and the mean vaf Industrial Ergonomics 34 (2004) 421443 439As shown in Fig. 10, if a > 0; the maximum of hais Solution 2; but if ao0; the maximum of ha isSolution 3. For a more precise comparison, theaverage ha of each solution must be solved.
Step 5: Solving the averages of ha:It can be observed in Fig. 10 that the curve
a ha levels out as the curve approacheseither side of positive innity and negativeinnity. Hence, we applied Simpsons rule andthe mean value theorem (see Fig. 11) to solve theaverages of ha:
10050
olution 1olution 2olution 3
weighted generalized mean solutions.
X
f(x)
iX Xi-1 Xi+1 X2nb
lue theorem to obtain the averages of ha:
Simpsons rule:Z ba
f x dxDDxff x0 4f x1 2f x2
? 2f x2n2 4f x2n1 f x2ng=3: 11
Let a=100, b=100; f x hj a w1aaj1 w2a
aj2 w3a
aj3 w4a
aj4
1=a:Z
characteristics: (see Fig. 13)
new
ARTICLE IN PRESS
S.-W. Hsiao, J.-R. Chou / International Journal o440Areaj 100
100hja da
Dhj100 4hj99 2hj98
? 2hj98 4hj99 hj100=3: 12
Mean value theorem:
_Z b
a
f x dx f c b a; 13
Z 100100
hja da hja100 100; 14
i.e.
hja Areaj=200; 15
where hja represents the ha of the jth solution, j
1; 2; 3:By applying formulas (12) and (15), we obtained
results as shown in Table 9.Step 6: Determining the optimal solutionAccording to the comparison of Table 9, we
found that Solution 3 was superior to Solution 2 interms of the conventional weighted means, buttheir averages were very close. For a more credibleevaluation, the weighted generalized means wasdetermined, and the result showed that Solution 3was better than Solution 2. Accordingly, we choseSolution 3 as the optimal solution in this casestudy.
Table 9
Operational results of the decision-making model
Area ha E
Solution 1 134 0.67 0.6499
Solution 2 138 0.69 0.7028
Solution 3 142 0.71 0.7038
E Pn
wi ai; wiX0;Pn
wi 1i1 i16. Conclusions
Creativity as a tool plays an important role innew product development. It should be integratedinto a product design process. In this research, wedeveloped a creativity method called the sensuousassociation method (SAM), which can be em-ployed to encourage designers potential to pro-duce innovative ideas quickly. Subsequently, weproposed a creativity-based design process inte-Flyidesigratwan, R.O.C. in 2002. We have authorized theng electric motor company to use the patentedgn in future production.ertyTaihown in Fig. 14. In addition, we patented thisstyling design through the Intellectual Prop-Ofce of Ministry of Economic Affairs,desias sased on engineering drawings of the conclusivegn, a prototype of the electric scooter was builtBand detachable battery module provides foroperational stability.(5)metal accessories in chromate cladding revealits modern technical aesthetics.The combination of a one-piece running-board(1) No front fender design gives people a newimpression of an electric scooter.
(2) The body outline designed with the goldensection proportion yields a scooter with aclassical and elegant appearance.
(3) A rounded and smooth conguration gives thelook of harmony and gracefulness.
(4) The digital luminescent instrument panel and5. Results and discussions
With the determination of the optimal solution,we made a 1/20 scale mock-up of the scooter frompolyurethane foam for previewing the 3D appear-ance. Additionally, the conclusive design wascompleted by constructing and modifying thedetailed contours according to the aestheticprinciple of golden section proportion (Elam,2001, pp. 542). The outcome of the appearancedesign is shown in Fig. 12, and has the following
f Industrial Ergonomics 34 (2004) 421443ing various systematic design methodologies
ARTICLE IN PRESS
nal oS.-W. Hsiao, J.-R. Chou / International Jourwith the SAM. The proposed design processincludes three essential stages: divergence, trans-formation, and convergence, whose essence is notthe sole performance of the stages and thedeveloped creativity tool but the coherent effortsamong each involved technique of the process.
Fig. 13. Illustration of the conclu
Fig. 12. Appearance design according to the aesthf Industrial Ergonomics 34 (2004) 421443 441To illustrate the procedures of implementingand verifying the practicality of the proposeddesign process, a case studyappearance designfor an electric scooterwas conducted step bystep. By applying evolutionary thinking, systema-tic structure analysis, the sensuous association
sive design characteristics.
etic principle of golden section proportion.
ARTICLE IN PRESS
w elec
nal omethod, and other systematic approaches to the
Fig. 14. The prototype of the ne
S.-W. Hsiao, J.-R. Chou / International Jour442design process, not only the problem space wassearched and identied, but also a large number ofsub-solutions involved in the solution space weregenerated, particularly the six sets of fundamentalconcept solutions in outer cover design.By integrating the applicable sub-solutions
among each subsystem, three sets of conceptsolutions for the electric scooter design weredeveloped, and then an optimal solution wasdetermined by using the decision-making modelbased on the weighted generalized mean method.With the determination of the optimal solution,the conclusive design was completed by construct-ing and modifying the detailed contours accordingto the aesthetic principle of golden sectionproportion. A prototype of the electric scooterwas also built.In conclusion, the nal work of the electric
scooter design has met our initial product devel-opment objectives. We hope that the research canprovide a new approach to the design process forindustries, and allow companies to take advantageof the creativity-based design process to achievethe goals of innovative product design.Acknowledgements
tric scooter for this design case.f Industrial Ergonomics 34 (2004) 421443The authors are graceful to the editor andreviewers for their helpful suggestions on revisingthis paper.
