Production control and operation assignment

on
Single Sampling Plan, OC Curve, AQL, AOQL,
LTPD

By -sabyasachi chowdhury
 (215110074)

Accepting sampling

Acceptance sampling, the third branch of statistical quality control, refers to the process of
randomly inspecting a certain number of items from a lot or batch in order to decide whether
to accept or reject the entire batch.
What makes acceptance sampling different from statistical process control is that acceptance
sampling is performed before or after the process, rather than during the process.
.
Examples include randomly testing a certain number of computers from a batch to make sure
they meet operational requirements, and randomly inspecting snowboards to make sure that
they are not defective.
Acceptance sampling is used when inspecting every item is not physically possible or would
be overly expensive, or when inspecting a large number of items would lead to errors due to
worker fatigue. This last concern is especially important when a large number of items are
processed in a short period of time.

Sampling Plans

A sampling plan is a plan for acceptance sampling that precisely specifies the parameters of
the sampling process and the acceptance/rejection criteria. The variables to be specified
include the size of the lot (N), the size of the sample inspected from the lot (n), the number of
defects above which a lot is rejected (c), and the number of samples that will be taken.

Single sampling plan :

There are different types of sampling plans. Some call for single sampling, in which a
random sample is drawn from every lot. Each item in the sample is examined and is labelled
as either “good” or “bad.” Depending on the number of defects or “bad” items found, the
entire lot is either accepted or rejected.

For example, a lot size of 50 cookies is evaluated for acceptance by randomly inspecting 10
cookies from the lot. The cookies may be inspected to make sure they are not broken or
burned. If 4 or more of the 10 cookies inspected are bad, the entire lot is rejected.
In this example, the lot size N =50, the sample size n =10, and the maximum number of
defects at which a lot is accepted is c =4. These parameters define the acceptance sampling
plan.

Double sampling plan :

Another type of acceptance sampling is called double sampling. This provides an opportunity
to sample the lot a second time if the results of the first sample are inconclusive. In double
sampling we first sample a lot of goods according to preset criteria for definite acceptance or
rejection.

Multiple sampling plan

In addition to single and double-sampling plans, there are multiple sampling plans.
Multiple sampling plans are similar to double sampling plans except that criteria are
set for more than two samples.

Operating Characteristic (OC) Curves

As we have seen, different sampling plans have different capabilities for discriminating
between good and bad lots. At one extreme is 100 percent inspection, which has perfect
discriminating power. However, as the size of the sample inspected decreases, so does the
chance of accepting a defective lot. We can show the discriminating power of a sampling
plan on a graph by means of an operating characteristic (OC) curve. This curve shows the
probability or chance of accepting a lot given various proportions
of defects in the lot.

OC functions of single sampling plans for lot proportion nonconforming

Figure shows a typical OC curve. The x axis shows the percentage of items that are defective
in a lot. This is called “lot quality.” The y axis shows the probability or chance of accepting a
lot. You can see that if we use 100 percent inspection we are certain of accepting only lots
with zero defects. However, as the proportion of defects in the lot increases, our chance of
accepting the lot decreases.
Regardless of which sampling plan we have selected, the plan is not perfect. That is, there is
still a chance of accepting lots that are “bad” and rejecting “good” lots. The steeper the OC
curve, the better our sampling plan is for discriminating between “good” and “bad.”

We can see that the steeper the slope of the curve, the more discriminating is the sampling
plan. When 100 percent inspection is not possible, there is a certain amount of risk for
consumers in accepting defective lots and a certain amount of risk for producers in rejecting
good lots.

Acceptable quality level (AQL) & Lot tolerance percent defective (LTPD):

There is a small percentage of defects that consumers are willing to accept. This is called the
acceptable quality level (AQL) and is generally in the order of 1–2 percent. However,
sometimes the percentage of defects that passes through is higher than the AQL. Consumers
will usually tolerate a few more defects, but at some point the number of defects reaches a
threshold level beyond which consumers will not tolerate
them. This threshold level is called the lot tolerance percent defective (LTPD).

