Professional Documents
Culture Documents
( Reaffirmed 2001 )
Indian Standard
SAMPLING INSPECTION PROCEDURES
PART II INSPECTION BY VARIABLES FOR PERCENT DEFECTIVE
Seventh Reprint AUGUST 1998
( Incorporating Amendment No. 1 )
UDC 519.271.3
: 620.111
(083)
(8 Copyright 1996
Gr8
June 1966
Indian Standard
SAMPLING INSPECTION
PART II INSPECTION
BY VARIABLES
PROCEDURES
FOR PERCENT
Committee,
DEFECTIVE
SMDC
Refwsent{ng
The Tata Iron & Steel Co Ltd, Jamshedpur
Chairman
DR A. V. SUKHATME
Membtrs
SHRI V. D. ACARWAL
SHRIJ. C. BANERJEE
SHRI A. K. BHATTACHARYA
SHRI M. N. BH~DE(dffernalc)
DR U. N. BHRANY
SHRI J. N. BURMAN(Alftmafc)
DR M. K. BOSE
SHRI S. K. DUTTA
STANDARDS
BUREAU
OF
INDIAN
&\NAK 1lEIAVAN,
9 BAI1ADUR SHAH %.41-.4R AfARCJ
NEW DFI11HI 110tl\i:!
otr pcgc 2)
Members
SHRI D. B. L.AHIRI
DR A. MAT-THAI (Ahnate)
SHRI N. MAJUMDAR
Rcpresenting
Indian Statistical Institute, Calcutta
Tables, SMDC
4 : P6
7
8
8
9
9
:o
11
12
13
15
15
16
17
18
FOR VARIA-
4
6
6
OR -
n .
U-L
19
Indian Standard
SAMPLING
INSPECTION
PART. II INSPECTION
BY VARIABLES
0.
PROCEDURES
FOR PERCENT
DEFECTIVE
FOREWORD
0.1 This Indian Standard was adopted by the Indian Standards Institution
on 1 October 1965, after the draft finalized by the Methods of Sampling
Sectional Committee
had been approved by the Structural and Metals
Division Council.
0.2 Part I of this standard dealing with inspection by attributes and by count
of defects had been issued earlier with a view to facilitating the wide-spread
use of sampling inspection in those situations where items can be classified
as defectives or non-defectives, satisfactory or non-satisfactory.
However,
this type ofinspection may require comparatively larger sample size and hence
in certain situations like the determination
of warp breaking strength of
cotton fabrics, it may become uneconomical due to the destructive nature or
prohibitive cost of testing.
In such cases, inspection by variables, wherein
quality is measured on a continuous scale like tensile strength of steel wire,
This standard has been prepared to
may be more useful and economical.
meet the growing demand for the use of sampling plans for inspection by
variables.
0.3 Further economy in sampling inspection may be obtained if the units in
tire lot are quite uniform in quality as a smaller sample may then be adequate
to represent the lot. This uniformity may be achieved by controlling
the
quality at the production stage itself and helpful guidance may be obtained
in this respect from IS : 397-1952*.
0.4 Sometimes the quality characteristics can be inspected both by attributes
and by variables, for example, the diameter of a shaft can be checked either
In such cases, a decision has to be
by gauging or by actual measurement.
Some of
made whether inspection should be by attributes or by variables.
the important considerations which provide the basis for a suitable choice
have been given in 4.2.3 of IS : 1548-1960t.
However, a more detailed
discussion of such considerations is given in 3 of this standard.
*Method for statistical quality control during production by the use of control chart
(s&w revised).
tManua1 on basic principles of lot sampling (tince stied).
AND SYMBOLS
2.0 For the purpose of this standard, the following definitions shall apply.
The symbols used in this standard including those for some of the terms
defined below are explained in Appendix A.
2.1 Sampling Inspection - Inspection in which only a portion of a lot is
inspected with a view to making a decision about accepting or rejecting
the lot.
2.2 Sampling Plan - A statement
for making decisions about the lot.
2.3 Item - Ultimate
be performed.
on which inspection
will
2.4 Lot - A collection of items from which a sample is drawn and inspected
to determine its acceptability.
2.5 Lot Size (N) 2.6 Sample
2.7 Sample
Size
2.8 Defective
requirements.
