Chapter 4

Acceptance sampling

4.1. Introduction
 Inspection of raw materials, semi-finished products, or

finished products is one aspect of quality assurance.

 When inspection is for the purpose of acceptance or

rejection of a product, based on adherence to a standard,

 The type of inspection procedure employed is usually

called acceptance sampling.

1
A typical application of acceptance sampling is as follows:

 A company receives a shipment of product from a supplier.

 This product is often a component or raw material used in

the company's manufacturing process.

 A sample is taken from the lot, and some quality

characteristic of the units in the sample is inspected.

 On the basis of the information in this sample, a decision is

2
made regarding lot disposition.
  Usually, this decision is either to accept or to reject

the lot. Sometimes we refer to this decision as lot

sentencing.

Generally, there are three approaches to lot sentencing:

1) Accept with no inspection;

2) 100% inspection-that is, inspect every item in the

lot, removing all defective units found and

3) Acceptance sampling.

3
Acceptance sampling is most likely to be useful in the
following situations:
 When testing is destructive

 When the cost of 100% inspection is extremely high.

 When 100% inspection is not technologically feasible or

would require so much calendar time that production

scheduling would be seriously impacted.
 When there are many items to be inspected and the

inspection error rate is sufficiently high that 100%

4
inspection
 When the supplier has an excellent quality history, and

some reduction in inspection from 100% is desired, .

 When there are potentially serious product liability risks,

and although the supplier's process is satisfactory, a

program for continuously, monitoring of the product is
necessary.

5
Advantages of Acceptance Sampling
When acceptance sampling is contrasted with 100%
inspection, it has the following advantages:
 It is usually less expensive because there is less

inspection.
 There is less handling of the product, hence reduced

damage.
 It is applicable to destructive testing.

6 Fewer personnel are involved in inspection activities


 It often greatly reduces the amount of
inspection error
The rejection of entire lots as opposed to the

simple return of defectives often provides a

stronger motivation to the supplier for quality
improvements.

7
 Disadvantages of Acceptance sampling
 There are risks of accepting "bad" lots and rejecting

"good" lots.
 Less information is usually generated about the

product or about the process that manufactured the

product.
 Acceptance sampling requires planning and
documentation of the acceptance sampling procedure
whereas 100% inspection does not.
8
 4.1.1 Basic concepts
 A sampling plan is a lot-sentencing procedure in which

the sample of n units is selected at random from the lot,

and the disposition of the lot is determined based on the
information contained in that sample.
a. Single –Sampling plan
b. Double –Sampling plan
c. multiple-Sampling plan
d. Sequential -Sampling
9
 Lot Formation: How the lot is formed can influence

the effectiveness of the acceptance-sampling plan.

There are a number of important considerations in
forming lots for inspection. Some of these are as follows:
a) Lots should be homogeneous

b) Larger lots are preferred over smaller ones

c) Lots should be conformable to the materials-

handling systems used in both the supplier and

10
consumer facilities.
Rectifying Inspection: Acceptance sample
programs usually require corrective action when
lots are rejected. This generally takes the form of
100% inspection or screening of rejected lots,
with all discovered defective items either
removed for subsequent rework or return to the
supplier, or replaced from a stock of known good
items.
11
  Average Sample Number (ASN): For a single
sampling LASP (n, c) we know each and every
lot has a sample of size n taken and inspected or
tested. For double, multiple and sequential
LASP's, the amount of sampling varies
depending on the number of defects observed.

12
  Random Sampling:: the units selected for inspection from the

lot should be chosen at random, and they should be

representative of all the items in the lot. The random-sampling
concept is extremely important in acceptance sampling. Unless
random samples are used,
 Lot Tolerance Percent Defective (LTPD):: The LTPD is a

designated high defective level that would be unacceptable to

the consumer.
 Operating characteristic (OC) function: OC is defined as

the probability of accepting the batch regarded as the function

13
of the unknown parameter.
 4.1.2 Guidelines for Using Acceptance
Sampling
 An acceptance-sampling plan is a statement of the

sample size to be used and the associated acceptance or

rejection criteria for sentencing individual lots.
 In general, the selection of an acceptance-sampling

procedure depends on both the objective of the

sampling organization and the history of the
14 organization whose product is sampled.
4.2. Acceptance Sampling for Proportion(Attribute)
Let
N = lot size
n = sample size
a) Single-sampling plans
Let
= proportion of defective items in the population
X= number of defectives in the sample
The procedure to arrive at a decision about the acceptance or
rejection of a lot is as follows:-
(i) Select a random sample of size n from the submitted

15 lot.
 ii) Inspect each and every unit included in the sample

and classify them as defective or non-defective.

