QARE Student Cycle 1

DEPARTMENT OF INDUSTRIAL ENGINEERING &
MANAGEMENT
Laboratory Manual

Quality Assurance &
Reliability Engineering
Laboratory

Edition
2011

LABORATORY MANUAL
Quality Assurance & Reliability Engineering
Laboratory (07IM61)
Dept. of IEM, R.V.College of Engineering
Mysore Road, Bangalore 560059.
Phone 91-080-8601700 Extn. 232 Fax 91-080-2915050

SEE: 3 hrs SEE Marks: 50 Marks
Class Work: 3 Hrs/Week/Batch CIE Marks: 50 Marks

Lab In-charge:

C K Nagendra Guptha
Kiran K

Approved By

Dr. A.V.Suresh

R.V. College of Engineering
Bangalore

Laboratory Certificate

This is to certify that Mr. / Ms. ___________________________________
has satisfactorily completed the course of experiments in Practical
__________________________________________________ prescribed by
University __________________________________________________ course
in the Laboratory of this College in the year _________________

Name of the Candidate _____________________________________________________
USN No. ____________________________ Examination Centre ___________________
Date of Practical Examination: _______________________________________________

Signature of the Teacher in-charge of the batch

Date: Head of the Department
Marks
Maximum Obtained

Dept. of I.E.M., R.V.C.E. Quality Assurance & Reliability Engineering Laboratory 4

DEPARTMENT OF INDUSTRIAL ENGINEERING & MANAGEMENT.
Quality Assurance & Reliability Engineering Laboratory 07IM61
SCHEME OF CONDUCT AND EVALUATION

CLASS: VI SEMESTER SEE Marks: 50
YEAR: 2011 CIE Marks: 50
SL.
No.
Expt.
No.

TITLE
Marks
Class
Work
Test
CYCLE I
1. QE 01

To test the Goodness of fit for the given quality characteristic using
Normal distribution
15

2. QE 02 To test the Goodness of fit for the given quality characteristic using
Poisson distribution
15

3. QE 03
QE04
To test the Goodness of fit for the given quality characteristic using
Binomial & Uniform distribution
30

4. QE 05 Experiments on correlation and Simple regression
15

5. QE 06 Conduction of Repeatability and Reproducibility studies for the given
measurement system
15

6. QE 07

Estimation of process variability using Demings funnel Experiment /
Quincunx Apparatus
15

7. QE 08 Developing Quality Function Deployment Matrix for a Product / Service 15
8. QE 09 Performing Quality Audit of a System 15
CYCLE II
9. QE 10
QE 11
Construction of control chart for attribute quality characteristics &
construction of control charts using SYSTAT / SQC PC IV software
30

10. QE 12

Assessing Process Capability of the given manufacturing process using
Normal probability paper method and process capability indices
15

11. QE 13 Exercises on Attribute sampling Plans-Single, Double and Multiple
sampling plans
15

12. QE 14 Conduction of Design of Experiments-Full Fractional approach for the
given quality characteristics for machining operation.
15

13. QE 15 Exercises to demonstrate Taguchis orthogonal Array technique through
DOE software
15

14. QE 16 Performing Failure Modes and Effects Analysis for a system 15
15. QE 17 Estimation of System Reliability using Reliability Software Package 15
TEST -- 50
TOTAL 180
(Min)
50

EVALUATION SCHEME:
CIE MARKS (Reduced to 50)

Proposed by Prepared by Approved by
1. C. K. Nagendra Guptha R. Shekar Dr.A.V.Suresh
2. Kiran. K


General Guidelines
The Students are required to strictly follow the scheme of conduction of the experiments.
The Students are hereby advised to prepare for the Experiments well in advance and the
timing of the lab slots are to be effectively utilized for the computational and analysis
aspects of the experiments.
The Students are required to bring in their Blue Books duly complete in all respects,
without which the students will not be allowed to do the experiments in the lab.
Submission of Lab Records complete in all respects is to be done in the next subsequent
Lab Class.
The Outline of the record is as follows:
Aim/Objective of the Experiment.
Procedure.
Inference.
Learning Experience.
Students must maintain strict academic discipline in the Laboratory.
Use of external hard disks (thumb drives) in the Lab without prior permission of the Lab-in-
charge will be viewed very seriously.
CIE Marks will be awarded on the overall assessment based on the discipline and the
fulfilment of the Lab requirements.


TABLE OF CONTENTS
SL.
No.
Expt.
No.
Date TITLE
Date of
Submission
Marks

TEST

TOTAL



Goodness of Fit Normal Distribution

Aim: To test the goodness of fit for the quality characteristic, width across the flats for the
Hexagonal Head Bolts manufactured with respect to normal distribution.

