Taguchi Loss Function

Taguchi
Loss
Function
1/6/17
Traditional Goal post View of

Quality
1/6/17
In
the
past,
quality
means
conformance to valid customer
requirement. That is as long as an
output fell within specification limits
around a target value (also called
nominal
value);
it
is
deemed
conforming, good or acceptable. This
is known as goal post definition of
quality.
1/6/17
1/6/17
Losses are the minimum until lower

specification limit (LSL) or upper
specification limit (USL) is reached.
Then
suddenly,
losses
become
positive and constant, regardless of
the magnitude of the deviation from
the nominal value.
1/6/17
Taguchi says that as soon as quality

characteristics depart from its exact
mean or target, performance starts
deteriorating with increased customer
dissatisfaction,
although
the
measurement may be well within the
specification limits. Thus departure from
mean is a quality loss, because the part
may require rework at extra time and
cost. Taguchi expresses this quality loss
as a quadratic function, known as
Quality Loss Function (QLF).
1/6/17
The quality loss is taken to mean a

cost to the customer (and the effect to
the market share) as well as the
company cost overheads through
scrap or rework and loss of productive
time. Hence this has been termed by
Taguchi as a loss to the whole
society.
1/6/17
Nominal the best: some quality

factors are such that the best value
would be the target value; neither
higher nor lower values are desired.
1/6/17
Bigger the better: some quality

factors are such that their values need
to be set as high as possible, if possible
at infinity.
1/6/17
9
Smaller the better: some quality

factors are such that their values
needed to be set as small as possible,
preferably zero.
1/6/17
10
Taguchi Loss Function

Equations
y = the measured value of the quality characteristic for a
particular item of product.
m = the nominal value, or target value for quality
characteristic.
A = the loss(cost) of exceeding specification limits (the cost to
scrap a unit of output)
d = the allowable tolerance from the nominal value that is
used to determine specification limits.
k = A/d2 = a constant.
= population standard deviation.
s = sample standard deviation, when is unknown.
= population mean, if it is unknown then y* is used.
y* = average or mean of the measured values.
1/6/17
11
Taguchi Loss Function

For Single Piece
Smaller
better
the
Bigger
better
the
Nominal
better
the
L(y)= k( y-m)2
L(y)= k(1/y)2
For Multiple Piece

(1)
(2)
L(y) = k(y-m)2 (3)
L(y)= k(2+(y*-m)2 ) (4)

L(y)=k/2 [1+32/ 2] (5)
L(y)= k[2+(y*-m)2
or
L(y)= k[s2+(y*-m)2 (6)
1/6/17
12
1/6/17
13
Design parameters
Specification
Material
S15C
External diameter
11.1
Internal diameter 8.72+.012

of the large hole
Internal diameter 7.95+0.12
of the small hole
Thick measure
19.1-0.1
Distance measure
47.60.1
Surface treatment
S4
H.L.
0.2-0.3
H.V.
550-750
1/6/17
14
1/6/17
15
1/6/17
16
medical company produces a part that has a hole

measuring 0.5" + 0.050". The tooling used to
make the hole is worn and needs replacing, but
management doesn't feel it necessary since it still
makes "good parts". All parts pass QC, but several
parts have been rejected by assembly. Failure costs
per part is $45.00 Using the loss function, explain
why it may be to the benefit of the company and
customer to replace or sharpen the tool more
frequently. Use the data below:
Measured
Value
0.459 | 0.478 | 0.495 | 0.501 | 0.511 | 0.527
0.462 | 0.483 | 0.495 | 0.501 | 0.516 | 0.532
0.467 | 0.489 | 0.495 | 0.502 | 0.521 | 0.532
0.474 | 0.491 | 0.498 | 0.505 | 0.524 | 0.533
0.476 | 0.492 | 0.500 | 0.509 | 0.527 | 0.536
1/6/17
17
Solution:
The
average of the points is 0.501 and the

standard deviation is about 0.022.
find k,
using L(x) = k * (x-t)^2
$45.00 = k * (0.550 - 0.500)^2
k = 18000
next, using the Average loss equation: L=k *
(s^2 + (pm - t)^2)
L = 18000 * (.022^2 + (.501 - .500)^2) = 8.73
So the average loss per part in this set is $8.73.
For the loss of the total 30 parts produced,
= L * number of samples
= $8.73 * 30
= $261.90
1/6/17
18
From
the calculations above, one can

determine that at 0.500", no loss is
experienced. At a measured value of 0.501",
the loss is $0.018, and with a value of
0.536", the loss would be as much as $23.00.
Even though all measurements were within
specification limits and the average hole size
was 0.501", the Taguchi loss shows that the
company lost about $261.90 per 30 parts
being made. If the batch size was increased
to 1000 parts, then the loss would be $8730
per batch. Due to variation being caused by
the old tooling, the department is losing a
significant amount of money.
1/6/17
19

Taguchi Loss Function

Uploaded by

Copyright:

Available Formats

Taguchi Loss Function

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Taguchi Loss Function

Uploaded by

Copyright:

Available Formats

Taguchi

Traditional Goal post View of

Losses are the minimum until lower

Taguchi says that as soon as quality

The quality loss is taken to mean a

Nominal the best: some quality

Bigger the better: some quality

Smaller the better: some quality

Taguchi Loss Function

Taguchi Loss Function

For Multiple Piece

L(y) = k(y-m)2 (3)

L(y)= k(2+(y*-m)2 ) (4)

Internal diameter 8.72+.012

medical company produces a part that has a hole

average of the points is 0.501 and the

the calculations above, one can

You might also like