Number
Collection
-
(n) -
of items in a lot.
from a lot.
of items in a sample.
the
ratio
of the number
of
of
.
IS : 2500 (Part II) - 1965
2.14 Range (R)The difference between the largest and the smallest
observations or test results in a sample (see also Appendix A).
NOTE 1 - If the sample size is less than 10, the range shall be calculated for the sample
as such. If, however, the sample size is 10 or more (in multiples of 5), the observations
shall be divided into subgroups of five each by taking them consecutively in the same
order as obtained.
The range of each subgroup shall then be separately determined.
2.15 Mean Range CR) - The mean of a set of ranges calculated for subgroups of five observations in the sample (see also Note 1 and Appendix A).
2.16 Process Average -The
average percent defective of the products
submitted by the producer for original inspection.
(Original inspection
is the first inspection for the particular quantity of product as distinguished
from the inspection of products which have been re-submitted after prior
rejeclion.)
2.17 Acceptable
Quality Level (AQL) - The maximum percent defective
that, for the purpose of sampling inspection, can be considered as a satisfactory process average.
NOTE2 - When a consumer designates some specific value of AQL, he indicates to the
producer that his (the consumers) acceptance sampling plan will accept the great majority
ofthe lots that the producrr submits, provided the process average level ofpercent defective
in thrse lots is not greater than the designated value of AQL.
Thus, the AQL is a
desigrlatcd value of percent defective that the consum+r indicates will be accepted most
of thr time (approximately 89 to 95 percent in this standard).
BETWEEN
THE
Is
3.6 Variables plans are based on the assumption that the distribution of the
quality characteristic is normal.
Hence it is important that these plans are
not used indiscrimiitely.
In case the assumption of normality is in doubt,
it is advisable to obtain the guidance of a competent statistician to ascertain
the feasibility OF application of these plans.
4. PRELIMINARIES
TO THE
SELECTION
OF VARIABLES
INSPECTION PLANS
4.1 Formation of Lots - A lot (see 2.4) should, as far as possible, consist
of items of a single type, grade, class, size, etc, produced under relatively
uniform conditions of manufacture by a single firm so that the items in the
lot are of uniform quality.
In such a case, the size of the sample to be
tested in the lot for a given protection would be small and inspection would
be economical.
Furthermore, a lot should, consistent with the conditions
of homogeneity, be as large as possible so that the incidence of cost of inspection per item is minimized.
A lot can be a stationary lot or a moving lot
subject to the convenience of stacking, access to each item in the lot and
identification.
Each lot shall be properly identifiable and each stationary
lot shall be presented as far as possible, in such a way that the inspector
8
Is:
2!ioo
(Part II)-196!5
will have easy access to all parts of the lot to select at random
making up the sample.
the items
Variability Known-When
the variability in the lot, that is, the
value of the lot standard deviation (u) is known beforehand either
from the past experience, control chart data or any other means.
b)
Variability
Unknown - Standard Deviation Method - When the variability in the lot is not known and is estimated from the sample
standard deviation (s).
Unknown - Range Method - When the variability
in
the lot is not known and is estimated from the sample range (R)
or sample mean range (H).
c) Variability
separately
for the
1) When one-sided specification limit, that is, either an upper specification limit (U) or a lower specification limit (L) is given.
2) When two-sided specification limits, that is, both the upper and the
lower specification limits are given.
4.2.1 For the same degree of protection, the sample size is minimum
for the variability known method and maximum for the range method and,
Moreover, from the administratherefore, the former is most economical.
tive and computational
points of view the variability known plans are
easiest to operate as they require the calculation of mean alone, further
computations remaining the same for all the three types of plans.
The
range plans come next as they require the calculation of both the mean
and the range and lastly the standard deviation plans which require the
calculations of both the mean and standard deviation, the latter being more
Though the variability known
cumbersome to calculate than the range.
plans are the simplest to operate, the requirement of a priori knowledge
of variability is a stringent one.
4.3 Drawing
of Samples
- The sampling inspection plans included
in the standard assume that the items constituting a sample are selected
from the lot at raxdum. Technically, a sample is said to have been selected
at random if the methcd of selection gives the same chance to every item
in the lot for being included in the sample.
Various methods available
for collecting a random sample including the approximation
methods
like systematic sampling with random start have been discussed in 3.3.2
of IS : 154%1960*.
