1. Explain what is QFD, (Quality Function Deployment)?

QFD is a method to aid in the determination and fulfillment of customer needs. It takes the qualitative customer input
and quantitatively expresses them as requirements. Further, the end product or service is evaluated on its ability to
meet those requirements. The overall structure of the method is made up of six sub-matrices which comprise the
House of Quality master matrix.
2. Explain the WHATs in a QFD matrix.
The WHATSs in the QFD matrix are the customer needs. These customer needs are the foundation of the product or
service and ultimately should all be fulfilled by the final product or service. In the QFD method the WHATs are used
to determine the HOWs.
3. Explain the HOWs in a QFD matrix.
The HOWs in the QFD matrix are the technical requirements of the product or service. These requirements should
meet or exceed all of the customer needs (HOWs) as determined by the initial step in the process. The customer
needs are generally qualitative (ie. the product must be comfortable) whereas the technical requirements translate
those qualitative requirements into quantitative measurable requirements (ie. the product must have down feathers).
How well an organization translates the customer needs into technical requirements is critical and can often be the
difference between success and failure. Not many other factors in the design and development of a product have
such profound consequence.
4. Explain the 1, or 3, or 9 interrelationship values in a QFD matrix.
The interrelationship matrix of the QFD provides a quantitative measurement between the customer needs (WHATs)
and the technical requirements (HOWs) of the product or service. The interrelationship of each customer need and
each technical requirement is rated by a numbering or symbol. A designation of 9 means the technical require has a
very strong relationship to the customer need. Whereas, a designation of 3 and 1 mean the technical requirement
has a medium and weak relationship respectively. If the technical requirement has absolutely no relationship to the
customer need then there is no designation and the cell of the matrix is left blank.
5. Explain how you calculate the technical priorities in the design target matrix.
The technical priorities are a product of the customer need (HOWs) ranking and the interrelationship value. Each
technical requirement (WHATs) receives a raw score. This raw score is often obscure in the fact that it is better and
more commonly represented as a percentage of the sum of all the raw scores. the technical requirements with the
highest percentages should be considered to have the highest priority by the design team.
6. Define Statistical Process Control (SPC).
SPC is a statistical method of separating variation resulting from special causes from variation resulting from natural
causes in order to eliminate the special causes and to established and maintain consistency in the process, enabling
process improvement.
7. Explain Control Charts for Variables, with a simple mathematical example.
The basic steps for developing control chart for data with measured values are these:
1. Determine sampling procedure. Sample size may depend on the kind of product, production rate, measurement
expense, and likely ability to reveal changes in the process. Sample measurements are taken in subgroups of a

specific size (n), typically from 3 to 10. Sampling frequency should be often enough that changes in the process are
not missed but not so often as to mask slow drifts. If the object is to set up control charts for a new process, the
number of subgroups for the initial calculations should be 25 or more. For existing processes that appear stable, that
number can be reduced to 10 or so, and sample size (n) can be smaller, say, 3 to 5.
Sampling size = 100.
2. Collect initial data of 100 or so individual data points in k subgroups of n measurements.
- The process must not be tinkered with during this time let it run.
- Dont use old data that may be irrelevant to the current process.
- Take notes on anything that may have significance.
- Log data on a data sheet designed for control chart use
3. Calculate the mean (average) values of the data in each sub group ( ).
4. Calculate the data range for each subgroup ().

5. Calculate the average of the subgroup averages . This is the process average and will be the centerline for the
chart.
6. Calculate the average of the subgroup ranges . This will be the centerline for the R-chart.

68
= = 2 = 5.667

7. Calculate the process upper and lower control limits, UCL and LCL respectively (using a table of factors, such as
the one shown in Figure 18-6). UCL and LCL represent the 130 limits of the process averages and are drawn as
dashed lines on the control charts.
= 100.067 + (0.31 5.667) = 101.82377
= 100.067 1.75677 = 98.31023
= 4 = 1.78 5.667 = 10.08726
= 3 = 0.22 5.667 = 1.24674

8. Draw the control chart to fit the calculated values.

9. Plot the data on the chart.

8. Explain control charts for attributes, with a simple mathematical example.

The p-Chart
Attributes data are concerned not with measurement but with something that can be counted. For example, the
number of defects is attributes data. Whereas the - and R-charts are used for certain kinds of variables data, where
measurement is involved, the p-chart is used for certain attributes data. Actually, the I-chart is used when the data
are the fraction of some set of process output. It may also be shown as the percentage defective. The points plotted
on a p-chart are the fraction (or percentage) of defective pieces found in the sample of n pieces.
Fraction Defective by Subgroup (p)

The p values in Figure 18-11 are derived by the formula = .

1

For the first subgroup: = 100 = 0.01

1

For the second subgroup: = 100 = 0.01

And so on
Process Average ( )
The Process Average is calculated by dividing the total number defective by the total number of pens in subgroups:
=

1 + 2 + +
42
=
= 0.0175
1 + 2 + +
2400

Control Limits ( and )

Due to the fact that this is the first time that these control limits will be calculated for the process they should be
considered trial limits. Special cases which are identified by data points outside the control limits must be addressed
and eliminated. The limits are then recalculated with the special cases removed. A control limit may be calculated to
be a negative number but in the real world control limits which are negative do not exist. As such, if a negative
number is calculated for a control limit that limit will be set to zero.
= +

3 (1 )

3 (1 )

= 0.0175 +

= 0.0175

30.0175(1 0.0175)
100
30.0175(1 0.0175)
100

= 0.0568

= 0.0218 (= 0)

9. Discuss and explain various continual quality improvement methods and tools.
Pareto charts, cause-and-effect diagrams, stratification, check sheets, histograms, scatter diagrams, flowcharts,
design of experiments and FMEA are all various tools which can be used for continual quality improvement. The data
gathered by these methods can be used to implement stricter control charts and identify the cause of special cases
to be eliminated from such control charts. These tools and methods may be used in conjunction with each other or on
their own depending on the need of the application.
10. Explain the way control charts could be used for quality improvements.
Control charts take a process which operates freely and sets a operationally acceptable range in order to removed
said freedom from the process. By using run data to create and monitor control charts process creep can be
controlled and eliminated. Thus, the process is more capable of producing a quality product time and time again.