INDUSTRIAL QUALITY CONTROL

CHAPTER 4
INSPECTION AND SAMPLING PLAN
1

ACCEPTANCE SAMPLING PLAN DECISIONS

o An inspection procedure used to determine whether

to accept or reject a specific quantity of materials
• Impact of TQM

o Basic procedure
 Take random sample
 Accept or reject, based on results
o Producer, or seller, is origin of the material or service
o Consumer, or buyer, is destination of the material or
service
o Sampling plans
2
 QUALITY AND RISK DECISIONS

Acceptable quality level (AQL) is the quality level desired by the

consumer

Producer’s risk (α) is the probability that a shipment having exactly

this level of quality will be rejected
 Rejecting a good (AQL) lot is a type I error
 Consumers also desire low producer’s risk because sending
good materials back to the supplier disrupts the consumer’s
production processes
 Most often the producer’s risk is set at 0.05, or 5 percent

QUALITY AND RISK DECISIONS

Lot tolerance proportion defective (LTPD), the worst level the

customer can tolerate

Consumer’s risk, (β ) is the probability a shipment having

exactly this level of quality (the LTPD) will be accepted
 Accepting a bad (LTPD) lot is a type II error
 A common value for the consumer’s risk is 0.10, or 10
percent

4
 SINGLE-SAMPLING PLANS

 States the sample size, n, and the acceptable

number of defectives, c
 The accept-reject decision is based on the results of
one sample taken at random from a large lot
 If the quality characteristic of the sample passes the test
(defects ≤ c), accept the lot
 If the sample fails (defects > c) there may be complete
inspection of the lot or the entire lot is rejected
 A good lot could be rejected if the sample includes an
unusually large number of defects
 A bad lot could be accepted if the quality in the sample is
better than in the lot

DOUBLE-SAMPLING PLANS
• Two sample sizes, (n1 and n2), and two acceptance
numbers (c1 and c2)
• Take a random sample of relatively small size n1, from a
large lot
• If the sample passes the test (≤ c1), accept the lot
• If the sample fails (> c2), the entire lot is rejected
• If the sample is between c1 and c2, then take a larger
second random sample, n2
• If the combined number of defects ≤ c2 accept the lot,
otherwise reject

6
 SEQUENTIAL SAMPLING PLANS
• Results of random samples of one unit, tested one-by-one, are
compared to sequential-sampling chart
• Chart guides decision to reject, accept, or continue sampling,
based on cumulative results
• Average number of items inspected (ANI) is generally lower
with sequential sampling

SEQUENTIAL SAMPLING CHART

8–

7–
Number of defectives

6–
Reject
5–

4–
Continue sampling
3–

2–

1– Accept
0– | | | | | | |
10 20 30 40 50 60 70
Cumulative sample size
8
OPERATING CHARACTERISTIC CURVE
• Perfect discrimination between good and bad lots requires
100% inspection
• Select sample size n and acceptance number c to achieve the
level of performance specified by the AQL, α, LTPD, and β
• Drawing the OC curve
• The OC curve shows the probability of accepting a lot Pa, as a
dependent function of p, the true proportion of defectives in
the lot
• For every possible combination of n and c, there exists a
unique operating characteristics curve

OPERATING CHARACTERISTIC CURVE

1.0
Ideal OC curve
α
Probability of acceptance

Typical OC curve

β
AQL LTPD
Proportion defective
10
 TABLE G1

11

TABLE G1 (CONT)

12
 TABLE G1 (CONT)

13

CONSTRUCTING AN OC CURVE

EXAMPLE 1
The Noise King Muffler Shop, a high-volume installer of replacement
exhaust muffler systems, just received a shipment of 1,000 mufflers.
The sampling plan for inspecting these mufflers calls for a sample size n
= 60 and an acceptance number c = 1. The contract with the muffler
manufacturer calls for an AQL of 1 defective muffler per 100 and an
LTPD of 6 defective mufflers per 100. Calculate the OC curve for this
plan, and determine the producer’s risk and the consumer’s risk for the
plan.