*Manual on basic principles of lot sampling (sincerevised).
OF SAMPLING
PLANS
5.1 Classification
of Sampling
Plans
5.1.1 A QL Plans
- The comprehensive set of sampling plans classified
in terms of AQL provided in the standard includes the following:
a) Variability known plans (see Table 2 on P 16);
b) Variabiiity unknown plans, standard deviation method (see Table 3
on P 17) ; and
c) Variability unknown plans, range method (see Table 4 on P 18).
5.1.1.1 The choice of a sampling plan for a particular product requires
the decision on the quality requirements;
if AQL plans are being used,
it is necessary to choose an appropriate
AQL.
In selecting an AQL value
a compromise may have to be struck between the quality desired and the
quality attainable.
If the AQL is superior to the quality that cau be
attained under usual production conditions, an excessive amount of product
will be rejected; on the other hand, if AQL is not exacting enough, an
excessive amount of inferior products may be accepte.2.
The value of
AQL may be specified by agreement between the parties concerned, giving
*Manual
on basic
principles
of lot sampling
(since rcuised).
10
known
b) Variability
unknown
plans,
c) Variability
unknown
plans,
These
plans;
standard
range
deviation
method;
and
method.
in Appendix
B.
5.2 Inspection
Level - In order to determine
which of the sampling
plans given in the standard
are to be used in a particular
case, having
decided
upon the quality requirement,
it is necessary
to decide upon the
Inspection
Level.
The term Inspection
Level is used to designate
the
relative amount
of inspection
one is required
to do.
A higher inspection
level means relatively more inspection
and consequently
lesser risk of accepting lots of quality worse than the chosen AQL.
That is to say, the higher
the inspection
level, the greater is the protection
against acceptance
of low
quality lots; but then the cost of inspection
goes up.
It may, therefore,
be necessary
to strike a compromise
between
a large sample which gives
a reliable estimate of the lot quality and a small sample which reduces the
inspection
cost.
Such a compromise
can be arrived at by proper selection
of the inspection
level.
5.2.1 Table 1 (see P 15), which is auxiliary
to Tables 2, 3 and 4 gives
five inspection
levels with the sample
size in code letters
(see 5.2.2)
Inspection
level I calls for the smallest
sample size, thereby
minimizing
the cost of inspection.
Inspection
level V gives relauvely
the largest
sample size, thereby
lessening both the risk of accepting
bad quality lots
and rejecting
good quality lots but increasing
the cost of inspection.
For
majority
of products
under normal
conditions
of acceptance
inspection?
a reasonable
compromise
between
the high inspection
costs amd the ri:.k
involved
may be achieved
by taking the sample size corresponding
to the
inspection
level IV.
5.2.2 Table 1 gives code letters B, Cl, D, . , . . , to indicate
the sample
sizes under different
inspection
levels for varying
lot sizes.
Tt is in terms
of a code letter selected from Table 1 that a plan will be selected from Tables
2 to 4. These code letters are known as sample size code letters designating
as they do a particular
sample size.
For example,
J denotes
a ;ample
11
(see Table
5.3.1
5.3.1.1 For one-sided specijcatiqn limits - On the basis of the AQL and
the sample size code letter chosen, the value of the sample size (a) and a
suitable factor (k) shall be obtained from Table 2. The mean of the n test
results shall then be calculated and the lot shall be declared as acceptable if:
a) the value of the expression
fication limit (U) is given;
b) the value of the expression
fication limit (L) is given.
(E+ko)
<
OR
(Z-J/o)>
L, when
AQL (%)
0.15
0.25
040
the
lot shall
Q the maximum
lower
be
speci-
declared
value specified
:
0.65
1.00
1.50 2.50
4.00
6.50
10.00
Upper Limit
of
-&
0.152 0,158 0.165 0.175 0.184 0.194 0.206 0.223 0.243 0.270 0.304
1
(Z--k o)>L.
5.3.2
Variability Unknown -
L, when
qG&ation
lot shall
limits-The
Q the maximum
(Z++ks)<U,
(E--ks)>L.
5.3.3
be declared
if:
Variability Unknown -
value specified
combination
of the chosen
and
Range Method
or (Z+kR)
Q U, when the
OR
b) the value of the expression (a-kR)
or (R--kR)
lc ver specification limit (L) is given.
