BY:

Pavankumar B Pol
What is Acceptance Sampling?

Contributions of Dodge and Romig to acceptance sampling:

Acceptance sampling is an important field of statistical quality control that was popularized by
Dodge and Romig and originally applied by the U.S. military to the testing of bullets during World War II.
If every bullet was tested in advance, no bullets would be left to ship. If, on the other hand, none were
tested, malfunctions might occur in the field of battle, with potentially disastrous results.

Definition of Lot Acceptance Sampling:

Dodge reasoned that a sample should be picked at random from the lot, and on the basis of
information that was yielded by the sample, a decision should be made regarding the disposition of the
lot. In general, the decision is either to accept or reject the lot. This process is called Lot Acceptance
Sampling or just Acceptance Sampling.

"Attributes" (i.e., defect counting) will be assumed:

Acceptance sampling is "the middle of the road" approach between no inspection and 100%
inspection. There are two major classifications of acceptance plans: by attributes ("go, no-go") and by
variables. The attribute case is the most common for acceptance sampling, and will be assumed for the
rest of this section.

A point to remember is that the main purpose of acceptance sampling is to decide whether or
not the lot is likely to be acceptable, not to estimate the quality of the lot.

Scenarios leading to acceptance sampling:

Acceptance sampling is employed when one or several of the following hold: 

 Testing is destructive

 The cost of 100% inspection is very high

 100% inspection takes too long

Acceptance Quality Control and Acceptance Sampling:

It was pointed out by Harold Dodge in 1969 that Acceptance Quality Control is not the same as
Acceptance Sampling. The latter depends on specific sampling plans, which when implemented indicate
the conditions for acceptance or rejection of the immediate lot that is being inspected. The former may
be implemented in the form of an Acceptance Control Chart. The control limits for the Acceptance
Control Chart are computed using the specification limits and the standard deviation of what is being
monitored.

An observation by Harold Dodge:

In 1942, Dodge stated:

"....basically the "acceptance quality control" system that was developed encompasses the
concept of protecting the consumer from getting unacceptable defective product, and encouraging the
producer in the use of process quality control by: varying the quantity and severity of acceptance
inspections in direct relation to the importance of the characteristics inspected, and in the inverse
relation to the goodness of the quality level as indication by those inspections."

To reiterate the difference in these two approaches: acceptance sampling plans are one-shot
deals, which essentially test short-run effects. Quality control is of the long-run variety, and is part of a
well-designed system for lot acceptance.

An observation by Ed Schilling:

Schilling (1989) said:

"An individual sampling plan has much the effect of a lone sniper, while the sampling plan
scheme can provide a fusillade in the battle for quality improvement."

Control of product quality using acceptance control charts:

According to the ISO standard on acceptance control charts (ISO 7966, 1993), an acceptance
control chart combines consideration of control implications with elements of acceptance sampling. It is
an appropriate tool for helping to make decisions with respect to process acceptance. The difference
between acceptance sampling approaches and acceptance control charts is the emphasis on process
acceptability rather than on product disposition decisions.
What kinds of Lot Acceptance Sampling Plans (LASPs) are there?

LASP is a sampling scheme and a set of rules:

A lot acceptance sampling plan (LASP) is a sampling scheme and a set of rules for making
decisions. The decision, based on counting the number of defectives in a sample, can be to accept the
lot, reject the lot, or even, for multiple or sequential sampling schemes, to take another sample and
then repeat the decision process.

Types of acceptance plans to choose from:

LASPs fall into the following categories:

 Single sampling plans:. One sample of items is selected at random from a lot and the disposition
of the lot is determined from the resulting information. These plans are usually denoted as (n,c)
plans for a sample size n, where the lot is rejected if there are more than c defectives. These are
the most common (and easiest) plans to use although not the most efficient in terms of average
number of samples needed.

 Double sampling plans: After the first sample is tested, there are three possibilities:

1. Accept the lot

2. Reject the lot

3. No decision

If the outcome is (3), and a second sample is taken, the procedure is to combine the results of
both samples and make a final decision based on that information.

 Multiple sampling plans: This is an extension of the double sampling plans where more than
two samples are needed to reach a conclusion. The advantage of multiple sampling is smaller
sample sizes.

