CHAPTER 5

CONCEPT GENERATION

Developed in collaboration with Gavin Zau.

EXHIBIT 1 An exist ing hand-held nailer lor the proless ional market. (Courtesy SencoProducts .)

The president of a tool manufacturing cornpany i:ommissioned a team lo

develop a new hand-held nailer for the high-end consumer market. One of sev­eral currently available pneurnatic tools for the professional market is shown inExhibit 1. The mission of the team was to co nsider broadly alternative productconcepts, assuming only that the tool would employ conventional nails as thebasic fastening technology. After identifying a set of customer needs and estab­lish ing target product sp ecifi cations, the team faced the following questions:

• What existing solution concepts, if any, could be successfully adapted forthis application?

• What new concepts might satisfy the established needs and specifications?

• What methods can be used to facilitate the concept generation process?

THE TA5K OF CONCEPT GENERATION IN PRODUCT DEVELOPMENT

A product concept is an approximate description of the technology, workingprincipIes, and form of the product. It is a concise description of how the prod­uct will satisfy the customer needs. A concept is usually expressed as a sketch oras a rough three-dimensional model and is often accompanied by a brief textu­al description. The degree to which a product satisfies customers and can be suc­cessfully commercialized depends to a large measure on the quality of the under­lying concepto A good concept is sometimes poorly implemented in subsequentdevelopment phases, but a poor concept can rarely be manipulated to achievecommercial success. Fortunately, concept generation is relatively inexpensiveand can be done relatively quickly in comparison to the rest of the developmentprocess. For example, concept generation had typically consumed less than 5percent of the budget and 15 percent of the development time in previous nail­er development efforts. Because the concept generation activity is not costly,there is no excuse for a lack of diligence and care in exccuting a sound conceptgeneration methodology.

The concept generation process begins with a set of customer nceds and tar­get specifications and results in a set of product concepts from which the teamwill make a final selection. The relation of concept generation to the other con­cep t development activities is shown in Exhibit 2. In most cases, an effectivcdevelopment tearn will generate hundreds of concepts, of which 5 to 20 wiIlmerit serious consideration during the concept selection activity.

Good concept generation leaves the team with confidence that the full spaceof alternatives has been explored. Thorough cxploration of altcrnatives early inthe development process greatly reduces the likelihood that the team wiIl stum­ble upon a superior concept late in the development process or that a competi­tor will introduce a product with dramatically better performance than the prod­uct under development.

78

CHAPTER 5 : CONCEPT GENERATlON 79

elopment

on i ¡ment Identify Establish Generate Seleet a

~Reline- Customer ...... Target ~ Produet ~ Product

Specifications;

Needs Specilieations Coneepts Coneeptt

l_. , ·······l..·....

PlanDev

Analyze Perform 1:Remaining

. PlanCompetitive , Eeonomie ~ Development~

Products Analysis ¡

Projeel ;

N~"l«" ..., " -..""'," .....,..., " f

Miss!State

.11------------- CONCEPT DEVELOPMENT ------------+-

EXHIBIT 2 Concept generation is an integral part 01 the concept development phase.

Structured Approaches Reduce the Likelihood 01Costly Problems

Common dysfunctions exh ib ited by development teams during concept genera­tion in elude:

• Consideration of only one or two alterna tives, often proposed by the mostasse rt ive members of the team

• Failure to consider carefuJly the usefulness of concepts employed by otherfirms in related and unrelated products

• Involvement of only one or two people in th e process, resulting in lack ofconfidence and commitment by the rest of the team

• In effective integration of promising partial solutions

• Failure to consider en tire categories of solu tions

A structured approach to co ncep t generation reduces the incidence of th eseproblems b y encouraging th e gathering of information from many di sparateinformation sou rces, by guidin g th e team in th e thorough exp lora tio n of alter­natives, and by providing a mechanism for integrating partial solutions. A struc­tured methodology also provides a step-by-step procedure for those members ofthe team who may be less ex perie nced in d esign-intensive activities, aJlowingthem to participate actively in th e process.

A Five-Step Methodol,ogy

This chap te r presents a five-step concept generation methodology. The method­ol ogy, outlined in Exhibit 3, breaks a complex problern into simpler subprob­lems. Solution concepts are then id entified for the subproblems by exte rna l andinternal search procedures. Classification trees and concept combination tablesare th en used to exp lo re systematicaJly the space of solution concepts and to

80 PRODUCf DESIOf\¡ AN(; OEVt:lU".,.tNT

1. Clarlfy theproblem.

• Understanding• Problem

decomposition• Focus on criticaí

subproblems

2. Searchexternally.

• Lead users• Experts• Patents• Literature• Benchmarking

3. Searchinternally.

• Individual• Group

• Classification tree• Combination table

Integrated Solutions

5. Reflect on thesolutions andthe process.

• Constructivefeedback

EXHIBIT 3 The five-step concept generation methodology.

integrate th e subproble m solu tio ns into a total solution. Finally, the team takesa ste p back to refl ect o n th e valid ity and applicabili ty of the re sults, as well as onthe proc ess used.

This chap te r will foll ow the re commended methodology an d will describ eeach of the five ste ps in detail . Although we present the methodology in a linearsequence , co ncep t ge n e ration is almost always iterative . Lik e our other develop­merit methodologies , these steps are intended to be a baseline from whichproduct d evelopment teams can d evelop an d refine their own unique problern­solving style .

CHAPTER 5 : CONCEPT GENERATlON 81

Our presentation oí rhc methodology is focused primarily on the overall con­cept for a new product; however, the methodology can and should be userl atseveral difTerent points in the development process. The process is useful notonly for overall product conccpts but for concepts for su bsystems and specificcomponents as well, Also note that while the examplc in this chapter involvcs arclatively technical product, the same basic approach can be applied to nearlyany product.

STEP 1: CLARIFY THE PROBLEM

Clarifying the problem consists of developing a general understanding and th enbreaking the problem down into subproblems if necessary.

The mission statement for the product, the customcr needs list, and the pre­liminary product specification are the ideal inputs to the concept generationprocess, although often these pieces of information are still being refined as th econcept generaúon phase begins. Ideally the team has been involved in both theidentification of the customer needs and in the setting of the target productspecifications. Those members of the team who were not involved in these pre­ceding steps should become familiar with the processes used and their resultsbefore concept generation activities begin. (See the chapters "IdentifyingCustomer Needs" and "Establishing Product Specifications.")