5.3.3.2 For two-sided spec&ation limits -The
Acceptable if:
or GL
the maximum
or (Z+kR)
or (R-k@
5.4 Normal,
<
>
as
value
combination
of
U, and
L.
5.4.0 When the quality of the submitted lots shows significant shifts,
it is desirable to make appropriate changes in the sampling plans.
If the
quality deteriorates, it is necessary to tighten the inspection; if the quality
improves, it may be desirable to relax the inspection.
5.4.1 Notmal InsfJection- Inspection under a sampling plan that .is in
force for a particular product and producer is called Normal Inspection.
It may be continued as long as the quality of the product submitted is better
than or equal to the chosen AQL.
The consistency in maintaining this
13
5.4.2.2 From the tables given in this standard the choice of a plan for
tightened inspection shall be made in the following manner :
Retain the same sample size code letter as before but refer to an AQL
a step lower than the AQL used for normal inspection.
For example,
if the AQL used for normal inspection is 4.0 percent for the code letter
K (in Tables 2, 3 or 4), the AQL to be used for tightened inspection shall
be 2.5 percent for the same code letter.
5.4.2.3 In certain cases such tightened inspection may lead to an
increase in the size of the sample.
For example, if an AQL of 0.65 percent
has been used for normal inspection for the sample size code letter D, the
AQL and the code letter to be used for tightened inspection shall be 0.40
and E respectively.
5.4.3 Reduced Insrpection- If the quality of the submitted lots is consistently
better than the AQL chosen, reduced inspection may be resorted to either
by selecting lower sample size code letter than the one used for normal
inspection or by relaxing the AQL.
Since the former approach leads
to economics in inspection it is preferable, unless there is an agreement
to the contrary, to reduce inspection by changing over to a plan with a
lower sample size code letter than the one adopted for normal inspection.
5.4.3.1 The following criteria may be applied for changing over from
normal to reduced inspection and vice versa:
a) If none out of 10 consecutive lots has been rejected while on normal
inspection, change over to reduced inspection.
14
If a lot is rejected, and if at the same time the rejected lot is preceded
by less than 10 lots accepted on reduced inspection, change over
to normal inspection.
5.4.3.2 From the tables given in this standard, the choice of a plan
for reduced inspection may be made in the following manner :
Retain the same AQL as before but refer to the sample size code
letter one step lower than that used for normal inspection.
For example,
if the code letter used for normal inspection is J with an AQL of 2.5
percent (in Tables 2, 3 or 4), then the code letter to be used for reduced
inspection may be H for the same AQL.
3.4.3.3 In certain cases suitable plans for reduced inspection given
in 5.4.3.2
may not be available.
Reduced
inspection may then be
resorted to by choosing the sample size code letter one step lower and the
AQL one step higher than that used for aormal inspection.
For example,
if the code letter D and the AQL of 0.65 percent have been used for normal
inspection, then the code letter and AQL to be used for reduced inspection
may be C and 1 .O percent respectively.
6. TABLES
TABLE
AND
SAMPLE
ILLUSTRATIVE
EXAMPLES
LEVELS
AND
INSPECTIONLEVELS
*
III
IV
%ZE
I
I
II
Size
( s a m pl e
2to
9 I,
16 ,,
8
15
25
50
26 >,
100
51 >,
150
101 ,,
300
151 ,,
500
301 ,,
501 ,,
1000
1001 ,,
3000
3001 ,, 10000
10OOlandabove
B
B
B
B
B
B
B
B
C
D
E
F
B
B
B
B
B
B
C
D
E
F
G
H
J
K
15
Letters)
Code
B
B
B
B
C
D
E
F
G
,
II
B
B
C
D
E
F
G
J
K
L
M
C
D
E
F
G
J
K
L
M
M
N
TABLE2
v-
SINCLESAMIUNG
PLANS
?DR
lTY
VAKUBIL
KNOWN
MHKOD
--_irr
-1
4i 1 4 -c4
Cl
(Umues 5.1.1,
5.1.1.1.
