Sunteți pe pagina 1din 41

STATISTICAL QUALITY

CONTROL
STATISTICS
Study of
•Collection
•Presentation
•Analysis
•Interpretation
of data
QUALITY

The extent to which the products meet


the specifications
CONTROL

Involves
•Assessing product quality
•Assessing process deviations
CENTRAL TENDENCY OF DATA
Mean (average)
x/n
Median M
The central value after arranging data
in ascending or descending order
Mode
The value occurring most frequently
MEASURES OF VARIATION

Range Xmax - Xmin


Standard deviation

2
(X - X)
n-1
2
Variance SD
COEFFICIENT OF VARIATION

SD
X

Used when the averages of two sets of


data are far apart
The average and standard deviation of two
sets of data are contradictory
FREQUENCY CURVE
frequency

parameter
TYPES OF VARIATION

Random variation
The variation due to inherent characteristics
of the process
Can not be eliminated
Assignable variation
The variation due to some controllable
causes
Can be identified and eliminated from the
process
PROCESS UNDER STATISTICAL
CONTROL

A process is said to be in a state of


statistical control if it is governed only
by random causes of variation and there
is no assignable variation in it
NORMAL CURVE
frequency

68.26%
95.45%
99.73%
STANDARD NORMAL VARIATE

(X – X) / SD

•The normal curve using standard normal


variate instead of original data is called
standard normal curve
•The standard normal curve has a mean of
0 and a standard deviation of 1
CHARACTERISTICS OF NORMAL CURVE
•It does not touch x axis on either side
•It is symmetrical
•The total area under the normal curve is
1.0
•The mean is at the point of highest
frequency
•Mean = median = mode
CONTROL CHARTS

A tool for monitoring and thereby controlling


the process characteristics

Control charts for variables X - R chart


Control charts for attributes p chart
c chart
GENERAL LAYOUT OF A CONTROL
CHART
USL
+ 3 SD UCL

- 3 SD LCL
LSL
PRE REQUISITES FOR A CONTROL CHART

•Define the characteristic


•Frequency of sample
•Sample size
•Place of control chart
X - R chart

Controls within sample and between samples


variation X – chart R chart
X X / n
M
Central line R / m

M
Central line X / m
M

UCL D4 R
LCL D3 R
UCL X + A2 R
LCL X - A2 R
n A2 D3 D4 n A2 D3 D4

2 1.880 0 3.267 14 0.235 0.328 1.672

3 1.023 0 2.574 15 0.223 0.347 1.653

4 0.729 0 2.282 16 0.212 0.363 1.637

5 0.577 0 2.114 17 0.203 0.379 1.622

6 0.483 0 2.004 18 0.194 0.391 1.608

7 0.419 0.076 1.294 19 0.187 0.403 1.597

8 0.373 0.136 1.864 20 0.180 0.415 1.585

9 0.337 0.184 1.816 21 0.173 0.425 1.575

10 0.308 0.223 1.776 22 0.167 0.434 1.566

11 0.285 0.256 1.744 23 0.162 0.443 1.557

12 0.266 0.283 1.717 24 0.157 0.451 1.548

13 0.249 0.307 1.693 25 0.153 0.459 1.541


P chart (fraction defective chart)

p number defective / number inspected


p total number defective / total inspected
Control limits
Central line p
UCL p + 3 p (1-p)
n
LCL p -3 p (1-p)
n
c chart (defects chart)

Sample size is defined


Sampling frequency is defined
c c/n
Control limits
Central line c
UCL c + 3 c
LCL c - 3 c
Interpretation of control chart

Normal behaviour
Interpretation of control chart

One point out above


Investigate the cause of poor performance
Interpretation of control chart

.
One data point out below
Investigate the cause of improvement
Interpretation of control chart

..

Two data points near upper limit


Investigate cause of poor performance
Interpretation of control chart

. .
Two data points near lower limit
Investigate the cause of improvement
Interpretation of control chart

. .. ..

Five successive data points above central line


Investigate the cause of sustained poor performance
Interpretation of control chart

.. .
. .

Five successive points below central line


Investigate the cause of sustained improvement
Interpretation of control chart

. . .
. . . ..
. .

Five successive data points on increasing or


decreasing trend
Investigate the cause of progressive change
Interpretation of control chart

Erratic behaviour
Investigate
Interpretation of control chart

Sudden change in level


investigate
Process capability

The ability of a process to produce within


specifications
Process capability index

Cp (USL – LSL) / 6 SD

Cpk (USL – X) / 3 SD
(X – LSL) / 3 SD
(which ever is lower)
Process is capable if
Cp and Cpk are greater than 1
Cp and Cpk greater than 1
Cp greater than 1
Cpk less than 1
Cp and Cpk less than 1
ACCEPTANCE SAMPLING

Deciding the acceptability or otherwise of a lot


on the basis of inspecting a sample
Sampling inspection is necessary if

•The inspection is destructive


•The inspection is costly
•The inspection takes too much time
•The lot size is too large
SAMPLING RISK
sample
Good bad
good

Producer’s risk
lot

Supplier’s risk
bad
Acceptable quality level

The limit of defectives % acceptable in


the lot
A SAMPLING PLAN INCLUDES

•Lot size
•Sample size
•Acceptance number
•At a specific AQL
Probability of acceptance

Lot defective %
c
N
N

S-ar putea să vă placă și