1

Lesson-1

INDEX NUMBERS
Structure
1.0 Objectives
1.1 Introduction
1.2 Meaning of Index Numbers
1.3 Types of Index Numbers
1.4 Concept of Price Relatives
1.5 Problems in the Construction of Index Numbers
1.6 Summary
1.7 Further Readings
1.8 Model Questions
1.0 Objectives
After going through the lesson you should be able to:
 Define Index number.
 Explain the role of Index numbers.
 Explain the distinction between Index number and other Measures of Central
tendency.
 List at least four types of Index Numbers.
 List the issues in the construction of index numbers.
1.1 Introduction
Index numbers are commonly used statistical device for measuring the combined
fluctuations in a group related variables. If we wish to compare the price level of consumer
items today with that prevalent ten years ago, we are not interested in comparing the prices
of only one item, but in comparing some sort of average price levels. We may wish to compare
the present agricultural production or industrial production with that at the time of
independence. Here again, we have to consider all items of production and each item may
have undergone a different fractional increase (or even a decrease). How do we obtain a
composite measure? This composite measure is provided by index numbers which may be
defined as a device for combining the variations that have come in group of related variables
over a period of time, with a view to obtain a figure that represents the ‘net’ result of the
change in the constitute variables.
1.2 Meaning of Index Numbers
Index numbers are statistical measures designed to show changes in a variable or
group of related variables with respect to time, geographic location or other characteristics
such as income, profession, etc. A collection of index numbers for different years, locations,
etc., is sometimes called an index series.
 2

According to Bowley, “Index numbers are used to measure the changes in some
quantity which we cannot observe directly”. For example, changes in business activity in
a country are not capable of direct measurement but it is possible to study relative changes
in business activity by studying the variations in the values of some such factors which affect
business activity, and which are capable of direct measurement.
Index numbers may be classified in terms of the variables that they are intended to
measure. In business, different groups of variables in the measurement of which index
number techniques are commonly used are:
(i) price,
(ii) quantity,
(iii) value and
(iv) business activity.
Thus, we have index of wholesale prices, index of consumer prices, index of industrial
output, index of value of exports and index of business activity, etc. Here we shall be mainly
interested in index numbers of prices showing changes with respect to time, although
methods described can be applied to other cases. In general, the present level of prices is
compared with the level of prices in the past. The present period is called the current period
and some period in the past is called the base period.
1.3 Types of Index Numbers
Following types of index numbers are usually used:
Index number may be simple or composite.
Simple Index Number:
A simple index number is a number that measures a relative change in a single
variable with respect to a base.
Composite Index Number:
A composite index number is a number that measures an average relative changes in
a group of relative variables with respect to a base.
Price index Numbers:
Price index numbers measure the relative changes in prices of a commodities
between two periods. Prices can be either retail or wholesale.
Quantity Index Numbers:
These index numbers are considered to measure changes in the physical quantity of
goods produced, consumed or sold of an item or a group of items.
Mathematically, an economic index number is an average of the many different
changes in price or quantity of the goods and services included in the index.
There are many different valid ways of determining this average and therefore many
different possible estimates of average price and quantity changes. It is therefore important
that index numbers are calculated in a way which is appropriate for their intended use.
 3

Points to Remember
 An index number is a statistical device (measure) to express average changes in
related variables.
 Most widely used index number in practical statistics in price index number.
 A price index number measures relative changes in average prices over time.
 An index number is specialized average. It can be simple/unweighted or weighted.
1.4 Concept of Price Relatives
The simplest example of an index number is a price relative. It is defined as a ratio of
the price of a single commodity in a given period called current period (P1) to its price in some
past period called base or reference period (P0).
P1
Price Relative (PR) =
P0
Index numbers are designed to measure the magnitude of economic changes over time
using the concept of price relatives. Because they work in a similar way to percentages they
make such changes easier to compare. Briefly, this works in the following way.
Suppose that a cup of coffee in a particular coffee cost 75p in 1995. In 2002, an
identical cup of coffee cost 99p. How has the price changed between 1995 and 2002? The
particular time period of 1995 which we've chosen to compare against, is called the base
period.
The variable for that period, in this case the 75p, is then given a value of 100,
corresponding to 100%.
The index can then be calculated for the later period of 2002 as a proportionate
change as follows:
99 99
index number =  100, so the index number   100  132
75 75
The index number shows us that there has been a price increase of 32% since the
base period. An index number for a single price change like this is called a price relative.
Rule for finding the price relative
If we let P0 be the price in the base period and let P1 be the price in the later period, then
P1
the price relative for the price change between these periods is given by
P0
A price index number is a sort of average of these individual price relatives and
measures the 1 price changes of all commodities collectively.
Thus, in our day to day life we are faced with situations wherein index numbers are
required m one form of the other. Is it not desirable, therefore that we gain systematic
information about these crucial numbers? Hope you are convinced and we are in agreement.
So let us proceed with the construction, formulation and limitations of some of these Indexes.
 4

1.5 Problems/Issues in the Construction of Index Numbers

Let us discuss the various problems/issues that arise in the construction of a
construction of a price index number for any country. The problems may be enumerated as
follows:
(a) Purpose of the index.
(b) Choice of the base period.
(c) Choice of commodities.
(d) Collection of data.
(e) Method of combining data.
(f) Choice of weights.
(g) Interpretation of the index.
(a) Purpose of the Index
The purpose for which the index number is being constructed should be clearly and
unambiguously stated, since most of the later problems will depend upon the purpose. For
instance, if we want to construct an index number for measuring the change in the general
price level, we have to take the wholesale prices of finished products, intermediate products,
agricultural products, mineral products, etc. For example, one consumer price index
number cannot be suitable for indicating the changes in consumer prices in different
localities and for different classes of people. Similarly, one index number of industrial
production cannot be suitable for different localities and different sectors of industry.
(b) Choice of the Base Period
Suppose we want to compare the price levels of two time-periods, say the price level of
1990 with that of 1980. We call the year 1990 the current period and the year 1980 the base
period. The base period thus constitutes the basis of comparison. The price level of the base
period is arbitrarily taken as 100 and the price level of the current period is expressed
relative to that.
 The base period should be a normal period in the recent past. It should be a normal
period; i.e., the prices of that period should not be subject to a boom or a depression
or effects of catastrophes like wars, floods, famines, etc.
 It is also desirable to select a base period which is not too far in the past, for then we
may not get comparable figures. Market conditions, i.e. tastes and habits of people,
may undergo some change, resulting in the replacement of old goods by new ones.
Thus we find that when a base, on being used for a number of years, becomes a period
in the remote past, it is to be shifted to a period in the recent past for subsequent
comparisons.
 The base period should not be too short or too long. It should not be too short, e.g. a
single day or week, because the prices for too short period are highly unstable and
unreliable. Again, it should not be too long, e.g. six years, for the average price for that
period may smooth out some important fluctuations.
 The base period should be a period for which reliable figures are available and
preferably a period for some economic importance for the country concerned. For
 5

example, the year 1980 may have some economic importance for India, being the
inaugural year of India’s sixth five-year plan.
(c) Choice of Commodities
It is practically impossible to include the prices of all commodities of an economy in
constructing a price index number. The reason is that it involves too much time, money and
labour. Most of the practical uses of index numbers require that current index numbers
should be available quickly. As such, we are to take suitable sample commodities so that the
index number may reflect the change in the price level, keeping in mind the purpose it is to
serve.
The selection of commodities should be by judgement sampling, and not by random
sampling, since to make the index representative of the price fluctuations we have to select
the important and relevant commodities. Different groups of commodities display different
patterns of prices and movements. So the commodities are to be classified into groups
showing similar patterns of price fluctuations, and a number of commodities representative
of each of these groups are to be selected. For constructing a wholesale price index number
to measure the change in the general price level, the commodities may be classified according
to the following scheme:

Commodities

Unmanufactured articles Manufactured articles

Agricultural Farm Mineral

products products products Forest Others
products

Semi- Finished
Food Non-food
manufactured or products
articles articles
intermediate
goods

The quality of the selected commodities should not vary much from period to period,
and no commodity should disappear from the market. Reliable figures should also be
available for the selected commodities.
The exact number of commodities included in the index should depend on the purpose
of the index. Thus, if we construct an index number based on a few commodities most of
which are food-items (which are known to be highly sensitive to price changes), the index
may be useful for certain purposes, but cannot be used for measuring the change in the
general price level. No rigid rue, however, can be laid down for the number of commodities to
be included. But it may be stated that the number should neither be too large nor too small.
 6

(d) Collection of Data

Makers of price index numbers take great pains to collect the necessary data in each
period for all the commodities included in the index number. The price of a commodity at a
particular period of time will vary from one market to another and also for different grades.
So we are to collect prices of a commodity from a number of representative markets for a few
important grades of the commodity. Each of these prices referred to as a price quotation. In
the case of wholesale price index numbers we are to collect wholesale prices of commodities,
and for cost of living index numbers retail prices are required. As in all other cases of
collection of statistical data, here too ulmost care should be taken to get accurate data.
(e) Method of Combining Data
The price fluctuations of different commodities are reflected in the price relative. We
want to represent the change by means of a single number. So we are to consider some
means of combining these individual price fluctuations. Although different commodities may
have peculiar characteristics in their price fluctuations, it has been empirically found that,
taken as whole, the distribution of price relatives is bell-shaped with a marked central
tendency in combining the different price relatives. Price relatives for individual items are
calculated to include individual price movements. Then at the next stage, these price
relatives are averaged to produce an index number. The resulting figure is taken as an
indicator of the overall movement in the prices. We shall illustrate the procedure with the
help of an example.
Amongst the various measures of central tendency, the arithmetic mean and the
geometric mean of price relatives are generally used.
Let us denote by P0i the price of i th commodity in the base period and by P1i the price of the
commodity in the current period.
If we use arithmetic mean of price relatives for constructing the index number, then
k

( P 1i / P0i )
l01  i 1
k
Where I01 is the index for the current period and  denotes summation over the k
commodities. This is simple or unweighted index using the arithmetic mean of price
relatives.
Similarly, the formula for the index number using the simple geometric mean of price
relatives will be:
1/ k
 k 
l01    P1i / P0i 
 i 1 
In the same way, simple harmonic mean, median or mode of price relatives may be used.
So far we have considered some kind of average price relative to get the index number.
We can also get a simple index number by comparing the simple aggregate of actual price for
the current period with that for the base period. Symbolically,
 7
k

P 1i
l01  i 1
k

Pi 1
0i

This is called a simple aggregative index.