Suppose the number of defectives in the sample is X,
the criteria for the acceptance or rejection of lot is,

a) If X , the lot is accepted. In this case all the

defective items in the sample are replaced by

non-defectives and the lot is release for marketing.

16
 b) If X>c, the lot is rejected. In this situation, inspect
the whole lot and replace all defectives found in the lot
by non defectives and release the lot for sale.

Case I: N is Large (Consider N Infinite)

The distribution of the number of defectives X in a
random sample of n items is binomial with parameters n
and .
P(X=x defectives) = P(x) = .

17
  The OC function is defined as the probability of

accepting the lot/batch for the given.

 OC(𝜃)= = P(X c

= =
 OC(𝜃) = +

+ + ….+
 OC(𝜃) = + + +

….+
18
 Note that
 This OC function is a decreasing function

 Its value at 𝜃=0, OC(0) =1

 Its value at 𝜃=1, OC(1) =0

19
 I. (Fix c, let c=0): Suppose we want to be at least 95%
certain of rejecting a batch that has 5% of defectives, how
large should we take n?
• c =0 corresponds to taking a sample of n items and

accepting the batch only if there are no defects in the

sample. That is all items are good.
OC()=, if c = 0, then
OC()= =

20
P(rejecting) ≥0.95  P(accepting) ≤ 1-0.95, that is: OC(𝜃) 0.05
Since OC(  ) =(1-  )n , for c=0
 OC(0.05) = (1-0.05)n
 (0.95)n  0.05
By taking natural logarithm on both sides we get,
n*log0.95  log 0.05
= n  log 0.05 \log 0.95 =58.4

21
  n must be an integer so choose n=59, we can use the

rule “take a sample of 59 objects and accept the

batch if every object is good otherwise reject if there
is defective”.
 If we want to be even more certain, we might use the

condition “be at least 99% certain of rejecting a

batch with 2% of defectives “ (c=0)

22
Solution:
  2% , OC(  )  0.01
 OC(0.02) = (1-0.02)n  0.01
 (1-0.02)n  0.01 . By taking natural logarithm on both sides we get,
n*log(0.98)  log 0.01

 n  log 0.01 / log 0.98 =227.9,

that is, n  228 since n must be an integer,
so that we can use the rule “take a sample of 228 observation
and accept the batch if every object is good and reject if there is  1 defective”.

23
24
 II. (fix n): If we take n =100, then what value should we
take for c if we want to make sure at least 95% of
rejecting a batch which has 5% of defectives?
We choose c to solve OC (0.05)  0.05 , try c= 0, 1, 2, 3… each in turn.
c= 0, OC(0.05) = P(X ≤ 0) = P(X=0)
=(1-0.05)100 = (0.95)100 = 0.00592
c=1, OC(0.05) = P(X  1) = p(X=0) + P(X=1)

100 
= (0.95) 100
+   (0.05)1 (0.95)99 = 0.037
1 

25
c=2, OC(0.05) =P(X  2) = P(X=0) + p(X=1) +P(X=2)

100  99 100 
OC(0.05) = (0.95) 100
+   (0.05) (0.95) +   (0.05)2 (0.95)98 = 0.11
1

1  2 
< 5%, 𝑐 = 0, 1
Therefore, OC(0.05) = ൜
> 5%, 𝑐 = 2, 3, 4, … , 100
c = 0 and c = 1 both satisfy the constraint.

26
 c=0 c=1
Buyer’s risk Lower Higher (we might accept a bad lot)
Producer’s risk Higher (we might reject a good lot) Lower

In practice we would probably take c=1. First we put a

limit on the buyer’s risk. Then we minimize the
producer’s risk subject to that limit. (This procedure is
similar to hypothesis testing).

27
 III. (Fix ): A buyer tests a sample of n fuses from
infinite batch (population) of fuses. She will accept the
batch if the sample contains ≤4% defectives. She is
considering by taking a sample of n = 25, 50, 75, 100
etc fuses. How large should “n” be if she wants to be
90% of certain of rejecting a batch with 10% of
defectives? i.e. if c/n=0.04, find the value of c and n so
that OC (0.1) 10%.

