QUALITY FUNCTION DEPLOYMENT

Introduction
What is QFD?
QFD is a system for designing a product or service based on customer demands that involves all
members of the producer or supplier organisation. In Japanese, deployment refers to an extension
or broadening of activities and hence Quality Function Deployment means the responsibilities for
producing a quality item must be assigned to all parts of a corporation. It is sometimes referred to as
the most advanced form of Total Quality Control, Japanese style. The system can be understood by
defining each of the terms in Quality Function Deployment within the context of QFD.
Quality - Meeting Customer Requirements
Function - What Must Be Done - Focusing the attention
Deployment - Who Will Do It, When
Scope of QFD
QFD theory started in 1972 at Mitsubishis Kobe Shipyard when they began using a matrix that put
customer demands on the vertical axis and the methods by which they would be met on the
horizontal axis. This was recognised almost immediately as a major breakthrough. Since that time the
Japanese have developed the system to encompass other areas of a secondary nature such as
improved communications between the design and manufacturing departments, considering the
function of the product, potential failure modes, possible new technologies and cost reduction.
Who Uses QFD?
Current Japanese users of QFD concepts include Nissan, Toyota, Komatsu, Nippondenso and
Honda. In the United States of America users include Ford, GM, Chrysler, DEC, TI, 3M, HP,
AT&T Bell Labs, NovAtel, Xerox, Exxon and Dow.
The Purpose of QFD
The purpose of QFD is three fold. Firstly, it allows us to get higher quality products to market faster
and at a lower cost. Secondly, we will achieve customer driven product design and, finally, it will
provide a tracking system for future design or process improvements.
The results we can expect by carrying out the QFD studies are many:
Better understanding of customer needs
Improved organisation on development projects
Improved introduction to production
Fewer design changes late in development
Fewer manufacturing start-up problems
Reputation for being serious about quality
Increased business
Documented product definition based on customer requirements
The System
QFD achieves these results by breaking down customer requirements into segments and identifing
means for achieving each segment. QFD also involves all parts of a company and facilitates
simultaneous product and process design. Finally, it integrates the use of other quality tools such as
Taguchi Methods. The QFD process is as follows:
Customer Requirements
Design Requirements
Component Characteristics
Operations Requirements
Working Procedures

The Customer Requirements

In order for any company to stay in business they must sell their products/services and be able to
rely on repeat business. This can only be done with a backbone of continual satisfaction on behalf of
the customer. In turn, customer satisfaction can only be achieved if we can fulfill the customers
requirements. Hence, the gathering and use of customer requirements is the foundation of QFD. It
follows, therefore, that the first area that must be understood in QFD is that of customer
requirements. As QFD is a technique for translating the customer requirements into a product
design, we must be able to gather these requirements. It is essential, however, that the requirements
we gather are the true customer requirements. We as engineers and managers are so close to our
product that our level of expectation and our values are far removed from the average customer.
Hence, we cannot speak for todays customers. In order to understand the area of customer
requirements we must understand three distinct factors:
The Customer Chain
What do we mean by Customer in QFD
Obtaining Customer Requirements
The Customer Chain
It is necessary in order to obtain as full a picture as possible that we understand who our customer
actually is. For some businesses we may only have one or two types of customer, but for the majority
of todays businesses there will be a chain of customers. This raises the question, what do we mean
by a customer chain? By customer chain we mean that for a simple product there could be more than
one customer. For a simple product the chain may be short i.e. an owner, an operator and a user.
The examples below show simple chains but it is conceivable that they could become much longer
and more complex.
Example

Manufacturers
Breakfast Cereal
Aircraft Seat

Product Customer
Supermarket - Parent who buys Child who eats
Aircraft Manufacturer - Aircraft lease company - Airline - Passenger

If a company has a very complex customer chain it is possible to use a system for its identification
similar to that used for supplier chain analysis. The system for customer chain analysis involves
simply identifying who is the companys immediate customer and who is the end user. Thus, by
asking questions of each it is possible to fill in the customer chain between them.
Internal Customers
There is one type of customer which when considering QFD we must be very careful of, namely, the
internal customer. The reason for this comes from why customer requirements are so important. By
satisfying them we stay in business because people buy our products/services. As such it must always
be the external customers we satisfy as it is they who pay for our products/services. Therefore, if we
include internal customers requirements which then lead to a compromise on an external customer
requirement, we have not satisfied the external customer. The correct answer would be to satisfy the
external customer and then solve the associated internal problems. In other words, we become
customer driven.
Example