Apparatus: Inspection Gauge, Sample bolts and Measuring Guide.

Form No. 01

Theory:
A normal distribution is a continuous probability distribution. It is characterized by a bell
shaped, symmetric distribution curve. It extends from - to +, the area under the curve is 1.

( )
( )

=
2
2
2
2
2
1
o
u
to
x
x f exp
u = Average = Shift parameter,
o = Standard Deviation = Scale parameter.

Procedure:

The experiment is conducted using Bolt heads produced from a production process. The quality
characteristic measured is width across the flats.

1. The given test sample of bolts, is taken from a group of bolts placed in different boxes

2. Each bolt is taken individually and with the marked face adjacent to the inner vertical
face of bar, it is traversed along till it comes to its maximum traverse position. It should
be ensured that the bolt head always touch the bottom peak of the gauge.

3. The measuring guide is then traversed on the gauge, keeping the shorter arm on the
outside vertical face of the bar till it touches the bolt body. (The traverse direction is
opposite to that of the bolt.)

4. Using this as a reference, the width across flats is gauged by reading the class in which
it falls.

Experiment
QE01


5. The value is tabulated in its appropriate class as shown in the Observation sheet. Chart
of scale calibration is used to determine the upper specification and lower specification
of the class interval.

6. The steps repeated for the remaining bolts.

7. The mean and standard deviation of the sample data are calculated using the formula.
The calculations are shown below.

8. The standard normal ordinate Z = x-u / o is calculated, where x is the upper
specification limit of the class interval, u is the mean and o is the standard deviation of
the sample. From the normal tables the probability corresponding to the Z is obtained.
This is the cumulative probability. Individual probability for each class interval are
determined. Expected frequency is given by the formula Pi X EOi.

9. The
( )
i
i i
E
E O
2

gives the Chi Square calculated value.

10. From the table Chi square table value is noted for the specified degrees of freedom
and confidence level. (DOF = n-k-1, where n = No. of class intervals of pooling, k =
No. of parameters estimated, 6-2-1=3) confidence level o= 0.5.

11. The data fits the normal distribution provided ; cal less than ;tab,

Experimental Setup Inspection Gauge

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15


GRIFFIN MADE IN ENGLAND

Scale
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Class
Name
A
B
C D E F G H I J K L M
N
O
W
I
D
T
H

A
C
R
O
S
S

F
L
A
T

(
m
m
)

19.8700
19.9496 19.9496
20.0292 20.0292
20.1088 20.1088
20.1884 20.1884
20.2680 20.2680
20.3476 20.3476
20.4272 20.4272
20.5068 20.5068
20.5864 20.5864
20.6660 20.6660
20.7456 20.7456
20.8252
20.8252
20.9048 20.9048
20.9844 20.9844
21.0640

CHART OF SCALE CALIBRATION FOR INSPECTION GAUGE


Form No. 01
DEPARTMENT OF INDUSTRIAL ENGINEERING AND MANAGEMENT.
R. V. COLLEGE OF ENGINEERING, BANGALORE 560 059.
OBSERVATION SHEET FOR
GOODNESS OF FIT TEST OF INDUSTRIAL PRODUCTION
(Normal Distribution)
Expt. No: Date :
Source of production: Straddle Milling Product Name : Hexagonal Bolt
Variable Checked:

Sl. No.

Class
Range Class
Mid Point

Observed Frequency

Total
Lower
Limit
Upper
Limit



TABLE FOR COMPUTATION OF SAMPLE STATISTICS
Class
Name
Range on
Scale
Width across flats range for
scale class
Class mid
point X
mi

(mm)
( )
h
A X
U
mi
i
=

O
i
O
i
x U
i
O
i
x U
i
2

(XL) Lower
(mm)
(XU) Upper
(mm)
D 3 4
E 4 5
F 5 6
G 6 7
H 7 8
I 8 9
J 9 10
K 10 11
L 11 - 12
Total


h = Class width = 0.0796 A = Assumed mean = 20.3874

Mean = A
O
U O
h
i
i i
+
|
|
.
|
\
|

= u

Std. Dev. =
2
2
|
|
.
|
\
|
=
i
i i
i
i i
O
U O
O
U O
h o


CALCULATION OF ; STATISTIC FOR GOODNESS OF FIT =
o =

Class
RANGE
X
u

Z

CUM P
i

(E P
i
)

P
i

O
i
Pooled Oi E
i
= P
i
EO
i

Pooled Ei
(O
i
E
i
)
2

E
i

D 3 4

E 4 5

F 5 6

G 6 7

H 7 8

I 8 9

J 9 10

K 10 11

L 11 - 12



- Cumulative Pi = P ( X s Xu)
=
( )
|
.
|
\
| s
o
u
u
X Z
P


Note: On Pooling of frequencies: The accuracy of ; test will be better whenever it is ensured
that the expected frequency is larger than 5. In order to guarantee this, O
i
, and E
i
are pooled
suitably.