We have to multiply each formula by 100 to express the index in the percentage form.
However, this factor is generally omitted from the formula and introduced at the last stage.
Suppose that we are concerned with the price movement in the entire fruit-basket
instead of apples alone. Table below contains relevant information on units, prices and
quantities in the basket for two periods, called 0 and t.
Table Data on units prices and quantities, period o and t.
Item Unit Prices (Rs.) Price
Relative
Pt/p0
Period Period
0 t
(p0i) (pti)
(1) (2) (3) (4) (8)
Apples Kg. 10 15 1.5
Banana Doz. 5 6 1.2
Grapes 1/4th kg. 10 8 0.8
Citrus 6ps. 15 15 1.0
Pine-apple Ps. 8 4 0.5
Pomegranate Kg. 14 28 2.0
Guava 5kg. 40 30 0.75
Total 102 106 7.75
Now let us try to sum up the situation through an index.
The simplest way out would be to take arithmetic mean of the price relatives as :
k

( P 1i / P0i )
l01  i 1
…….(3)
k
where the subscript i is for the ith commodity and k is the number of commodities
under consideration: From the data in Table 3 if we sum up the last column, we get
7


i 1
Pti / Poi  7.75

where k=7 (the number of items being considered) and

 8

7.75
I 01   1.1071
7
(f) Choice of Weights
The commodities included in the index number are not all of equal importance. For
instance, in constructing a wholesale price index number of India, ‘rice’ should have greater
importance than ‘tobacco’. So the problem of weighting the different commodities included in
the index number according to their importance deserves attention. If we ignore weights, we
shall not get an unweighted or a simple index, but an inappropriately weighted index.
It is also easily seen that in the simple average of price relatives, each relative
influences the index number according to its percentage of increase or decrease over the base
period. The influence which a commodity exerts on the simple aggregative index depends on
the price per unit in which it is quoted.
Thus we must adopt a system of weighting for the price relatives or prices that will
truly reflect the importance of each commodity. Since our index should not depend on the
units on which the prices or quantities are reported, we shall weight the price relatives by
values and the prices by quantities. The quantity used for determining the weight may be the
quantity of the commodity produced, marked or sold, imported or exported. The price and
quantities required for the weights may relate either to the base period or to the current
period.
If wi be the weight attached to the price relative for the i th commodity, then the
weighted arithmetic mean of the price relatives is given by
p1i
ki1 wi
p0i
l01 
ki1 wi
The weighted geometric mean by
1/ ki1 wi
 k  p wi 
l01     1i  
 i 1  p0i  
 

Similarly, if wi is the weight attached to the price of ith commodity, then the weighted
aggregative index is given by
 ki 1 p1i wi
l01 
 ki 1 p0i wi
Now let us consider some particular weighted index numbers of prices.
If we take wi  q0i , the base-period quantities, then we get
 9

ki1 p1i q0i

l01 
ki1 p0i q0i
which is known as Laspeyre’s formula.
If we take wi  q1i , the current-period quantities, then we get

ki1 p1i q1i

l01 
ki1 p0i q1i
which is known as Paasche’s formula.
If we take wi   q1i  q0i  / 2 , the average of current-period and base-period quantities, then we
get

ki1 p1i  q1i  q0i 

l01 
ki1 p0i  q1i  q0i 
which is known as Marshall-Edgeworth formula.
Irving Fisher tested a large number of formulae and selected the following formula, which is
obtained by crossing Laspeyre’s and Paasche’s formulae geometrically:

ki1 p1i q0i ki1 p1i q1i

l01  
ki1 p0i q0i ki1 p0i q1i

This is known as Fisher’s ideal index number, because it satisfies certain tests of
consistency which Irving Fisher considered appropriate.
In the majority of countries, the index numbers are computed using Laspeyre’s
formula or its equivalent, the weighted arithmetic mean of price relatives, the weights being
the base-period values. The formula is simple to calculate and the necessary data may be
easily obtained. The other most commonly used formula is the constant-weight aggregative or
constant-weight arithmetic mean of price relatives is generally used in view of the difficulty
involved in its calculation. Formulae involving current-period quantities are also not
frequently used, since it is difficult to obtain these figures.
(g) Interpretation of the Index
The interpretation will depend on the purpose of the index number. The wholesale
price index number measures the change in the general price level from the base period to
the current period, while the cost of living index number compares the amounts of money
required to purchase the same basket of goods and services for the two periods.
Generally, the index numbers are expressed as percentage form and /oc, the index
number for the base period, is taken as 100. Thus, the statement "The wholesale price index
number for India during 1989-90 with 1981-82 as base is 165.7” means that, as compared
with the price level during the year 1981-82, the price level during the year 1989-90
increased 1.657 times.
 10

Check You Progress (CYP) -1

Please use the space given below to answer the question. And after giving your answers match your
answers with the suggested answers given at the end of the lesson.
1. Why index numbers are said to be Specialized Averages?
-------------------------------------------------------------------------------------------------------------------------------------
-------------------------------------------------------------------------------------------------------------------------------------
2. Which index number is said to be 'Ideal' Index Number?
-------------------------------------------------------------------------------------------------------------------------------------
-------------------------------------------------------------------------------------------------------------------------------------

Example
The table below gives the wholesale prices (p) and quantities produced (q) of a number of
commodities in Delhi. Calculate Laspeyre's, Paasche's, the Edgeworth-Marshall and Fisher's
Ideal' index numbers for the year 1985, with the year 1982 as base.
Commodity 1982 1985
P q P q
Rice 277,92 1.1 366.67 6.2
Wheat 176.25 06.0 186.58 116.9
Jowar 151.00 4.2 182.57 5.5
Barley 121.83 2.4 181.25 1.0
Bajra 156.75 13.1 155.75 6.1
Gram 273.00 1.0 498.83 0.6

Let P01, q01, and P11, q11 denote the prices and quantities for 1982 and 1985, respectively.
Then
6

P q
i
01 01 = 22.241.229

P q
i
11 01 = 23,921.776

P q
i
01 11 = 24399.034

P q
i
11 11 = 26,519.314

The wholesale price index according to the four formulae will then be as follows:
23921.776
Laspeyre's formula l01  100  107.56
32241.339
 11

26519.314
Paasche's formula l01  100  108.69
24399.034
5.441.08
Edgeworth — Marshall formula l01  100  108.15
46640.263
Fishers 'ideal formula l01  107.56 108.69  108.12
1.6 Summary
Index numbers are commonly used statistical device for measuring the combined
fluctuations in a group related variables. It is a composite measure which helps in
understanding the variations that have come in group of related variables over a period of
time. Index numbers may be simple or composite. Though the use of index number is not
tied with any particular sphere of human activity or a particular branch of scientific pursuit,
it is the field of business and economics which has the most numerous index series in use.
Index number may be simple or composite. We can construct Price index Numbers and
Quantity Index Numbers. Price index numbers measure the relative changes in prices of a
commodities between two periods. In contrast, Quantity Index Numbers are considered to
measure changes in the physical quantity of goods produced. In this lesson we have
discussed various problems/issues that arise in the construction of a construction of a price
index number for any country. In this lesson we have precisely discussed very popular index
numbers such as Laspeyre's, Paasche's and Edgeworth-Marshall.
1.7 Further Readings
1. A.M. Goon, M. K. Gupta, B. Dasgupta: Fundamentals of Statistics Vol.1.
2. John E. Freund and Frank, I Williams: Modem Business Statistics.
3. S. C. Gupta, and V.K. Kapoor : Fundamentals of Applied Statistics.
1.8 Model Questions
1. What is an Index Number? Discuss the importance of the Price Relatives.
2. List important cautions to be taken in the 'choice of the base period' in the
construction of an Index Number.
3. For the data given in the following data relating to different articles find price and
quantity index numbers for 1990 taking 1989 as base year using Laspeyre's ,
Paasche's, the Edgeworth-Marshall and Fisher's 'ideal' index numbers formula.
Year Article 1 Article 2 Article 3 Article 4
Price Price Qty Price Qty Price
Qty Qty
1990 5.00 5 7.75 6 9.63 4 12.50 9
1989 6.50 7 8.80 10 7.75 6 12.75 9
4. Construct the wholesale price index number for 1969 and 1970 from the data given below.
Use 1968 as the base year.
 12

Commodity Wholesale price in Wholesale price in Wholesale price in Rs.

Rs. Per quintal Rs. per quintal per quintal
1968 1969 1970
Rice 140 160 190
wheat 120 130 140
Linseed 100 105 108
Gur 75 80 90
Cotton 250 270 300
Tobacco 400 420 450

Suggested Answers to CYP

1. Because these measure difference variables having different behavior.
2. Fisher's.

*****
 13

Lesson-2

TESTS OF INDEX NUMBERS

Structure
2.0 Objectives.
2.1 Introduction
2.2 Time Reversal Test
2.3 Factor Reversal Test
2.3.1 Examples
2.4 Chain Index
2.5 Relative Merits and Demerits of Chain-base and Fixed-base Methods
2.6 Errors in the Index Numbers
2.7 Summary
2.8 Further Readings
2.9 Model Questions
2.0 Objectives
After going through the lesson you should be able to:
 State the two tests of consistency for Index Numbers.
 Distinguish between Time reversal test and Factor reversal test.
 Explain the meaning of index numbers.
 Define Chain Index.
 Distinguish between Chain based Index and Fixed based Index.
2.1 Introduction
In the previous lesson a score of index number formulae, with their relative merits and
demerits have been discussed. Each formulation, indeed, offers something positive to suit a
particular situation. But there is, no index number formula that is relevant for all situations.
Therefore, in any practical application if one is clear about the objective for which an index
number is being computed only then one would be able to make the right choice.
However, the literature offers certain mathematical tests of consistency as a guide to a
sound choice. It is asserted that an 'ideal' index ought to satisfy those tests. Though the
logical validity of those tests is not an accepted proposition, yet, it is good to acquaint
ourselves with the tests.
On the tests of adequacy of Index Number Formulae. The mathematical tests of
consistency for index numbers are inspired by the logic of analogy. It is contended that a
 14

proposition that holds true for a single commodity should also hold for a group of
commodities considered as a composite.
Tests for index numbers.
Irving Fisher considered two tests of consistency which a price index number should
satisfy, viz. the time reversal and the factor reversal test.
2.2 Time Reversal Test
According to this, any formula to be accurate should be time-consistent; that is, we
should get the same picture of the change in the price level if the base period and the current
period be interchanged. Consider, a particular commodity, say rice. If the price of rice is
doubled from 1978 to 1989, then the price relative for the period ’89 and ’78 as base is 2.00,
while that ’78 and ’89 as base will be 0.50. Thus one is the reciprocal of the other and the
product is 1. This obviously true for each individual price relative and, according to the time
reversal test, it should be true for the index number. In symbols, this test says
l01  l10  1
Let us elaborate this point. Consider the following hypothetical data for a single
commodity, called A.
Price of A Quantity Bought
(Rs. per unit) (units)
Period 0 Period 1 Period 0 Period 1
(P0) (P1) (qo) (q1)
100 200 20 20
P1 200
Now P01   2
P0 100
P0 100
Now P10    0.5
P1 200
One can see that
1
P01 
P10
i.e. P01 × P10 =2×0.5=1

There is hardly anything unusual about it. If the price in period 1 has doubled in
comparison to period 0 then one would always say that the prices of this commodity in
period 0 was just half its price in period1.
This is the genesis of the Time Reversal test, which demands that, even when we are
dealing with several commodities, an interchange of the time subscripts of an index number
formula should result in a formula that is reciprocal of the original one. In other words, if P 01
and P10 are computed with some index number formulation, then P01 should be equal to l/P10
for the satisfaction of the time-reversal test.
 15

Let us apply this test to the three foremost contenders viz. Laspeyre's price index,
Paasche's index, and Fisher's formula:
Let there be two periods 0 and t representing base and current period respectively.
Σi pti q0i
Now P0t(Laspeyre) =
Σi p0i q0i
and
Σi p0i qti
Pt0 (Laspeyre) =
Σi pti qti
Since the entities involved have undergone a change, in general
Σpti q0i Σp0i qti
 1
Σp0i q0i Σpti qti
Hence, the Laspeyre index number will not satisfy the time-reversal test, in general.
Σi pti qti
Similarly, P0t (Paasche) = ;
Σt p0i qti
Σi p0t q0i
Pt0 (Paasche) = and given the normal movement of prices and quantities,
Σt pti q0i
ingeneral, would not obtain Pi0 (Paasche) as the reciprocal of Poi (Paasche). So, even this
popular index number, in general, does not satisfy their time reversal test.