5.2.1,
5.3.1.1
AaaPrAmm
SAWLE
I7.8
5.2.2,
coru
0.10
nY
0.15
rk-
0.25
0.40
QuAUlY
065
and lZ.m&s
mtd 5.4.2.2,
1 and
3)
LWBL
l-Ml
1.50
2.50
4.00
nk-k-Yr:Yn
k-
6.50
rk
1oaO
nk
E
F
G
2.39
1*56
I.36
1.25
1.09
0.936
0.755
0.573
1.42
1.33
1.17
1.01
oa25
0.641
0.728
1.94
1.81
I.69
I.56
I.44
1.28
I.11
0.919
%I9
2.07
I.91
lacl
1.69
1.53
1.39
1.20
0991
0.797
2.30
2.14
2.05
la6
I.78
I.62
1.45
I.28
1.07
11
0.877
246
2.34
2.23
2.08
1.95
I+0
I.63
1.49
10
1.31
12
1.11
14
0.906
2.49
2.37
2.25
2.13
1.96
1.83
10
1.70
11
1.51
13
1.34
15
1.13
17
0.924
2.50
2.36
2.26
2.13
10
1.99
I1
I.86
12
1.72
13
1.53
19
I.33
18
I.15
21
0.942
2.54
2.45
2.29
10
2.16
II
2.01
12
I+8
14
1.75
15
I.56
18
I.38
20
I.17
24
0.964
2.54
10
2.45
11
2.51
12
2.18
13
2.03
I4
1.69
15
1.75
18
1.57
20
1.38
23
1.17
27
O-965
II
2.59
12
2.49
13
2-35
I4
2.21
16
2.07
17
1.93
19
1.79
22
1.61
25
I.42
29
I.21
33
0.995
T-3
3XNGLEMMPLING
v-
(Clmw5.1.1,5.1.1.1,5.2.1,
@lo
0.15
0.25
I t::
2.42
2.24
2.32
2.20
E
2151
;:30
241
::;8
22:;
2-60
2.45
2.45
2-50
2.24
Et
2135
UNKNOWN-3TANDARD
andExmnpk4)
1.00
1.50
2.50
4.00
6.50
0.765
lO*OO
T
1.4s
1.53
1.34
140
1.12
1.17
1.24
Y-t?
1107
Z:X:::
:6?
O-675
::g
1.79
::g
1.65
1.33
1.41
1.47
:::;
1.30
3.955
1.03
1.09.
0.755
0.828
0.886
::z
1.66
1.69
I*72
1.73
1.51
1.53
1.55
i:;
1.12
1.14
1.15
0.917
8:Z
1.76
1.76
1.80
1.57
::g
1.42
1.18
1.18
1.21
8:E
1.00
i:;
::g
Use first sampling plan below the arrow, that b. both sample size as well as k value.
acecds lot size, cvuy item in the lot shall be inspected.
TABLE4
8Amaamv~FLANsFoRvUFNLNOWN-MBTHOD
SMCL8
(C1aw.m5.1.1,5.1.1.1,5.2.1,5.2.2,5.3.3.1
amf5.4.2.2, andExmnplrt2aad4)
_0.10
0.15
0.25
0.40
0.65
1.00
1.50
2.50
03!39
1.05
I.01
o&3
0.502
0450
0431
0401
0.364
0.352
0.272
0.613
0.755
0.792
0.569
0.703
0.738
0.525
0.650
0.684
::%
0.610
0.405
0.507
0.536
0.336
0.424
0.452
0.266
0.341
0.368
0.779
0647
0.654
0.658
0.571
0.577
0.581
0.484
0.490
0.494
0.398
8%
0.591
0.598
0.503
0.510
0.415
0.42 I
;:g
8:::
0.848
E:
0.723
0.730
0.734
0.92 1
0.931
0.860
0.893
0.803
0.812
0.746
0.754
:::
t :::
;:z
!:E
0.978
0.988
-i-
0.587
0.525
0.498
*
0.659
0.811
0.850
O-896
1040
0.598
0.565
+
0.702
0.863
0.903
0.951
0.959
0.964
6.50
0.:51
0.614
0.916
0.958
4.00
EE
EC!
Use first sampling plan below the arrow, that i, both ample
or exceeds lot size, every itan in the lot shall be inspected.