Conclusion:



Goodness of Fit - Poisson Distribution

Aim: To determine the goodness of fit for Poisson distribution for the defectives in a sample
drawn from a lot.

Apparatus: Sampling Gadget, Beads - Red (920 Nos.) & Yellow.(200 Nos.)

Form No. 02

Theory:

The Poisson distribution is a discrete distribution bounded at 0 on the low side and unbounded
on the high side. The Poisson distribution is a limiting form of the Hyper-geometric
distribution.

The Poisson distribution finds frequent use because it represents the infrequent occurrence of
events whose rate is constant. This includes many types of events in time or space such as
arrivals of telephone calls, defects in semiconductor manufacturing, defects in all aspects of
quality control, molecular distributions, stellar distributions, geographical distributions of
plants etc. It is an important starting point in queuing theory and reliability theory. Note that the
time between arrivals (defects) is exponentially distributed, which makes this distribution a
particularly convenient starting point even when the process is more complex.

( )
!

X
e
X P
x
u
u
=

Where u is the average number of outcomes occurring in the given time interval or specified
region.

The Poisson distribution is applicable when the conditions such as N= 10n, where N is the lot
size and n is the sample size and P proportion defective should be less than 0.1 are satisfied.

Procedure:

Calculate the No. of red and yellow beads to suit the specified proportion.
Mix the beads thoroughly in the top portion of the sampling gadget.
Mix thoroughly and draw a sample of beads as per the sample size given.
Observe the number of defectives (Yellow beads) in the sample.
Use form No: 2 to record the frequency of occurrence of defectives.
Experiment
QE02


Repeat the procedure of drawing the sample and observing the number of defectives in the
sample as many times as specified in the class.
The observed frequencies w.r.t the No. of defectives are tabulated in form 2.
The calculations for X
2
cal are done using a suitably designed spread sheet and the
calculations are shown in the tabulation.
i
i i
e
e O
cal
2
2
) (
= ;

X
2
tab can be obtained by referring to Chi square distribution and the degrees of freedom.
Table value for X
2
tab read off.
The degrees of freedom can be determined as Dof = number of classes after grouping
number of parameters estimates -1.
If X
2
cal < X
2
tab then conclude that the sample data provides evidence that the random
variable (no. of defectives in a sample of n items) follow a Poisson distribution. Otherwise
it does not fit the Poisson distribution.
It is desirable to pool the classes on the basis of the expected frequency (If the expected
frequency is less than 5 pool the observed frequency and calculate the Chi Square value).

Experimental Setup:

Box
Tray
Red beads Yellow beads


DEPARTMENT OF INDUSTRIAL ENGINEERING AND MANAGEMENT

GOODNESS OF FIT TEST OF POISSON DISTRIBUTION

Expt. No. : Date :
Sample Size : Quantity in lot :
Proportion Defective : Quantity Defective:

Sl.
No.
No. of
Defectives
Observed Frequency

Total

Signature Staff In-Charge



Tabulation:

Sl.
No.
No. of
Defects
Observed
Frequency
O
i

Pooled
Observed
Frequency
Cum.
Probability
Individual
Probability
Expected
Frequency
e
i

Pooled
Expected
Frequency
( )
i
i i
e
e O
2

Specimen Calculation:

For Sl.No. 1



Inference:



Goodness of Fit - Binomial Distribution

Aim: To determine the goodness of fit for Binomial distribution.

Apparatus: Dies

Form No. 02

Theory:
( ) ( )
x n
x
p p
x
n
x p

|
|
.
|
\
|
= 1

( )! !
!
x n x
n
x
n
=
|
|
.
|
\
|

The Binominal distribution is a discrete distribution bounded by [0,n]. Typically, it is used where
a single trial is repeated over and over, such as the tossing of a coin. The parameter, p, is the
probability of the event, either heads or tails, either occurring or not occurring. Each single trial is
assumed to be independent of all others. For large n, the Binomial distribution may be
approximated by the Normal distribution.

The Binomial distribution has had extensive use in games, but is also useful in genetics, sampling
of defective parts in a stable process, and other event sampling tests where the probability of the
event is known to be constant or nearly so.

