Scribd
Upload
4 views4 pages

45 - SQC

Statistical Quality Control (S.Q.C.) is a method developed during World War II to manage variations in product quality in large-scale manufacturing. It distinguishes between chance variations, which are natural and unavoidable, and assignable variations, which can be controlled and eliminated. S.Q.C. employs tools like control charts for process control and sampling inspection for product control to maintain quality within specified limits.

Uploaded by

Gaurav Sisode
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
4 views4 pages

45 - SQC

Statistical Quality Control (S.Q.C.) is a method developed during World War II to manage variations in product quality in large-scale manufacturing. It distinguishes between chance variations, which are natural and unavoidable, and assignable variations, which can be controlled and eliminated. S.Q.C. employs tools like control charts for process control and sampling inspection for product control to maintain quality within specified limits.

Uploaded by

Gaurav Sisode
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

STATISTICAL QUALITY CONTROL

Consumer goods are being produced on a large scale with the new industrial
techniques. In any manufacturing process it is not possible to produce goods of
exactly the same quality. Some variation in the quality of the product is
inevitable. To ensure that the quality of the product continues to be in the
specified range of variability some scientific method is required.
This is the origin of Statistical Quality Control. These methods were developed
during Second World War. During this war, millions of parts of complex
machines were produced on a large scale. They had to confirm to rigid
specification limits and initially a large number of them were rejected. By the
applications of S.Q.C. method it was possible to cut down the rejects to a very
large extent. As a result the importance of S.Q.C. in industry was realized and
when the war ended it became a standard technique in all major industries.
Consider an industrial process of manufacturing threaded nuts of some specific
diameter. When thousands of such nuts are produced daily the scientific
measurements of their diameter will always show a variation. The causes of this
variation can be broadly divided into two categories:
1. CHANCE VARIATIONS: Certain small variations are natural in any
manufacturing process and cannot be prevented under any circumstances. These
variations are due to large number of causes beyond human control.
There may be a small variation in the machine settings due to fluctuations in
electric current. It is possible to ensure that the variation due to such causes is
within the specification limit. This type of variation is called variation due to
chance causes or allowable variation.
2. ASSIGNABLE / PREVENTABLE VARIATIONS: The other type of
variation which may occur in the quality of the goods can be attributed to causes
which can be detected. e.g. faulty machine settings, low quality of raw material,
improper handling of machines, inefficiency of the operator, etc. As against the
chance variation these variations can be controlled and eliminated. Hence this
type of variation is called variation due to assignable causes or preventable
variations.
When these variations are present a large number of items go beyond the
specified limits and hence the production becomes uneconomic. In such a case,
the process is said to be out of control.

©Dr. Ashwini Deshpande Page 1 of 4


Thus in S.Q.C. the aim is to control the manufacturing process so that the
proportion of items beyond specified limits is not very large. This is known as
Process Control.
The other type of problem is to ensure that lots of manufactured goods do not
contain large proportion of defective items. This is known as Product Control or
Lot Control.
These two are distinct problems because even when the process is in control so
that the proportion of defective items over a long period is not very large, an
individual lot may not be of satisfactory quality.
Process Control is achieved mainly through the technique of Control Charts
whereas Lot Control is achieved through Sampling Inspection.
By quality of the product we mean any characteristic of finished product or of an
intermediate product.
Many quality characteristics are measurable quantitatively. For e.g.: length of a
screw, chemical composition of a drug, life of an electric bulb, etc. Most of them
are continuous variables.
Sometimes the quality characteristics cannot be measured and is expressed as an
attribute. In such cases the items may be classified as defective or non-defective.
The quality of the product of an individual process may be determined either by
inspecting each item produced i.e. 100% inspection or by inspecting a sample
from the product.
But 100% inspection is a costly method. It is also found to be less accurate in
actual practice. The very routine of inspecting a large number of items brings
down the efficiency of the inspecting personal. Quality control by sampling
inspection is therefore an acceptable routine.
Thus S.Q.C. is a statistical method of sampling and tests of significance to ensure
that the quality of the product of the industrial process is maintained at a desired
level i.e. to ensure that the process remains under control.
It also finds out whether the variations in the quality of the product are within the
limits of chance variation.

©Dr. Ashwini Deshpande Page 2 of 4


SHEWART’S CONTROL CHARTS
A control chart is a basic tool of S.Q.C. It essentially is a graph in which the
results of inspection of a sample of the product are plotted from time to time.
Whether a particular sample is of required quality or not is determined from it on
the basis of some statistic calculated from it. The setting up of control chart
commences with the drawing of random samples from the production process at
regular intervals. Then some statistic of interest such as sample mean or sample
range or proportion of defectives in the sample is calculated for each sample. If
no assignable causes are present in the process and the variations are purely
random then the variations in the values of statistic will follow a certain
statistical distribution which can be suggested by theory. If X is the statistic
under inspection and µ and σ are its mean and standard deviation respectively.
Then under the assumption of normality of the distribution of X,
P(µ - 3σ < X < µ + 3σ) = 0.9973
If X is not normally distributed then by Chebyshev’s inequality:
P(|X - µ| < k) ≥ 1- (σ/k)2
Putting k = 3σ,
P(|X - µ| < 3σ ) ≥ 1-1/9
P(µ - 3σ < X < µ + 3σ) ≥ 8/9
Thus, (µ ∓ 3σ) limits cover the maximum range for the statistic X for normal as
well as non-normal population. The control chart for X then consists of:
1. Upper control limit (UCL): µ+3σ
2. Central line (CL): µ
3. Lower control limit (LCL): µ-3σ

The graph is as shown below:

UCL

CL

LCL

©Dr. Ashwini Deshpande Page 3 of 4


If the process is under statistical control then we expect each value to fall within
the control limit. If any value falls outside the control limit, it is an indication of
the presence of an assignable cause. As long as the value continues to lie within
the control limit the process is under control. As soon as sample point falls
outside the control limits it is the signal indicating that variation in the product is
no longer only chance variation. A control chart gives such a signal as soon as
something goes wrong with the process thus avoiding wastage of time and
money.
Lack of control is also shown by non-random variation within the control limits.
This may be:
1. Cyclical Movements
2. Bulging of points near L.C.L. or U.C.L.
3. A run of 7 or 8 points.
There are two types of control charts:
1. Control charts for variables: These charts are used for the control of those
characteristics which can be measured or expressed in specific units of
measurements. These variables are of continuous type and hence can be
regarded to follow Normal distribution. The control charts are:
i) Control charts for mean (X̅) and range (R)
̅) and s.d. (S)
ii) Control charts for mean (X

2. Control charts for attributes:


i) Control charts for fraction defectives (p-chart)
ii) Control charts for number of defectives (np-chart or d-chart)
iii) Control charts for number of defects (c-chart)

XXXXX

©Dr. Ashwini Deshpande Page 4 of 4

You might also like