STANDARD
(Claw
OF
DEVIATION
B
C
D
3
4
5
20
O-15
0.25
METHOD
ACCEPTABLE
5.10
FOR VARUBKITY
&-
040
065
QIJALXTY
L
1.00
LEVEL
1.50
2.50
4.00
6.50
10.00
0.436
0.453
0.475
0.502
0.339
0,353
0.374
0.399
0.432
0.472
0.294
0.308
0.323
0.346
0.372
0408
0.452
0.242
0.253
0.266
0280
0.295
0.318
0.345
0.381
0.425
0.214
0.224
0.235
0.248
0.261
0.276
o-298
0.324
0.359
0.403
0.195
0.202
0.211
0.222
0.235
0.248
0.262
0.284
0,309
0.344
0.386
o-190
0.197
0.206
0.216
0.229
0.242
0.255
0.277
0.302
0.336
0.377
0.238
0.251
0.213
0.297
0.331
0,372
25
0.187
0.193
0.203
0.212
0.225
30
0.185
0.192
0.201
0.210
0.223
0.236
0.249
0.270
0.295
0.328
0.369
35
0.183
0.189
0.198
0.208
0.220
0.232
0.2.45
0.266
0.291
0.323
0.364
nf
40
0.182
0.188
0.198
0.207
0.219
0232
0.245
0.266
0.290
0.323
0.363
0.204
0.317
0.356
50
0.178
0.164
@lQ4
0.203
0214
0.227
0.241
0.261
TN&E6
TElEuPPEnLIMITFoRTxiEvALuEoF
FOR VAmmnrrY
UNKNOWN-RANGE
METHOD
(Claurc 5.3.3.2,
and lZwmplr4)
f
sAyPLEslzs
CDDB
SAxFEL"
ACCEPTABLE
Quam
LEVEL
.5
l_JnTEa
0.10
0.15
0.25
3-----_--
4-----
10
15
0.40
0.65
l*OO
1.50
2.50
4.00
6.50
lO*oo
0.833
0.865
0.907
0.950
0.756
0.763
0.836
0.891
0.965
1.056
0.730
0.764
Oaol
0.857
0.913
I.011
1.118
0695
0.727 0.765
@I304
0.8s
0.910
0905
1.086
1.209
3,529
0.553
0.579 0.610
0.642
0.677 o-730
0.793
0.876
0.977
0.477
0.493
0.517
0.542 0.572
0.602
0.637 0.688
0.746
0.830
0.928
25
0447
0.463
0486
0.509 0.537
0.567
o+oo
0.707 0.785
0.879
30
0442
0.457
0.480
0.503
0.699
0.776
0.870
35
0.438
0.454
0.476
0.499 0.527
0.556
0.771
0.864
40
0.432
0.447
0469
0.492 0.519
0.684
0.761
O-852
50
0.426
0.441
0463
0486
0.542
0.676
0.752
0.843
0.531 0.560
0.503
0.649
0.593 0.642
0.573 0.621
IS t 25oo(Pmtn)-l965
likamjlt 1:
The specified minimum yield point for certain steel castings is 40.0
Suppose lots containing 400 items are submitted for inspection.
kg/n&.
A single sampling variables plan with inspection level III and an AQL
Suppose the variof 2.5 percent is adopted for the purpose of inspection.
ability (u) is known to be 1.O kg/mm*.
Reference to Table 1 then gives the sample size code letter F corresponding
to which Table 2 shows the sample size 5 and factor k equal to 1.39.
From the lot, 5 castings shall then be selected at random and their yield
points determined.
Suppose the yield points for the 5 castings are 421,
39.9, 40.7, 43.2 and 42.6.
The mean (3) is then found to be 41.7.
Also, the value of the expression (Z--ka) ccmes out to be 41.7- 1.39 x 1.O
Since this value is greater than the lower specification limit
=40.31.
of40.0, the lot shall be accepted.
The LTPD
Example 2 :
i) The specified maximum resistance of a certain electrical component
is 660 ohms. Suppose lots containing 150 components are submitted for
inspection. If it is agreed to use inspection level IV, an AQL of one percent
and single sampling variables plans for variability unknown (standard
deviation method), then reference to Tables 1 and 3 gives the sample size
10 and the factor k equal to 1.72.
Suppose the resistances (in ohms) of the 10 components selected at random
from the lot are 639, 640, 650, 647, 662, 637, 652, 643, 657 and 649.
Then
f=647.6,
s=
and
(639~-647*6)*+(640-647.6;*+.