Procedure:

Toss the 6 dies for 70 times.
Count the number of times the occurrence of 1 (You may choose occurrence of any number) and
record the observed frequency in the form No. 2
Determine the Binomial probability using the density function
( ) ( )
x n
x
p p
x
n
x p

|
|
.
|
\
|
= 1

Calculate the expected frequency.
Determine the Chi square table value with Degrees of Freedom = (No. of Class intervals No. of
parameters estimated 1) and confidence level o = 0.05.
If Chi Square calculated value is less than the Chi Square table value data fits binomial
distribution.
Experiment
QE03
x = 0, 1, , n,
p = probability of the event occurring
n = number of trials



Experimental Setup:




GOODNESS OF FIT TEST OF BINOMIAL DISTRIBUTION

Expt. No. : Date :

Sl.
No.
No. of
Defectives
Observed Frequency

Total

Signature of Staff in-charge



Tabulation:

Sl.
No.
No. of
Defects
Observed
Frequency
O
i

Pooled
Observed
Frequency O
i

Individual
Probability
Expected
Frequency
e
i

Pooled
Expected
Frequency
( )
i
i i
e
e O
2


For Sl.No. 1

Inference:



Goodness of Fit - Uniform Distribution

Aim: To determine the goodness of fit for Uniform distribution.

Apparatus: Dies

Form No. 18

Theory:

( )
1
1
+
=
min max
x p

The Discrete Uniform distribution is a discrete distribution bounded on [min, max] with constant
probability at every value on or between the bounds. Sometimes called the discrete rectangular
distribution, it arises when an event can have a finite and equally probable number of outcomes.
Note that the probabilities are actually weights at each integer, but are represented by broader bars
for visibility.

Procedure:

- Toss the dies for 120 times.
- Count the number of times the occurrence of 1, 2, 3, etc., and record the observed
frequency in the form No. 18
- Determine the probability of uniform distribution (Since there are six faces, the probability
of occurrence of any one face is equal to 1/6)
- Calculate the expected frequency (Probability of occurrence X Total No. of Observations,
1/6 X 120 = 20)
- Determine the Chi square table value with Degrees of Freedom = No. of Class intervals
No. of parameters estimated 1 and confidence level o = 0.05..
- If Chi Square calculated value is less than the Chi Square table value data fits uniform
distribution.

Experimental Setup

Experiment
QE04
X = min, min+1, max
Min = minimum x
Max = maximum x


Rolling Dies

GOODNESS OF FIT TEST OF UNIFORM DISTRIBUTION

Expt. No. : Date :

Signature Staff in-Charge

Face
No.
Probability P
i
Observed Frequency

Total



Tabulation:

Sl.
No.
Face No. Observed
Frequency
O
i

P
i
Expected
Frequency
e
i
= Pi x EO
i

( )
i
i i
e
e O
2


Inference:



Simple Linear Regression

Aim: To build Simple Linear Regression model and find the dependent variable with help of
independent variable.

Apparatus: LVDT, slip gauge set.

Theory:
Regression analysis is a technique for the modeling and analysis of numerical data consisting of
values of a dependent variable (also called response variable or measurement) and of one or more
independent variables (also known as explanatory variables or predictors). The dependent variable
in the regression equation is modeled as a function of the independent variables, corresponding
parameters ("constants"), the parameters are estimated so as to give a "best fit" of the data. Most
commonly the best fit is evaluated by using the least squares method.

Regression can be used for prediction (including forecasting of time-series data), inference, and
hypothesis testing, and modeling of causal relationships.

Procedure:

1. Slip gauges are cleaned.
2. Slip gauge dimension is measured the with the help of LVDT.
3. LVDT readings are tabulated for 20 slip gauges.
4. Simple linear regression model is built.
a. Calculations are done as shown in the table.
5. The graph is plotted.
Experiment
QE05


Department of Industrial Engineering and Management
R.V. College of Engineering, Bangalore - 59
OBSERVATION SHEET
Date: Name: Reg. No.
Lab: Class: Expt No.

Title: Simple Linear Regression model

Observation: -
Sl no Slip gauge LVDT
reading
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

Signature of the Staff in-charge


Calculation
Sl no Xi Yi (Xi-x) (yi-y) ((Xi-x)- (yi-y)) (Xi-x)
2

X= y=
n =no of observations

Regression Equation

Equation for st line: y=a+bx
b=
a=y-bx
Excepted graph:

X Y slope


Inference:


Repeatability and Reproducibility Studies
Aim: To determine the repeatability and reproducibility for a given measurement system.

Apparatus: Micrometer

Form No. 16

Theory:
Measurement system analysis, QS 9000, Measurement variation stability, Bias, Linearity,
Repeatability, Reproducibility.

Repeatability:

Standard deviation for repeatability
2
[R(Ave)] range Average
d
e
= o d
2
is a control chart Constant

Equipment variation (EV) = Average range[R(Ave)] X k
1
(k1 4.56 for 2 trials)

Reproducibility:

Standard deviation for reproducibility
2
[Xdiff] difference Average
d
o
= o

Appraiser variation (AV) = ( )
|
|
.
|
\
|

nr
EV
Average
2
2
2
k [Xdiff] difference
k
2
= 3.65 for 2 appraisals
n = number of parts
r = number of trials.