Σi pti q0i Σi pti qti

However, P0t (Fisher) = 
Σi p0t q0t Σi p0i qti
and

Σt p0t qti Σi p0i q0i

Pt0 (Fisher) = 
Σi pti qti Σt pti q0i
Thus, the time-reversal test is easily met by this formulation.
2.3 Factor Reversal Test
The value of commodity is the product of the price per unit and the number of units of
the commodity produced. The value of all commodities will be the sum of these products for
various commodities. Thus the ratio of values for the two periods (0 and 1 say)gives the value
index  lv  :

 i p1i q1i
lv 
 i p0i q0i
 16

According to this test, if the price and quantity factors in the price index formula l p  
be interchanged so that a quantity index formula lq   is obtained, then the product of these
two indices should give the value index. Symbolically, one should have
l p  lq  lv
Fisher’s ‘ideal’ formula is the only price index which satisfies factor reversal test.
For this formula,

ni p1i q0i ni p1i q1i

lp  
ni p0i q0i ni p0i q1i

ni q1i p0i ni q1i p1i

lq  
ni q0i p0i ni q0i p1i
while
 i p1i q1i
n

lv 
ni p0i q0i
where n denotes the number of periods
Obviously, l p  lq  lv for this formula.

This harmless looking test renders Laspeyres' index and Paasche's index undesirable,
for, the two formulae tail to satisfy this test. And, quite expectedly, the Fisher's index number
formula does satisfy this Factor reversal test. So if we give credence to these tests then, out of
the stated index number formulae, only Fisher's index comes good. Because; even the
Marshall-Edgeworth formula meets just the time reversal test and not the factor reversal test,
whereas fisher's version satisfies both the tests.
But that is no sufficient ground for accepting Fisher's index number to the total
exclusion of either Laspeyre's or Paasche's index. After all, Laspeyres and Paasche index
offered legitimate and direct interpretation of what was being measured. There is an obvious
trade off between technical superiority and logical content; one can make the choice
accordingly.
2.3.1 Example
From the following classic decades old data relating to prices of some commodities
show that the arithmetic mean of price relatives does not satisfy the time reversal test
1950 1960
Hats 5.00 6.00
Ties 1.00 1.50
Shoes 8.00 8.00
 17

Solution:
items 1950(p0) 1960(p1)

hats 5.00 6.00 1.2 .833

ties 1.00 1.50 1.5 .677
Shoes 8.00 8.00 1.0 1.000
Total 3.7 2.510
1 p1 3.7
P01 =  = =1.2333
k p0 3
1 p0 2.51
P10 =   =0.837
k p1 3
Hence P01  P10=1.2333  0.837  1
Hence time reversal test is not satisfied by Arithmetic mean of price relatives.
Another Example
For the data given in the following table relating to different articles show that the Time
reversal test and factor reversal tests are satisfied by Fisher’s Ideal test (Take 1989 as base
year)
Year Article 1 Article 2 Article 3 Article 4
Price Qty Price Price Qty. Price Qty.
Qty.
1990 5.00 5 7.75 6 9.63 4 12.50 9
1989 6.50 7 8.80 10 7.75 6 12.75 9

Solution:
Article P0 Q0 P1 Q1 P1Q0 P0Q0 P1Q1 P0Q1

1 5.00 5 6.50 7 32.50 25.00 45.50 35.00

2 7.75 6 8.80 10 52.80 46.50 88.00 77.50
3 9.63 4 7.75 6 31.00 38.52 46.50 57.78
4 12.50 9 12.75 9 114.75 112.50 114.75 112.50
Total 231.05 222.52 294.75 282.78

231.05
P01(La) =  100=103.8
222.52
 18

282.78
Q01(La) =  100=127.8
222.5
294.75
P01(Pa) =  100=104.23.8
282.78
294.75
Q01(Pa) =  100=127.57
231.05
231.05 294.75
P01(F) =   100 = 104.1
222.52 282.78
282.78 294.75
Q01(F) =   100 =127.24
222.52 231.05
231.05 294.75 282.78 222.52
Now P01(F)  P10(F) =    1
222.52 282.78 294.75 231.05
231.05 294.75 282.78 294.75
P01(F)  Q10(F) =   
222.52 282.78 222.52 231.05
294.75
= = V01= value index
222.52
Hence Fisher’s ideal index number satisfies the time reversal and factor reversal tests.
Remark
In the above context (i.e. making point-to-point or binary comparisons) if the time gap
is very large or if development is proceeding at a rapid pace then our understanding of the
change, through index numbers, can be erroneous in spite of a 'good' formula. This is so
because a selection of x commodities, involved in two periods of time, may be devoid of
sameness. A strict comparison calls for binary articles and not a numerically same collection
of the unique items.
Thus, we face a strange situation: While these index number formulae require
homogeneity, the fast changing situations offer variety and heterogeneity. There lies the
source of Homogeneity error. We can always reduce this error by choosing fairly close base
year and constructing complete 'chains'; But practical situations involve comparisons of
prices spreading over several years. An immediate response would be to fix a base year called
period 0, and go in for pairwise comparisons like (0,1), (0;2), (0,3) and so on. In other words,
given a spread of prices over periods 0, 1, 2……, n, we can construct an index number series
I0I , I02, I03,...,I0n with each entity summing up the price situation, in its respective period,
relative to the base year prices.
However, if we so desire, we can make pair-wise comparisons without fixing a base i.e.,
we can construct the series I0I, I12, I23……..,In-1,n . Such a move would be a safeguard against
the problem of out of dateness of fixed weights, particularly so, when the time-series is quite
long. However, if we so desire, we can make pair-wise comparisons without fixing a base i.e.,
we can construct the series P01, P12, P23,……..Pt-1 t. Such a move would be a safeguard against
the problem of out of dateness of fixed weights, particularly so, when the time-series is quite
long.
 19

At time, this process is extended to form a chained index which is discussed in the
next section.
2.4 Chain index
The index numbers we have considered so far are of the fixed-base type; that is, the
base period with which we compare the other time periods remains fixed with the progress of
time. We have also noted that with the passage of time new commodities enter the market
and old ones disappear; besides, the quality of the commodities may undergo a change. Also,
the relative importance of various commodities, being dependent on the tastes and habits of
the consumers, changes. If an index number is needed for comparing successive time
periods – says, the periods 0, 1, 2, ..., n – it is not necessary to use a fixed base 0. We may
use the previous period as base for comparing any time period and construct what are called
link-indices. There is no change in the method of calculation; only the base period changes
for each comparison and in each case it is the previous period.
The symbol used for such an index for comparing the prices of period k with those of
period (k-1) is lk 1,k

Thus we construct link indices: I01 I12, I23, . . . In-1,n

By multiplying successive links, i.e. by chaining, we obtain the chain indices as shown
below:
I01
I`02 = I01 X I12
I`03 = I01 X I12 X I23

I`0n = I01 X I12 X . . . . .X In-1,n

In this method, there is no fixed base period. The year immediately preceding the one
for which price index have to be calculated is assumed as the base year. Thus, for the
year2004 the base year would be 2003, for 2003 it would be 2002 for 2002 it would be 2001
and so on. In his way there is no fixed base. It goes on changing
Example:
Find index numbers for the following data taking 2005 as base year.

Years 1999 2000 2001 2002 2003 2004

Price 18 21 25 23 28 30

Solution:
Link Relatives
Year Price Pn Chain Indices
= 100
Pn1

18
1999 18 100  100 100
18
 20

21 100 116.67
2000 21 100  116.67  116.67
18 100

25 116.67 119.05
2001 25 100  119.05  138.9
21 100

23 138.9  92
2002 23 100  92  127.79
25 100

28 127.79 121.74
2003 28 100  121.74  155.57
23 100

30 155.57 107.17
2004 30 100  107.14  166.68
28 100

These chain indices will not in general be equal to the corresponding fixed-base
indices unless the formula used meets the so-called circular test.
Stated symbolically, the test is
l01  l12 .... ln1n  ln0  1
The time reversal test reversal test l01  l10  1 is a particular case of the circular test. Thus if a
formula satisfies the circular test, then
l01  l12 .... ln1n  1/ ln0  l0n
Base Shifting Formula
The base period can be shifted to any convenient subsequent period if the formula satisfies
the circular test, since lkn can be calculated from the following relation, which follows from
the circular test-
l0 n
lkn 
l0 k
Consider the following example: An index number series is given (with Base 1990=100),
recast the series by shifting the base to 1995 (i.e. 1995=100).
S.No. Original Index New Index Number
Number Series Series (1995=100)
(1990-100)
1 1990 100 100
 100  66.67
150

2 1991 110 110

 100  73.33
150
 21

3 1992 115 110

 100  76.67
150

4 1993 122 122

100  81.33
150

5 1994 140 140

100  91.33
150

6 1995 150 150

100  100
150

7 1996 200 1200

100  133.33
150

8 1997 105 105

100  70.00
150

9 1998 180 180

100  120.00
150

10 1999 170 170

100  113.33
150

11 2000 190 190

100  126.67
150

12 2001 185 185

100  123.33
150

Remarks
 Such a procedure is valid in the case of an aggregate index with fixed weights and
does not hold for a series based on changing weight formulations. However, in practice
this clause is usually ignored and the same procedure is applied for shifting the base,
irrespective of the underlying formulation.
 A useful application of this procedure (of base shifting) found where we need to
combine two or more overlapping series of index numbers, for the same entity but
different base year, for obtaining a single continuous series encompassing all the
fragments. In technical jargon this exercise of combining is called splicing. The
exercise is akin to base shifting because for a single continuous series we ought to
have the same base but the fragments had different base years, so the process of
combining would necessarily involve base shifting as well.
2.5 Relative Merits and Demerits of Chain-Base and Fixed-Base Methods
 We have seen that the fixed-base index numbers become more and more inaccurate as
the distance between the base period and the current period increases. As the chain-
base index numbers are based on a number of link-indices, each of which is expected
 22

to be quite accurate, it is claimed that the chain-base index numbers are more
accurate than the fixed-base ones, so far as long-term comparison is concerned.
 Also, a chain index fully utilises the information regarding prices and quantities of all
the intervening periods between the base period and the current period, whereas a
fixed-base index makes use of data concerning the base period and the current period
only.
 Fixed – base index numbers are generally easier to calculate and are more easily
understood by users of index numbers than chain-base index numbers.
 The chief advantage of this method is that the price relatives of a year can be
compared with the price level of the immediately preceding year. Businessmen mostly
interested in comparison of this type rather than in comparison relating to distant
past.
 Yet another advantage of the chain base method is that it is possible to include new
items in an index number or to delete old times which are no more important. In fixed
base method it is not possible.
 The practical advantage of a chain index is that the sample of commodities and/or the
set of weights may be kept quite up-to-date in any index number. However, any
change in the set of commodities or in the set of weights will upset the circular test.
Some authorities, on the other hand, hold that since a chain index is obtained by
multiplying a number of link-indices, it may involve a cumulative error, although none has
put forward any convincing proof for the existence of such error.
2.6 Errors in the Index numbers
The index numbers thus constructed will be subject to different types of errors. The
errors are generally classified as :
(a) formula error.
(b) sampling error and
(c) homogeneity error.
The formula error arises out of the choice of a particular formula in the construction of
an index number. There cannot be any universally accepted formula which can measure the
price changes with exactitude, and hence each formula is subject to some error inherent in
the formula.
The sampling error arises from the selection of certain commodities out of the
complete list of binary commodities, i.e. the commodities which are marketed in
approximately the same quality in the current and base periods. Naturally, the sampling
error decreases with an increase in the sample number of commodities included in the
construction of the index number.
The third type of error is homogeneity error. This error arises from the fact that index
numbers are calculated from the data on binary commodities, whereas they should be based
on ail the commodities marketed in the base period and the current period, including both
binary and unique commodities. Since with the passage of time many old commodities
disappear from the market and new commodities appear, the homogeneity error increases as
the gap between the base period and the current period increases.
 23

Self Assessment Questions

Please use the space given below to answer the question. And after giving your answers
match your answers with the suggested answers given at the end of the lesson.
1. What is Time Reversal Test?