As stated before, the challenge was to "design a better hand-held nailer." Thescope ofthe design problem could have been defined more generally (e.g., "fas­ten wood together") or more specifically (e.g., "improve the speed of the exi st­ing pneumatic tool concept"). Sorne of the assumptions in the tearn's mi ssionstatement were:

• The nailer will use nails (as opposed to adhesives, screws, etc.) .

• The nailer will be compatible with nail magazines on existing tools.

• The nailer will nail into wood.

• The nailer will be hand-held.

Based on the assumptions, the team had identified the customer needs for ahand-held nailer. These included:

• The nailer inserts nails in rapid succession.

• The nailer fits into tight spaces.

• The nailer is lightweight.

• The nailer has no noticeable nailing delay after "tripping" the too!.

The team gathered supplemental information to clarify and quantifv th eneeds, such as the approximate energy and speed of the nailing. These basi cneeds were subsequentiy translated into target product specifications. The targetspecifications included the following:

82 PROOUCT OESIGN ANO OEVELOPMENT

• Nail lengths from 50 millimeters to 75 millimeters

• Maximum nailing energy of 80 joules per nail

• Nailing forces of up to 2,000 newtons

• Peak.nailing rate of 1 nail per second

• Average nailing rate of 4 nails per minute

• Ability to insert nails between standard stud/joists (368 millimeteropening)

• Tool mass less than 4 kilograms

• Maximum trigger delay of 0.25 seconds

Decompose a Complex Problem into Slmpler Subproblems

Most design challenges are too complex to solve as a single problem and can beusefully divided into several simpler subproblems. Dividing a problem into sim­pler subproblems is called problem decomposition. There are many schemes bywhich a problem can be decomposed. Here we demonstrate a functional decorn­position but then list several other approaches that are frequentIy useful.

The first step in decomposing a problem functionally is to represent it as asingle black box operating on material, energy, and signal flows , as shown inExhibit 4(a) . Thin solid lines denote the transfer and conversion of energy, thicksolid lines signify the movement of material within the system, and dashed Unesrepresent the exchange of control and feedback signals within the system. Thisblack box represents the overall function of the producto

The next step in functional decomposition is to divide the single black boxinto subfunctions to create a more specific description of what the elements ofthe product do in order to implement the overall function of the product. Eachsubfunction can generally be further divided into even simpler subfunctions.The division process is repeated until the team members agree that each sub­function is simple enough to work with. A good rule of thumb is to createbetween 3 and 10 subfunctions in the diagram. The end result, shown in Exhibit4(b), is a function diagram containing subfunctions connected by energy, mate­rial, and signal flows.

Note that at this stage the goal is to describe the functional elements of theproduct without implying a specific technological working principle for theproduct concepto For example, Exhibit 4(b) includes the subfunction "isolatenail." This subfunction is expressed in such a way that it does not imply any par­ticular physical solution concept, such as indexing the "stick" of nails into a slotor breaking a nail sideways off of the stick. The team should consider each sub­function in turn and ask whether it is expressed in a way that does not imply aparticular physical solution principIe.

There is no single correct way of creating a function diagram and no singlecorrect functional decomposition of a product. A helpful way to create the d ía-

CHAPTER 5 : CONCEPT GENERATION 83

Input

Energy (?)

Malerial (nails)

Signal (1001 "Irip")

. Hand-HeldNai!er--- ....~-

" .... ... ,_ . --~. ',- ....~ -'

(a)

Output

Energy (?)

Malerial (driven nail)

Signa! (?)

Slore or Convert

Energyaccepl energy lo

extemal translatlo nal

~energy energy"'"" ,- ~ -

Nails---+l~

"Trip" 01 Tool - -

Slorenails

SensaIrip

- -.-

Isolalenail

Trigger1001

(b)

Apply ~ DrivenIranslalional I-+-, ~~ Nailenergy lo nail ¡

EXHIBIT 4 Function diagram 01 a hand-held nailer arising lrom a lunclional decomposition: (a) overall"black box"; (b) relinemenl showing sublunctions.

gram is to quickly create seve ral d raft'> and then wo rk to refine th em into a sin­gle diagram that the te am is comfortable with. Sorne useful techniques fOI" get­ting started are:

• Create a function di agram o f an existing product.

• Create a function d iagram based on an arbitrary product con ce p t alreadygenerated by the team 01' based o n a known su bfu nctio n technology. Besure to generalize the diagram to the appropriate level of abstraction.

• Follow o ne of the flows (fo r exam ple, material) an d d etermine what op e ra­tions are required. The d e tails of th e other flow s can be derived by thinkingabou t their co n n ec tions to the initial flow .

Note that the function diagram is typically not unique. In particular, sub­functions can often be ordered in different ways to produce different fun ctiondi agrams. Also note that in sorne applications th e m aterial , e n e rgy, and signa lflow s are difficult to identify. In these cases, a simple list of the subfunctions ofth e product, without connections between them, is often su fficie n t.

84 PRODUCT DESION ANO OEVELOPMENT

Functional decomposition is most applicable to technical products, but it canalso be applied to simple and apparently nontechnical products. For example,an ice cream scoop has material flow of ice cream being separatcd, formed,transponed. and deposited. These subfunctions cou ld form the basis of a prob­lem decomposition.

Functional decomposition is only one of several possible ways to divide a prob­lem into simpler subproblems. Two other approaches are described below:

• Decomposition by sequence o/user actions: For example, the nailer problemmight be broken down into three user actions: moving the tool to the grossnaiiing position , positioning the tool precisely, triggering the tool. Thisapproach is often useful for products with very simple technical functionsinvolving a lot of user interaction.

• Decomposition by key customerneeds: For the nailer, this decomposition mightinclude the following subproblems: fires nails in rapid succession, fits intight places, and has a large nail capacity. This approach is often useful forproducts in which form, and not working principIes or technology, is theprimary problem. Examples of such products include toothbrushes (assum­ing the basic brush concept is retained) and storage containers.

Focus Initial Efforts on the Critical Subproblems

The goal of all of these decomposition techniques is to divide a complex prob­lem into simpler problems such that these simpler problems can be tackled in afocused way. Once problem decomposition is complete, the team chooses thesubproblems that are most critical to the success of the product and that are like­ly to drive the overall solution to the problem. This approach involves a con­scious decision to defer the solution of sorne of the subproblems. For example,the nailer team chose to focus on the subproblems of storing/accepting energy,converting the energy to translational energy, and applying the translationalenergy to the nail. The team felt confident that the nail handling and triggeringissues could be solved after the energy storage and conversion issues wereaddressed. The team also d eferred most of the user interaction issues of the tool.The team believed that the choice of a basic working principIe for the tool wouldso constrain the eventual form of the tool that they had to begin with the coretechnology and then proceed to consider how to embody that technology in anattractive and user-friendly formo Teams can usually agree after a few minutes ofdiscussion on which subproblems should be addressed first and which should bedeferred for later consideration.