28
Possible solutions to check: n=25, c=1; n=50, c=2; n=75, c=3
and n=100, c=4. Therefore, for c =0, n=25, OC(0.1) implies
that OC(0.1)= = P(X=0)+P(X=1)

29
The OC function for the 3 sampling schemes is tabulated below.
c= 1, n=25 c=2, n=50 c=3, n=75
𝜃 0.01 0.97 0.986 0.993
0.03 0.83 0.811 0.812
0.04 0.74 0.677 0.647
0.05 0.64 0.541 0.480
0.1 0.27 0.111 0.05
0.2 0.027 0.001 7x10-5

30
 With an increasing proportion of defectives, the probability
that the lot is accepted decrease.
The ideal OC function is very close to 1 for small values of
and very close to 0 for values of greater than an acceptable
quality limit.

31
 Case 2: N is small (consider N is finite)
Suppose a lot consists of N (consider N is finite) items

 𝜃 = proportion of defectives

 M= total number of defectives

n = sample of size

𝜃 = (𝜃 can only take the values 0, ,, , …., 1)

 X= number of defective in the sample has hyper-

geometric distribution.

32
i. The probability of getting x defectives in the sample is:

 M  N  M 
  
 x  n  x 
P(X=x) = , where x  n , n -x  N  M
N
n 
 
ii. The probability of accepting the lot having not more than c defectives in the
sample of n items is,
σ 𝑐𝑥=0 ൫𝑀 ൯൫𝑁−𝑀 ൯
𝑥 𝑛−𝑥
P(X≤c) =
൫𝑁
𝑛൯

33
 Example : Suppose N=1000, c=0 (i.e. we have a batch
of size 1000. We use the rule “accept if lot/batch the
sample contains no defectives”.)
Choose n so that we are 95% certain of rejecting a batch
with 5% defectives
We want to choose n so that OC(0.05)  0.05
𝑀
 𝑁
=   5% , N =1000, M=N  =50.

OC(X =0\M=50) = P(Every item in a sample of size n is not defective / M=50)

= 950\100  949 \ 999  948 \ 998  ...  950  (n  1) \ 100  (n  1) .
ቀ50 ቁቀ 950 ቁ
0 𝑛−0
OC (0.05) = P(X = 0) =
൫1000
𝑛 ൯

34
 We want to choose n large enough to make sure this

product is less than or equal to 0.05.

This is solved by substitution and a solution can’t be

obtained in a closed form.

The solution is n≥57.
Example: Suppose the batch size is N= 40 and we take
c=0, then find the value of n so that OC (0.05)  5% .
For this value of n calculate the OC functions
(for 𝜃 = 0.025, 0.05, 0.075, …)
35
M=N  =0.05  40  2 .
We want to solve: OC(0.05) = P(accept the batch/M=2)  0.05 .
P(No defective in the sample) = P(1st item is good )  P(2nd item is good/1st item is good)  P(3rd
item is good/1st and 2nd item are good)  ……  P(nth item is
good /1st ,2nd ,3rd ….(n-1)th item are good).
38 37 36 38  n  1
=    ... 
40 39 38 40  n  1
ቀ2ቁቀ40−2ቁ ൫38
0 𝑛 −0 𝑛൯ 38! 40! ሺ40−𝑛 ሻ(39−𝑛 )!
= = = ൗ =
൫40
𝑛൯ ൫40
𝑛 ൯ 𝑛! 38−𝑛 ! 𝑛! 40−𝑛 !
ሺ ሻ ሺ ሻ 39𝑥40

We want to choose n so that this is ≤ 0.05.

Then by substitution and inspection, for n = 29, 30, 31 OC(0.05) equals 0.071, 0.058, 0.046,
respectively.
Therefore, we need to choose n =31 (the minimum sample size)

36
 b) Double-stage sampling plan
 A number of extensions of single-sampling plans for

attributes are useful.

 These include double- sampling plans, multiple-

sampling plans, and sequential-sampling plans were

invented to give a questionable lot another chance.
 A double-sampling plan is a procedure in which, under,

certain circumstances, a second sample is required

before the lots can be sentenced.
37
  Double sampling is the results of the first sample are not

conclusive with regard to accepting or rejecting, a second

sample is taken.
Suppose a sample of size n is taken. Let X=number of
defectives in the sample.
• If X≤c1, accept the lot (batch) of items.