14
 CONSTRUCTING AN OC CURVE
SOLUTION
Values for the Operating Characteristic Curve with
n = 60 and c = 1
Comments
Proportion Pa= P(d ≤ c)
Defective np
(p) with d: true defects USE TABLE G1
0.01 (AQL) 0.6 0.878 α = 1.000 – 0.878
= 0.122
(producer’s risk)
0.02 1.2 0.663
0.03 1.8 0.463
0.04 2.4 0.308
0.05 3.0 0.199
0.06 3.6 0.126 β = 0.126
(LTPD) (consumer’s risk)
0.07 4.2 0.078
0.08 4.8 0.048
0.09 5.4 0.029 15

0.10 6.0 0.017

CONSTRUCTING AN OC CURVE

1.0 –
0.9 –
0.878 α = 0.122
Probability of acceptance

0.8 –
0.7 – 0.663
0.6 –
0.5 – 0.463
0.4 –
0.308
0.3 –
0.199
0.2 –
0.126 0.078
0.1 – 0.048
β = 0.126 0.029
0.0 – | | | | | | | | | | 0.017
1 2 3 4 5 6 7 8 9 10
(AQL) (LTPD)
Proportion defective (hundredths)
16
 APPLICATION 1

A sampling plan is being evaluated where c = 10 and n = 193. If

AQL = 0.03 and LTPD = 0.08. What are the producer’s risk and
consumer’s risk for the plan? Draw the OC curve.

SOLUTION
Finding α (probability of Finding β (probability of
rejecting AQL quality) accepting LTPD quality)

p= 0.03 p= 0.08
np = 5.79 np = 15.44
Pa = 0.965 Pa = 0.10
α = 0.035 (or 1.0 – 0.965) β= 0.10

17

APPLICATION 1
1.0 –
α = 0.035
Probability of acceptance

0.8 –

0.6 –

0.4 –

0.2 –

0.0 –| | | | | |
β| = 0.10
| | | |
0 2 4 6 8 10
Percentage defective

18
EXPLAINING CHANGES IN THE OC CURVE

• Sample size effect

• Increasing n while holding c constant increases the
producer’s risk and reduces the consumer’s risk

Producer’s Risk Consumer’s Risk

n (α ) (β )

60 0.122 0.126

80 0.191 0.048
100 0.264 0.017

120 0.332 0.006

19

EXPLAINING CHANGES IN THE OC CURVE

1.0 –
0.9 – n = 60, c = 1
Probability of acceptance

0.8 –
0.7 – n = 80, c = 1
0.6 –
0.5 – n = 100, c = 1
0.4 –
n = 120, c = 1
0.3 –
0.2 –
0.1 –
0.0 – | | | | | | | | | |
1 2 3 4 5 6 7 8 9 10
(AQL) (LTPD)
Proportion defective (hundredths)
20
EXPLAINING CHANGES IN THE OC CURVE
• Acceptance level effect
• Increasing c while holding n constant decreases the
producer’s risk and increases the consumer’s risk

Producer’s Risk Consumer’s Risk

c (α ) (β )

1 0.122 0.126

2 0.023 0.303

3 0.003 0.515

4 0.000 0.706
21

EXPLAINING CHANGES IN THE OC CURVE

1.0 – n = 60, c = 1
0.9 –
n = 60, c = 2
Probability of acceptance

0.8 –
0.7 – n = 60, c = 3
0.6 – n = 60, c = 4
0.5 –
0.4 –
0.3 –
0.2 –
0.1 –
0.0 – | | | | | | | | | |
1 2 3 4 5 6 7 8 9 10
(AQL) (LTPD)
Proportion defective (hundredths)
22
 AVERAGE OUTGOING QUALITY

• AOQ is the expected (or Average) proportion of defects

that a particular sampling plan would allow to pass
through (Outgoing Quality) inspection
• Rectified inspection – defects found during the sampling
process are removed and reworked or replaced with
conforming material

p( Pa )( N − n )
AOQ =
N
where
p= true proportion defective of the lot
Pa = probability of accepting the lot
N= lot size
n= sample size 23

AVERAGE OUTGOING QUALITY

• Rejected lots are subjected to 100% inspection

• Average outgoing quality limit (AOQL) is the maximum
value of the average outgoing quality over all possible
values of the proportion defective
• Different sampling plans have different AOQs and AOQLs

24
 CALCULATING THE AOQL
EXAMPLE 2
Suppose that Noise King is using rectified inspection for its single-sampling
plan. Calculate the average outgoing quality limit for a plan with n = 110, c = 3,
and N = 1,000. Use Table G.1 to estimate the probabilities of acceptance for
values of the proportion defective from 0.01 to 0.08 in steps of 0.01.