STEP 2: SEARCH EXTERNALLy

External search is aimed at finding existing solutions to both the overall prob­lem and to the sub p roblem s identified during the problem clarification step.

CHAPTER 5: CONCEPT GENERATlON 85

While external search is the second ste p in the concep t generati on methodol o­gy, this se quen tial labeling is deceptive; external search occurs continuaIlythroughout the development process. Implemen ting an existing so lu tio n is usu­aIly quick er and cheape r than d eveloping a new so lu tio n. Liberal use of exi stingso lu tio ns aIlows th e team to focus its creative e nergy o n the cr itical subproblcmsfor wh ich th ere are no existing solutions. Furthe rrn o re , a co nventio na l sol utionto one su bp roble m ca n frequen tly be com bined with a novel so lutio n to anoth­el' subproblem to yield a su perio r overalI design oFo r this reason ex te rn al searchincludes d etailed evaluation not only of directly com pe ti tive products but also oftechnologies used in products with rel ated subfunctions.

The exte rnal search for solutions is essentialIy an information-gatheringprocess. Available time and res ources ca n be optimized by u sing a n expand-and­focus strategy: first expand th e scope of th e searc h by broadly gathering informa­tion that might be related to the problem and th en Jocus the scope of the searchby exploring the promisin g directions in m ore d etail. Too much of eithe rapproach wiII make th e exte rnal searc h inefficient,

There are a t least five good ways to gather information from external sources:lead u se r interviews, expen consultation, patent search es, literature searches ,and competitive benchmarking.

Interview Lead Users

While identifying cus to mer needs, th e team may h ave sought o u t 0 1' encoun­tered lead u sers. Lead usa s are those u sers of a class o f product who experienceneeds m onths 01' years before th e m ajority of th e m arketplace and sta n d to ben­efit substantialIy from a product innovation (vo n Hippel , 1988 ) . Frequentlvthese lead users wiIl h ave already in vented so lu tio ns to meet th eir n eeds. This ispanicularl y true among highly te chnical user co m m u n ities, su ch as those in th emedical 0 1' scie n tific fields. Lead users may be sought out in the m arket for whi chthe team is developing the new product, 01' th ey may be found in markets forproducts implementing so me of the su bfu nctio ns of the product,

In the hand-held nailer case , th e nail er team co ns u lted with th e building con­tractors from the PBS tel evision se ries This Old House in order to so lic it new con­ce p ts. These lead u sers made many inte resting observations about the weakness­es in existing tools, but in this case did not provide many new product concepts,

Consult Experts

Experts with knowledge of one 01' m ore of the subproblems not o n ly can provideso lution con ce p ts directl y but also ca n redirect th e se arch in a m ore fruitful arca.Experts may include professionals a t firms manufacturing related products, pro­fessional cons ulta n ts, un iversity faculty, and technical representatives of suppli­ers. These p eople can be found by calIing universitie s, by caIling companies, andby looking up authors of articles, While finding exper ts can be hard work, it isalmost always less time-eonsuming th an re-ereating ex isting knowlcdge.

86 PAODUCT DeSIGN ANO OEVf.LOPMUcT

Most experts are willing to talk on the telephone or meet in person for anhour or so without charge. In general, consultants will expect to be paid for timethey spend on a problem beyond an initial meeting or telephone conversation.Suppliers are usually wilIing to provide several days of effort without direct corn­

pensation if they anticípate that someone will use their product as a componentin a designo Of course , experts at directly competing firms are in most casesunwilling to provide proprietary information about their product designs. Agood habit to develop is to always ask people consulted to suggest others whoshould be contacted. The best information often comes from pursuing these"second generation" leads.

The nailer design team consulted dozens of experts, including a rocket fuelspecialist at Morton-Thiokol, electric motor researchers at MIl', and engineersfrom a vendor of gas springs. Most of this consultation was done on the tele­phone, although the engineers from the spring vendor made two trips to visit theteam, at thcir company's expense.

Search Patents

Patents are a rich and readily available source of technical information contain­ing detailed drawings and explanations of how many products work. The maindisadvantage of patent searches is that concepts found in recent patents are pro­tected (17 years in the United States), so there may be a royalty involved in usingthem. However, patents are also useful to see what concepts are already protect­ed and must be avoided or licensed. Concepts con tained in foreign patents with­out global coverage and in expired patents can be used without payment of ro y­alties.

The formal indexing scheme for patents is difficult for novices to navigate .Fortunately, several databases (e .g., Lexis/Nexis, Dialog) contain the actual textof all patents. These text databases can be searched electronically by key words.Key word searches can be conducted effectively with only modest practice andare remarkably effective in finding patents relevant to a particular product.Copies of U.S. patents including iIlustrations can be obtained from the U.S .Patent and Trademark Office for a few dollars (call 703-557-3158) .

A U.S. patent search in the area of nailers revealed several interesting con­cepts. One of the patents described a motor-driven double-flywheel nailer. Oneof the ilIustrations from this patent is shown in Exhibit 5. The design in thispatent uses the accumulation of rotational kinetic energy in a flywheel which isthen suddenly converted into translational energy by a friction clutch. The ener­gy is then delivered to the nail with a single impact of a drive pino

Search Published Literature

Published literature includes journals; conference proceedings; trade maga­zines; government reports; market, consumer, and product information; andnew product announcements. Literature searchcs are therefore very fertilesources of existing solutions.

/26

CHAPTER 5 : CONCEPl OENERATlON 87

EXHIBIT 5 Concepl from motor-driven double-flywheel nailer palenl (U.S. Patent 4,042,036). The accornpa­nying texl describing the patent is nine pages long.

Electronic database searches are frequently the most efficient way to gatherinformation from published literature. These databases are available from on­line sources or on mass sto rage devices (e .g., CD-RüM) in the library, Currcntly,most databases store only abstracts of articIes and not the full text and diagrams.A follow-up search for an actual article is often needed for complete informa­tion. The two main difficulties in conducting good database searches are deter­mining the key words and limiting the scope of the search . There is a trade-offbetween the need to use more key words for complete coverage and the need torestrict the number of "matches" to a manageable number.