The designers at a small boatyard are designing fishing boats. The owner, the external customer, has
asked for an aluminium wheelhouse. Production, the internal customer, is saying they want anything
but aluminium in the vessel because they cannot work aluminium. The designers compromise and
specify a wooden wheelhouse. The result is that the external customer is not satisfied. What the
designers should have done was to specify the aluminium wheelhouse then considered the associated
internal problem. This will automatically happen in QFD if the external customers only are
considered as the internal problems will be highlighted and then dealt with at a later stage.

What is meant by Customer?

In QFD, when we refer to the customer in terms of gathering their requirements, it is not sufficient
just to consider our customer chain. We must consider the individual parts of that chain in more
detail. It is not sufficient, if we want a full picture, just to consider those people who bought our
product; we must look further.
We should consider the following people:
Those who bought our products
Those who bought competitors products
Those who switched to our competitor
Those who are satisfied
Those who are not satisfied
Obtaining Customer Requirements
In the domain of market research there are two types of data, namely, primary and secondary. For
the purposes of QFD we consider in the main primary data techniques. The main reason for this is
as follows:
Secondary data is that which is not directly gathered from field work.
Examples of secondary data are:
Government produced statistics
Company accounts
Commercially produced reports
Business news in newspapers and trade magazines
Although this type of data can be useful to parts of QFD, for customer requirements it is necessary
that the requirements are gathered in the customers own language. That is to say that the words used
in the recording of the requirements should be the actual words used by the customer. The
techniques used to get this type of information are generally referred to as Field Research Techniques
and they produce what is called primary data. Field research techniques can be broken down into
those using questionnaires and those using non-questionnaire techniques. Questionnaire techniques
involve the use of a set of questions which respondents are asked to answer. The type of questions
asked can be very formal and structured, requiring a simple yes, no or dont know response or
they can be semi-structured, requiring an explanation of a customers action.
Before listing the questionnaire techniques, there must be one word of warning about using them in
the context of QFD. As the use of them is to produce the customers requirements, both technical
and emotional, we must be careful not to structure the questionnaires in such a way that they lead the
customer into a particular response or in fact, miss an area of requirement completely.
Questionnaire methods consist of three basic types:
Postal Questionnaires

Postal Questionnaires are impersonal. The respondent is left to fill in a questionnaire alone. It is the
most structured of the questionnaire techniques, and its main advantage is its comparative low cost an interviewer is not required to put the questions directly to a respondent - hence, it is possible to
send the questionnaire to a large number of potential respondents. However, the respondent is under
no obligation or pressure to answer, so may simply throw away the questionnaire or leave it a long
time before replying. As a result, response rates are generally low and replies may take a long time to
come in. As a result, respondents may not represent a true cross section of the population.
Telephone Interviews

Telephone interviewing, because it involves an interviewer putting questions to respondents, often

has the advantage of drawing out answers which are given instantly. It is more expensive than a
postal survey because a trained interviewer must be employed and telephone calls can be costly.
Face-to-face Interviews

Respondents meet an interviewer face to face. As a result, the procedure can be more informal than
the other two techniques, with time to elaborate on the meaning of questions and the reasons that
the answers were given. This is, however, a very costly method in time and money. For nonquestionnaire techniques there are three basic categories:

 Discussion techniques
Diary panels
Audit techniques
For QFD, because we are considering products at the design stage we are interested in only one of
these above categories. The category of interest is the discussion technique. Of this type there are
various different methods.
Clinics

The clinic method is very useful for the purposes of QFD as it provides a variety of feedback
information. A clinic takes the form of grouping customers at a venue where they may see the
product in question, or a mock up, and its competition. The group will encompass all the types of
people referred to previously, those who bought your products and those who did not. The feedback
that these people give can thus be noted throughout the period of the clinic. Much use of clinics has
been made by the car manufacturers. They use clinics to expand on the information gathered from
postal surveys. Selected respondents are asked to bring their cars to a venue where the manufacturer
will also have examples of the competition. The customers can then discuss their likes and dislikes
and compare the competition.
Focus Groups