9
. . . . . +(649-647.6)
= @%3=8.09
The value .of the expression (Z+ks) =647.6 + 1.72 x 869 =661.5.
Since
this value is greater than the maximum specification limit of 660 ohms,
the lot shall be rejected.
ii) In the above example, if it is intenrsd to use the single sampling
variables plan for variability unknown (range method), then reference to
Tables 1 and 4 gives the sample size as 10 and factor k equal to 0.703.
21
IS : 25oo(PartII)-1965
Taking the above values of sample results, it is found that :
2 =647.6
Range of first 5 results -662 -639 =!3
Range
of last 5 results=657-637
=20
23+20
2
=21.5
Since
660 ohms,
The LTPD
value associated with the above plans can be read from Fig. 6
Thus lots containing approximately 19.5 percent defectives
would be accepted only 5 percent of the times by the above plans.
(see P 34).
Exaniple
3:
to Tables
k equal to 1.38.
&=m
of 0.243. Hence the lot is rejected straightway without drawing
items for sample inspection.
Exampk
the 20
4:
equal to 0.284.
in 15 containers selected
63.5, 66.0, 65.0, 68.5, 69.5, 66.5, 67.0, 62.5, 66.0, 67.5, 64.0, 69.0, 70.0,
66.0, 66.5
22
Is : 2!m(PartlI)-1965
Then
compute
997.5
mean
(2) = 15=66.5
(63.5-66.5)*+.
and s=
. . . . . +(66.5-665)s
14
2.19
= 1.
GL
=0.219
=66.5-
which
limit of 0.284.
1.47 x 2.19=63.3
1.47
2.19 =
speci-
equal to 0.688.
Range
=6.0
Range
=5.0
Range
-62.5
6.0+5.0+6.0
3
=5.7
_
Applying
given in 5.3.3.2,
-&
=5<
=0.57
which
.i.s
limit of 0.688.
2+k~=66~5+0.610~
limit,
5.7=70.0
is equal
and 3-k~=66~5-0~610~5~7=63~0
specification limit.
is
to the maximum
greater
than
the
specification
minimum
APPENDIX
x,+x*.
(I=
(x1-
_g*+ . . . . + (xx- @a =
J
s
.N
(x12+. . . . . . +x2)
JJv
--Jvp
sample standard deviation; if x1, x2,. . . . . . . . , x,, are the n measurements of items in the sample, then :
s=
. . .+x,
n
(x1 -a>*+.
. . . . . + (xn -3)2_
(n-l)
- J-
(x,2+. . . . . . +x,*)---nz*
n-l
measurements of items in a
range; if x1, x,, . . . . . . . . x,, are the
sample, arranged in the ascending order of magnitude, then :
R=x,-x,
mean range; if R,, &, . . . . . . . . R, are the ranges of m sub-groups
of five observations each (so that sample size n=5 m), then:
R=
R1+R,.
. . . . . . .R,,,
m
<
>
for
single
Is :25oo(PartlI)-1965
APPENDIX
(Clause 5.1.2)
FORMULA3 AND TABLES FOR CONSTRUCTING
SAMPLING
AQL-LTPD
VARIABLES PLANS
ONE-SIDED SPECIFICATION LIMITS
B-O. STIPULATIONS
SINGLE
FOR
VARXABILITY
KNOWN
METHOD
52.
VARIABILITY
METHOD
UNKNOWN - STANDARD
DEVIATION
9.96 TG oa9
(awe
B-O)
0.00
0.01
0.02
0.03
0*04
0.05
2.326
2.054
l+l81
I.751
1645
1.555
1476
1.405
1.341
1.282
I.227
1.175
1.126
la80
I.036
0.994
0.954
0915
0.878
0.842
0.806
0.772
0.739
0.706
0.674
0.643
0.613
0.583
0.553
o-3
0.524
0.496
0.468
0440
0.412
0.385
0.358
0.332
0.305
0.279
0.4
0.253
0.228
0.202
0.176
0.151
0.126
0. loo
0075
0+50
0*025
0.5
OMO
FB 4OTlON4
0.06
0.07
008
0.09
-0m5
-0.050
-0.075
-O*lOO
-@126
-0*151
-0.176
-0.202
-0.228
0.6
-0.253
-0.279
-@305
-0.332
-0.358
-0.385
-0.412
-0440
-0.468
-0.496
0.7
-0.524
-0.553
-0ci83
-0.613
-0643
-0.674
-0.706
-0.739
-0.772
-0al6
0.8
-0a42
-0.878
-0.915
-0.954
-0.994
-1.036
-la80
-1.126
-1.175
- 1.227
0.9
- 1.282
-1.341
- 1.405
- 1.476
- 1.555
- 1645
- 1.751
-I*881
-2.054
-2.326
Is : 2!5oo(PartII)-1965
found out.
