EXPERIMENT NO.

Object: to conduct process capability study for a machine in the workshop.

Theory

Process capability:
Process capability may be defined as the minimum spread of a specific measurement
variation which will include 99.7 % of the measurements from the given process. In other
words, process capability 6 since, 6G is taken as a measure of spread of the process, which is
also called natural tolerance; process capability study is carried out to measure the ability of
the process to meet the specified tolerances.
By this study, it becomes possible to know the percentage of the products which will be
produced within 3 limits on either side of the mean X
A process capability analysis consist
1. Measuring the process capability to find out either the process is inherently capable of
meeting the specified tolerance limits.
2. Discovering why a process capable is falling to meet specifications.
When making the study it is important to minimize the effect of factors such as unnatural
material variation, process adjustment etc. hence homogeneous material should be used , no
process adjustments should be made during the study, trained operators should be allowed to
be perform the work. A number of samples are then taken over a period of time . each sample
consists of a consecutively made pieces.

The analysis is done in the following manner.

1. Calculate the average X and range T of each sample.

2. Calculate the grand average X, these measures the..
3. Calculate control limits and plot X and R charges
This measures the stability of the process, i.e. the extent to which is changes with time.

4. Calculate the process capability 6a2=6??
The measures the piece to piece variability of the process.

Possible relationship of a process in control to upper and lower specification limits.

When a controlled process must meet two specification limits on individual values, upper
specification limit and lower specification limits, the possible situations may be grouped int
these general classes as described below.

1. (Xmax-Xmin)>60
Where X max = upper specification limit
Xmin= lower specification limit
In this case the spread of the process (62 is considerably less than the difference between the
upper specification limit and lower specification limit. The first situation is shown in fig.
The frequency curves a, b, c, d and e shows various positions in which the process might be
centered.

Conclusion:

i) With any position a, b, c, practically all the products manufactured will meet
specifications as the process stays in control.
ii) It may be considered economically advisable to permit X to go out of control if it
does not go too far, i.e. the distribution may be allowed to move between positions
b and c. this may avoid the cost of frequent machine setups and the delay due to
hunting for assignable causes of variation that will not be responsible for
unsatisfactory product.
iii) If (Xmax-Xmin)/6 ratio is considerably large, frequency of control chart may ve
reduced.
iv) If there is an economic advantage to be gained by tightening the specification
limits, it may be considered.
With the process in position d, some product will fall above the upper specification limit in
position c some product will fall below the lower specification limit. In both cases, it is
absolutely necessary to change the centering of the process, bringing it close to position a.

2. (Xmax - Xmin) < 6. In this case the spread of the process (6) to appreciably greater
than the difference between the specifications limits as shown in fig.
Conclusion: in this type of situation defective parts will always be there, therefore, the
remedy will be
i) Increase the tolerance.
ii) Reduce the dispersion, by making fundamental changes in the production
methods, machines used etc.
iii) Suffer and sort out the defectives, if it is economical than making the fundamental
changes, (iv) it is still important to maintain catering of the process.
3. (Xmax-Xmin) = 6 2

Conclusion :

i) It is necessary to maintain centering of process.

ii) It is advisable to increase tolerance if they are tighter than is really necessary.
iii) Reduce dispersion if it is economical.
Importance of process (machine) capability study to solving quality problems:

The information obtained form process capability it of great importance in solving quality
problems as follows:-

1. The design engineer, knowing the capability of the process and the available
equipment, has more rational basis while setting the specifications.
2. The planning engineer can assign the jobs with more tight tolerances to the most
capable machines and that with wider tolerance to the less precise machines.
3. The tool designer can spot the places there tooling improvement must be made to
maintain the process capability.
4. The capability information helps the foreman to decide which machines may require
overhaul.
5. The machine set up learns which machine requires the most attention to set up and
which one needs only normal care
6. The machine operates and inspector can decide which machines need closest with in
production.
7. While purchasing tool provides a means to actual performance of equipment with
manufacturing claim.
 EXPERIMENT NO. 6

Object: to design a sampling scheme based on OC curves.

Theory

The operating curves for an attribute sampling plan is a graph of fraction defectives in a lot
against the probability of acceptance. For any fraction defective p in a submitted lot, the OC
curve shows the provability Pn that such a lot will be accepted by the sampling plan. In a
single sampling plan three parameters are specified.

N = lot size from which the samples are down

n = sample size

C = acceptance number.