• If X≥d1, reject the batch

• If c1 <X <d1, take a further sample of size m and let

Y= number of defectives in this sample; then accept the

38 batch if X+Y ≤ c2, otherwise reject the lot (batch).
 Disadvantages of Double-stage sampling plan
 It may require more total inspection than would be

required in a single-sampling plan that offers the same

protection.
 it is administratively more complex, which may

increase the opportunity for the occurrence of

inspection errors.

39
 Example: Consider double – stage binomial sampling
scheme with n= 50, cl = 0, d1=2. m = 150, c2 = 3. That is
to mean “Take a sample of size 50, then
 If it contains no defectives accept the batch.

 If it contains two or more defectives reject the batch.

 If it contains exactly one defective take a further

sample of size 100 and accept the batch if this sample

has less than or equal to two defectives”
Calculate the OC function.
40
 OC(𝜃) = P(Accept the lot/𝜃)=P(Accept at the first stage/𝜃) +
P(Accept at the second stage/𝜃)
P(Accept at the first stage/𝜃) =
= (1-𝜃)50.

P(Accept at the second stage/𝜃) = P(X=1and Y≤2/𝜃)

Since X~Bin(50, 𝜃) and Y~ Bin(150, 𝜃)
50 1 49 σ 2 150 𝑦 150−𝑦
P(Accept at the second stage/𝜃) = ቀ ቁ𝜃 (1 − 𝜃) 𝑦=0 ൬𝑦 ൰𝜃 (1 − 𝜃)
1
150  149 2
= 50𝜃(1-𝜃)49[(1-𝜃)150 +150 𝜃 (1-𝜃)149 + 𝜃 (1-𝜃)148]
2
150  149 2
Hence, OC(𝜃) =(1-𝜃)50 + 50𝜃(1-𝜃)49[(1-𝜃)150 +150 𝜃 (1-𝜃)149 + 𝜃 (1-𝜃)148]
2

41
 The OC function for the general case (n, m, c1, d1, c2), X=x and

Y≤c2 –x (i.e X+Y≤c2) is:

OC() =
+ *.
Examples:
1. Find the OC function when n = 100, c1 = 1, d1 = 4, m = 300

and c2 =6.
Solution:

42
 OC() =
+ *.
OC() =
+ P()*
+ P()*.

43
 2. Suppose we intend to use the plan with n =40, c 1 = 0,

d1 = 2, c2 = 1, then how should we choose m if we want

OC(0.1) < 5%.
Solution:
OC() =

+ *.
= +

44
 OC() = +
= + 4< 0.05
= < ¼ (0.05 -
= (39 + m) > ln(¼ (0.05 - )/ln(0.9) = 44.9
m > 5.9, take m = 6.
The Average Sample Number (Average total inspected)
 In single sampling, the size of the sample 'inspected from the

lot is always constant, whereas in double sampling, the size

of the sample selected depends on whether or not the second
sample is necessary.
45
  The probability of drawing a second sample varies with the

fraction defective in the incoming lot.

 With complete inspection of the second sample, the average

sample size in double sampling is equal to the size of the

first sample times the probability that there will only be one
sample, plus the size of the combined samples times the
probability that a second sample will be necessary.
For a two stage sampling scheme “n” items are sampled with
probability P(X≤c1) +P(X≥d1), “n+m” items with probability

P(c1<X<d1).
46
 Hence, the average sample number (average total inspected)
is given as
• ASN = n(P(X≤c1) +P(X≥d1)) + (n+m)P(c1<X<d1)

= n + m (P(c1<X<d1)), where n=number of units

inspected in the first stage and m= number of units inspected
in the second stage.
• Note this ASN is a function of 𝜃 (proportion of defectives).

• If the above equation is evaluated for various values of lot

fraction defective 𝜃, the plot of ASN versus 𝜃 is called an

47 average sample number curve.
Example: Find the ASN for the two stage sampling scheme
with n=100, m=200, cl = 3, d1=5, c2 = any value.
ASN = n + m P(c1<X<d1) = 100 + 200(P(X=4))
100  4
= 100 + 200   𝜃 (1- 𝜃)96
4 
6 100  x
If d1 =7, ASN=100 + 200    𝜃 (1- 𝜃)100-x
x4 x 

48
 4.3. Comparing and Choosing Sampling schemes
 The Rejectable Quality Level (RQL) is the value of for

which the quality of the lot is unacceptable.

