Acceptance Sampling

Accepting or rejecting a lot

(of parts, components, etc.)
based on the inspection of a sample
drawn from it

03-10-2017 1
 Acceptance Sampling

● Accept/reject entire lot based on sample results

● Created by Dodge and Romig during WWII
● Not consistent with TQM of Zero Defects
● Does not estimate the quality of the lot

03-10-2017 2
 What is acceptance sampling?
Lot Acceptance Sampling
A SQC technique, where a random sample is taken from
a lot, and upon the results of appraising the sample, the lot
will either be rejected or accepted
A procedure for sentencing incoming batches or lots of
items without doing 100% inspection
The most widely used sampling plans are given by Military
Standard (MIL-STD-105E)…
But Discontinued from 1995 Feb, … 2005 Feb….2008
Feb..
Presently ANSI Charts are used.
03-10-2017 3
 What is acceptance sampling?

• Purposes

1) Determine the quality level of an incoming

shipment or at the end of production
2) Judge whether quality level is within the level that
has been predetermined

But! Acceptance sampling gives you no idea about

the process that is producing those items!

03-10-2017 4
 Another area of quality control and improvement

• Closely connected with inspection and testing of product

• Inspection can occur at many points in a process

Acceptance Sampling: the inspection and classification of a

sample of units selected at random from a larger batch or lot
and ultimate decision about disposition of the lot –
LotDisposition or LotSentencing

Two common points of sampling inspection

• When parts are received

• After production

03-10-2017 5
 Why use acceptance sampling?
• Can do either 100% inspection, or inspect a sample of a few
items taken from the lot
• Complete inspection
• Inspecting each item produced to see if each item meets the
level desired
• Used when defective items would be very detrimental in
some way
Why not 100% inspection?

Problems with 100% inspection

• Generally very expensive
• Can’t use when product must be destroyed to test it
• Handling by inspectors can induce defects
• Inspection may be very tedious so defective items may slip
through even 100% inspection
03-10-2017 6
 Types of sampling plans
• Sampling by attributes vs. sampling by variables
Attributes (“go no-go” inspection)
Defectives-product acceptability across range
Defects-number of defects per unit
Variable (continuous measurement)
Usually measured by mean and std deviation

• Incoming vs. outgoing inspection

• Rectifying vs. non-rectifying inspection
• What is done with nonconforming items found
during inspection
• Defectives may be replaced by good items
03-10-2017 7
 Sampling Plans
• Acceptance sampling: Form of inspection applied to
lots or batches of items before or after a process, to
judge conformance with predetermined standards
• Sampling plans: Plans that specify lot size,
sample size, number of samples, and
acceptance/rejection criteria
• Single-sampling
• Double-sampling
• Multiple-sampling and Sequential sampling

03-10-2017 8
 A Lot-by-Lot Sampling Plan

• Specify the plan (n, c) given N

• For a lot size N, determine
the sample size n, and
the acceptance number c.
• Reject lot if number of defects > c
• Specify course of action if lot is rejected

03-10-2017 9
 Quality Definitions
• Acceptance quality level (AQL)
The smallest percentage of defectives that will make the lot definitely acceptable.
A quality level that is the base line requirement of the customer

• RQL or Lot tolerance percent defective (LTPD)

Quality level that is unacceptable to the customer

Remember !
You are not measuring the quality of the lot, but, you are to sentence the lot to either
reject or accept it

Sampling involves risks:

Good product may be rejected
Bad product may be accepted
Because we inspect only a sample, not the whole lot!

03-10-2017 10
 Lots should be such that …
• produced on the same machines, by same operators, from
common raw materials, at approximately the same time period
• Larger lots are better than smaller lots These are more
representative of overall quality
• Lots should be conformable to the material handling systems and
personnel

The most common and easiest plan to use but not most efficient
in terms of average number of samples needed
• One sample drawn from the lot and 100% inspected
• Single sampling plan
N = lot size
n = sample size (randomized)
c = acceptance number
d = number of defective items in sample
Rule: If d ≤ c, accept lot; else reject the lot
03-10-2017 11
Producer’s & Consumer’s Risks due to
mistaken sentencing of lot

• TYPE I ERROR = P(reject good lot)

α or Producer’s risk
5% is common

• TYPE II ERROR = P(accept bad lot)

β or Consumer’s risk
10% is typical value

Producer’s risk Consumer’s risk

Risk associated with a lot of acceptable quality rejected
• Alpha α
= Prob (committing Type I error)
= P(rejecting lot at AQL quality level)
Receive shipment, assume good quality, actually bad
quality
• Beta β
= Prob(committing Type II error)
= Prob (accepting a lot at RQL quality level)

03-10-2017 12
 The Operating Characteristic or OC Curve

The OC curve indicates a sampling plan’s ability to

discriminate between good and bad lots

• It is a graph of the % defective (p) in a lot or batch vs. the

probability that the sampling plan will accept the lot
• Shows probability of lot acceptance Pa as function of lot quality
• It is based on the sampling plan
• Curve indicates discriminating power of the plan
• Aids in selection of plans that are effective in reducing risk
• Helps to keep the high cost of inspection down

