SPC

Statistical Process Control
Statistical : Collection, Analysis & Implementation of

Numerical data (Facts & Figures).
Process : A Combination of man, machine, material, method &
equipments for producing the desired product & Service
Control : Performance comparison with standard and taking
necessary action for achieving the end result.
is to influence and to control the quality at the time of
manufacturing.
Statistical Process Control (SPC) focuses on controlling
the manufacturing process to prevent defects rather than detect
them.

Purpose and Application of SPC
BENEFITS OF SPC
Less downtime
Fewer production interruptions
Less trouble at next operation
Improved performance
More involved operators
Fewer complaints
Fewer complaints
Fewer Defectives
Savings in time
Increased Capacity
Less Scrap.
Sources of variation which are built into the process and
will be caused by such problems as
Types of variation
Inherent variation (chance cause) :
These are the source of random variation , the extent of
which can be measured and monitored. The level of variation
will continue unless something special occurs.
Environment
power Slight variation in raw-material / machine,
lack of human perfection in reducing instruments etc.,
Sources of variation which will be due to specific
identifiable causes such as variation in
Raw material
Tool wear
Setting
State of maintenance
Batch of defective raw material
Untrained operator &
faulty setup etc.,
Special cause variation (assignable cause)
CHANCE CAUSES ASSIGNABLE CAUSES
Man
1)Normal errors in handling material and
machine slides
1)Carelessness
2)Mood
3)Untrained operator
Machine 1)Normal play in Machine slides 1)Excessive play (old machine)
2)Poor maintenance
Material 1)Hardness within Tolerance 1)Defective material
2)Hard spots
3)Blow holes
4)Mix-up of material
Method 1)Small variation in job clamping 1)Wrong speed and feed
2)Bush oversize
3)Clamps broken
4)Wrong drawings
Environment 1)Temperature change 1)Checking of Hot components
with cold instruments
Defect Any non-conformance of an item with respect to the
specification for a characteristic or dimension
Defective: Any item containing one or more defects.
The basic element of SPC is data analysis, and the primary
document is control charts.
SPC involves control followed by improvement. Processes are
initially brought under control by identifying and eliminating
the ASSIGNABLE CAUSES of Variation.
A controlled situation is one when the process is operating only
under the influence of CHANCE CAUSE(Inherent).
Statistics is the art of making decisions about a process
based on an analysis of information obtained from a
process.
Statistical methods provide means by which the output is
studied and evaluated.
Statistics is quite simply the collection, collation and use
of data.
Statistics
Average ( ) or Mean :

It simply means sum of all the individual observed data divided
by the no. of observations.
X1 + X2 + X3 + - - - - + Xn
n
n = no. of Observations.

Example : Observations = 20, 24, 26, 28, 43, 18

n = 6 20+24+26+28+43+18

X =
X
X =
6
= 26.5
Range (R) :
Range is measure of the variation in a set of data. It is
calculated by subtracting the lowest value in the data set
from the highest value in that same set.
R = X max X min

X max = 43, X min = 18
R = 43 18 = 25
Observations = 20, 24, 26, 28, 43, 18
Assuming Normal distribution as the pattern of variation
for measured characteristics, we know that Average + 3 / - 3
comprises almost all (99.73%) of the observations.
Hence 6 sigma is used as a measure of process capability.
Lack of statistical control results in a variability larger than 6
sigma.
Thus 6 Sigma is valid for statistically stable process and
sigma is estimated as
= / d2

R
Standard Deviation () :
Standard deviation is measure of the of the process
output or the of a sampling statistic from the process (eg.
Of subgroup averages) denoted by the greek letter (Sigma).
n 2
(Xi )
i = 1
n
Where,
= Standard deviation, Xi = Observed value
= Average , n = No. of Observations
X
=
X
spread
spread
Indices
Tolerance
Process Capability
= =
USL - LSL
6 / d2 R
Cp = Potential Capability
The Capability Index is defined as the ratio of the
Specification spread or Tolerance to the Process Variation
The capability index is defined as:

The capability index show how well a process is able to
meet specifications. The higher the value of the index, the more
capable is the process:

C
p
< 1 (process is unsatisfactory)
1 < C
p
< 1.6 ( process is of medium relative capability)
C
p
> 1.6 (process shows high relative capability)
Cpk = Achieved Capability
3
Min of Or
- LSL
3
Cpk =
USL
X
X
Where, = / d2 and the actual average is R X
Chart Factor 2 3 4 5 6 7 8 9 10
A2 1.881 1.023 0.729 0.577 0.483 0.419 0.373 0.337 0.308
F2 1.88 1.187 0.769 0.691 0.549 0.509 0.432 0.412 0.363
d3 0 0 0 0 0 0.076 0.136 0.184 0.223
d4 3.267 2.575 2.282 2.115 2.004 1.924 1.864 1.816 1.777
B3 0 0 0 0 0.03 0.118 0.185 0.239 0.284
B4 3.267 2.568 2.266 2.089 1.97 1.882 1.815 1.761 1.716
d2 1.128 1.693 2.059 2.326 2.534 2.704 2.847 2.97 3.078
Table of factors for computing control chart limits
Subgroup Size
X
X
median
R
S
CONSTANT
Inferences based on Cpk :

1. Cpk = 2.00 represents a very capable process. Such an index is
achieved only when you are able to produce within 50% of
tolerance.

2. Cpk = 1.33 is considered to be a minimum capability
requirement and most industries demand a Cpk of 1.33 as
minimum criterion. This is achieved if you are able to produce
75% of the tolerance.

Inferences based on Cpk :

3. Cpk = 1.00 represents a just capable process and the process
can give good results only if it is perfectly centered. Even a
small deviation from the mid of specification will result in non
conformance. Here 100% of the tolerance is used. Not a very
desirable solution.

4. Cpk < 1.00 means that the process not capable of meeting the
specified tolerance. In this case, rework and rejection will be
inevitable.
X
UCL
-
5
10
15
20
25
30
1
X
4
8
9
.
9
5

-

4
9
1
.
9
5
4
9
1
.
9
5

-

4
9
3
.
9
5
4
9
3
.
9
5

-

4
9
5
.
9
5
4
9
5
.
9
5

-

4
9
7
.
9
5
4
9
7
.
9
5

-

4
9
9
.
9
5
4
9
9
.
9
5

-

5
0
1
.
9
5
5
0
1
.
9
5

-

5
0
3
.
9
5
5
0
3
.
9
5

-

5
0
5
.
9
5
5
0
5
.
9
5

-

5
0
7
.
9
5
5
0
7
.
9
5

-

5
0
9
.
9
5
5
0
9
.
9
5

-

5
1
1
.
9
5
LSL
USL
X
USL
X +3S
X -3S
X -1S
X -1S X -2S
X -2S

SPC

Uploaded by

Copyright:

Available Formats

SPC

Uploaded by

Document Information

Original Title

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

SPC

Uploaded by

Copyright:

Available Formats

Statistical Process Control

Statistical : Collection, Analysis & Implementation of

You might also like