Handbooks cataloging technical information can also be very useful refer­ences for external search. Examples of such engineering references are Mark 'sStandard Handbook o/ Mechanical Engineeri.ng, Perry 's Chemical Engineers' Handbook,and Me chanisms and Mechanical Devices Sourcebook.

• ~ftOOUCT DESIGN ANO OEVELOPMENT

Thc nailer tearn found several useful anides related lo the subproblems,induding articles on energy storage describing flywheel and battery technolo­gies. In a handbook they found an impact tool mechanism that provided a use­fuI energy conversion concept.

Benchmark Related Products

In the con text of concept generation, benchmarking is the study of existing prod­ucts with functionality similar to that of the product under development or tothe subproblems on which the team is focused. Benchmarking can reveal exist­ing concepts that have been implemented to solve a particular problern, as wellas information on the strengths and weaknesses of the competition.

At this point the team wiIl likc1y already be familiar with the competitive anddosely reIated products. Products in other markets, but with related functional­ity, are more difficult to find. One of the rnost useful sources of this informationis the Thomas Register ofAmerican Manufacturers, a directory of manufacturers ofindustrial products organized by product type. Often the hardest pan of usingthe Thomas Register is finding out what related products are actually called andhow they are cataloged.

For the nailer, the dosely related products included a single-shot gunpowder­actuated tool for nailing into concrete, an electrical solenoid-actuated tacker, apneumatic nailer for factory use, and a palm-held multi-blow pneumatic nailer.The products with related functionality (in this case, energy sto rage and co nve r­sion) included air bags and the sodium azide propellant used as an energysource, chemical hand warmers for skiing, air rifles powered by ca rbon dioxidecartridges, and portable computers and their battery packs. The team obtainedand disassembled most of these related products in order to discover the gener­al concepts on which they were based, as well as other more detailed informa­tion, including, for example, the names of the suppliers of specific components.

External search is an important method of gathering solution concepts. Skillin conducting external searches is therefore a valuable personal and organiza­tional asset. This ability can be developed through careful observation of th eworld in order to develop a mental database of technologies and through thedevelopment of a network of professional contacts. Even with the aid of person­al knowledge and contacts, external search remains "detective work" and is co m­pleted most effectively by those who are persistent and resourceful in pursuingleads and opportunities,

STEP3:SEARCHINTERNALLV

In ternal search is the use of personal and team knowledge and creativity to gen­erate solution concepts. The search is internal in that alI of the ideas to emergefrom this step are created from knowledge already in the possession of the team.This activity may be the most open-ended and creative of any in new productdevelopment. We find it useful to think of internal search as a process of retriev­ing a potentially useful piece of information frorn ories memo!)' and then adapt-

CHAPTHI ~ CONCEPT GENERATIO~ 89

in g that information to the p roblem at hand. This process can be carried out hyindividuals working in isol ation or by a group of people working together.

Four guidelines are useful for improving both individual and group internalsearch:

Suspendjudgment. In most aspects of daily life , success d epends on an ab ili­ty to quickly evaluate a se t of alternatives and take action. For exarnple ,none of us would be very productive if deciding what to wear in the morn­ing or what to eat for breakfast involved an extensive period of generatingalternatives before m aking ajudgment. Because most decisions in our day­t<H1ay uves nave implications of onIy a few minutes or hours, we are accus­tomed to making d ecisions quickly and moving o n. Concept generation isfundamentalIy different. We have to live with the consequences of productcon cep t decisions for years. As a result, suspending evalu a tio n for the d aysor weeks required to generate a large set of alternatives is critical to success.The imperative to suspend judgment is frequentIy translated into the rulethat during group concept generation sessions no criticism of concepts isalIowed. A better approach is for individuals perceiving weaknesses in con­cepts to channel any judgmental tendencies into suggestions for improve­ments or alternative concepts.

2 Generate a lot o/ ideas. Most experts believe that th e more ideas a team gen­erates, the more likely the team is to explore fulIy the so lu tio n space .Striving for quantity lowers th e expec ta tions of quality for any particularidea and therefore m ay encou rage people to share id eas th ey may otherwiseview as not worth mentioning. Further, each idea ac ts as a stirnulus forother ideas, so a large number of ideas h as th e potential to stimulate eve nmore ideas.

3 Welcome ideas that may seem infeasible. Ideas whi ch initialIy appear infeasiblecan often be improved, "d e bugged ," or "repaired" by other members of th eteam. The more infeasible an idea, the more it stretches the bouridaries o fthe solution space and encourages the team to think of the limits of possi­bility. Therefore, infea sibl e ideas are quite valuable and their expressionshould be encouraged.

4 Use graphical and physical media. Reasoning about physical and geometricinformation with words is difficult. Text and verbal language are inherent­ly inefficient vehicles for describing physical e n tities . Whether working as agroup or as an individual , ab u nd an t sketching surfaces should be available .Foam, c1ay, ca rd board , a nd other three-dimensional media mar also beappropriate aids for p roblems requiring a d eep understanding of form a ndspa tial relationships.

Both Individual and Group Sessions Can Be Useful

Formal studies of group and individual problem solving suggest that a set of peo­pIe working alone for a period of time will generate more a nd better conceptsth an the same people wo rking together for th e sam e tim e period (McGra th.

90 PRODUCT DESIGN AND DEVElOPMENT

1984). This finding is contrary to the actual practices of th e many firms that per­form most of th eir concept generation activities in group sessions. Our informalobservations confirm the formal stud ies , and we believe that team membersshould spend at least sorn e of th eir co nce p t generation time working alone. Wealso bclieve that group sess io ns are critica l for building co nsensus, communicat­ing informati on, and refining co ncep ts. In an id eal se tt ing, each individual onthe team would spend severa l hours working alo ne and then the group wouldget together to di scuss and improve th e co nc ep ts generated by individuals.

However, we also know th at th ere is a practical reason for holding group con­cept generation sessions: it is one way to guarantee that the individuals in thegroup will devote a certain amount of tim e to the task. Especially in very intenseand demanding work environments, without sched u ling a meeting, few peoplewill allocate several hours for concentrated individual effort on generating newconcepts. The phone rings, people interrupt, urgent problems demand atten­tion. In certain environments, sched uled group sessions may be the only way toguarantee that enough attention is paid to the concept generation activity.