Repeatability & Reproducibility = R & R =
2 2
AV EV +

Part-to-part variation:

PV = Range of sample average X k
3
(k
3
1.62 for 10 parts)
Experiment
QE06


Total variation:
2 2
m p
TV o o + =

o
p
= part to part standard deviation
2
[Rp(Ave)] average part of Range
d
= (d
2
constant)

o
m
= measurement system standards deviation
=
2 2
o e
o o +

% Repeatability and Reproducibility = % R & R =
100
&
TV
R R

Procedure:

- Obtain a sample of 10 parts that represents the expected range of process variation.
- Number the parts 1 to 10 so that they are not visible to the appraisers, who are referred to
as say A & B.
- Appraiser A: Measure 10 parts in a random order and enter the results. Appraiser B:
Measure the same 10 parts in another random order without seeing the others readings.
- Repeat the cycle using a different random order for measurement.
- Input the values into the Repeatability and Reproducibility software
- Obtain the result and draw the inference

Analysis R & R
Error

<10 % The measurement is acceptable

10 30% May be acceptable based upon importance of application cost of gauge, cost
repairs.
> 30% The measurement system needs improvement.

Repeatability > Reproducibility Gauges
Reproducibility > Repeatability Appraiser.



R.V. College of Engineering, Bangalore 560 059.

Gauges Repeatability and Reproducibility Data Sheet

Equipment
Description
Part Number
Equipment
Number
Characteristics
Specification Date
Least Count Location
Sl.No. Appraiser A Appraiser B Part Average
X
p
I Trial II Trial Range I Trial II Trial Range
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Averag
e X
Ra (Ave) Rb(Ave)
(Ave) Xa (Ave) Xb Rp
(Ave)

X diff R(Ave)
Equipment variation (EV) % EV
Appraiser variation (AV) % AV
Repeatability & Reproducibility
(R&R)
% R&R
Part variation (PV) % PV
Total variation (TV)
Remarks



Inference:


Demings funnel Experiment

Aim: to demonstrate the losses caused by tampering with real world processes at our places of
work.

Theory:
The funnel experiment is a mechanical representation of many real world processes at our places
of work. The aim of the experiment is to demonstrate the losses caused by tampering with these
very same processes. The primary source of this tampering is the use of Management by Results,
reactions to every individual result.
In the experiment, a marble is dropped through a funnel, and allowed to drop on a sheet of paper,
which contains a target. The objective of the process is to get the marble to come to a stop as close
to the target as possible. The experiment uses several methods to attempt to manipulate the
funnels location such that the spread about the target is minimized. These methods are referred to
as rules.
Funnel Experiment: Rule 1
During the first setup, the funnel is aligned above the target, and marbles dropped from this
location. No action is taken to move the funnel to improve performance. This rule serves as our
initial baseline for comparison with improved rules.
The results of rule 1 are a disappointment. The marble does not appear to behave consistently. The
marble rolls off in various directions for various distances. Certainly there must be a better (smart)
way to position the funnel to improve the pattern.

Experiment
QE07


During rule 2, we examine the previous result and take action to counteract the motion of the
marble. We correct for the error of the previous drop. If the marble rolled 2 inches northeast, we
position the funnel 2 inches to the southwest of where it last was.
A common example is worker adjustments to machinery. A worker may be working to make a
unit of uniform weight. If the last item was 2 pounds underweight, increase the setting for the
amount of material in the next item by 2 pounds.
Other real examples of Rule 2 include periodic calibrations. One checks a meters measurement
against a known standard, and adjusts the meter to compensate for the error against the standard.
Many automated feedback mechanisms perform this adjustment continuously. Other examples
include taking action to change policies and production levels based upon on last months budget
variances, profit margins, and output.
We also see this when setting next years goals and targets based upon last years levels.