 The Acceptable Quality Level (AQL): is the value of for

which the quality of the batch can be regarded as good.

 We want OC(RQL) to be small and we might fix a small

number, and insist [OC(RQL) < ].

And also, we want OC(AQL) close to 1 and we might fix a
small number, (example =5%) and insist [OC(AQL) > 1 - ].

49
 • Consider single-stage sampling scheme:

 If we are given either inequality (OC(RQL) < or

OC(AQL) > 1 – ), we can always choose n and c so that a

scheme’s OC function satisfies these inequalities.
 There are infinite number of solutions to these

inequalities. Any scheme that satisfies these constraints

is called an admissible scheme.
 As there are infinite number of admissible schemes we

will often choose the cheapest one (smallest value of n)

50 or use some other criterion.
 Example: Choose n and c so that (use the acceptance sampling
tables) (Assume N is infinite)
 OC(0.1)<10%

 OC(0.01)>90%

Solutions
From table 1 From table 2
n= 25 c=0 c≥1 no solution
n= 50 c≤ 1 c≥1 n=50, c=1
n= 70 c≤ 3 c ≥2 n=70, c=2 or 3
n= 80 c≤ 4 c ≥2 n=80, c=2,3 or 4
n= 100 c≤ 5 c ≥2 n=100, c=2,3,4 or 5

51
Table 1: Rejectable Quality Level

Choose c to be less than or equal to the tabulated value

 0.05 0.1 0.2

  0.01 0.05 0.1 0.01 0.05 0.1 0.01 0.05 0.1
n= 25 0 0 1 2
50 0 0 1 1 3 5 5
70 0 0 1 2 3 6 8 9
80 0 1 1 3 4 7 9 10
100 0 1 1 3 4 5 10 13 14
Note: the blank spaces in the top left-hand corner mean it is impossible to design the
required scheme for these values of n,  and  .

52
Table 2: Acceptable Quality Level

Choose c to be greater than or equal to the tabulated value

 0.01 0.05 0.1

  0.01 0.05 0.1 0.01 0.05 0.1 0.01 0.05 0.1
n= 25 2 1 1 4 3 3 6 5 4
50 3 2 1 7 5 5 10 9 8
70 3 2 2 8 7 6 13 11 10
80 3 2 2 9 7 7 15 13 12
100 4 3 2 11 9 8 18 15 14

53
Average Outgoing Quality (AOQ)

A common procedure when sampling and testing is non-

destructive, is to 100% inspect rejected lots and replace all
defectives with good units.
 P(Defective) = P(defective batch was accepted)

+P(defective batch was rejected)

= P(defective batch was accepted) P(accepted)
+P(defective batch was rejected)P(rejected)
= OC()+0
54 =OC()
This is called the average outgoing quality
[AOQ() = OC(]
Note that AOQ ()< OC( for < 1.

The AOQ is a function of and we want this to be small, but we

can not minimize it as is unknown.

Average Outgoing Quality Limit (AOQL): A plot of the AOQ
(Y-axis) versus the incoming lot (X-axis) will start at 0 for
= 0, and return to 0 for = 1 (where every lot is 100%
inspected). In between, it will rise to a maximum.

55
 This maximum, which is the worst possible long term AOQ, is
called the average outgoing quality limit (AOQL) and
AOQL =.

56
 Example: Suppose we want to design a one-stage
sampling plan with infinite N and c = 0 such that
OC(AQL) > 80% where AQL = 0.01 and AOQL < 3%,
then what values of n satisfy these two constraints?
Solution:
• First, since c = 0, OC () = (1 - . Hence,

(1 – > 0.8 implies 0.99n > 0.8

n ln(0.99) > ln(0.8), then n =22.

57
 Second AOQ( = OC( = (1 - , as c = 0

Here we need to maximize AOQ and equating to zero by

differentiation as:
AOQ() = ( (1 - ) = 0, gives = .
Therefore, the AOQ attains its maximum at = .
AOQL = ( (1 - ) = (1 - =

58
Choose n so that < 0.03, solve by trial and error to give n
12.
Therefore, take n to be any value between 12 and 22.

59
 Example: Suppose we want to design a one-stage sampling
plan (with infinite N) and then
(i) find n so that AOQL < 2% if c=0
(ii) find AOQL if n=25 and c = 0

Solution:
(i) c = 0 implies OC ( = (1 - , then AOQ () = (1 -
Find its unique maximum by differentiating and equating to
zero, then AOQ() has a maximum at = .