Focus Groups are small groups of about eight to twelve people. These groups should be a cross
section of the population selected at random who agree to come to a venue for a one or two hour
discussion. These discussions should be led by someone experienced at such things who is capable of
moving the discussion in the right direction so that all issues are covered.
There is one other method for acquiring customer information which is not covered in the above
field techniques. This final technique is one that the Japanese have shown that they are exceptionally
good at: it is called Listening.
Listening

Listening takes the form of engineers and managers simply listening at trade shows and outside retail
outlets. What they must listen for is simply the comments of customers, potential customers and past
customers about the particular product in which they are interested.
The example below shows the customer requirements as gathered from the customer chain for the
aircraft seat example:
Passenger

Comfortable, does not give you bum ache.

Enough leg room.
Lumbar support does not hurt short person.
Does not burn.
Hole in tray for coffee cup.
Comfortable seat belt.
Seat belt feels safe.
Magazines can be easily removed from rack.
Comfortable when you recline.
Does not hit person behind when you recline.
Back can be adjusted to any position and does not slip.
Not dirty.
Nice colour.
Does not soak up a spilt drink.
Does not make you sweat.
Arm rests not too narrow.
Arm rest folds right away.
Airline

Easy/cheap to refit.
Easy to clean.
Easy to move around.
Fit more passengers in.

Passengers feel comfortable.

Fabric in certain colours.
Fabric stain resistant.
Meets regulatory requirements.
Long life.
Low operating cost.
Cheap to service.
Aircraft Manufacturer

Cheap.
Light weight.
Durable.
Easy to fit in plane.
Safe in crash.
Meets regulatory requirements.

Technical And Regulatory Requirements

There are some requirements which the customers will not identify. These requirements will be either
technical or regulatory requirements. Regulatory requirements are such things as government
legislation, safety requirements, quality standard requirements, classification requirements etc.
Technical requirements occur because the company may have some specific plans for the new
product. For example, complete automated assembly, single monocoque construction etc. All these
types of requirements should be included in the requirement side of the chart so that total
satisfaction can occur i.e. customer, company, regulatory. Earlier, it was mentioned that we had to be
careful of internal customers. The technical and regulatory section of the chart is the only place in
which they should appear. The technical and regulatory requirements are added at the bottom of the
chart as will be shown later. In the case of the aircraft seat example, the technical and regulatory
requirements are as follows:
FST 1000 - fire safety standard
16G load case - safety standard
Max allowable deflection interface to floor rails

Organising Customer Requirements

There are three reasons behind organising the customer requirements:
To organise the raw customer data into groups of associated data to simplify using the QFD
charts at a later stage.
Because the customer requirements are listed in the language of the customer, the process of
organising the data allows the QFD team to reach a common understanding of customer wants.
The customer research techniques by the nature of their sampling and the customer responses
may not highlight all the customer needs. The process of organising listed below gives the team the
chance to surface areas and requirements which the customer has not talked about.
In organising the customer requirements there are three categories into which they can be put. These
are:
Primary: Are the very basic customer wants. At this level all the requirements should give an overall
product view. For example, Toyota has only four primary categories for describing a new car: Good
at first glance, easy ride and handling of car, feels safe when driving, long life.
Secondary: These requirements are the primary level in more detail and are in fact the headings for
groups of tertiary level requirements
Tertiary: These are the requirements in their most detailed level. In fact, when developing these
organisation levels with the customer requirements we start at the tertiary level and work outwards
towards the primary level. The reason for this is that QFD should always be customer driven and,
therefore, we should start with the information the customer has given us. If the customer is a
manufacturer of industrial goods, that manufacturer can generally identify its own primary

requirements. The buying public, however, often mentions secondary or tertiary requirements when
asked what it wants.
The Process of Organisation
Step 1:

All the customer requirements should be written on to pieces of card, or post it notes. Each piece
of card should contain only one of the requirements.
Step 2:

The cards should now be pinned to a board or spread out on a large table where all members of the
team can see them easily.
Step 3:

The team should now start to group cards together that have some relationship. It will be the
discussions during this step which lead to reaching a common understanding of customer wants. It is
quite possible that certain requirements appear to belong to more than one group. If this is the case,
at this stage, repeat cards should be written and put with all possible groups.
Step 4:

Once the cards have been grouped the team should write title cards which best describe the category
for the cards. The title cards should be as specific as possible. Occasionally there will be cards within
the group which could easily act as title cards for the group. If this is the case it should become a title
card. Once this has been done the cards have effectively been grouped into secondary (title) and
tertiary (group) requirements. At this stage the team should discuss any doubling up of cards from
the step and decide to which single group they should belong. At the end of the organizing process
there should be no repeated requirements. It is possible that confusion will arise as to whether a
requirement should be a secondary requirement or tertiary requirement. The team can alleviate this
problem by applying a simple test.
Test: The Why Question

Of the requirement the team is unclear about being secondary or tertiary; ask the question, Why
does the customer want this? An experienced team should be able to answer this question. If the
answer is simply the original requirement restated in a different format, then the requirement is
tertiary. If on the other hand, the answer is more detail about the requirement then the requirement
is secondary.
Step 5:

Once the secondary requirements have been found, the team should discuss each one in turn to
surface any possible tertiary requirements which could also be included.
Step 6:

Now steps 3, 4, and 5 should be repeated for the secondary requirements in order to get primary
requirements and possible missing secondary requirements. If new secondary requirements surface,
these will have to be discussed to obtain the underlying tertiary requirements. It may even be
necessary to go back to the customers to find their tertiary requirements for the identified area.
Step 7:

The primary, secondary and tertiary requirements can now be drawn in a tree diagram. This is shown
in the example below for an aircraft seat. There is one word of warning for the QFD team when they
are discussing additional requirements. The team should take care not to introduce engineering
terminology. The team should continually try to think like the customer at this stage.
On the diagram below, the following legend has been used with respect to the previously mentioned
customer chain:
P Passenger Requirement
A Airline Requirement
AM Aircraft Manufacturer Requirement
* QFD Team Developed Title

Aircraft seat example of organizing customer requirements.

Entering the Customer Requirements onto the QFD Chart

Once all the requirements have been gathered and organised, they can be entered on the QFD chart
on the left hand side as shown on the next page.

Customer and technical & regulatory requirements.

Establishing
Customer
Importance
Customer Competitive Comparisons

Ratings

&

Although these are two distinct areas with different objectives, they can be carried out together on
the same customer survey. Hence, they have been linked together in this section. After the customer
requirements have been gathered it is necessary to quantify two separate issues:
How important each of the characteristics is to the customer. It will definitely be the case that the
customer will attach more importance to certain requirements than others. It is important to QFD
that we are able to reflect these different importance ratings.
How the customer rates how well our company performs on each of his/her requirements against
the best of our competition.
It is necessary to have both of the above pieces of information so that the team can use them in the
analysis of the QFD chart to focus in on the sources available to the company for competitive
advantage.
Any of the survey techniques mentioned in the customer requirement section can be used to gather
customer importance ratings and competitive comparisons. For many products, e.g. cars, it is
possible to obtain lists of owners from government registers for both a companys products and its
competitors. For those that lists cannot be found the process becomes much harder.
An example of the type of form used in the clinic is shown as below. The final possibility is in fact
listening. A wealth of competitive comparison information can be acquired by engineers and
managers listening and taking note of what customers say at exhibitions, trade shows and retail
outlets. Comments like I like the boot but it isnt as big as the Toyota we looked at yesterday if
noted, can prove invaluable.
Entering on the QFD chart
The above is shown entered on the QFD chart for the aircraft seat example on the two pages
following the clinic example

Distribution of importance points.

Customer importance ratings and comparisons.

Customer Complaint History

There is one other area which adds to the useful information we can gather for use in the QFD
study. Most businesses today have records of customer complaints. The reason for complaint
histories being useful to the QFD study is that often the complaint history cannot be related to a
specific customer want. This is because when the service call report was written the engineer will
have translated the customers complaint language into technical language and hence the related
customer want is lost.

The identification of the customer complaints and their inclusion in the QFD chart allows the QFD
team to identify the nature of the customer complaint and the time frame over which it arises. For
example, take a car door. The customer may want a door which is easy to open and close. We as
manufacturers have designed the door such that this is the case. When we look at the complaint
history however, we find that there have been a great many complaints in this area. On investigation
the team finds that the hinge design becomes very stiff after certain conditions combined with a
certain amount of wear.

Customer complaint history.