The nailer team used both individual effort and group sessions for internalsearch. For example, during o ne particular week eac h member was assigned oneor two subproblems and was expected to develop at least ten solu tion concepts.This divided the concept generati on work among all members. The group thenmet to discuss and expand on th e individually generated concepts. The morepromising concepts were investigated further.

Hints for Generating Solution Concepts

Experienced individuals and teams can usually just sit down and begin generat­ing good co nce pts for a subp roblem . O ften these people have developed a set oftechniques th ey use to stirnulate their thinking, and these techniques havebecome a natural part of their problem-solving proce ss. Novice product devel­opment professionals may be aided by a se t of hints th at stirnu la te new ideas orencourage relationships among ideas. VanGundy (1988), von Oech (1983) , andMcKim (19 80) give d ozens of helpful suggestions. H ere are sorn e hints we havefound to be h elpful:

• Make analogies. Experienced d esigners always ask th emselves what otherd evices so lve a related p roblem. Frequently th ey will ask themselves if thereis a natural or biological an alogy to the problem. They will think aboutwhether their problem exists a t a much larger or sm aller dimensional scalethan that whi ch th ey are consid ering. They will ask wh at devices do sorne­thing similar in an unrelated area of applicati on. The nailer team, whenposing th ese questions, realized tha t construc tio n pile drivers are similar tonailers in some respects . In foll owing up on th is idea, th ey developed theconcept of a multiblow tool.

• Wzsh ami uumder. Beginning a th ought or co m me n t with "1 wish we could. . ." or "1 wonder what would happen if . . ." helps to stimu late oneself or

CHAPH.f\ 5 CO"~U'1 Ut~"~"A'ION 91

the group to consider new possibilities. These questions cause reflection onthe boundaries of the problem. For example, a member of the nailer team,when confronted with the required length of a raíl gun (an electromagnet­ic device for accelerating a projectile) for driving a nail, said, "1wish the toolcould be I meter long." Discussion ofthis comment led to the idea that per­haps a long tool could be used like a cane for nailing decking, allowingusers to remain on their feet.

• Use related stimuli. Most individuals can think of a new idea when presentedwith a new stimulus. Related stimuli are those stimuli generated in the con­text of the problem at hand. For exarr.ple, one way to use related stimuli isfor each individual in a group session to generate a list of ideas (workingalone) and then pass the list to his or her neighbor. Upon reflection onsomeone else's ideas, most people are able to generate new ideas. Otherrelated stimuli inelude customer needs statements and photographs of theuse environment of the product.

• Use unrelated stimuli. Occasionally, random or unrelated stimuli can beeffective in encouraging new ideas. An example of such a technique is tochoose, at random, one of a collection of photographs of objects, and thento think of sorne way that the randomly generated object might relate to theproblem at hand. In a variant ofthis idea, individuals can be sent out on thestreets with an instant camera ro capture random images for subsequent usein stimulating new ideas. (This may also serve as a good change of pace fora tired group.)

• Set quantitative goals. Generating new ideas can be exhausting. Near the endof a session, individuals and groups may find quantitative goals useful as amotivating force. The nailer team frequently issued individual concept gen­eration assignments with quantitative targets of lOto 20 concepts.

• Use the galiery method. The gallery method is a way to display a large numberof concepts simultaneously for discussion. Sketches, usually one concept toa sheet, are taped or pinned to the walls of the meeting room. Team mern­bers circulate and look at each concept. The creator of the concept mayoffer explanation , and the group subsequently makes suggestions forimproving the concept or spontaneously generates related concepts. Thismethod is a good way to merge individual and group efforts.

Exhibit 6 shows sorne of the solutions the nailer team generated for the sub­problems of (1) storing or accepting energy and (2) delivering translationalenergy to a nail.

STEP 4: EXPLORE SYSTEMATICALLY

As a result of the external and internal search activities, the team will have col­lected tens or hundreds of concept fragmenls--solutions to the subproblems.Systematic exploration is aimed at navigating the space of possibilities by orga-

(a) Solutlons lo Subproblem01 Storlng or Acceptlng Energy

(b) Solutlons to Subproblem 01 ApplylngTranslatlonal Energy to Nall

• Selt-regulating chemical reactionemitting high-pressure gas

• Carbide (as tor lanterns)

• Combusting sawdust trom job site

• Gun powder

• Sodium azide (air bag explosive)

• Fuel-air combustion (butane , propane ,acetylene, etc.)

• Compressed air (in tank or tromcompressor)

• Carbon dioxide in tank

• Electric wall oullet and cord

• High-pressure oilline (hydraul ics)

• Flywheel with charg ing (spin-up)station

• Battery pack on tool, belt, or tloor

• Fuel cell

• Human power : arms or legs

Single Impact

Mulliple Impacls(tens or hundreds)

Mull iple Impacls(hundreds or Ihousands)

• Melhane trom decomposing organicmalerials

• "Burning" like lhal ot chemical handwarmers

• Nuclear reactions

• Cold tusion

• Solar eleclr ic cells

• Solar-steam conversion

• Stearn supply line

• Wind

• Geolherm al

Push

Twist-Push

I

1EXHIBIT 6 Sorne of Ihe solutions lo Ihe subproblems 01 (1) storing or accepting energy and (2) delivering

translalional energy to a na i!.

92

CHAPU:" S CO"<-t.~1 utNE"'''''10N 93

nizing and synthesizing these solution Iragments. The naile r team focused o nthe energy storage , conversion, and delivery subproblems and had generateddozens of concept fragments for each su bp ro blem . One approach to organizin gand syn thesizing th ese fragments would be to considel' all of th e possible corn­binations of the fra gments associat ed with each su bproblem ; however, a littlearithmetic reveals the impossibility of this approach. Given the three subprob­lems on whi ch the team focused and an average of 15 fragments for ea ch su b­problem, the team would have to co nsid er 3,375 combinations of fragmen ts (15X 15 X 15). This would be a daunting task for even the most enthusiastic team.Furthermore, the team would quickly discover that many of the combinations donot even make sense. Fortunately, there are two specific tools for managing thiscomplexity and organizing the thinking of the team: the concept cLassification treeand th e concept combination tableoThe ~Iassification tree helps the team divide th epossible solutions into independent categories. The combination table guidesthe team in selectively considering combinations of fragments.

Concept Classification Tree

The concept c1assification tree is used to divide the entire space of possible solu­tions into several distinct classes which wilI facilitate comparison and pruning.An example of a tree for the nailer example is sh own in Exhibit 7. The branch­es of this tree correspond to different energy so u rces.