2. Besides formula error, which other errors could creep in the Index numbers?

2.7 Summary
Each formula in the array of index number formulae carries their relative merits and
demerits. Each formulation, indeed, offers something positive to suit a particular situation.
But there is, no index number formula that is relevant for all situations- Therefore, in any
practical application if one is clear about the objective for which an index number is being
computed only then one would be able to make the right choice. Irving Fisher considered two
tests of consistency which a price index number should satisfy, viz. the time reversal and the
factor reversal test. In this lesson we have also discussed the chair-based index number,
wherein we use the previous period as base for comparing any time period and construct
what are called link-indices. And finally we had discussed the Relative merits and demerits of
chain-base and fixed-base methods.
2.8 Further Readings
1. A.M. Goon, M. K. Gupta, B. Dasgupta: Fundamentals of Statistics Vol.1.
2. John E. Freund and Frank, I Williams: Modem Business Statistics,
3. S. C. Gupta, and V.K. Kapoor : Fundamentals of Applied Slatistics.
2.9 Model Questions
(1) Show that for the following series of fixed base index numbers the chain base
index numbers are same as fixed base index numbers.
Year 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Index 100 120 122 116 120 120 137 137 141 143
Number
(2) From the data given below examine if Fisher's price index number satisfies the -
Time Reversal Test and the Factor reversal test.
Commodity 2005 2009
P q P q
Rice 277.92 1.1 366.67 6.2
Wheat 176.25 106.0 186.58 116.9
Jowar 151.00 4.2 182.57 5.5
Barley 121,83 2.4 181.25 1.0
 24

Bajra 156.75 13.1 155.75 6.1

Gram 273.00 1.0 498.83 0.6

(3) Show that Fisher's index number lies between Laspeyre's and Paasche's index
number.
Suggested Answers to SAQ
1. In the context of index numbers, in Time Reversal test, the formula should be
time-consistent. Same result should be there with interchange of base period and
current period.
2. Sampling errors, and homogeneity error.

—
 25

Lesson-3

COST OF LIVING INDEX NUMBER AND

USES OF INDEX NUMBERS
Structure
3.0 Objectives
3.1 Introduction
3.2 Construction of Consumer Price Index Numbers
3.3 Examples
3.4 Real Income
3.5 Uses of Index Numbers
3.6 Summary
3.7 Further Readings
3.8 Model Questions
3.0 Objectives
After going through the lesson you should be able to:
 Define Consumer Price Index Number.
 Compute Consumer Price Index Numbers.
 Define real Income.
 Enlist at least four uses of Index numbers.
3.1 Introduction
A cost of living index number (or a consumer price index number) measures the
relative change in the amount of money required to produce equivalent satisfaction in two
different situations. The cost of living index number always relates to a designated group of
people, e.g. the menial class of people in Calcutta. In practice, this index is constructed by
comparing the consumer (retail) prices, for the two situations, for a fixed set of goods and
services representing the consumption level (or the level of living) of the given group of people.
A consumer price index (CPI) is a measure estimating the average price of consumer goods
and services purchased by households. A consumer price index measures a price change for
a constant market basket of goods and services from one period to the next within the same
area (city, region, or nation).It is a price index determined by measuring the price of a
standard group of goods meant to represent the typical market basket of a typical urban
consumer. It is one of several price indices calculated by most national statistical agencies.
The percent change in the CPI is a measure estimating inflation.
The cost of living index should cover the food, clothing, fuel and lightening, house-rent
and miscellaneous groups. Each group should include a representative sample of the items
in consumption.
 26

A separate index number is to be published for each of the major groups and a general
index for all the groups combined. In calculating this index, weights are to be used
proportional to the relative importance in consumption of the items in a group and also of the
different groups. For each item, there will be number of price quotations covering different
brands and markets. The price relative of an item is the simple average of the price relatives
for the different quotations of the item.
A group index is a weighted average of the price relatives of the different items of the group,
the weights being proportional to their consumption expenditure. The general index is, in its
turn, the weighted average of the group indices, the weights being proportional to the
consumption expenditure on the different groups.
The question of determining the list of items to be priced and their weights is very important.
The items should represent the consumption level of the given group of people. This is found
by means of family-budget enquiry. On the basis of this enquiry, a list of items
representing the level of living can be determined. An obvious criterion for the selection of
items is their importance. For a satisfactory picture of the price movements, all types of items
having characteristic price movements should be included. Weights, which are proportional
to consumption expenditure, are also determined from the family-budget enquiry.
With a change in the consumption pattern, there arises a need for a new study of consumer
purchases. Even in the normal course of events, economic changes sometimes outmode the
old consumption pattern. As a result of wars and economic upheavals, very great changes
occur in the consumption pattern. Such changes in the pattern necessitate the undertaking
of a fresh family-budget enquiry, on whose basis the list of items and the corresponding
weights have to be modified.
Consumer price index number measures the changes in the prices paid by the consumers
for purchasing a special “basket” of goods and services during the current year as compared
to the base year. The basket of goods and services will contain items like (1) Food (2) House
Rent (3) Clothing (4) Fuel and Lighting (5) Education (6) Miscellaneous like washing,
transport, newspaper etc.
Consumer price index number is also called cost of living index numbers or retail price index
number.
3.2 Construction of Consumer Price Index Numbers
The following steps are involved in the construction of consumer price index numbers.
(1) Class of People:
The first step in the construction of consumer price index (CPI) is that the class of
people should be defined clearly. It should be decided whether the cost of living index
number is being prepared for the industrial workers, middle or lower class salaried people
living in a particular area. It is therefore necessary to specify the class of people and locality
where they reside.
(2) Family Budget Inquiry:
The next step in the construction of consumer price index number is that some
families should be selected randomly these families provided information about food, clothing,
house rent, miscellaneous etc. The inquiry include questions on family size, income, the
quality and quantity consumed and the money spent on them and the weights are assigned
in proportions to the expenditure on different items.
 27

(3) Price Data:

The next step is to collect the data on retail prices of the selected commodities for the
current period and the base period these prices should be obtained from the shops situated
in that locality for which the index numbers are prepared.
(4) Selection of Commodities:
The next step is the selection of the commodities to be included. We should select
those commodities which are mostly used by that class of people.
3.3 Examples
1. Suppose it is required to determine the index number of agricultural area in West
Bengal for the year 1985-86 with the triennium ending crop year 1971-72 as base. For this,
it is necessary to obtain individually the indices for the groups: foodgrain: cereals, pulses;
non-foodgrain: oilseeds, fibre, plantation crops, spices, vegetables and miscellaneous crops.
The indices of the different groups, except cereals, along with percentage weights are given
below:
Group Index number 1985-86 Weight
Foodgrain:
(i) Cereals - 77.97
(ii) Pulses 64.05 9.40
Non-foodgrain :
(i) Oil seeds 219.23 2.42
(ii) Fibre 154.05 7.17
(iii) Plantation crop 110.38 1.26
(iv) Spices 472.84 0.14
(v) Vegetables and miscellaneous 144.22 1.64
crops

The basic data for food group (cereals) are given below:
Articles Index number 1985-86  li  Weight  wi 

Rice 101.78 71.28

Aus 60.45 11.42
Aman 102.30 57.02
Boro 257.44 2.84
Wheat 92.51 4.71
Barley 25.19 0.99
Maize 115.30 0.70
 28

Ragi 160.32 0.12

Small millets 75.39 0.17

We obtain the index for foodgrain (cereals) by using weighted arithmetic mean of the
indices of the articles of cereals from the above table.
Index number for foodgrain (cereals)
Wi Ii 15082906
=   101.06
Wi 149.25
Applying the weights to the group indices, taking the index for foodgrain (cereals) as
calculated, we have finally the index number of agricultural area in West Bengal (for 1985-86)
with base: triennium ending crop year 1971-72 = 100) as
I = [101.06 x 77.97 + 64.05 x 9.40 + ... + 144.22 x 1.64 ] / 100
10558.591
  105.59
100
2. The following data relate to the group indices and the corresponding weights (shown in
brackets) for the menial class cost of living index numbers in Calcutta:
Year Food Clothing Fuel and light House Miscellaneous
rent
(71.28) (2.89) (9.27) (9.87)
(6.69)
1978 370.1 423.3 469.1 110.0 279.2
1979 387.2 440.4 469.8 115.8 287.1
1980 394.0 432.9 352.0 116.9 285.1
1981 396.7 551.4 366.0 116.9 291.8
1982 380.2 504.2 336.8 116.9 283.6

Calculate the general cost of living index for each of the above years.
Sol.
IW
Formula used: CI 
W
For year 1978

Group Weights (W) I IW

Food 71.28 370.1 26380.728
Clothing 2.89 423.3 1223.337
Fuel and Light 9.27 469.1 4348.557
House Rent 6.69 110.0 735.90
 29

Misc. 9.87 279.2 2755.704

Total 100.00 35444.226

IW 35444.226
CI    354.44226
W 100
For year 1979
Group Weights (W) I IW
Food 71.28 387.2 27599.616
Clothing 2.89 440.4 1272.756
Fuel and Light 9.27 469.8 4355.046
House Rent 6.69 115.8 774.702
Misc. 9.87 287.1 2833.677
Total 100.00 36835.797

IW 36835.797
CI    368.35797
W 100
For year 1980
Group Weights (W) I IW
Food 71.28 394.0 28084.32
Clothing 2.89 432.9 1251.081
Fuel and Light 9.27 352.0 3263.04
House Rent 6.69 116.9 782.061
Misc. 9.87 285.1 2813.937
Total 100.00 36194.439