Establishing Priorities
This stage of the QFD study is in fact optional, but is the area which allows the company to make
major improvements quickly. In some cases where the information mentioned in the previous
sections has been entered on the chart it will point to an area of major customer dissatisfaction. That
is to say a customer requirement which is seen as being highly important by the customer we as a
company are the worst of our competitors. This may also be backed up by a very large customer
complaint history. If this is the case, the QFD team may decide to begin a study of the particular
issue concerned before continuing with the QFD study.

Customer priorities.

Establish Engineering Characteristics

The starting point for QFD is positive statements of Customer Requirements expressed in the
customers own language. This data is then organised into logical groups and your competitive
position, as perceived by your customers, is determined against these requirements. The next step is
to determine the Engineering Characteristics which must be optimised to assure customer
satisfaction. The marketing domain tells us what to do, the engineering domain tells us how to do it.
The Customer Requirements are listed down the left hand side of the chart. The QFD team then
lists, along the top, those Engineering Characteristics that are likely to affect one or more of the
Customer Requirements. Engineering Characteristics should describe the product in measurable
terms and should directly affect customer perceptions. This translation of Customer Requirements
into language meaningful to a designer is a very important step in the QFD process and deserves
considerable study and development. If the step is performed correctly the customers voice will be
carried through. If not, one of the major objectives of the study, the customers voice, will have been
lost. Teams should avoid ambiguity in interpretation of the Engineering Characteristics. They must
avoid hasty decisions that a Customer Requirement is satisfied by using current quality control
measurement practices. If you do what you have always done, you will get what you have always got.
This is a time for systematic, creative and exhaustive analysis of each of the requirements, a time for
brainstorming. This process need not be conducted by the QFD team alone. Subgroups may be
formed with members brought in from outside the core team, but each subgroup should be led by a
team member. Brainstorm each secondary, and its associated tertiary requirements, to generate
Engineering Characteristics. As many ideas as possible should be generated, both established and
novel approaches, any aspect of engineering that affects at least one of the Customer Requirements.
However, there is one overriding requirement that must be rigidly enforced. Each Engineering
Characteristic must be measurable. If is not measurable, it cannot be optimised and so cannot be
used to assure customer satisfaction. From the study of the aircraft seat the following was generated.

STEP 1 Brainstorming
Secondary Requirement

Comfortable Upright

Tertiary Requirements

Engineering Characteristics

Arm rest folds right away

Arm rest wide enough
Enough leg room
Doesnt give you bum ache
Doesnt hurt the shoulders of
tall person
Lumbar support doesnt hurt
back of short person

Width of gap between seat backs

Depth of gap between seat backs
Width of arm rest
Thickness of seat back
Profile of seat back (front and back)
Pan height
Pan profile
Lumps of adhesive left on surface of structure
Position of upholstery attachments
Foam thickness
Foam hardness

STEP 2 - Grouping data

These ideas must now be grouped under three headings, and as a check that all the Engineering
Characteristics are measurable, the units of measurement must be defined.
Human Requirements
Process Requirements
Engineering
Units of
Characteristics
Measurement
Train assembly staff to attach
Wipe excess adhesive off seat Seat gap width
mm
covers on the back and side
structure during assembly.
Seat gap depth
mm
offset only.
Grind off any lumps of set
Arm rest width
mm
adhesive after assembly.
Seat back thickness
mm
Seat back profile
mm (x, y, z)
Seat pan profile
mm (x, y, z)
Seat pan height
mm
Foam thickness
mm
Foam hardness
Vickers

STEP 3 - Entry onto QFD chart

Once the Engineering Characteristics are established, they must be entered along the top of the QFD
chart as shown below.