 Sequential sampling plans: . This is the ultimate extension of multiple sampling where items are
selected from a lot one at a time and after inspection of each item a decision is made to accept
or reject the lot or select another unit.

 Skip lot sampling plans:. Skip lot sampling means that only a fraction of the submitted lots are
inspected.
Definitions of basic Acceptance Sampling terms:

Deriving a plan, within one of the categories listed above, is discussed in the pages that follow.
All derivations depend on the properties you want the plan to have. These are described using the
following terms:

 Acceptable Quality Level (AQL): The AQL is a percent defective that is the base line requirement
for the quality of the producer's product. The producer would like to design a sampling plan
such that there is a high probability of accepting a lot that has a defect level less than or equal to
the AQL.

 Lot Tolerance Percent Defective (LTPD): The LTPD is a designated high defect level that would
be unacceptable to the consumer. The consumer would like the sampling plan to have a low
probability of accepting a lot with a defect level as high as the LTPD.

 Type I Error (Producer's Risk): This is the probability, for a given (n,c) sampling plan, of rejecting
a lot that has a defect level equal to the AQL. The producer suffers when this occurs, because a
lot with acceptable quality was rejected. The symbol is commonly used for the Type I error
and typical values for range from 0.2 to 0.01.

 Type II Error (Consumer's Risk): This is the probability, for a given (n,c) sampling plan, of
accepting a lot with a defect level equal to the LTPD. The consumer suffers when this occurs,
because a lot with unacceptable quality was accepted. The symbol is commonly used for the
Type II error and typical values range from 0.2 to 0.01.

 Operating Characteristic (OC) Curve: This curve plots the probability of accepting the lot (Y-axis)
versus the lot fraction or percent defectives (X-axis). The OC curve is the primary tool for
displaying and investigating the properties of a LASP.

 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. In this
case, all rejected lots are made perfect and the only defects left are those in lots that were
accepted. AOQ's refer to the long term defect level for this combined LASP and 100% inspection
of rejected lots process. If all lots come in with a defect level of exactly p, and the OC curve for
the chosen (n,c) LASP indicates a probability pa of accepting such a lot, over the long run the
AOQ can easily be shown to be:

where N is the lot size.

 Average Outgoing Quality Level (AOQL): A plot of the AOQ (Y-axis) versus the incoming lot p (X-
axis) will start at 0 for p = 0, and return to 0 for p = 1 (where every lot is 100% inspected and
 rectified). In between, it will rise to a maximum. This maximum, which is the worst possible long
term AOQ, is called the AOQL.

 Average Total Inspection (ATI): When rejected lots are 100% inspected, it is easy to calculate
the ATI if lots come consistently with a defect level of p. For a LASP (n,c) with a probability pa of
accepting a lot with defect level p, we have

ATI = n + (1 - pa) (N - n)
Where N is the lot size.
 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. For any given
double, multiple or sequential plan, a long term ASN can be calculated assuming all lots come in
with a defect level of p. A plot of the ASN, versus the incoming defect level p, describes the
sampling efficiency of a given LASP scheme.

The final choice is a tradeoff decision:

Making a final choice between single or multiple sampling plans that have acceptable properties
is a matter of deciding whether the average sampling savings gained by the various multiple sampling
plans justifies the additional complexity of these plans and the uncertainty of not knowing how much
sampling and inspection will be done on a day-by-day basis.

How do you choose a Single Sampling Plan?

Two methods for choosing a single sample acceptance plan:

A single sampling plan, as previously defined, is specified by the pair of numbers (n,c). The sample size is
n, and the lot is rejected if there are more than c defectives in the sample; otherwise the lot is accepted.

There are two widely used ways of picking (n,c):

1. Use tables (such as MIL STD 105D) that focus on either the AQL or the LTPD desired.

2. Specify 2 desired points on the OC curve and solve for the (n,c) that uniquely determines an OC
curve going through these points.
Choosing a Sampling Plan: MIL Standard 105D:

The AQL or Acceptable Quality Level is the baseline requirement

Sampling plans are typically set up with reference to an acceptable quality level, or AQL . The
AQL is the base line requirement for the quality of the producer's product. The producer would like to
design a sampling plan such that the OC curve yields a high probability of acceptance at the AQL. On the
other side of the OC curve, the consumer wishes to be protected from accepting poor quality from the
producer. So the consumer establishes a criterion, the lot tolerance percent defective or LTPD. Here the
idea is to only accept poor quality product with a very low probability. Mil. Std. plans have been used for
over 50 years to achieve these goals.