The classification tree provides at least four important benefits:

1 Pruning o/ less prornising branches: If by studying the c1assification tree th eteam is able to identify a solution approach th at does not appear to havemuch merit, th en this approach can be "p ru ned" and th e team can focus itsattention on the more promising branches of the tree. Pruning a branch ofthe tree requires sorne evaluation ami judgment and sh ould thereforc bedone carefulIy, but the reality of product development is th at there ar e lim­ited resources and that focusing the available resources on the most prornis­ing directions is an important success factor. For the nailer team , thenuc1ear energy so u rce was pruned from cons idera tio n . Although th e teamhad identified sorne very intriguing nuc1ear d cvices for use in powering ar ti­ficial hearts, th ey felt that th ese devices would not be economically practi­cal for at least a d ecade and would probably be hampered by regulato ryrequirements indefinitely.

2 Identifirotion o/ independent approaches lo the problem: Each branch of th etrce can be considered a different approach to solving the overalI problem.Sorne of these approaches may be almost com ple tely independent of eachother. In these cases, the team can cleanly divide its efforts among two 01'

more individuals 01' task forces. When two approaches both look promising,this division o f effo rt can reduce the complexity of the concep t generationactivities. It also may engender sorne healthy competition among theapproaches under consideration . The nailer team found that both the

94 PROOUCT OESIGN ANO OEVElOPMENT

Sto re orAcceptEnergy

Fuel-Air Systems

Explosive Systems

WallOutlet

Battery

Fuel Cell

Nuclear

EXHIBIT 7 A c1assification tree tor the nailer energy source concept fragments.

chem ical/exp losive branch an d th e elec tri ca l branch ap peare d quitep romising. They assigned th ese two approaches to two different sub teamsan d pursu ed th em independently for seve ral wecks.

3 Exposure of inappropriate emphasis un certain branches: Once the tree is con­structe d, the team is able to refl ect quickly on wheth er the effo rt applied toeac h branch has been appropriatcl y allocated. The n ailer team recognizedthat th ey had applied very little effo rt to thinking abou t hydraulic energysources an d conversio n tech nolo gies . This recognition guided them tofocus o n this branch of th e tree for a few d ays.

4 Refinement of the problem decomposition for a particular branch: Sometimes aproblem decomposition can be usefully tail ored to a particular approach tothe problem. Consider th e branch of th e tree co rrespond ing to the electri­ca l energy sou rce . Based o n ad d itional investigation of the nailing proccss,the team d etermined that the in stantaneous power delivered during th enailing process was about 10,000 watts for a few milliseconds and so exceedsthe power wh ich is available from a wall ou tle t , a battery, or a fuel ccll (ofreasonable size , cost, and m ass). They concluded, therefore , that e nergyrnust be accu m ulate d ove r a su bsta n tial period of the nailing cycle (say 100

CHAPTER 5 . CONCEPT GENERATlON 95

ElectricalEnergy Convert

Energy loTranslational

Energy

AccumulaleTranslalional

Energy

ApplyTranslalional

EnergytoNail

Energy. Applled to Nail

EXHIBIT 8 A new problem decomposition assuming an electrical energy source and the accumulalion 01energy in the mechanical domaín.

milliseconds) and then suddenly released to supply the required instanta­""VU:> PV W"l lV ~l iV" the uail. This quick analysis led the team to ad d a sub­function (Uaccumulate translational energy") to their function diagram (se eExhibit 8) . They chose to add the subfunction after the conversion of elec­tri cal energy to mechanical energy, but briefly considered the possibility ofaccumulating the energy in the clectrical domain with a capacitor. This kindof refinement of the function diagram is quite common as the team makesmore assumptions about the approach and as more information is gathered.

The classification tree in Exhibit 7 shows the alternative solutions to the cner­gy source subproblem. However, there are other possible trees. The team mighthave chosen to use a tree classifying the alterriative solutions to the energy deliv­ery subproblem, showing branches for single impact, multiple impact, or push­ing. Trees can be constructed with branches corresponding to th e solution frag­ments of any of the subproblems, but certain classifications are more useful. Ingeneral, a subproblem whose solution highly constrains the possible solutions tothe remaining subproblems is a good candidate for a classification tree. Forexample, the choice of energy source (electrical, nuclear, pneumatic , etc.) co n­strains whether a motor or a piston-eylinder can be used to convert the energyto translational energy. In contrast, the cho ice of energy delivery mechanism(single impact, multiple impact, etc.) does not greatly constrain the solutions tothe other subproblems. Reflection on which subproblem is likely to most highlyconstrain the solutions to the remaining subproblems will usually lead to one ortwo clear ways to construct the classification tree.

Concept Combination Table

The concept combination table provides a way to consider combinations of solu­tion fragments systematically. Exhibit 9 is an example ofa combination table thatthe nailer team used to consider the combinations of fragments for the electri­cal branch of the classification tree. The columns in the table correspond to thesubproblems identified in Exhibit 8. The entries in each column correspond tothe solution fragments for each of these subproblems derived from external andinternal search. For example, the subproblem of converting electrical energy totranslational energy is the heading for the first column. The entries in this columnare a rotary motor with a transmission , a linear motor, a solenoid, and a rail gun.

Potential solutions to the overall problem are formed by co m bin in g one frag-

96 PROOUCT OESIGN ANO OEVELOPMENT

ConvertElectrlcal Energy to

Translatlonal EnergyAccumulate

Energy

ApplyTranslationalEnergy to Nail

Rotáry motor withtransmission

I Linear motor

¡;¡¡;;;;:S;;¡¡;..,'" • f ~,

I Spring 11J., L..~ ..._~:;;¡;;",.s§ .,,·.n "' ;' ~N " "''''' > ' " _ ¡ ""~ ...."a'o;;'

IN=~:::::':s .* ~

I Single impact tt>" ~ ;¡: ..¡¡;;• ..,. lZi?",D _ k1: ",...,,:¡'.

Multiple impacts

EXHIBIT 9 Concepl combinalion lable lor lhe hand-held nailer.

ment from each column. For the nailer example, there are 24 possible cornbi­nations (4 X 2 X 3) . Choosing a combination of fragments does not lead spon­taneously to a solution to the overalI problem. The combination of fragmentsmust usualIy be developed and refined before an integrated so lu tio n emerges .This development may not even be possible or may lead to more th an one sol u­tion, but at a minimum it involves additional creative th ought. In so rne ways, thecombination table is simply a way to make forced associatio ns among fragmentsin order to stim ulate further creative thinking; in no way does the mere act ofselecting a combinatio n yield a comple te solution.