IW 36194.439
CI    361.94439
W 100
For year 1981
Group Weights (W) I IW
Food 71.28 396.7 28276.776
Clothing 2.89 551.4 1593.546
Fuel and Light 9.27 366.0 3392.820
House Rent 6.69 116.9 782.061
Misc. 9.87 291.8 2880.066
 30

Total 100.00 36925.269

IW 36925.269
CI    369.25269
W 100
For year 1982
Group Weights (W) I IW
Food 71.28 380.2 27100.656
Clothing 2.89 504.2 1457.138
Fuel and Light 9.27 336.8 3122.136
House Rent 6.69 116.9 782.061
Misc. 9.87 283.6 2799.132
Total 100.00 35261.123

IW 35261.123
CI    352.61123
W 100
3.4 REAL INCOME
Whenever there is a change, both the content and the form undergo a change. For
instance, the per capita income of India was Rs. 245.50 in 1980-81 and in 1995-1996 it
stood at Rs. 2595,60. Does it mean that our command over goods and services have gone up
ten times? In monetary terms yes, but we are also aware that prices have also gone up very
many times. So in order to get at the 'real' situation we need to discount this inflated picture
by the extent of price rise. This is done as follows:
Current Money Income/Earnings
Real Income/Real Wage = 100
Price Index
So, whenever any entity expressed in nominal units produces a money illusion, we can
resort to this procedure of deflating by using the relevant price deflator.
If we look up any Economic Survey we will find that National Income (or other variants
of national prosperity like Cross National Product, Net National Product or Net Domestic
Product) series are always reported at, both, current and constant prices. This way one can
isolate the 'real' from the 'monetary'
.Example1. The total wages and the number of workers employed in jute mills around
Calcutta are given below:
Year Total wages Number of workers
(Rs. Lakhs) (000)
1948 2076 319
1949 2453 306
1950 2246 291
 31

1951 2231 272

1952 2552 275
Calculate the average nominal wages and real wages for the jute textile workers, using
the general cost of living indices for the menial class people of Calcutta.
Sol.
Total wages
Average Nominal wages =
No. of Worker
Average wages
Real wages = 100
Cost of living index
Year Total wages Number of Cost of Average Real wages
(Rs. Lakhs) workers Living index wages
(000)
1948 2076 319 354.44226 650.7836 183.607
1949 2453 306 368.35797 801.6339 217.623
1950 2246 291 361.94439 771.8200 213.242
1951 2231 272 369.25269 820.2205 222.129
1952 2552 275 352.61123 928.0000 263.179

2. The daily wages rate of agricultural labourers in the village Chenchery distt.
Trivendrum and village Panur distt. Caunancre of the state Kerala is given below for number
of years. Also are given the consumer price index no. for agricultural labourers of Kerala with
the base 1960/61
Year Daily wages village Chenchery CPI in percentage Daily wages village Panur
1957-58 1.34 92 N.A.
1958-59 1.44 96 1.42
1959-60 1.75 99 1.50
1960-61 1.75 100 1.50
1961-62 1.98 104 1.58
1962-63 2.00 105 1.62
1963-64 2.02 108 1.75
1964-65 2.38 144 1.96
1965-66 2.77 178 2.25
1966-67 3.42 203 2.25
1967-68 4.58 221 2.38
1968-69 4.58 223 2.96
1969-70 4.67 221 3.00
1970-71 4.50 231 3.75
 32

Obtain series of index numbers to represent the daily wages of agricultural labourers
in the two villages with base 1960/61 and then deflate these series to obtain the index
numbers for real wages.
Sol.
Daily wages for the concerned year
Index number for a particular year = 100
Daily wages for the base year
Index number for concerned year
Index number for real wages = 100
C.P.I. of the conerned year
For village Chenchery
Year Daily wages CPI in Series of I.no I.no. for Real
village Chenchery percentage wages
1957-58 1.34 92 76.57143 83.22981
1958-59 1.44 96 82.28571 85.71429
1959-60 1.75 99 100 101.0101
1960-61 1.75 100 100 100
1961-62 1.98 104 113.1429 108.7912
1962-63 2.00 105 114.2857 108.8435
1963-64 2.02 108 115.4286 106.8783
1964-65 2.38 144 136 94.44444
1965-66 2.77 178 158.2857 88.92456
1966-67 3.42 203 195.4286 96.27023
1967-68 4.58 221 261.7143 118.4228
1968-69 4.58 223 261.7143 117.3607
1969-70 4.67 221 266.8571 120.7498
1970-71 4.50 231 257.1429 111.3173

For village Panur

Year Daily wages CPI in Series of I.no I.no. for Real
village Panur percentage wages
1957-58 N.A. 92 - -
1958-59 1.42 96 94.66667 98.61111
1959-60 1.50 99 100 101.0101
1960-61 1.50 100 100 100
1961-62 1.58 104 105.3333 101.2821
1962-63 1.62 105 108 102.8571
 33

1963-64 1.75 108 116.6667 108.0247

1964-65 1.96 144 130.6667 90.74074
1965-66 2.25 178 150 84.26966
1966-67 2.25 203 150 73.89163
1967-68 2.38 221 158.6667 71.79487
1968-69 2.96 223 197.3333 88.49028
1969-70 3.00 221 200 90.49774
1970-71 3.75 231 250 108.2251

Self Assessment Questions

Please use the space given below to answer the question. And after giving your answers
match your answers with the suggested answers given at the end of the lesson.
1. What is 'Cost of Living Index number'?
----------------------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------------------
------------------------------------------
2. What is Real Income?
----------------------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------------------
------------------------------------------

3.5 Uses of Index Numbers

Before concluding this lesson, let us look at the uses of Index Numbers.
Though the use of index number is not tied with any particular sphere of human
activity or a particular branch of scientific pursuit but it, is the field of business and
economics which has the most numerous index series in use.
Again, at the Micro level, a business magnet wishes to feel the pulse of the economy
and judge the prospects as well as the performance of his enterprise prior to his undertaking
fresh investment decision. Relevant indices of costs, prices and production provide him the
necessary information.
Similarly, a consumer feel concerned about the 'purchasing power of money' and at the
negotiation table a worker wants to impress upon the management that the current wages
are not sufficient for his sustenance. In both the cases a concrete yardstick, capable of
causing change, is required. It is in situations, like these that one looks for the 'price index'
or the 'Cost of Living Index'.
Following are the main uses of index numbers
1. Index numbers serve as a barometer for measuring the value of money. With the
help of index number we can easily make a comparison in the value of money in
different years.
 34

2. Importance for the Govt,: The change in the value of money has a direct effect on
the life of the public. So the Govt. adopts the fiscal and monetary policy according
the results of index numbers.
3. Throws Light on Economic Condition: Index numbers are very helpful in
comparing the economic conditions of a particular group at two different periods.
4. Consumption Standard:
If we want to know the true consumption standard of a class in a locality we can
compute the consumption index number.
5. Fixation of Wages: The money wages can be revised according the proportionate
charge in the cost of living. The cost of living index number guides the Govt. and
the executives for the fixation and revision of wages.
6. Importance for the producer: Price Index Number indicates the producer that he
should expand the production or he should reduce the production, if price level is
rising it means profit rate is high.
7. Analysis of Industry: If we want to judge the prospects of manufacturing
concerns, the investment index number can be constructed, to know the net yield
of the industrial sector.
3.6 Summary
• Index numbers are used in the fields of commerce, meteorology, labour, industrial,
etc.
• The index numbers measure fluctuations during intervals of time, group
differences of geographical position of degree etc.
• They are used to compare the total variations in the prices of different
commodities in which the units of measureme its differs with time and price etc.
• They measure the purchasing power of money. :

• They are helpful in forecasting the future economic trends.

• They are used in studying difference between the comparable categories of

animals, persons or items.
 Index numbers of industrial production are used to measure the changes in the
level of industrial production in the country.
 Index numbers of import prices and export prices are used to measure the changes
in the trade of a country.
 The index numbers are used to measure seasonal variations and cyclical variations
in a time series.
3.7 Further Readings
1. A.M. Goon, M. K. Gupta, B. Dasgupta: Fundamentals of Statistics Vol.1.
2. John E. Freund and Frank, I Williams: Modem Business Statistics.
3. S. C. Gupta, and V.K. Kapoor: Fundamentals of Applied Statistics.
 35

3.8 Model Questions

(1) The following table gives group index numbers and corresponding group weights
with regard to cost of living for a given year. Construct the overall cost of living
index for the year.
Groups Index number Weight
Food 350 5
Fuel and Lighting 220 1
Clothing 230 1
Rent 160 1
Miscellaneous 190 2
How is the overall index changed if all the group index numbers are increased by 10
and all the weights doubled ?
(2) What are the uses and limitations of a cost of living index number? Describe in
details how it is constructed in general.
(3) An enquiry into the budgets of middle class families in a family gave the following
information;
Expenses Food Rent Clothing Fuel Others
30% 15% 20% 10% 25%
Prices in Rs. In 1990 100 20 70 20 40
Prices in Rs. In 1995 90 20 60 15 55
What are the changes in the cost of living figures of 1995 with that of 1990?
(4) Deflate the per capita income shown in the following table to get the Real income.
year 1985 1986 1987 1988 1989 1990 1991 1992
Cost of living index 100 110 120 130 150 200 250 350
Per capita income (Rs.) 65 70 75 80 90 100 110 130

Suggested Answers to SAQ

1. A measure to scale the relative change the amount of money required to produce
equivalent satisfaction in two different situations.
2. It is adjustment of current income with proce index. As formula is 100 x (Current
Income/Proce Index)

____
 36

SECTION - B

Lesson-4

STATISTICAL QUALITY CONTROL :

AN INTRODUCTION
Structure
4.0 Objectives
4.1 Introduction
4.2 Defining Quality
4.3 Statistical Quality Control (SQC)
4.4 Chance and Assignable Variations
4.5 Process Control and Lot Control
4.6 Statistical Basis of Control Charts: 3  and Probability Limits
4.7 Variable and Attributes in SQC
4.8 Levels of Understanding SQC
4.9 Uses of SQC
4.10 Summary
4.11 Further Readings
4.12 Model Questions
4.0 Objectives
After going through the lesson you should be able to :-
– Define quality.
– State three functions of SQC.
– Enlist the types of variations which are prime focus in SQL.
– Distinguish between process control and product control.
– Enlist four uses of SQC.
4.1 Introduction.
Statistical Quality Control (SQC) has had its origin of manufacturing industry and it is
in the manufacturing industry that the SQC finds its important applications. The feature of
manufacturing work which makes it possible to utilize statistical methods in an effective
manner is the presence of large numbers of repetitive operations. If in any process, there are
no repetitive operations, then it is unlikely that statistical methods will have any useful role
to play in increasing the efficiency of such a process. At any rate the tools so far developed
for SQC will not be applicable. Thus in composing a poem or painting or any other work of
art, statistics has very little help to render. But in the making of nuts and bolts of ball-
 37