The U.S. Department of Defense Military Standard 105E

Military Standard 105E sampling plan:

Standard military sampling procedures for inspection by attributes were developed during
World War II. Army Ordnance tables and procedures were generated in the early 1940's and these grew
into the Army Service Forces tables. At the end of the war, the Navy also worked on a set of tables. In
the meanwhile, the Statistical Research Group at Columbia University performed research and
outputted many outstanding results on attribute sampling plans.

These three streams combined in 1950 into a standard called Mil. Std. 105A. It has since been
modified from time to time and issued as 105B, 105C and 105D. Mil. Std. 105D was issued by the U.S.
government in 1963. It was adopted in 1971 by the American National Standards Institute as ANSI
Standard Z1.4 and in 1974 it was adopted (with minor changes) by the International Organization for
Standardization as ISO Std. 2859. The latest revision is Mil. Std 105E and was issued in 1989.

These three similar standards are continuously being updated and revised, but the basic tables
remain the same. Thus the discussion that follows of the germane aspects of Mil. Std. 105E also applies
to the other two standards.

Description of Mil. Std. 105D

Military Standard 105D sampling plan:

This document is essentially a set of individual plans, organized in a system of sampling

schemes. A sampling scheme consists of a combination of a normal sampling plan, a tightened sampling
plan, and a reduced sampling plan plus rules for switching from one to the other.
AQL is foundation of standard:

The foundation of the Standard is the acceptable quality level or AQL. In the following scenario,
a certain military agency, called the Consumer from here on, wants to purchase a particular product
from a supplier, called the Producer from here on.

In applying the Mil. Std. 105D it is expected that there is perfect agreement between Producer
and Consumer regarding what the AQL is for a given product characteristic. It is understood by both
parties that the Producer will be submitting for inspection a number of lots whose quality level is
typically as good as specified by the Consumer. Continued quality is assured by the acceptance or
rejection of lots following a particular sampling plan and also by providing for a shift to another, tighter
sampling plan, when there is evidence that the Producer's product does not meet the agreed-upon AQL.

Standard offers 3 types of sampling plans:

Mil. Std. 105E offers three types of sampling plans: single, double and multiple plans. The choice
is, in general, up to the inspectors.

Because of the three possible selections, the standard does not give a sample size, but rather a
sample code letter. This, together with the decision of the type of plan yields the specific sampling plan
to be used.

Inspection level:

In addition to an initial decision on an AQL it is also necessary to decide on an "inspection level".

This determines the relationship between the lot size and the sample size. The standard offers three
general and four special levels.

Steps in the standard:

The steps in the use of the standard can be summarized as follows:

1. Decide on the AQL.
2. Decide on the inspection level.
3. Determine the lot size.
4. Enter the table to find sample size code letter.
5. Decide on type of sampling to be used.
6. Enter proper table to find the plan to be used.
7. Begin with normal inspection; follow the switching rules and the rule for stopping the inspection
(if needed).
Choosing a Sampling Plan with a given OC Curve
Sample OC curve:

We start by looking at a typical OC curve. The OC curve for a (52 ,3) sampling plan is shown below.

Number of defectives is approximately binomial:

It is instructive to show how the points on this curve are obtained, once we have a sampling plan
(n,c) - later we will demonstrate how a sampling plan (n,c) is obtained.

We assume that the lot size N is very large, as compared to the sample size n, so that removing
the sample doesn't significantly change the remainder of the lot, no matter how many defects are in the
sample. Then the distribution of the number of defectives, d, in a random sample of n items is
approximately binomial with parameters n and p, where p is the fraction of defectives per lot.