Exhibit 10 shows a sketch of a concept arising from the combination of thefragments "solenoid," "spring," and "múltiple impacts. " Exhibit 11 shows sornesketches of concepts arising from the combination of the fragments "ro ta rymotor with transmission," "sp ring," and "single impact." Exhibit 12 shows aske tch of a concept arising from th e co mbinatio n of "rotary motor with trans­mission, " "spring," and "multiple impacts ." Exhibit 13 shows sorne sketches ofconcepts arising from the cornbination of "lin ear motor," "moving mass," "an dsingle impact. "

Two guidelines make the concept combination process easier. First, if a frag­ment can be eliminated as being infeasible before combining it with other frag­rnents, then the number of com binations th e team needs to consider is dramat­icalIy reduced. For exampIc, if the team could determine that the rail gun wouldnot be feasible under any condition, they could reduce the number of co m bi­nations from 24 to 12 . Second, the concept combination table should be con­centrated on the subproblems th at are coupled. Co upled subproblems are those

CI'4Al"ru.. s c..Ollft,..tt-' úENE.RATION 97

ConvertElectrical Energy to

Translatlonal EnergyAccumulate

Energy

ApplyTranslationalEnergy to Nall

Single impactSpringRotary motor withtransmission

Linear motor Multiple impacts

Solenoid Push nail

Rail gun J

EXHIBIT 10 In this solution concept, a solenoid compresses a spring and then releases il repealedly inorder to drive the nail wilh multiple impacls.

whose solutions can only really be evaluated in combination with the solutionsto other subproblems. For example, the choice of the specific electrical energyso urce to be used (e.g., battery versus wall outlet), although extremely critical, issomewhat independent of the choice of energy conversion (e.g., motor versussolenoid). Therefore, the concept combination table does not need to contain acolumn for the different types of electrical energy sources. This reduces thenumber of combinations the team must consider. As a practical matter, conceptcombination tables lose their usefulness when the number of columns exceedsthree or four.

98 PROOUCT OESIGN ANO OEVELOPMENT

ConvertElectrical Energy to

Translational Energy

Rotary motor with ~transmission P--+--!

Linear motor

I Raíl gun t""' ;;"""'-"'<"';$,."".0"""'"'_''''1 ' >' :''0,,",' ''''''''''

AccumulateEnergy

Spring

ApplyTranslatlonalEnergy to Na"

Single impact

Multiple impacts

EXHIBIT 11 Multiple solut ions arising Irom the combination 01a motor with transmission, a spring, and singleimpact. The motor winds a spr ing, accumulating potential energy which is then delivered lo Ihenail in a single blow.

Managing the Exploration Process

The c!assificati on trce and combinati on tables are tools that a team can usesomewhat flexibly. They are simple ways to organize thinking and guide the ere­ative cnergies of the team. Rarely do teams generare only one classification treean d one concept combination tableo More typi caIly the team wiIl create severalalternative c!assification trees and several concept combination tabIes.Intersp ersed with this exploratory activity may be a refining ofthe original prob-

ConvertElectrical Energy to

Translational Energy

Rotary motor with ¡,............f---ltransmission

Linear motor

Solenoid

AccumulateEnergy

.ullWCt'" OENERATION

ApplyTranslationalEnergy to Na/l

Single impact

Multiple impacts

Push nail ~' _ ..~ .~ ... - ,, · .. cc.,",,"·__~...:;:O;::;:;~

99

EXHIBIT 12 Solut ion from the combination of a motor with transmission, a spring, and multiple impacts. Themotor repeatedly winds and releases the spring , storing and delivering energy over several blows.

lem d ecompositi on or th e pursui t of ad d itio nal internal or exte rnal sea rc h. Theexp loration ste p of concep t generatio n usu all y acts more as a guide for fu r thercrea tive thinking than as th e final step in th e process.

Recall th at a t the b eginning of th e pmcess th e team ch ooses a few subprob­lems o n whi ch to focus attention. Eve n tu ally the team must return to address allof the subp roble rns . This usu ally oc curs afte r th e team has narrowed th e rangeof alternatives for th e critical subproblems. The nailer team narrowed its alter­natives to a few chemical and a few electric con ce p ts and then refined them byworking out the user interface , industri al d esign, and co n figu ra tio n issues. Oneof the resulting concep t d escriptions is shown in Exhibit 14.

100 PAODUCT DESIGN ANO OEVELOPMENT

ConvertElectrical Energy to

Translational EnergyAccumulate

Energy

ApplyTranslationalEnergy to Nail

Rotary motor withtransmission

Linear motor

Spring

Moving mass

Single impact

Multiple impacts

Solenoid

Aail gun

PHA5l0lINE....RMOTOR.

r1I1I1I1I1 lINE.....'"1 MOTOR

II

IIII I

'ni~I

O " /NAll S ? cou.?'é1 / .

1/ '

I Push n~ il !:~? , ~,.;am~: ,~." b~~..,.. t;: "",<.!'

IIWRA,PPEDM

L1"""R./,\O'lQR

EXHIBIT 13 Solutions frorn the combination ot a linear motor, a moving mass, and single impacto A linearmotor accelerates a massive hammer, accumulating kinetic energy which is delivered to thenail in a single blow.

STEP 5: REFLECT ON THE SOLUTIONS AND THE PROCESS

Although the refleetion ste p is plaeed here al the end for eonvenience in pre·sentation, refl ection should in fact be pcrformed throughout the whole process.Questions to ask include : Is the team developing confidenee that the solutionspace has been fulIy exp lo red? Have ideas from everyone been aeeepted andintegrated in the process? Are there alternative ways to deeompose the problem?Have external sourees been thoroughly pursu cd? Are there alternative functiondiagrams?

Thc nailer team members diseussed whether they had focused too much

I

~ ... "P Ifl'< s I.O",c.tl"l GENERATION 101

EXHIBIT 14 One of several refined solution eoneepts. (lIIustration eourtesy of Produet Genesis Ine.)

a tte n tio n on the energy storage and conversion issu es in the tool while ignoringth c user interface and overall configuration. They decided that the energy issuesremained at the core of th e problem and that their decision to focus on theseissues first was justified. Thcy also wondered if they had pursued too manybranches ofthe classification tree. Initially they had pursued electrical, chemical,and pneumatic concepts before ultimately settling on an electric concept. Inhindsight, the chemical approach had some obvious safety and customer per­cep tion shortcomings (they were exploring the use of explosives as an energysource) . They decided that although they liked some aspects of the chemicalsolution they should have eliminated it from considerati on earlier in the process,allowing more time to pursue so me of the more promising branches in greaterdetail.