SOLUTION
Use the following steps to estimate the AOQL for this sampling
plan:
Step 1: Determine the probabilities of acceptance for the desired
values of p. These are shown in the following table. However, the
values for p = 0.03, 0.05, and 0.07 had to be interpolated because
the table does not have them. For example, Pa for p = 0.03 was
estimated by averaging the Pa values for np = 3.2 and np = 3.4, (or
0.603 + 0.558)/2 = 0.580.
25

CALCULATING THE AOQL

Proportion Probability of
Defective (p) np Acceptance (Pa)

0.01 1.10 0.974

0.02 2.20 0.819

0.03 3.30 0.581 = (0.603 + 0.558)/2

0.04 4.40 0.359

0.05 5.50 0.202 = (0.213 + 0.191)/2

0.06 6.60 0.105

0.07 7.70 0.052 = (0.055 + 0.048)/2

0.08 8.80 0.024

26
 CALCULATING THE AOQL

Step 2: Calculate the AOQ for each value of p.

For p = 0.01: 0.01(0.974)(1000 – 110)/1000 = 0.0087

For p = 0.02: 0.02(0.819)(1000 – 110)/1000 = 0.0146
For p = 0.03: 0.03(0.581)(1000 – 110)/1000 = 0.0155
For p = 0.04: 0.04(0.359)(1000 – 110)/1000 = 0.0128
For p = 0.05: 0.05(0.202)(1000 – 110)/1000 = 0.0090
For p = 0.06: 0.06(0.105)(1000 – 110)/1000 = 0.0056
For p = 0.07: 0.07(0.052)(1000 – 110)/1000 = 0.0032
For p = 0.08: 0.08(0.024)(1000 – 110)/1000 = 0.0017

27

CALCULATING THE AOQL

1.6 – AOQL
Average outgoing quality (percent)

Step 3: Identify the largest

AOQ value, which is the
1.2 – estimate of the AOQL. In this
example, the AOQL is 0.0155
at p = 0.03.
0.8 –

0.4 –

0– | | | | | | | |
1 2 3 4 5 6 7 8
Defectives in lot (percent) 28
 APPLICATION 2
Demonstrate the model for computing AOQ
Management has selected the following parameters:

AQL = 0.01 α = 0.05

LTPD = 0.06 β = 0.10
n = 100 c=3
What is the AOQ if p = 0.05 and N = 3000?

SOLUTION
p= 0.05
np = 1000(0.05) = 5
Pa = 0.265

AOQ =
(0.05 )(0.265 )(2900 )
= 0.0128 29
3000

SOLVED PROBLEM
An inspection station has been installed between two production
processes. The feeder process, when operating correctly, has an
acceptable quality level of 3 percent. The consuming process, which
is expensive, has a specified lot tolerance proportion defective of 8
percent. The feeding process produces in batch sizes; if a batch is
rejected by the inspector, the entire batch must be checked and the
defective items reworked. Consequently, management wants no more
than a 5 percent producer’s risk and, because of the expensive
process that follows, no more than a 10 percent chance of accepting a
lot with 8 percent defectives or worse.

a. Determine the appropriate sample size, n, and the

acceptable number of defective items in the sample, c.
b. Calculate values and draw the OC curve for this inspection
station.
c. What is the probability that a lot with 5 percent defectives 30
will be rejected?
 SOLVED PROBLEM

SOLUTION
a. For AQL = 3 percent, LTPD = 8 percent, α = 5 percent, and β = 10
percent, use Table G.1 and trial and error to arrive at a sampling
plan. If n = 180 and c = 9,

np = 180(0.03) = 5.4
α = 0.049

np = 180(0.08) = 14.4
β = 0.092

Sampling plans that would also work are n = 200, c = 10;

n = 220, c = 10; and n = 240, c = 12.

31

SOLVED PROBLEM
b.The following table contains the data for the OC curve. Table
G.1 was used to estimate the probability of acceptance.

Proportion Probability of c or
Defective (p) np Less Defectives (Pa) Comments
0.01 1.8 1.000
0.02 3.6 0.996
0.03 (AQL) 5.4 0.951 α = 1 – 0.951 = 0.049
0.04 7.2 0.810
0.05 9.0 0.587
0.06 10.8 0.363
0.07 12.6 0.194
0.08 (LTPD) 14.4 0.092 β = 0.092
0.09 16.2 0.039
0.10 18.0 0.015
32
 SOLVED PROBLEM
1.000 0.996
1.0 —
c. According to the table, the
α = 0.049 probability of accepting a lot
0.9 — 0.951
0.810 with 5 percent defectives is
Probability of acceptance (Pa)

0.8 — 0.587. Therefore, the probability

0.7 — that a lot with 5 percent defects
0.6 — 0.587 will be rejected is 0.413, or 1 –
0.5 —
0.587

0.4 — 0.363
0.3 —
0.2 — 0.194
0.092
0.1 — 0.039
β = 0.092
0— | | | | | | | | | | 0.015
1 2 3 4 5 6 7 8 9 10
(AQL) (LTPD) 33
Proportion defective (hundredths)(p)