A possible flaw in rule 2 was that it adjusted the funnel from its last position, rather than relative
to the target. If the marble rolled 2 inches northeast last time, we should set the funnel 2 inches
southwest of the target. Then when the marble again rolls 2 inches northeast, it will stop on the
target. The funnel is set at an equal and opposite direction from the target to compensate for the
last error.
We see rule 3 at work in systems where two parties react to each others actions. Their goal is to
maintain parity. If one country increases its nuclear arsenal, the rival country increases their
arsenal to maintain the perceived balance.
If drug enforcement increases, prices rise due to increased demand, and drug runners have
incentive to go to further lengths due to increased price.
A common example provided in economics courses is agriculture.
A drought occurs one year causing a drop in crop output. Prices rise, causing farmers to plant
more crops next year. In the next year, there are surpluses, causing the price to drop. Farmers plant
less next year. The cycle continues


In an attempt to reduce the variability of the marble drops, we decide to allow the marble to fall
where it wants to. We position the funnel over the last location of the marble, as that appears to be
the tendency of where the marble tends to stop.
A common example of Rule 4 is when we want to cut lumber to a uniform length. We use the
piece we just cut in order to measure the location of the next cut.
Other examples of Rule 4 include:
- Brainstorming (without outside help)
- Adjusting starting time of the next meeting based upon actual starting time of the last
meeting
- Benchmarking, in order to find examples to follow
- A message is passed from one person to the next, who repeats it to another person, and so
forth.
- The junior worker trains the next new worker, who then trains the next, and so forth.


Tampering
Rules 2, 3, and 4 are all examples of process tampering. We take action (dont just stand there -
do something!) as a result of the most recent result.
Rule 2 leads to a uniform circular pattern, whose size is 40% bigger than the Rule 1 circle. This is
because the error in distance from the funnel is independent from one marble drop to the next. In
positioning the funnel relative to the previous marble drop, we add the error from the first drop (by
repositioning the funnel) to the second drop (the error in the marble).
The standard deviation of adding n independent random variables is the square root of n times the
standard deviation of the individual. So the combined standard deviation is 1.4 times the original
standard deviation. Note, this statistical principle is a standard question that appears on every
Certified Quality Engineer exam in some form or another.

The problems of Rule 2 are corrected with dead bands in automated feedback mechanisms and
better calibration programs. We wait for a certain error to build up before taking action. But how
is the dead band determined? A control chart provides the answer. Plot the results on a control
chart, and recalibrate (or give a feedback signal) when a statistically significant change is detected.
Program dead bands approximate the control chart action.
Rules 3 and 4 tend to blow up. In rule 3, results swing back and forth with greater and greater
oscillations from the target. In rule 4, the funnel follows a drunken walk off the edge of the table.
In both cases, errors accumulate from one correction to the next, and the marble (or system)
heads off to infinity. Rules 3 and 4 represent unstable systems, with over-corrections tending to
occur.
Conclusion
Schemes to control the location of the funnel should be control chart based. In addition, we may
have to think outside of the box to fix this system. If we lowered the height of the funnel, we
would fundamentally reduce the variation in the process. If we added more layers of cloth or paper
to cushion the marbles landing, then the marble would roll less. The impact of these changes
would be detected by the control chart, and would prove whether or not an improvement did
occur.


Galtons Quincunx Apparatus

Aim: to illustrate the Law of dispersion, or Central Limit Theorem.

Theory:
Sir Francis Galton (18221911) invented the Quincunx in the 1870s to demonstrate the law of
error and the normal distribution. He was a prolific inventor, scientist, and mathematician and was
knighted in 1909. Beyond the Quincunx, Galton conceived the standard deviation, invented the
use of the regression line, and was the first to describe the phenomenon of the regression toward
the mean.
The quincunx is such a mechanical device and was first publicly demonstrated at the Royal
Institute in February 1874 (Stigler, 1986). The device consists of two parallel, vertical planes,
between which are many horizontal rows of equally spaced pins, with alternate rows offset by half
the pin spacing; see figure 1. At the top is a funnel into which lead shot (small spherical balls of
lead) is poured, and at the bottom a row of compartments to collect the shot. In Galtons own
words (the following quotes are from Natural Inheritance), when lead shot is dropped into the
device it scampers deviously down through the pins in a curious and interesting way; each of
them darting a step to the right or left, as the case may be, every time it strikes a pin.
One purpose of this device is to illustrate the Law of dispersion, or central limit theorem, in that
the cascade [of shot] issuing from the funnel broadens as it descends, and, at length, when
collected at the compartments at the base, approximates the Binomial or Normal distribution.
Galton explains this as follows:
The principle on which the action of the apparatus depends is that a number of small and
independent accidents befall each shot in its career. In rare cases, a long run of luck continues to
favor the course of a particular shot towards either outside place, but in the large majority of
instances the number of accidents that cause Deviation to the right, balance in a greater or less
degree those that cause Deviation to the left. Therefore most of the shot finds its way into the
compartments that are situated near a perpendicular line drawn from the outlet of the funnel, and
the Frequency with which shots stray to different distances to the right or the left of that line
diminishes in a much faster ratio than those distances increase.