The LTPD is the upper limit of the percentage of defective items consumers are willing to
tolerate. Consumer’s risk is the chance or probability that a lot will be accepted that contains
a greater number of defects than the LTPD limit. This is the probability of making a Type II
error—that is, accepting a lot that is truly “bad.” Consumer’s risk or Type II error is generally
denoted by beta.

Producer’s risk is the chance or probability that a lot containing an acceptable quality level
will be rejected. This is the probability of making a Type I error—that is, rejecting a lot that
is “good.” It is generally denoted by alpha.

Average Outgoing Quality Level

As we observed with the OC curves, the higher the quality of the lot, the higher is the chance
that it will be accepted. Conversely, the lower the quality of the lot, the greater is the chance
that it will be rejected. Given that some lots are accepted and some rejected, it is useful to
compute the average outgoing quality (AOQ) of lots to get a sense of the overall outgoing
quality of the product.

Assuming that all lots have the same proportion of defective items, the average outgoing
quality can be computed as follows:

Where
Pac _ probability of accepting a given lot
p _ proportion of defective items in a lot
N _ the size of the lot
n _ the sample size chosen for inspection
Usually we assume the fraction in the previous equation to equal 1 and simplify the
Equation to the following form:

AOQ = (Pac) p
=pa p (N-n)/N

Problem:1

Let us consider a single sampling plan with N=2000,n=50,c=2.The incoming quality of the
batch is 2 percent non conforming. Find the average outgoing quality level.

Solution:

From the Poisson cumulative distribution table, the probability of using the lot using the
sampling plan is .920

The average outgoing quality is,

AOQ= (.920)(0.02)(2000-50)/2000
=0.0179

Thus if batches come in 2 % non conformity the average outgoing quality level is 0.0179.

Problem:2

Find a single sampling plan that satisfies a producer’s risk of 5% for lots that are 1.5% non
conforming.

Solution:

We are given alpha = 0.05 and p1 =AQL=0.015.

If we choose a acceptance number c=1, we get np1 =0.355, the sample size is

N=0.355/p1 =0.355/0.015=23.67=24

If our acceptance number is 3, we have np1 =1.366 and the sample size is

N=1.366/0.015=92

If our acceptance number is 6, we have np1 =3.286 and the sample size is

N=3.286/0.015=220
Problem 3

Let’s say that we want to develop an OC curve for a sampling plan in which a sample of n _
5 items is drawn from lots of N _ 1000 items. The accept/reject criteria are set up in such a
way that we accept a lot if no more than one defect (c _ 1) is found.

Solution:

Let’s look at the partial binomial distribution in Table 6-2. Since our criteria require us to
sample n = 5, we will go to the row where n equals 5 in the left-hand column. The “x”
column tells us the cumulative number of defects found at which we reject the lot. Since we
are not allowing more than one defect, we look for an x value that corresponds to 1. The row
corresponding to n =5 and x =1 tells us our chance or probability of accepting lots with
various proportions of defects using this sampling plan.

For example, with this sampling plan we have a 99.74% chance of accepting a lot with 5%
defects. If we move down the row, we can see that we have a 91.85% chance of accepting a
lot with 10% defects, a 83.52% chance of accepting a lot with 15% defects, and a 73.73%
chance of accepting a lot with 20% defects. Using these values and those remaining in the
row, we can construct an OC chart for n =5 and c =1

Problem :
Let’s go back to our initial example, in which we sampled 5 items (n _ 5) from a lot of 1000
(N =1000) with an acceptance range of no more than 1(c = 1) defect. Here we will construct
an AOQ curve for this sampling plan and interpret its meaning

solution :

For the parameters N =1000, n =5, and c= 1, we can read the probabilities of Pac from Figure.
Then we can compute the value of AOQ as AOQ = (Pac) p.

Figure shows a graphical representation of the AOQ values. The AOQ varies, depending on
the proportion of defective items in the lot. The largest value of AOQ, called the average
outgoing quality limit (AOQL), is around 15.85%. We can see from Figure that the average
outgoing quality

The AOQ for n =5 and c = 1

will be high for lots that are either very good or very bad. For lots that have close to 30% of
defective items, the AOQ is the highest. Managers can use this information to compute the
worst possible value of their average outgoing quality given the proportion of defective items
(p). Then this information can be used to develop a sampling plan with appropriate levels of
discrimination.