03-10-2017 13
 Acceptable Quality Level (AQL)
Percentage of defective items a customer is willing
to accept from you (a property of mfg. process)
• Lot Tolerance Percent Defective (LTPD)
Upper limit on the percentage of defects a
customer is willing to accept ( a property of the
consumer)
• Average Outgoing Quality (AOQ)

Average of rejected lots and accepted lots

• Average Outgoing Quality Limit (AOQL)

Maximum AOQ for a range of fractions defective

03-10-2017 14
03-10-2017 15
03-10-2017 16
 OC Curve helps visualize producer’s and
consumer’s points
• Type I and Type II decision errors correspond logically to the two
decision points on the oc-curve.
• Type I error -- Wrongful Rejection
A type I error is associated with the producer's point -- to reject
AQL LTPD when the true value of the quality characteristic is AQL. The risk
of rejecting an AQL lot is the producer's risk (α = alpha risk)
• Type II error -- Wrongful acceptance
A Type II error is to accept when the true value of the quality
characteristic is RQL -- at the consumer's point. The risk of
accepting a lot, if it is an RQL lot, is the consumer's risk (ß =
beta risk).

03-10-2017 17
Types of OC Curves

• Type A
Gives the probability of acceptance for an individual lot coming from
finite production
• Type B
Give the probability of acceptance for lots coming from a continuous
process or infinite size lot

The Ways of Calculating OC Curves

• Binomial distribution (Type B)
• Hypergeometric distribution (Type A)
Pa = P(r defectives found in a sample of n)
• Poisson formula
• P(r) = (np)r e-np)/ r!
• Larson nomogram

03-10-2017 18
 A Poisson formula can be used
P(r) = ((np)r e-np) /r! = Prob(exactly r defectives in n)

Poisson is a limit
Limitations of using Poisson
n ≤ N/10 total batch
Little faith in Poisson probability calculation when n is quite small
and p quite large.
For Poisson, Pa = P(r ≤ c)

03-10-2017 19
 Guidelines for choosing Producer’s and Consumer’s decision points
(AQL and RQL)

To choose the Producer's Point (AQL) in practice

• Lots at the producer's point quality level (AQL) should be
account historical quality levels and the consequence of returning
• Choose the producers risk of rejecting a lot that is of AQL quality.

To choose the Consumer's Point (RQL) in practice

• Lots at the consumer's point quality level (RQL) should be
consequence of the consumer’s accepting a lot at RQL quality
• Choose the consumers risk of accepting a lot that is of RQL

03-10-2017 20
 Properties of OC Curves
• The acceptance number c and sample size n are most important
factors in defining the OC curve
• Decreasing the acceptance number (c) is preferred over increasing
sample size (n)
• The larger the sample size n the steeper is the OC curve (i.e., it
becomes more discriminating between good and bad lots)

As sample size (n)increases,

the OC curve shape will approach the
Effects of n on OC curves ideal OC curve ~ more discriminating

• Note that c is kept in proportional

to n

03-10-2017 21
 Larson Nomogram

Minitab

03-10-2017 22
 How to Read the ANSI Tables for
Inspections Based on Random
Sampling

How to read the “ANSI tables”, aka “AQL tables”

Source: Mil-Std 105E, replaced by commercial standards:

ISO2859, ANSI/ASQ Z1.4-2003, NF06-022, BS 6001, DIN 40080.

03-10-2017 23
 Widely used settings:
 Normal severity
 Level II

Acceptance Quality Limit (AQL)

How to decide if the batch is accepted or rejected?

Widely used settings for consumer goods are:

 0% for critical defects (totally unacceptable: a user might

get harmed, or regulations are not respected).
 2.5% for major defects (these products would usually not be
considered acceptable by the end user).
 4.0% for minor defects (there is some departure from
specifications, but most users would not mind it).

03-10-2017 24
Example

 Order size: 40,000 pcs of memory sticks, to be made in 1

batch by the supplier.

 Single sampling plan.

 Normal severity, level II.

 AQLs: critical 0, major 1.5%, minor 4.0%.

Step 1: the Code Letter

03-10-2017 26
Step 2: Sample Size and AQLs

03-10-2017 27
 This Special Case
The ISO2859-1 standard says:
 If this procedure leads to different sample sizes for different
classes of nonconformities or nonconforming items, the sample
size code letter corresponding to the largest sample size
derived may be used for all classes of nonconformities or
nonconforming items, when designated or approved by the
responsible authority.
 Conclusion (for Single Plan)

Your code letter is “N”, so you will have to draw 500 pcs
randomly from the total lot size.
Here are the limits: the products are accepted if NO
critical defects, and NO MORE than 14 major defects,
and NO MORE than 21 minor defects are found.
Examples:
 If you find 0 critical defect, 17 major defects and 12
minor defects, the products are rejected.
 If you find 0 critical defect, 10 major defects and 21
minor defects, the products are accepted.
Same Example, Double Plan

 For simplicity, let’s only look at major defects.

 Code letter is unchanged.

03-10-2017 31
03-10-2017 32
03-10-2017 33
03-10-2017 34