Figure 1: Schematic of Galtons Quincunx Device

The Quincunx Model WD-7 (see figure 1) is an example of an off-the-shelf Quincunx board and is
comprised of a vertical board with 10 rows of pins. Beads are dropped via a funnel into the top of
the board. As they descend through the board the beads will bounce either to the left or right as
they encounter each row of pins. The pins may be arranged such that a bead will come in contact
with 4, 6, 8, or 10 rows. As each bead leaves the final row of pins, it is captured in one of several
bead-wide bins which may be numbered for reference by the user. After a sufficient number of
beads have been dropped, the height of the beads in the bins begins to resemble the classic bell-
shaped curve. In reality, the distribution of beads is binomial; however, the normal distribution is
approximated when n, the number of rows of pins, is large (n = 4, 6, 8, or 10).


Quality Function Deployment

Aim: to conduct quality function deployment study on given product

Theory:
Quality Function Deployment, or QFD, is a method used to identify critical customer attributes
and to create a specific link between customer attributes and design parameters. Matrices are used
to organize information to help marketers and design engineers answer three primary questions:

- What attributes are critical to our customers?
- What design parameters are important in driving those customer attributes?
- What should the design parameter targets be for the new design?

The organizing framework for the QFD process is a planning tool called the "house of quality".
Working as a team, design engineers and marketers first establish critical customer attributes for
the product. These attributes become the rows of the central matrix of the house of quality. The
team may group attributes into broader categories in order to simplify planning and analysis.

Steps in building house of quality
Example problem
Conduct a QFD study for handlebar stem for mountain bikes.

Step 1. List the customer requirements
QFD starts with list of goals / objectives. This list is often
referred as WHATs that the customer needs or expects in
particular Product.

Experiment
QE08


P
r
i
m
a
r
y

S
e
c
o
n
d
a
r
y

T
e
r
t
i
a
r
y

C
u
s
t
o
m
e
r

r
e
q
u
i
r
e
m
e
n
t
s

(
W
H
A
T
s
)

A
e
s
t
h
e
t
i
c
s

Reasonable cost
Aerodynamic look
Nice finish
Corrosion resistant
P
e
r
f
o
r
m
a
n
c
e

Light weight
Strength
Durable



Step 2. List Technical Descriptors
The next step in QFD is to come up with engineering characteristics or technical descriptors
(HOWs) that will affect one or more customer requirements. The technical descriptors make up
the ceiling or second floor of the house of quality

P
r
i
m
a
r
y

S
e
c
o
n
d
a
r
y

T
e
r
t
i
a
r
y

T
e
c
h
n
i
c
a
l

d
e
s
c
r
i
p
t
o
r
s

(
H
O
W
s
)

M
a
t
e
r
i
a
l

s
e
l
e
c
t
i
o
n

Steel
Aluminum
Titanium
M
a
n
u
f
a
c
t
u
r
i
n
g

p
r
o
c
e
s
s

Welding
Die casting
Sand casting
Forging
Powder metallurgy

Step 3 develop a relation matrix between WHATs & HOWs
The next step in building a house of quality is to compare the customer requirement & technical
descriptors and determine their respective relationship.
Then fill the relationship matrix with graphical symbols indicating the degree of influence
between each technical descriptor and each customer requirement like
A solid circle for strong relationship ( )
A single circle for medium relationship( )
A triangle represents a week relationship ( )


Step 4. Develop an inter relationship matrix between HOWs
The roof of the house of quality called the correlation matrix, is used to identify any inter
relationship between each of the technical descriptors. Symbols are used to describe the strength
of the interrelationship

- A solid circle for strong positive relationship ( )
- A single circle for positive relationship( )
- A x represents negative relation ship (x)
- An asterisk representing a strong negative relationship

The symbols describe the direction of correlation. In other words a strong positive
interrelationship would be a nearly perfectly positive correlation. A strong negative
interrelationship would be a nearly perfectly negative correlation.



Step 5-Competitive Assessments
The competitive assessments are a pair of weighted tables (or graphs) that depict item for item
how competitive products compare with current organization products. The competitive
assessment tables are separated into two categories, customer assessment and technical
assessment.

CUSTOMER COMPETITIVE ASSESSMENT

The customer competitive assessment is the block of columns corresponding to each customer
requirement in the house of quality on the right side of the relationship matrix, The numbers 1
through 5 are listed in the competitive evaluation column to indicate a rating of 1 for worst and 5
for best.

Technical competitive assessment

The technical competitive assessment is often useful in uncovering gaps in engineering judgment.
When a technical descriptor directly relates to a customer requirement, a comparison is made
between the customer's competitive evaluation and the objective measure ranking.



Step 6 - Develop Prioritized Customer Requirements
The prioritized customer requirements make up a block of columns corresponding to each
customer requirement in the house of quality on the right side of the customer competitive
assessment. These prioritized customer requirements contain columns for importance to customer,
target value, scale-up factor, sales point, and an absolute weight.