bearings or textiles or chemicals or plastics or any other similar products, there is ample
scope for the use of statistical methods, because in modern Industry these products are
made by standardized operations which are repeated over and over again in a large number
of times. Thus, the first point to be emphasized in connection with SQC is the presence or
absence of repetitive operations.
Statistical quality control is based on the theory of probability and sampling and is
extensively used in all types of Industries particularly those which produce large number of
small items like, corks, pins, screws, nuts, bolts, motor parts, medicines plastic parts, etc.
In fact in all repetitive processes the SQC plays a very important and significant role.
The technique of statistical quality control is of recent origin. It was introduced by W.A.
Shewhart of the Bell Telephone company of U.S.A., in 1924. He gave the idea of process
control with the help control charts, about which we will discuss later. Almost about the
same time, Harold F. Dodge and Rowing also of the Bell Telephone company developed the
idea of product control with the help of sampling inspection plans. They used probability
theory and the theory of sampling to determine (i) whether the production process is in
control or not and (ii) whether the finished produce is of the specified quality and as such
whether it would be accepted or rejected by the prospective buyers. According to Eugene L.
Grant, “Statistical Quality Control is simply a statistical method for determining the extent to
which quality goals are being met without necessarily checking every item produced and for
indicating whether or not the variations which occur are exceeding normal expectations. SQC
enables us to decide whether to reject or accept.
Duncan describes it as, “It is called statistical quality control (SQC) because quality is
economically judged by sampling methods and statistical experiments are scientifically
designed to assess variation in quality, to determine limits of tolerance etc”.
4.2 Defining Quality
We may define quality in many ways. The traditional definition of quality is based on
the point that products end services must meet the requirement of those who use them.
(1) Quality means fitness for use. There are two general aspects of quality----Quality
of design and Quality of conformance.
All goods and services are produced in different levels of quality and these variations
in quality arc intentional. For example all automobiles have basic aim of safe transportation
still they differ with respect to their shape, size and performances. It is called Quality of
design.
Quality of conformance means how well the product adheres to the specifications
required by the design. This type of the quality is influenced by many factors including the
quality assurance systems involving process control lot control and acceptance sampling.
The term quality is often used without making clear whether it is quality of design or quality
of conformance.
(2) A modern definition of quality is that Quality is inversely proportional to
variability.
The smaller the variation the greater is the quality. Hence quality improvement is
achieved through reduction of variability in process and products. Lesser variability implies
reduction of waste.
 38

Example- An automobile company in U.S.A. performed a comparative study of a

transmission that was manufactured by a domestic plant or Japan plant. The one made
by Japan Company showed less variability in quality, was working more consistently and
thus resulted in lesser warranty costs and thus showed better quality
Statistics being the science of variability can thus provide tools for maintaining quality.
Investing in a Statistical quality control programme means indirectly making more money
as it leads to reduced scrap, rework, and a smaller fraction of conforming products and
services.
4.3 Statistical Quality Control (SQC)
Statistical Quality Control (SQC) is the term used to describe the set of statistical
tools used by quality professionals. SQC should be viewed as a kit of tools which may
influence decisions related to the function of 1) specification, 2) production and 3) inspection.
Statistical quality control can be divided into three broad categories:
1. Descriptive statistics are used to describe quality characteristics and relationships.
Included are statistics such as the mean, standard deviation, the range, and a
measure of the distribution of data.
2. Statistical process control (SPC) involves inspecting a random sample of the output
from a process and deciding whether the process is producing products with
characteristics that fall within a predetermined range. SPC answers the question of
whether the process is functioning properly or not.
3. Acceptance sampling is the process of randomly inspecting a sample of goods and
deciding whether to accept the entire lot based on the results. Acceptance sampling
determines whether a batch of goods should be accepted or rejected.
The tools in each of these categories provide different types of information for use in
analyzing quality. Descriptive statistics are used to describe certain quality characteristics,
Such as the central tendency variability of observed data. Although descriptions of certain
characteristics are helpful, they are not enough to help us evaluate whether there is a
problem with quality. Acceptance sampling can help us do this. Acceptance sampling helps
us decide whether desirable quality has been achieved for a batch of products, and whether
to accept or reject the items produced. Although this information is helpful in making the
quality acceptance decision after the product has been produced, it does not help is identify
and catch a quality problem during the production process. For this we need tools in the
statistical process control (SPC) category.
All three of these statistical quality central categories re helpful in measuring and
evaluating the quality of products or services.
According to Eugene L. “Statistical Quality control is simply a statistical method for
determining the extent to which quality goals are being met without necessarily checking
every item produced and for indicating whether or not the variations which occur are
exceeding normal expectations. SQC enables us to decide whether to reject or accept.
However, statistical process control (SPC) tools are used most frequently because they
identify quality problems during the production process. For this reason, we will devote most
of the chapter to this category of tools. The quality control tools we will be learning about do
not only measure the value of a quality characteristic. They also help us identify a change or
variation in some quality characteristic of the product or process. We will first see what types
 39

of variation we can observe when measuring quality. Then we will be able to identify specific
tools used for measuring this variation.
Variation in the production process leads to quality defects and lack of product
consistency. The Intel Corporation the world’s largest and most profitable manufacturer of
microprocessors understands this.
Therefore, Intel has implemented a program it calls “copy-exactly” at all its
manufacturing facilities. The idea is that regardless of whether the chips are made in Arizona,
New Mexico, Ireland, or any of its other plants, they are made in exactly the same way.
The reason is to minimize all variation.
Now let’s look at the different types of variation that exist.
Activity (1) to Check Progress
Notes :
 Please use the space given below to each question to give your answer.
 Please check your response with the Model answers given at the end of this lesson.
I. What is the modern definition of quality?
___________________________________________________________________________________
___________________________________________________________________________________
II. What is SQC?
___________________________________________________________________________________
___________________________________________________________________________________

4.4 Chance and Assignable Variations (Types of Variations: Random and

Assignable variations)
If we look at bottles of a soft drink in a grocery store, we will notice that no two bottles
are filled to exactly the same level. Some are filled slightly higher and some slightly lower.
Similarly, if we look at blueberry muffins in a bakery we will notice that some are slightly
larger than others and some have more blueberries than others. These types of differences
are completely normal. No two products are exactly alike because of slight differences in
materials, workers, machines, tools, and other factors. These are called common, or random,
causes of variation. Common causes of variation are based on random causes that we
cannot identify. These types of variation are unavoidable and are due to slight differences in
processing. An important task in quality control is to find out the range of natural random
variation in a process. For example, if the average bottle of a soft drink called Cocoa Fizz
contains 16 ounces of liquid, we may determine that the amount of natural variations
between 15.8 and 16.2 ounces. If this were the case, we would monitor the production
process to make sure that the amount stays within this range. If production goes out of this
range—bottles are found to contain on average 15.6 ounces—this would lead us to believe
that there is a problem with the process because the variation is greater than the natural
random variation.
The second type of variation that can be observed involves variations where the causes
can be precisely identified and eliminated. These are called assignable causes of variation.
 40

Examples of this type of variation are poor quality in raw materials, an employee who
needs more training or a machine in need of repair. In each of these examples the problem
can be identified and corrected. Also, if the problem is allowed to persist, it will continue to
create a problem in the quality of the product. In the example of the soft drink bottling
operation, bottles filled with 15.6 ounces of liquid would signal a problem. The machine may
need to be readjusted. This would be an assignable cause of variation. We can assign the
variation to a particular cause (machine needs to be readjusted) and we can correct the
problem (readjust the machine).
Similar variations can be observed in any series of actual measurement of
manufactured products. From the stand point of, SQC is necessary to distinguish two types
of variations — (a) variations that can be economically traced to some causes and (b)
variations which cannot be so traced and which are attributed to chance. The former
variations are said to be due to ‘assignable causes’.
A process is said to be in process control when it is operating in the presence of
chance variations alone in the absence of assignable variations.
4.5 Process Control and Lot control (Type of Control)
The statistical quality control may be achieved either through process control or
through lot control or product control.
There are two broad ways of controlling the quality of a product viz., process control
and product control.
(i) Process Control:- This is concerned with controlling the quality of the goods
manufactured in the process of production. Process control detects whether the production
process is going on in the desired fashion. In other words it controls quality of the goods to
be produced. It ensures that the machines are turning out the product of a requisite
standard. This is largely achieved through control chart devices.
(ii) Product Control:- This is concerned with classification of raw materials or finished
goods into say acceptable, non-acceptable or whether another sample has to be tested. It is
concerned with the inspection of goods already produced. This is largely achieved through
Acceptance sampling.
It is necessary to note that even when the process is under control individual products
may turn out to be non-acceptable. It is also not necessary that product control in the case
of inputs shall ensure that the process is under control. In fact, process control is concerned
with operations, machines and hands while product control is concerned with the quality of
the product turned out. The following chart shows various techniques used for SQC:
 41

 -Chart

Control charts are the principal tools used for process control. Control charts are used
to routinely monitor quality. If a single quality characteristic has been measured or
computed from a sample, the control chart shows the value of the quality characteristic
versus the sample number or versus time. In general, the chart contains a center tine that
represents the mean value for the in-control process. Two other horizontal lines, called the
upper control limit (UCL) and the lower control limit (LCL), are also shown on the chart.
These control limits are chosen so that almost all of the data points will fall within these
limits as long as the process remains in-control.
The figure below illustrates this.

Variations within these control limits are usually attributed to chance and variations
which go outside these control limits ate attributed to assignable causes. Technical action to
these assignable causes is taken whenever variation goes outside these control limits.
 42

Sometimes these out of control variations happen to be in a desirable direction and

then the technical action should be designed to perpetuate the causes which lead to these
variations.
4.6 Statistical Basis of Control Charts
Since a process or a property is out of statistical control when a value is outside the
control limits, quality control requires that the probability for such an event to occur be
small Since a process or a property is out of statistical control when a value is outside the
control limits, quality control requires that the probability for such an event to occur be
small.
Probability Limits
The control limits as pictured in the graph might be .001 probability limits. If so, and if
chance causes alone were present, the probability of a point falling above the upper limit
would be one out of a thousand, and similarly, a point falling below the lower limit would be
one out of a thousand. We would be searching for an assignable cause if a point would fall
outside these limits. Where we put these limits will determine the risk of undertaking such a
search when in reality there is no assignable cause, for variation.
Since two out of a thousand is a very small risk, the 0.001 limits may be said to give
practical assurances that, if a point falls outside these limits, the variation was caused be an
assignable cause. It must be noted that two out of one thousand is a purely arbitrary
number. There is no reason why it could not have been set to one out a hundred or even
larger. The decision would depend on the amount of risk the management of the quality
control program is willing to take. In general (in the world of quality control) it is customary
to use limits that approximate the 0.002 standard.
3sigma limits
The Shewhart control charts make use of 3 limits
In US, whether X is normally distributed or not it is an acceptable practice to base the
control limits upon a multiple of the standard deviation. Upper and lower control lines (UCL
and LCL) are positioned at three standard deviations above and below the mean. Usually this
multiple is 3 and thus the limits are called 3-sigma limits. If T is the statistic under
consideration then jt + 3, is called the upper 3 control limit (UCL) and jr - 3, is called the
lower 3 control limit (LCL). The charts using sigma limits are known as Shewhart Control
chart (Note that in UK, statisticians generally prefer to adhere to probability limits.)
Letting X denote d value of a process characteristic, if the system of chance causes
generates a variation in X that follows the normal distribution, the 0.001 probability limits
will be very close to the 3 limits. From normal tables we glean that the 3 in one direction is
0.00135, or in both directions 0.0027. For normal distributions, therefore, the 3 limits are
the practical equivalent of 0.001 limits. However, there is a small probability that a value
that falls outside d limit is still due to normal variation. This is called Type I error, with the
error being the chance of concluding that there are assignable causes of variation when only
normal variation exists. Another name for this is alpha risk (a) where alpha a refers to the
sum of the probabilities in both tails of the distribution that falls outside the confidence
limits.
 43

Rational Subgroups and Shewhart's Control Chart

The fundamental idea in the use of control charts is the collection of sample data
according to what Shewhart called the rational subgroup concept. They are to be so chosen
that if assignable causes are present, the chance for differences between subgroups will be
maximized. While the chance for differences within a group will be minimized i.e. product
within a subgroup may be supposed to belong to homogenous population and the differences,
if any, corresponding to different sub-group will indicate the presence of systematic variation.
The basis for selection of sub-groups is the time order of production.
Strategies for dealing with out-of control findings
If a data point falls outside the control limits, we assume that the process is probably
out of control and that an investigation is warranted to find and eliminate the cause or
causes. Does this mean that when all points fall within the limits, the process is in control?
Not necessarily. If the plot looks non-random, that is, if the points exhibit some form of
systematic behavior, there is still something wrong. For example, if the first 25 of 30 points
fall above the center line and the last 5 fall below the center line, we would wish to know why
this is so. Statistical methods to detect sequences or nonrandorn patterns can be applied to
the interpretation of control charts. To be sure, "in control" implies that all points are
between the control limits and they form a random pattern.