References
Archer, L.B., 1971. Technological Innovation: a methodology.
Inforlink, Frimley.
Bullinger, H.J., Lentes, H.P., Scholtz, O.H., 2000. Challenges
and chances for innovative companies in a global informa-
tion society. International Journal of Production Research
38 (7), 14691500.
Chaturvedi, K.J., Rajan, Y.S., 2000. New product develop-
ment: challenges of globalization. International Journal of
Technology Management 19 (7/8), 788805.
Colella, W.G., 2000. Market prospects, design features, and
performance of a fuel cell-powered scooter. Journal of
Power Source 86, 255260.
Cooper, R.G., 1996. Overhauling the new product process.
Industrial Marketing Management 25, 465482.
Cross, N., 1997. Descriptive models of creative design:
application to an example. Design Studies 18 (4), 427440.
Dorst, K., Cross, N., 2001. Creativity in the design process: co-
evolution of problemsolution. Design Studies 22 (5),
425437.
Elam, K., 2001. Geometry of Design: studies in proportion and
composition. Princeton Architectural Press, New York.
French, M.J., 1971. Engineering Design: the conceptual stage.
Heinemann, London.
Geschka, H., 1996. Creativity techniques in Germany. Creativ-
ity and Innovation Management (June), 8792.
Goel, P.S., Singh, N., 1998. Creativity and innovation in
durable product development. Computers & Industrial
Engineering 35 (12), 58.
Goldschmidt, G., 1997. Capturing indeterminism: representation
in the design problem space. Design Studies 18 (4), 441445.
Grifn, A., 1997. PDMA research on new product development
practices: updating trends, and benchmarking best prac-
tices. The Journal of Product Innovation Management 14,
429458.
Hall, A.D., 1968. A Methodology for Systems Engineering, 6th
Edition. Van Nostrand, Princeton, NJ, pp. 8591.
Heap, J., 1989. The Management of Innovation and Design.
Cassell, London.
Hubka, V., 1989. Principles of Engineering Design. Springer,
New York.
Jones, J.C., 1992. Design Methods. Van Nostrand Reinhold,
New York.
Liu, Y.T., 2000. Creativity or novelty? Design Studies 21 (3),
261276.
Luckman, J., 1967. An approach to the management of design.
Operational Research Quarterly 18 (4), 345358.
Lugt, R., 2000. Developing a graphic tool for creative problem
solving in design groups. Design Studies 21 (5), 505522.
Majaro, S., 1988. Managing Ideas for Prot. McGraw-Hill,
Maidenhead.
McAdam, R., McClelland, J., 2002. Sources of new product
ideas and creativity practices in the UK textile industry.
Technovation 22, 113121.
Pahl, G., Beitz, W., 1984. Engineering Design: a Systematic
Approach. Chapter 3: The design process, Design Council,
London.
Poolton, J., Ismail, H.S., Shahidipour, M., 2000. The new
products process: effective knowledge capture and utiliza-
tion. Concurrent Engineering: Research and Applications 8
(2), 133143.
Stevens, G., Burley, J., 1997. 3000 Raw ideas=1 commercial
success. Research Technology Management 40 (3),
1627.
Stevens, G., Burley, J., Divine, R., 1999. Creativity+Business
Discipline=Higher Prots from New Product Develop-
ARTICLE IN PRESS
S.-W. Hsiao, J.-R. Chou / International Journal of Industrial Ergonomics 34 (2004) 421443 443Information. Prentice-Hall International, London.
Krohe, J., 1996. Managing creativity. Across the Board
(September), 1622.
Lai, H.C., Liu, J.S., Lee, D.T., Wang, L.S., 2003. Design
parameters study on the stability and perception of riding
comfort of the electrical motorcycles under rider leaning.
Mechatronics 13, 4976.468.
Tso, C., Chang, S.Y., 2003. A viable niche marketfuel cell
scooters in Taiwan. International Journal of Hydrogen
Energy 28, 757762.
Villa, A., 1998. Product-process design and continuing innova-
tion. Robotics and Computer-Integrated Manufacturing 14,
393401.Klir, G.J., Folger, T.A., 1988. Fuzzy Sets, Uncertainty, and ment. Journal of Product Innovation Management 16, 455
A creativity-based design process for innovative product designRelevance to industryIntroductionTheoretical backgroundThe use of creativity techniquesSensuous association method
Outline of the creativity-based design processImplementation methodsDescription of the target productDivergent stage of the design processEvolutionary thinking of the target productSystematic structure analysis of the products componentsSearching the problem space
Transforming stage of the design processGenerating the solution space
Convergent stage of the design processFinding applicable sub-solutionsIntegration of sub-solutions for developing concept solutionsEvaluation of concept solutions for an optimal solution
Results and discussionsConclusionsAcknowledgementsReferences