IMPORTANCE TO CUSTOMER

Customer requirement by assigned rating. Numbers 1 through 10 & are listed in the importance to
customer column to indicate a rating of 1 for least important and 10 for very important. In other
words, the more important the customer requirement, the higher the rating.

TARGET VALUE

The target-value column is on the same scale as the customer competitive assessment (l for worst,
5 for best). This column is where the QFD team decides whether they want to keep their product
unchanged, improve the product, or make the product better than the competition.

The target value is determined by evaluating the assessment of each customer requirement and
setting a new assessment value that either keeps the product as is, improves the product, or
exceeds the competition. For instance, if lightweight has a product rating of 3 and the QFD team
wishes to improve their product, and then the target value could be assigned a value of 4.

SCALE-UP FACTOR


The scale-up factor is the ratio of the target value to the product rating given in the customer
competitive assessment. The higher the number, the more effort is needed. The scale-up factor is
determined by dividing the target value by the product rating given in the customer competitive
assessment. For instance, if lightweight has a product rating of 3 and the target value is 4, then the
scale-up factor is 1.3. The scale-up factor for designing a handlebar stem for a mountain bike is
shown in Figure. Note that the numbers for scale-up factor are rounded off in Figure.

SALES POINT

The sales point tells the QFD team how well a customer requirement will sell. The objective here
is to promote the best customer requirement. The sales point is determined by identifying the
customer requirements that will help the sale of the product. For instance, an aerodynamic look
could help the sale of the handlebar stem, so the sales point is given a value of 1.5. If a customer
requirement will not help the sale of the product, the sales point is given a value of 1.

ABSOLUTE WEIGHT

Finally, the absolute weight is calculated by multiplying the importance to customer, scale-up
factor, and sales point:

Absolute Weight = (Importance to Customer) (Scale-up Factor)(Sales Point)

The absolute weight is determined by multiplying the importance to customer, scale-up factor, and
sales point for each customer requirement. For instance, for reasonable cost the absolute weight is
8 x 1.3 x 1.5 = 16.



Step 7-Develop Prioritized Technical Descriptors
The prioritized technical descriptors make up a block of rows corresponding to each technical
descriptor in the house of quality below the technical competitive assessment; these prioritized
technical descriptors contain degree of technical difficulty, target value, and absolute and relative
weights. The QFD team identifies technical descriptors that are most needed to fulfill customer
requirements and need improvement

DEGREE OF DIFFICULTY
Degree of technical difficulty for implementing each technical descriptor, which is expressed in
the first row of the prioritized technical descriptors. The degree of technical difficulty, when used,
helps to evaluate the ability to implement certain quality improvements.

The degree of difficulty is determined by rating each technical descriptor from 1 (least difficult) to
10 (very difficult). For instance, the degree of difficulty for die-casting is 7, whereas the degree of
difficulty for sand casting is 3 because it is a much easier manufacturing process.

TARGET VALUE
A target value for each technical descriptor is also included below the degree of technical
difficulty. This is an objective measure that defines values that must be obtained to achieve the
technical descriptor. The target value for each technical descriptor is determined in the same way
that the target value was determined for each customer requirement.

ABSOLUTE WEIGHT


The last two rows of the prioritized technical descriptors are the absolute weight-and relative
weight. A popular and easy method for determining the weights is to assign numerical values to
symbols in the relationship matrix symbols.
=
=
n
i
j RijCi a
1

aj= row vector of absolute weight for the technical descriptor.(I=1,m)
Rij =weight assigned to the relationship matrix (I=1.,n,j=1.,m)
ci= column vector of importance to customer requirements(I=1..,n)
m = no of technical descriptor.
m = no of customer requirements.

For example the absolute weight for aluminum is
9*8+1*5+9*5+9*2+9*7+3*5+3*3=227

Relative weight

The relative weight for j
th
technical descriptor is given by replacing the degree of importance for
customer requirement with absolute weight for customer requirement it is.
=
=
n
i
Rijdi bj
1

b= row vector of relative weight for technical descriptor (j=1.,m)

di
=Column vector of absolute weight for customer requirements (I=..,n)

For example the relative weight for die-casting is

3*16+9*8+9*5+3*2+0*18+3*5+9*3=213



R.V. College of Engineering, Bangalore - 59
OBSERVATION SHEET
Date: Name: Reg. No.
Lab: Class: Expt No.

Title quality function deployment study

QARE Student Cycle 1

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

QARE Student Cycle 1

Hochgeladen von

Copyright:

Verfügbare Formate

DEPARTMENT OF INDUSTRIAL ENGINEERING &

Das könnte Ihnen auch gefallen