The values of k and n which determine
plan uniquely are given by the following formulae :
B-3. VARIABILITY
UNKNOWN-RANGE
ME? HOD
is given as before by
[R is to be calculated
if the sample size is less than 10. In case the sample size is 10 or more,
R, based on the sub-groups of five samples each, is to be calculated].
The
values of k and n which determine the desired sampling plan uniquely are
obtained as given in B-3.1.1.
B-3.1.1 The values of n and k are first obtained by using the formulse
given under B-2.
In case n comes out to be less than or equal to 9, k is
determined from k=Dk where the values of D are given below for the
various sample sizes :
Sample Size, n
Values of D
om6
0.590 8
0.485 7
0.429 9
0.394 6
0.369 8
0.351 2
0.336 7
Is f 25oo(PartlI)-1965
APPENDIX
(.JVotc
3 under
Clause5.0)
THE OPERATING
GHARACTERB
TIC
CURVES
G-1. Figures 1 to 11 (su P 29-39) give the complete set of operating characteristic curves for the sampling plans based on variability unknownstandard deviation method as given in this standard. These curves are
obtained by plotting the percent of lots expected to be accepted against
percent defectives in submitted lots. The curves given would also approximately hold good for the plans pertaining to the variability known and range
methods with the same sample size code letter.
G-2. For any specified value of the acceptable quality level (see 2.17),
all the OC Curves obtained for the different sample size code letters are
given in the same figure. Thus, there are 11 figures corresponding to the
11 values of AQL from 0.10 to 10.00 percent specified in the standard.
The sample size code letter corresponding to any plan has been indicated
on the relevant OC Curves in each of the figures given.
C&S. For any percent defective, the percentage of lots expected to be accepted
may be read from the OC Curve corresponding to the chosen sampling
plan. As ,an example, for a plan with the sample size code letter L
and AQL of 2.5 percent, the relevant OC Curve is given in Fig. 8. It may
then be seen fkom this OC Curve that 8 percent defective lots would be
accepted in about 27 percent of the cases for the plan under consideration.
28
PERCfNT MFECTlVfS
IN SUBMITTED LOTS
knownurdmagemethxt
withthe
100
90
60
70
Y
.
: -60
Y
SO
2s L
5
?O
ti
3
bi
3o
I?
20
10
(2
16
20
24
26
32
36
44
40
B2
B6
64I
64
cuwen would also approximately hold good for sam liig plans based on variability known sod raqe
Fro. 7
oc
CURVESFOP
h,PUNG
hA2tS
kD
ON STANDARD
&VIATlON
hfETHOD,
PERCENT DEFECTIVES
IN
SUBMITTED
LOTS
T,
Telephone
201010
8-77 00 32
Regional OMces:
Central
Eastern
Northern
: SC0 331336,
160022
337 86 62
60 38 43
Southern
MUMBAI
323 76 17
235 23 15
832 92 95
400093
Branch Oi7kes:
Pushpak,
Nurmohamed
Road, FARIDABAD
5501348
839 49 55
380001
Road,
751001
55 40 21
40 36 27
641037
21 01 41
8-28 88 01
121001
8-71 19 96
201001
5315 Ward No. 29, R. G. Barua Road, 5th By-lane, GUWAHATI 781003
54 11 37
C&5%,
20 10 83
E-52, Chitaranjan
Marg, C-Scheme,
1171418 B, Sarvodaya
500001
JAIPUR 302001
37 29 25
21 68 76
Seth Bhawan, 2nd Floor, Behind Leela Cinema, Naval Kishore Road,
LUCKNOW 226001
23 89 23
Patliputra Industrial
26 23 05
P. 0. Palayam,
695034
6 21 17
52 51 71
Approach,
32 36 35
P. 0. Princep Street,
27 10 85
400007
Square,
309 65 28
222 39 71
India
!*