If the sampling plan is: N = 100, n =5 , C = 2

It means take a random sample of 5 from a lot of 100, if the sample contains more than x
defectives, reject the lot otherwise accept the lot. By changing the parameters N, n and C
different sampling plans can be obtained. For different sampling plans the OC curves will
differ. To construct an OC curve, we should know the mathematical probability of accepting
lots with varying percent defectives. This can be obtained from the table of Poissons
distribution (given in Appendix)

1. Acceptable quality level (AQL) : it represents the maximum proportion of defectives

which the consumer finds definitely acceptable. AQL can also be defined as the
maximum percent defectives that for the purpose of sampling inspection can be
considered as satisfactory as a process average. It is the fraction defective that can be
 tolerated without any serious effect upon further procession or on customer or on
customer relations.
As an AQL, is an acceptable quality level, the probability of acceptance for an AQL,
lot should be high. In fact the producers safe point is termed as AQL.
2. Reject able quality level (RQL). It is also called as lot tolerance percent defective
(LTPI). This is a definition of unsatisfactory quality. It represents the preparation of
defectives which the consumer funds definitely unacceptable. As RQL is an
unacceptable quality level, the probability of acceptance for an RQL lot should be
low. The probability of accepting a lot at RQL level represents consumers risk.
3. Indifference quality level (IQL). This is a quality level somewhere between the AQL
and RQL. It is frequently defined as the quality level having a probability of
acceptance of 0.50 for a given sampling plan.
4. Average outgoing quality (AOQ). It represents the average % defective in the
outgoing products after inspection, including all accepted and all rejected lots which
have been 100% inspected and defectives replaced by non-defectives.
So, for a given fraction defectives, the lot accepted as a result of first sampling in
section will have a fraction defective P, the rejected lots are subjected to 100%
inspection and rectification ( defective articles are either replaced or corrected) the
AOQ will therefore he less than P).
Let
n= sample size,
N= lot size,
K = number of lots submitted for acceptance
P = fraction defective.
Po = probability of acceptance
(1-Po) = probability of rejection.
Then, the proportion of the lots accepted =pok
Proportion of the lots rejected = k( 1-Po)
Which are subjected to 100% inspection and rectification?
Number of defects originally present in k lots = k.N.p
Number of defectives in the outgoing lots
=p. pe.k(N-n)
Total number of defectives
PPo.k(N-n) + k(n)(0)

????????? ???????????
4OQ = ???
? ??
= PaP? ? ?
If the sample size is much less as compared to lot size.

AOQ = Pap

The calculation of average outgoing quality gives the expected quality in the long run. Over a
short period, the outgoing quality may be better or worse than the long nm average.
Any acceptance/ rectification plan guarantees that regardless of the incurring quality
submitted, the outgoing quality in the long run will not be worse than the AOQL.

For acceptance / rejection scheme the OC curve is used but for rectification scheme the curve
of AOQ plotted against P is used :

The time AOQ = p represents what would happen if there were no inspection. For any given
values of n and c, the AOQ curve falls below the line AOq = P. As p increases, the
proportion of rectified lots increases and hence the AOQ curve falls below the line.

AOQ = p

Characteristics of OC curve:

1. The operating characteristic (OC) curve of an acceptance sampling plan shows the
ability of the plan to distinguish between good lots and bad lots.
2. Sampling acceptance plans with same percent samples gives very different quality
protection. For sample, the curves ( fig. 46.12) show that lots which are 4% defective
will be accepted 81% of the time using a 10% sample form a lot of 100,42% of the
time using a 10% sample from a lot o 200, and less than 3% of the time by a 10%
sample from a lot of 1000, assuming an acceptance number of zero in all cases.
Obviously a producer making a product 4% defective would have a strong motive for
trying to have his product inspected in lots of 50 rather than in lots of 100.
3. Fixed sample size tends towards constant quality protection. It is the absolute size of
the sample rather than its relative size that determines the quality protection given by
an acceptance sampling plan.
4. The OC curves of plan with acceptance numbers greater than zero are superior to
those of comparable plans with acceptance number of zero.
5. With fixed value of N, and can, larger the value of n. the better is the ability of the
plan to discriminate between good and bad lots. Fig. 13 illustrates this point.

6. The larger the sample size and acceptance number, the steeper the slope of the OC
curve. Fig. shows that the larger sample size which protects the consumer against the
acceptance of relatively bad lots also give better protection to the producer against
rejection of relatively good lots.