The binomial distribution:

The probability of observing exactly d defectives is given by

The probability of acceptance is the probability that d, the number of defectives, is less than or equal to
c, the accept number. This means that
Sample table for Pa, Pd using the binomial distribution

Using this formula with n = 52 and c=3 and p = .01, .02, ...,.12 we find
Pa Pd
.998 .01
.980 .02
.930 .03
.845 .04
.739 .05
.620 .06
.502 .07
.394 .08
.300 .09
.223 .10
.162 .11
.115 .12
Solving for (n,c)

Equations for calculating a sampling plan with a given OC curve

In order to design a sampling plan with a specified OC curve one needs two designated points.
Let us design a sampling plan such that the probability of acceptance is 1- for lots with fraction
defective p1 and the probability of acceptance is for lots with fraction defective p2. Typical choices for
these points are: p1 is the AQL, p2 is the LTPD and , are the Producer's Risk (Type I error) and
Consumer's Risk (Type II error), respectively.

If we are willing to assume that binomial sampling is valid, then the sample size n, and the
acceptance number c are the solution to

These two simultaneous equations are nonlinear so there is no simple, direct solution. There is
however a number of iterative techniques available that gives approximate solutions so that
composition of a computer program poses few problems

Average Outgoing Quality (AOQ)

Calculating AOQ's

We can also calculate the AOQ for a (n,c) sampling plan, provided rejected lots are 100% inspected and
defectives are replaced with good parts.
 Assume all lots come in with exactly a p0 proportion of defectives. After screening a rejected lot,
the final fraction defectives will be zero for that lot. However, accepted lots have fraction defectivep0.
Therefore, the outgoing lots from the inspection stations are a mixture of lots with fractions defective p0
and 0. Assuming the lot size is N, we have.

For example, let N = 10000, n = 52, c = 3, and p, the quality of incoming lots, = 0.03. Now at p =
0.03, we glean from the OC curve table that pa = 0.930 and

AOQ = (.930) (.03) (10000-52) / 10000 = 0.02775.

Sample table of AOQ versus p

Setting p = .01, .02….12, we can generate the following table

AOQ p
.0010 .01
.0196 .02
.0278 .03
.0338 .04
.0369 .05
.0372 .06
.0351 .07
.0315 .08
.0270 .09
.0223 .10
.0178 .11
.0138 .12
Sample plot of AOQ versus p

A plot of the AOQ versus p is given below.

Interpretation of AOQ plot

From examining this curve we observe that when the incoming quality is very good (very small
fraction of defectives coming in), then the outgoing quality is also very good (very small fraction of
defectives going out). When the incoming lot quality is very bad, most of the lots are rejected and then
inspected. The "duds" are eliminated or replaced by good ones, so that the quality of the outgoing lots,
the AOQ, becomes very good. In between these extremes, the AOQ rises, reaches a maximum, and then
drops.

The maximum ordinate on the AOQ curve represents the worst possible quality that results from
the rectifying inspection program. It is called the average outgoing quality limit, (AOQL).

From the table we see that the AOQL = 0.0372 at p = .06 for the above example.

One final remark: if N >> n, then the AOQ ~ pa p.

The Average Total Inspection (ATI)

Calculating the Average Total Inspection

What is the total amount of inspection when rejected lots are screened?

If all lots contain zero defectives, no lot will be rejected.

If all items are defective, all  lots will be inspected, and the amount to be inspected is N.

Finally, if the lot quality is 0 < p < 1, the average amount of inspection per lot will vary between
the sample size n, and the lot size N.

Let the quality of the lot be p and the probability of lot acceptance is pa, then the ATI per lot is

ATI = n + (1 - pa) (N - n)

For example, let N = 10000, n = 52, c = 3, and p = .03 we know from the OC table that pa = 0.930.
Then ATI = 52 + (1-.930) (10000 - 52) = 753. (Note that while 0.930 was rounded to three decimal places,
753 were obtained using more decimal places.)

Sample table of ATI versus p

Setting p= .01, .02 ...14 generates the following table

ATI P
70 .01
253 .02
753 .03
1584 .04
 2655 .05
3836 .06
5007 .07
6083 .08
7012 .09
7779 .10
8388 .11
8854 .12
9201 .13
9453 .14
Plot of ATI versus p
A plot of ATI versus p, the Incoming Lot Quality (ILQ) is given below.

References:
http://www.itl.nist.gov/div898/handbook/pmc/pmc.htm

http://www.scribd.com

http://www.wikipedia.com