Engineering characteristics

Technical Competitive Comparisons

The next step is to evaluate how your company performs in comparison to its most serious
competitors. If your company continually evaluates its products performance competitively against
other products in the market place, much of the required data, along with the organisational structure
necessary to obtain it, will already exist. If this is not the case, you will have to obtain the products
(buy or rent) or get access to them (how?), to conduct the analysis. In many situations solution of this
problem will require some lateral thinking. Basically, you have to get hold of your competitors
products, tear them apart and compare them to yours.
A few typical examples are:
Consumer product: use it then reverse engineer it.
Computer software: try it out, benchmark it against your software.
A shop: go and buy something there.
The Engineering Characteristics already identified must be matched up and competitive rankings
determined on those specific points. In addition, you should always be on the look out for any other
areas in which your competition has the edge on you.
It is important that the QFD team work closely with the test personnel who will conduct the
competitive analysis. There will be cases where an Engineering Characteristic cannot be
competitively evaluated by a standard company test. In these instances it will be necessary for the
team to help design a new test. Competitive comparisons provide a company with the hard facts
about where its products stand technically in relation to its competitors products. This knowledge is
vitally important to the QFD team during the analysis stage of the study. It will rarely prove possible
to easily evaluate all of the Engineering Characteristics. Competitive comparisons should become an
ongoing process with team members striving to develop data for every characteristic. In many cases
this ideal situation will never be achieved.
Aircraft Seat Study
The method of classification will vary from product to product, as will methods of testing. Below is
shown an extract from the Technical Competitive Comparison performed on aircraft seats. You will
have to develop your own methods but it is strongly recommended that you lay out the data in a
form similar to that shown.

In this case of the seat back thickness the technical comparison involved only a simple measurement.
The average thickness of the three seat backs was rounded up and taken as the normal value. The 1-5
value was then calculated by the following equation:

i.e. the ratio of (the normal value) divided by (the measured value), multiplied by (the mid value in
the 1-5 scale). In the above, the driving force is to minimise the thickness. In a situation where the
driving force is to maximise the value of the Engineering Characteristic, the ratio must be inverted.
In the case of the seat back profile, the technical position could not be evaluated by a simple
measurement. It is valid to include profile as an Engineering Characteristic since the profile can be
measured in terms of its coordinate geometry (x y z) and its curvature at any particular point. The
problem is that there are no established rules for judging the comfort of a seat back from the
coordinate measurements.
In this instance the three different profiles were subjectively evaluated by customers and their relative
comforts judged. Based on these judgments the values shown were selected by the team members.
This is a typical example where further work is required - in this case to determine the comfort
parameters of a seat back profile. This is an ideal situation in which to conduct a set of Taguchi
experiments.
Other Sources of Data
As stated earlier, the method of classification will vary from industry to industry. This is also true of
other sources of relevant information. One commonly used source is that of the service history of
the product. Data concerning the number of service repairs performed on any particular Engineering
Characteristic (or groups of characteristics) should be collected, as should the cost of the repairs. In
the case of computer software the data may take the form of the number of fixes released and the
areas they covered. Think laterally about such sources of data. The one criterion is that it should
relate to the performance of one or more Engineering Characteristics. If such data is collected, it
should be entered on to the chart below the Technical Competitive Comparison and above the
Technical and Regulatory Requirements, as shown

Technical competitive comparisons and other sources of data

Establishing Links
This stage involves filling in the Relationships Matrix in the main body of the chart and the triangular
Conflicts Matrix at the top of the chart. The idea is to highlight relationships between Customer
Requirements and Engineering Characteristics, and conflict/supporting relationships amongst the
Engineering Characteristics. In addition to highlighting the relationships their relative strengths must
also be judged and a range of symbols is used to represent this.
Relationships Matrix
This matrix highlights the relationships between the Customer Requirements and the Engineering
Characteristics. It must also be remembered to include the Technical and Regulatory Requirements in
the correlation. The team members should seek consensus on these evaluations, basing them on
expert engineering experience, customer responses and data from statistical studies and controlled
experiments.
There will be varying degrees of the strength of the relationships and so a set of symbols is used to
identify the significance.
The symbols used are:

Looking at the first Customer Requirement, Arm rest folds right away, you see that this is strongly
related to width of arm rest, arm rest recess depth and arm rest recess width. In addition, the team
felt that there was a weak relationship to the height of back and a medium relationship with the back
thickness. Every intersection between a Customer Requirement and an Engineering Characteristic
should be carefully considered. However, it should be noted that in the example shown less than
20% of the available space has actually been used. This is not unusual. In any event, intersection
symbols should certainly occupy less than 50% of the available space.

Establishing links relationship matrix

Conflicts Matrix
This is the triangular matrix at the top of the chart. It is used to highlight relationships between the
Engineering Characteristics. As before, symbols are used to indicate the strength of the relationship
but in addition, they also indicate whether it is a positive supporting relationship or a negative
conflicting relationship. The symbols used are:

Establishing links - correlation matrix

Consider item 5 Profile of back.