Self Assessment Questions

Note:
• Please use the space given below to each question to give your answer.
• Please check your response with the Model answers given at the end of this lesson.
I. What is Process control?
------------------------------------------------------------------------------------
------------------------------------------------------------------------------------
------------------------------------------------------------------------------------
--------------------------------------------------------------------------
II. What is Product Control?
------------------------------------------------------------------------------------
------------------------------------------------------------------------------------
------------------------------------------------------------------------------------
--------------------------------------------------------------------------
4.7 Types of Control Charts – Variables and Attribute Control Charts
Control charts are one of the most commonly used tools in statistical process control.
They can be used to measure any characteristic of a product; such as the weight of a cereal
box, the member of chocolate in a box, or the volume of bottled water. The different
characteristics that can be measured by control charts can be divided into two groups:
variables and attributes.
Note: Variables are conventionally ‘quantitative’ or numerical characteristics of
interest. There aren’t always ‘quantitative’ or numerical. The variable ‘gender’ consists
 44

of two text values: ‘male’ and ‘female’. An attribute is a specific value on a variable.
For instance, the variable sex or gender has two attributes: male and female.
A control chart for variables is used to monitor characteristics that can be measured
and have a continuum of values, such as height, weight, or volume. A soft drink bottling
operation is an example of a variable measure, since the amount of liquid in the bottles is
measured and can take on a number of different values. Other examples are the weight of a
bag of sugar, the temperature of a baking oven, or the diameter of plastic tubing.
Variables control charts are X , R,  charts.
A control chart for attributes, on the other hand, is used to monitor characteristics
that have discrete values and can be counted. Often they can be evaluated with a simple yes
or no decision. Examples include color, taste or smell. The monitoring of attributes usually
takes less time than that of variables because a variable needs to be measured (e.g., the
bottle of soft drink contains 15.9 ounces of liquid). An attribute requires only a single
decision, such as yes or no, good or bad, acceptable or unacceptable (e.g., the apple is good
or rotten, the meat is good or stale, the shoes have a defect or do not have a defect, the light
bulb works or it does not work) or counting the number of defects (e.g., the number of broken
cookies in the box, the number of dents in the car, the number of barnacles on the bottom of
a boat).
Statistical process control is used to monitor many different types of variables and
attributes. Attribute is dealt by p charts, c charts and np charts.
Control Chart will be explained in the next two lessons under two main headings:
(a) Control charts for variables.
(b) Control charts for attributes.
The methods of construction, maintenance and interpretation of these charts will be
explained in the subsequent lessons.
Note : We will be using the Shewhart control charts which are based on the idea of
rational sub grouping and use 3 limits.
In the fourth lesson, another topic which we shall discuss n some detail is sampling
inspection. Whenever a large number of pieces of any product have to be inspected 100 per
cant inspection becomes time consuming a costly. Sampling inspection in such a situation
can provide adequate quality assurance with much less cost. Further in situations where the
test is destructive as in testing the resistance of an electric fuse or the life of electric bulb or
the taste of articles of food, sampling is the only alternative available.
Some of the commonsense ideas of sampling, e.g., taking a fixed percentage of the
items to be inspected can have important deficiencies. Statisticians have now developed
sampling inspection plans and also prepared the necessary tables to enable the use of such
plans being made by persons with no specialized knowledge of probability or statistical
theory. Sampling plans have been built up for various degrees of quality assurance and for
various percentages of defective incoming the material.
4.8 Levels of Understanding SQC
In a manufacturing company, government procurement agency or any other
organisation there are four levels for understanding SQC. These are :
 45

(1) Level of understanding mathematics on which control charts and sampling

tables are based and their relationship to many other tools for analysis of data
that have been developed by the mathematical statistician. He should be
familiar with the literature on the subject. One or two persons are sufficient on
the subject for SQC Programme.
(2) The second level is that of general understanding of the principles underlying
the various types of control charts and sampling tables. It calls for
understanding why these methods work, how to interpret the results and how
to decide, which method to use in a particular case.
(3) The third level is that of a broad understanding of the objectives and possible
uses of SQC, even though this understanding is not sufficiently detailed and
precise to permit close supervision of SQC work. This type of understanding is
particularly helpful at higher management levels.
(4) The fourth level calls merely for use of one or more of the techniques on a rule
of Thumb Basis. In any plant in which many applications are made there will
doubtless be a number of inspectors, some machine operators and possible one
or more clerks on this level.
The success of any SQC programme is likely to depend on the number of people in an
organisation who are interested in SQC, and on the distribution of these people among
various departments. The co-operations among departments is necessary for the fullest
benefits from SQC. The more persons in inspection supervision, production supervision,
methods engineering, tool engineering, engineering design and top management who
understand the basic principles of SQC, the better the opportunity for effective use of these
techniques we should follow the Motto of SQC is everybody’s Job.
4.9 Uses of SQC
SQC is an important technique which is used to assess the causes of variation in the
manufactured goods. It helps us to determine whether the quality standards are being met
without inspecting each unit produced in the process. A process is said to be in a state of
statistical control if it is operating in the presence of chance causes only and is free from
assignable causes of variation. The principle advantages of SQC are:
(1) SQC helps is control, maintenance and improvement in the quality standards.
(2) SQC provides a continuous inspection of the product at various stages of the
manufactured process.
(3) SQC eliminates the need of 100 percent inspection of the finished product and
is usually more efficient and less costly than 100 per cent inspection.
(4) SQC reduces waste of time and material to minimum by giving early warning
about the occurrence of defects.
(5) SQC is very simple techniques and can be operated by semi-skilled operators in
the factory.
(6) Rejections by buyers are reduced to nil.
(7) Savings in terms of the factors stated above mean less cost of production and
hence may ultimately lead to more profits.
 46

4.10 Summary
The technique of statistical quality control is of recent origin. It was introduced by W.A.
Shewhart in 1924. A process is said to be in a state of statistical control if it is operating in
the presence of chance causes only and is free from assignable causes of variation. In this
lesson after having defined the terms Quality and SQC, we have discussed the Types of
Variations, viz., Random and Assignable variations, which influence the quality. The
statistical quality control may be achieved either through process control or through lot
control or product control. To gauge that 3 and probability limits are the prerequisites. In
this lesson we have also discussed the scope of SQC in Variable and attributes.
4.11 Further Readings
1. John E. Freund and Frank, I Williams : Modern Business Statistics.
2. Montgomery, D.C. : Introduction to Statistics Quality Control; Wiley.
4.12 Model Questions
1. What is quality and SQC.
2. Describe the types of variation which are of prime focus in SQC.
3. Distinguish between process control and product control.
4. What is the difference between probability limits and 3 limits?
5. Why to use control chart for variables than the control chart for attributes?
6. What is SQC? Discuss its uses.
Suggested Answers SAQ
1(I) A modern definition of quality is that Quality is Inversely proportional to variability.
The smaller the variation the greater is the quality.
1(II) SQC (Statistical quality control) Statistical process control is a collection of tools that
when used together can result in process stability and variability reduction.
2 (I) Process control is concerned with controlling the quality of the goods manufactured in
the process of production
2 (II) Product Control is concerned with classification of raw materials or finished goods as
acceptable, non-acceptable or whether another sample has to be tested.

------
 47

Lesson - 5

CONTROL CHARTS FOR VARIABLES

Structure
5.0 Objectives
5.1 Introduction
5.2 Types of Control Charts
5.3 Control Limits, Specification Limits and Natural Tolerance limits
5.4 Steps in making Control Charts
5.4.1 Control Chart for Mean
5.4.2 Control Chart for Range
5.4.3 Control Chart for S.D.

5.5 Some Facts About X and R Charts

5.6 Examples
5.7 Process Capability
5.8 Summary
5.9 Further Readings
5.10 Model Questions
5.0 Objectives
After going through the lesson you will be able to :
 List two prerequisites for constructing the control charts.
 State the purpose of control charts when standards given or not given.
 State the relation between Control limits, Specification limits and Natural Tolerance
limits.
 List steps in the making of control charts.
5.1 Introduction
The control charts for variables are used in situations where the quality characteristic
which is to be controlled can be measured. Examples are dimensions of a ball-bearing,
hardness of a steel part, tensile strength of a metal rod SQC. per cent of impurities in a
chemical compound, weight of the contents of any container, time in seconds for the blow of
fuse, life in hours of electric bulb.
Since the quality characteristic is continuous variable, such a variable may be
supposed to be normally distributed. Using Shewhart control charts we like to see if the
process is in control, all the sub-groups will have the same  ’s and same  ’s. Since
appropriate statistics for  and  are X (sample mean) and S (sample S.D) and R though
 48

inferior measure of dispersion is simpler and easier to use, so control charts on variables are
commonly based on X and R charts.
For constructing the control charts, we need a fairly good idea of:
(a) The average level of the characteristics to be controlled and
(b) The limit of variations about this average level.
In general, these two quantities have to be estimated from data collected for a period of
time from the process in question.
5.2 Types of Control Charts
The control charts discussed in these chapters will be given for the following two types:
(a) Control charts with standards given. Here we like to find whether the observed
values of the statistics like X and R differ from the respective specified standard
values by amount greater than that can be attributed’ to the chance.
(b) Control charts with no standards given: Here we like to find whether the observed
values of the statistics for given samples vary amongst themselves by amount greater
than that can be attributed to chance.