60
 This implies that AOQL = and now choose n so that <
0.02.

n 16 17 18
⁡nn 0.022 0.021 0.019
(n + 1)n+1
Therefore, AOQL < 2% for all n values greater than or
equal to 18. That is if we take n = 18, we know
< 2% of items sold will be defective

61
 (ii) c = 0 and n =25, AOQL = =
, that is if we take n = 25, we know
< 1.4% of items sold will be defective.

62
4.4 Acceptance Sampling for variate (quantitative variables)
 This section summarizes acceptance-sampling techniques for

quantitative variables sampling plans, which can be used as

alternatives to attribute plans when measurement data are
available.
Single-stage sampling plan
 Since the quality characteristic is a variate, there will exist either a

lower specification limit (LSL), an upper specification limit

(USL), or both, that define the acceptable values of this
parameter.
.
63
  Suppose that the standard deviation 𝜎 is known

Let X be the measurement made on the variate. Three cases may be

considered for an item being defective.
1. “Defective” if X > USL
2. “Defective” if X < LSL
3. “Defective” if X <LSL or x >USL
Assume
 Batch (lot) size N is large (can be thought of as infinite)

 Each item in the batch has a measurement made on it, X i, X1, X2, X3

… are independently and identically normal with mean µ and

 Let 𝜃=probability of being defective (As N is large 𝜃 can be thought
64
of as the proportion of defective in the batch)
  USL       USL 
Case 1: 𝜃 =1 -   =  
     
 LSL   
Case 2: 𝜃 =  
  
 LSL       USL 
Case 3: 𝜃 =   +  
     
Note:  ( z )  P( Z  z ) is the standard normal cumulative distribution.
Cases 1 and 2 are similar (one-sided). In each case there is a one-to-one
relationship between µ and 𝜃, and 𝜃 can take any value between 0 and 1.
Case 3 is two-sided. We will look case 2.

65
Method: (Case 2)
 Fix n and k
 Take a sample of size n
1
 Measure Xi, X n   X i and X n ~ N(µ,  2 / n )
n
 If X n > LSL+k  , then accept the batch (lot).
Otherwise, reject the batch.
 Note that k need not be an integer.
The operating characteristic (OC) function is defined as the probability of
accepting a batch (lot), but this can be regarded as either a function of µ or 𝜃.

66
67
  LSL   
We have seen (under case 2), 𝜃 =  .
  
LSL  
This implies that    =
1


Hence, P(Accepting batch/𝜃) = OC𝜃(𝜃)

=  n   1 ( )  k 
Example: A process makes springs which should be able to support a
weight of 5kg
“Defective”= “X <5kg”. X is the maximum weight the spring can
support. Assume strength springs are iid ~ N(µ, 0.52 ).
We want to choose an acceptance sampling scheme so that
68
OC(0.05) < 0.05 and OC𝜃(0.01) > 0.95. Find the values n and k
Solution:
 
OC(0.05) < 0.05   n   1 (0.05)  k   0.05

[ Note :  (1.645)  P( Z  1.645)  0.05   1 (0.05)   1.645]

  
n 1.645  k   0.05

 n (1.645  k )   1 ( 0.05)

 n (1.645  k )   1.645

1.645
 k 1.645  …. (i)
n

69
 There are an infinite number of solutions. We are likely to find the
solution with smallest value of n. From the graph (k versus n) of
these two inequalities (i and ii) the intersection point gives the
minimum value of n.
70
To get the intersection point solve,
1.645
k 1.645  and
n
1.645
k  2.33 
n
 2k = 2.33+1.645  k=1.987  n=23
Hence we take n=23 , k=1.987
This means if we use the rule “Take a sample of 23 items, and
accept the batch if X 23 > LSL+1.987 

 X 23 > 5+1.987*0.5  X 23 >6

Decision rule: “If sample mean is greater than 6 kg, accept the batch.
If sample mean is less than 6 kg , reject the batch.”
71
 Advantages & Disadvantages of sampling by quantitative
variables.
Advantages
 Usually sampling by varitaes requires sample sizes than an
equivalent sampling by attributes(it is important in about
distructive sampling).
 It is easier to draw inferences about the batch we are
sampling by variates.
Disadvantages
 We need the normality assumption to hold
 It involves measuring, this is usually more costly than
classifying an item as good/bad
 It can not used if we are dealing with quality that cannot be
measured.
72