The matrix is completed by moving up to the beginning of the angled column, then moving along
this column to the top right. Again, every intersection should be carefully considered. The first
intersection is with the seat back thickness. This has been highlighted as a strong negative
relationship since they are in conflict. The aim is to have as thin a back as possible, but this would
result in a flat back, not a comfortable moulded profile. Moving further along a positive relationship
has been shown against Fatigue performance. This is because a curved profile will increase the
rigidity of the back and improve its structural performance. Similar arguments have been used to
assign symbols to the rest of the boxes where necessary. Negative symbols show where a tradeoff
situation exists that must be resolved. These will be considered in the next stage when the target
values for the Engineering Characteristics are established.

Establishing technical targets and ratings

Once the relationships (correlations and conflicts) have been established, the next stage is to
determine target values for the engineering characteristics, a rating of the difficulty of achieving that
target and an importance rating for the engineering characteristics.
Engineering Targets
These engineering targets are very different from the normal design specifications with which most
people are familiar. Design specifications reflect the engineers judgment and knowledge of prior
problems and manufacturing capabilities. Engineering targets on the other hand only reflect what is
needed to assure customer satisfaction. They bear no relation to what currently can be achieved. In
choosing target values you should consider several things. Firstly, the ratings and absolute values
determined in the technical competitive comparison. Secondly, the importance ratings of the
customer requirements you are trying to satisfy along with the associated data from the customer
competitive comparisons. Thirdly, you should consider any relationships, positive and negative,
highlighted in the conflicts matrix.
Consider the 6th engineering characteristic from the aircraft seat study:
Seat back thickness
The technical competitive comparison gave the following:
Company A(us) 28 mm
B 36 mm
C 25 mm
Looking through the customer requirements that are affected by the back thickness, we see there is a
strong relationship with Enough leg room. This is rated highly with an importance of 8. The
customers see us as performing equal best, but there is room for improvement. Further down we see
a strong relationship with Fit more passengers in. This is rated with an importance of 5 and the
customers see us as being equal worse. Next look at the conflicts and bear in mind that we must not
compromise the strength and stiffness of the back and that we must still be able to accommodate the
hinge pivot in the back. Balancing all this, it was decided to match the best performer, and aim to
reduce the back thickness to 25 mm. The airline could then have the choice of giving the passengers
extra leg room or of shunting the seats closer together to get an extra row in. A similar process was
used to determine target values for all the engineering characteristics. When we come to the analysis
stage and study the completed chart, a few of the values may have to be slightly modified. The
number requiring modification will depend upon how well this stage was performed.
Degree of Technical Difficulty
A key factor that influences the analysis stage is the level of technical difficulty expected to be
encountered in achieving the engineering characteristics target values. Since each study represents a
unique situation, the method of determining the level of difficulty will have to be adapted to suit each
QFD study. The areas to be considered may include design, purchasing, manufacturing, packaging,
suppliers etc. Each QFD team will have to develop its own approach and decide on the degree of
technical difficulty. In the aircraft seat study a scale of 1-5 was used with 5 being most difficult. All
the functions affected by the proposed design changes were included in the determination of the

value. In the case of the seat back thickness, only the design department was affected by the
proposed reduction to 25 mm. However, it was felt that a major structural redesign would be
required to achieve the desired thickness and so it was rated with a value of 5.

Establishing engineering targets and degree of technical difficulty

Engineering Characteristic Importance Rating

These ratings are a combination of the customer importance rating and the strength of the
relationships between the customer requirements and the engineering characteristics.
Values are assigned to the correlation symbols. The recommended values are:

The ratings are calculated by summing down each column the product of the customer importance
rating and the value assigned to the correlation symbol. As an example consider the first column in
the aircraft seat example, the Width of the armrest. The absolute importance rating of 114 is
calculated from:
(5 x 9) + (5 x 9) + (9 x 1) + (5 x 3) = 114
The relative value is simply the absolute value expressed as a percentage of the total.

Engineering characteristics importance ratings.