SUMMARV

A product concept is an approximate description of th c technology, workingprincipies, and form of the product. The degree to which a product satisfies cus-

102 PRODUCT DESIGN ANO OEVELOPMENT

tomers and can be successfu Ily com mercialized depends to a large measure onthe quali ty of th e underlying concepto

• The concept generation process begins with a set of customer needs andtarget sp ecifications and re sults in a set of product concepts from which theteam will m ak e a final selection.

• In most cases, an effective devel opment team will generate hundreds ofconcepts, ofwhich 5 to 20 wiIl merit serious consideration during the sube­quent concept sclecti on activity.

• The concept generation methodology presented in this chapter co nsists 01"five steps:

1 Clarib the problem. Understand the problem and d ecompose it into sim­pler subproblems.

2 Search externally. Gather information from lead users, experts, patents,published Iiterature, and related products .

3 Search internally. Use individual an d group methods to retrieve and adaptthe kn owledge of the team.

4 Explore systematically. Use cIassification trees and combination tables toorganize the thinking of th e team an d to syn thesize solution fragments.

5 Reflect on the solutions and the process. Identify opportunities for improve­ment in subsequent iterations or future projects.

• AIthough co nce p t generation is an inherently creative process, teams canbenefit from using a structured m ethodology. Su ch an approach allows fuIlexplorati on of th e design space an d reduces th e chance of ove rsigh t in thetypes of solution concepts co nsid e red . It also acts as a map for those teammembers who are less experienced in design problem solving.

• Despite the linear presentation of the co ncep t generation process in thischapter, the team will likeIy return to eac h step of the process several times.It eration is particularly com mo n when th e team is developing a radicallynew producto

• Professionals who are good a t co n cep t gene ration seem to always be in greatdemand as team m embers. Contrary to popular opinion, we beIieve con­ce p t generation is a sk ill th at can be learned an d deveIoped.

REFERENCES AND BIBLlOGRAPHY

Pahl and Beitz were the driving for ce behind structured design methods in Germany.We adaptomany 0 1" th eir ideas for fun ctional de composition.

Pahl, Gerhard, and Wolfgang Beitz, Engineering Design , K. WaIlace (ed.), Springer­Verlag, New York, 1984. ISBN 0-387-13601-0.

Hubka and Eder have written in a detailed way about systematic concept generationfor technical products.

EXERCISES

CHAPTEA 5 : CONCEPT GENEAATION 103

Hubka, Vladimi r, and W. Emst Ed cr, Theory of Technical Systems: A Total ConceptTheory [or Engineering Design, Springcr-Verlag, New York, 1988. ISBN 0-387-17451­6.

von Hippcl reports on his empirical research on the sources of ncw product con­cep ts. His central argument is that lead users are the innovators in many markets.

von Hippel, Eric, The Sources of Innovation, Oxford University Press, New York,1988. ISBN 0-19-504085-6.

VanGundy presents dozens of methods for problem solving, many of which aredireetly applieable to product eoneepl generalion .

VanGundy, Arthur B., jr., Techniques of Structured Problem Solving, seeond edition,Van Nostrand Reinhold, New York, 1988 . ISBN 0-442-28847-6.

Author von Oeeh provides dozens of good ideas for improving individual and grouperealive performance.

von Oeeh, Roger, A Whack 071 the Side of the Head: How to Unlock Your M ind [orInnovation, Wamer Books, New York, 1983. ISBN 0-446-38000-8.

MeKim presents a holistic approaeh lo developing creative thinking skills in individ­uals and groups.

McKim, Robert H., Experiences in Visual Thinking, second edition, Brooks/ColePublishing, Monterey, CA, 1980. ISBN 0-818-50411-0.

MeGrath presents studies eomparing the relative performance of groups and indi­viduals in generaling new ideas.

MeGrath,joseph, E., Groups: Interaction and Performance, Prenlice-Hall, EnglewoodCliffs, Nj, 1984. ISBN 0-13-365700-0.

Engineering handbooks are handy sources of informalion on standard teehniealsolutions. Three good handbooks are:

Avallone, Eugene A., and Theodore Baumeister III (eds.), Mark's StandardHandbook of Mechanical Engineering, ninth edition, McGraw-Hill, New York , 1987 .ISBN 0-070-04127-X.

Green, Don W. (ed.), Perry's Chemical Engineers ' Handbook, sixth edition, MeGraw­Hill, New York, 1984 . ISBN 0-070-49479-7.

Chironis, Nieholas P., Mechanisms and Mechanical Deoices Sourcebook, McGraw-Hill,New York, 1991. ISBN 0-07-010918-4.

The Thomas Register is a eategorized listing of manufacturers of all types of industrialproduets.

Thomas Register of American Manufacturers, Thomas Publishing Co mpany, NewYork, published annually. ISBN 0362-7721.

Decompose the problem of designing a new barbecue grill. Try a functionaldeeomposilion as well as a deeomposition based on the user interaclions with theproduct,

2 Generate 20 concepts for the subproblem "prevent fraying of end of rope" as partof a system for cutling lengths of nylon ro pe from a spool.

104 PRODUCT OESIGN ANO OEVELOPMENl

3 Prepare an external seareh plan for th e problem o f perrnan ently applying serialnumbers to plastie produets.

THOUGHT QUESTIONS

1 What are th e prospeets for eomputer sup por t for eoneept ge nera tio n activitiesiCa n you think of any eomputer tools that would be especi alIy helpful in thisprocess?

2 What would be the rel ative advantages and disadvantages of involving actual cus­tomers in th e concept generation process?

3 For whai l} p.:.suf p i-0 düct.> -.•ou l.l the initial focus of the concept generation activoity be on th e form and user interface of the product and not on the co re technol­ogy? Describe specific examples.

4 Co uld you apply the five-step methodology to an everyday problem like choosingthe food Io r a picnic?

5 Consider th e task of generating new co nce p ts for the problem of d ealing withleaves on a lawn. How would the assumptions and problem decomposition differfor a manufacturer of plastic bags, from those of a manufacturer of lawn tools andequipment, from those of a company responsible for maintaining golf coursesaround the world? Should the context of the firm dictate the way concep t genera­tion is approached?