Analysing The QFD Chart

As the QFD chart nears completion, the data that has been accumulated and organised must be
analysed and, where necessary, target values finalised. The analysis of the chart will identify several
key areas:
Areas in which you hold the lead but must maintain your position.
Areas where you can gain a competitive advantage.
Areas where you lag behind and must catch up with your competitors.
There are no set procedures for analysing the chart. The team must work together allowing free and
open discussion. They must systematically examine each customer requirement and all the associated
data. A few specific areas which must be looked at will be described and a fuller picture will be given
by the worked example. Initially check each row and column for strong relationships. If none are
shown it deserves detailed attention. In the case of a column it shows that an Engineering
Characteristic has been introduced that is not needed. It is not uncommon to find Engineering
Characteristics that are historic or represent the voice of the engineer and have no real relationship
with a customer requirement. In the case of a row it means that a Customer Requirement may not
have been adequately satisfied. If this occurs, check your correlations to make sure that you have not
missed something. If necessary, run a brain storming session to generate Engineering Characteristics
to satisfy the customer requirement. This situation is particularly serious if it concerns a technical or
regulatory requirement. These must be satisfied otherwise you may not be able to make the product
or, if you do, it may not meet legal requirements. A QFD study is all about finding out what your
customers want and assuring that features are built into the end product to ensure that these wants
are satisfied. This should be borne in mind while analysing the chart. Start with the customer
requirements and systematically examine your competitive position and your capability for
improving. Work through the Customer Requirements in the order of their Importance Rating - the
most important first. In the aircraft seat study the highest ranked customer requirement is that of

Light Weight. Looking at the competitive comparison we see that we are rated the best but
Company C is in a fairly close second place. Of the Engineering Characteristics that affect the
weight, two are strongly related and three have a medium correlation. The degree of technical
difficulty is 4 for the two strong correlations and 2, 4, 4 for the others. There are no areas which can
be easily tackled to reduce weight, but this is seen as the most important aspect of the seat. We have
the lightest seat in the market and this position must be maintained. Looking at Doesnt give you
bum ache as the next customer requirement, we see that this has an importance of 8 and all the seats
are competitively ranked equal with a rating of 3. There is clearly an opportunity here to gain a
competitive advantage if the technical problems can be overcome. The strong relationships are with
pan profile, foam hardness and foam thickness. The foam can easily be changed (technical difficulty
= 2) but the pan profile has a difficulty of 4 associated with it. If the pan profile can be improved we
can take the lead. Further through the analysis we see that we are rated as the worst performer for
short person lumbar comfort. This has an importance rating of 7. The factor which strongly affects
this is the back profile, along with foam hardness and foam thickness of secondary importance. This
is clearly an area in which we must improve. The technical difficulty of changing the foam is only 2.
This maybe the route forward to tackle the problem in the short term. Longer term the back profile
must be modified to make it more comfortable. As stated earlier, the target values are the levels of
performance required for each engineering characteristic to fulfill customer expectations. They must
not be based on any level of capability. If during the analysis stage it becomes apparent that the
above has not been honoured, then the target value must be modified.

CHECK ROWS AND COLUMNS FOR STRONG RELATIONSHIPS

AREAS IN WHICH WE HOLD THE LEAD AND MUST MAINTAIN OUR POSITION

AREAS WHERE WE CAN GAIN A COMPETITIVE ADVANTAGE

AREAS WHERE WE LAG BEHIND AND MUST IMPROVE

Next Steps
The first level QFD study has now been completed. The early stages were concerned with finding
out what your customers actually want in the product. We then moved on to developing engineering
characteristics that could be used to assure customer satisfaction. The next stages involved
correlating the two, highlighting conflicting characteristics and setting target values. During the
setting of target values it became clear that we did not know enough about some aspects of the
design i.e.. the seat back and pan profiles. This is one of the strengths of QFD. It highlights areas of
weakness which are important to customers. Once a problem is identified it can be tackled. This
point comes up again during the final stages. The analysis of the chart shows areas of importance
that must be addressed. QFD in itself does not solve engineering problems for you. It is a forum for
gathering and organising all the data relevant to a design. It focuses attention on what the customer
wants and forces you to set target values for engineering characteristics to assure customer
satisfaction. In this context QFD shows you what to do, not how to do it. A large amount of effort
must go into preparing a first level study. An equal commitment should be made to solving any
engineering problems highlighted. The best problem solving techniques should be employed; the use
of designed experiments is strongly recommended. There are two other benefits of QFD that may
not be so obvious. Firstly, it acts as a facilitator for a multi-discipline team approach to product
design. Secondly, it is an excellent documentation medium for recording options considered and
decisions made during product design.