Control charts are established on the basis of past data and then continued into the
future by extending the average lines as well as the control limits. Periodic samples are then
taken from the process and successive points are plotted on the chart, when a large number
of points go out of control either on high side or on low side, it is an indication that revision
of the control limits is called for. This can happen either due to changes in circumstances
such as changed quality of raw materials or in atmospheric conditions or some other factors.
Due to the elimination of assignable causes the process may shift to a better level of
performance.
5.3 The Relation between Control Limits, Specification Limits and Natural
Tolerance Limits
At the outset it may be noted that there is no connection between the Control limits
and the Specification limits.
Control limits are drawn by the natural variability of the process (measured by the
process Standard deviation  ) i.e. by the Natural tolerance of the process. We have generally
in Shewhart control charts Upper Natural Tolerance Limit (UNTL) as 3  above the mean and
Lower Tolerance Limit (LNTL) as 3  below the mean.
 49

Specification limits are determined externally. They may be set, by the

management, the manufacturing engineers, the customers or by the product developers
/designers.
One should have knowledge of inherent process variability when setting specifications
but there is no mathematical or statistical relationship between control limits and
specification limits.
Tolerance limits define the range of variation that can be permitted from engineering
considerations of a nut that has to fit on a bolt; the screw threads cannot be allowed to differ
by more than a small magnitude. This permissible small magnitude is the tolerance limit.
Control limits on the other hand, represent what the machine is capable of producing. If the
control limits are within the tolerance limits, then all will be well and most of the product will
be found satisfactory. If on the other hand, the control limits are wider than specification
limits, even after the removal of all possible assignable causes, then it is practically certain
that a percentage of the product will be found unsatisfactory. In this situation there are two
alternatives:
(a) Make a radical change in the production process by bringing in new machinery
or using new materials or some such step.
or
(b) Allow the process to continue and then sort out the good product from the bad
and accept only the good product and scrap the bad product.
A process is said to be in a state of statistical control, when all the points on the
control chart lie within control limits. If the process is in a state of statistical control and if
the average quality of the product conforms to average specified, and if further the control
limits are within tolerance limits, then there is no need for acceptance inspection.
X , R chart provide information in three matters, which are required for taking action.
These are
1. Basic variability of the quality characteristics.
2. Consistency of performance.
3. Average level of the quality characteristic.
No production is good enough to produce all items of a product exactly alike. Some
variability is unavoidable. The amount of his variability will depend on various
characteristics of the production process, such as machines, the material and the men. We
have to see whether the basic variability of the process is so great that it is impossible to
make all the products within the specifications limits. When the control chart shows that
this is true and then the specifications cannot be changed, the alternatives are either to
make a fundamental change in the production process that will reduce its basic variability,
or to face the facts that will always be necessary to sort the good product.
Control charts tell us when to leave a process alone as well as when to take action to
correct trouble. The elimination of assignable causes of erratic fluctuations is described as
bringing a process under control and is responsible for many of the cost savings resulting
from statistical quality control.
If the natural tolerance range is narrower than the specified tolerance, and even
though the process is under control, showing a consistent pattern of variability, the product
 50

may be unsatisfactory because the average level of quality characteristic is too low or too
high. This is also disclosed by the control charts.
This is corrected by a change in machine setting or it may need a programme of
Research and development work. Once the control chart shows that a process is brought
under control at a satisfactory level and with satisfactory limits of variability, one may feel,
confident that the product meets specification. This suggests the possibility of basing
acceptance procedures on the control charts, using it to determine whether this happy state
of affairs is continuing or not?
5.4 Steps in making control charts:
After going through the example of metal knobs, it will be desirable to know the steps
and theory in the construction of X , R and  charts and how do we find the UCL and LCL
in each chart for practical purpose.

5.4.1 Control Chart For Mean ( X Charts)

Standards are given
Assuming that all the observations (x) in each sub-group are mutually independent.
We have,
E ( x)  

2 
V ( x)  , x 
n n
If the values of  and  are specified as x and  ' , then the control chart for X will
be given by :
'
LCL  x  3  x  A '
n
CL  x
'
UCL  x  3  x  A '
n
3
Where A  and A depends upon n, and is tabulated
n
If sample mean falls outside of these limits, it implies that process mean is no longer equal to
.
Standards are not given

Let there be m groups each having sample of size n. Let xi be successive subgroups
means. The pooled mean is defined as


m
xi
x i 1
m
 51

Where m = no. of subgroups

n = subgroup size
si : sample S.D.’s and
Ri : Sample ranges
Let s and R be the mean sample standard deviation and mean ranges


m
s
s i 1 i
m


m
Ri
R i 1
m
Then E ( x )  , E  s   c2 ,

E (R)  d2 (values for a normal variables)

where c2 and d2 are functions of n.

We have estimate of  and  given as
  x,
s R
 or  
cn dx
(Two estimates of )
Remark: In case we use estimate of  based on range R we get X — R chart and
incase we use the estimate of  based on s (sample S.D), we use X -s Chart

( x  s) chart
s
LCL  x  3  x  An s
cA n

CL  x
s
UCL  x  3  x  An s
cA n
3
Where An  (Tabulated for different values of n)
CA n

 X  R  chart
R
LCL  x  3  x  A2 R
dx n
 52

CL  x
R
UCL  x  3  x  A2 R
dx n
To draw the control chart

Plot the values of the sample statistic X for all the samples in the control chart. If all
the points fall within the control limits without forming a pattern, the process mean is said to
be under control.
5.4.2 Control Chart For Range
(Process variability may be monitored by plotting values of the sample range R on a control
chart)
There is well known relationship between the range of a sample from a normal
R
distribution and the S.D. of that distribution. The random variable W  is the relative

range.
The parameters of the distribution of W are function of sample size n
E(W )  d2
R
 E    d2
 
 E(R)  d2
R
(Note that E (R)  d2 and estimator of  is  
d2
R
So, estimator of  is  
d2

R
Remark: In case R is not known (or is estimated as R ) then  
d2
Control limits must be based on within sample variability only.
Remark:

When setting up x and R control charts, it is best to begin with the R chart. Because
the control limits on the x chart depends on the process variability unless process variability
is in control, these limits will not have much meaning.
Control Chart For Range (R CHART)
As we have seen sample range is related to process standard deviation
We have
E(R)  dx
 53

And  R  S.D. of R  d2

Where d2 is known function of n

R
Note: we have relative range W 

S.D. of W   w  d2
Hence  R  S.D. of R  d2
When standards are given
Let  ' be the standard value of  . Then the chart will have values as
LCL  d2 ' 3du  D1 '
CL  d2 '
UCL  d2 ' 3du  D2 '
Where
D1  d2  3d2
D2  d2  3d3
So, R - chart is
LCL  D1 '
CL  d2 '
UCL  D2 '
The values of D1, D2, as well as the values of d2 are tabulated for differ-art values of n.
When  is not given, we estimate  R  S.D. of R as :
 R  S.D. of R = d2
R
Estimate of E ( R)  d2  d 2 R
d2
Hence the parameter of R with 3   limits are:

R
LCL  R  3d R  R  3d3
d2

CL  R
R
UCL  R  3d R  R  3d3
d2
If we let
 54

d2
D2  1  3
d2
d2
D2  1  3
d2
So, R - chart is

LCL  RD2

CL  R
UCL  RD2
Where, D3 & D4 are tabulated for different values of n.
5.4.3 Control Chart For S.D.

Although x and R charts are widely used, it is occasionally desirable to estimate the
process standard deviation directly instead of indirectly through the use of the range R. This
leads us to use control charts for x and S where S denotes the sample standard deviation.

Generally x and S charts are preferable to their more familiar counterparts, x and R
charts, when either
(1) The sample size n is moderately large, say n > 10 or 12.
(2) The sample size n is variable.
Standards are given
We know E(s s )   2

 ( x  x) 2
n
yn1 i
Where s 2 
n 1
Now s is not unbiased estimator of 
E(s)  cA

S.D. of s   1  c43
Let  ' be standard value of  , then
LCL  c A ' 3 ' 1  c43  B2 '
CL  c A '

UCL  c A ' 3 ' 1  c43  B4 '

Where

B5  cA  3 1  c42
 55

B6  cA  3 1  c42
Values of B5 and B6 ate tabulated fir various sample sizes
Standards are not given
If no standard is given for ,
s
 by  
c4
Estimate
Then
s
LCL  s  3 1  c42  Ba s
ca

CL  s
s
LCL  s  3 1  c42  Ba s
c4
Where
3
Bi  1  1  c 2i
cd
3
Bi  1  1  c 2i
cd
Values of B3 and B4 are tabulated for various sample sizes

Self Assessment Questions

Please use the space given below to answer the question. And after giving your answers
match your answers with the suggested answers given at the end of the lesson.
1. Broadly, in SQC, Control Charts are divided into how many Types?

2. What is the relationship between Control Limits and Specifications limits?

.

5.5 Some Facts About X and R Charts

It is best to begin with the construction of an R-chart because the limits of X -chart
will depend upon the estimate of the process S.D. and this should properly be made from the
R-chart. If the R-chart shows that the variability of the process is in control, such an
estimate may be derived from the average of the simple ranges. If R-chart shows that the
variability of the process is generally out of control, it is better not to set up an X -chart until
 56

control of process variability has been attained. If however, only one or two points are outside
control limits and if assignable causes can be found for these points, we might venture to set
up limits for the X -chart on the basis of remaining points.
The R pattern is read first and from this it is possible to identify many causes directly.
The X pattern is read in the light of the R-chart, and this makes it possible to identify other
causes. Finally, the X pattern and R pattern are read jointly which gives still further
information.
5.6 Examples
The thickness of a printed circuit board is an important quality parameter. Data on
board thickness (in mm) are given here for 20 samples of 4 boards each.
Sample No. x1 x2 x4 x5
1 6.0 9.0 10.0 15.0
2 10.0 4.0 6.0 11.0
3 7.0 8.0 10.0 5.0
4 8.0 9.0 6.0 13.0
5 9.0 10.0 7.0 13.0
6 12.0 11.0 10.0 10.0
7 16.0 10.0 8.0 9.0
8 7.0 5.0 10.0 4.0
9 9.0 7.0 8.0 12.0
10 15.0 16.0 10.0 13.0
11 8.0 12.0 14.0 16.0
12 6.0 13.0 9.0 11.0
13 16.0 9.0 13.0 15.0
14 7.0 13.0 10.0 12.0
15 11.0 7.0 10.0 16.0
16 15.0 10.0 11.0 14.0
17 9.0 8.0 12.0 10.0
18 15.0 7.0 10.0 11.0
19 8.0 6.0 9.0 12.0
20 14.0 15.0 12.0 16.0
Set up X -R control chart. Is the process in statistical control?
Solution: We first construct the R-chart in order to see whether the process variations are in
control or not. For given data n=4, m=20, R =6.25, X =10.375, D3=0 and D4=2.28
The values of sample ranges Ri and sample means X 1 for i=20 are first calculated as
shown below. The control limits for R-chart are:
 57

LCL  RD2  0

CL  R  6.25
UCL  RDA  14.2628
Range (R) chart

Control Chart

Sigma level 3
Ri 15 11 10 13 13 12 16 10 12 16 16 13 16 13 16 15 12 15 12 16
10 7.75 7.5 9 9.75 10.7 10.7 6.5 9 13.5 12.5 9.75 13.2 10.5 11 12.5 9.75 10.7 8.75 14.2
x1

Since, all the points Ri lie within limits therefore the process variability is in control.

Now we construct ( X - R) chart i.e. X chart using R for estimating  . The control limits are:
LCL  x  A2 R  5.821268

CL  x  10.375

UCL  x  A2 R  14.928732 , Where A2=.729 is tabulated.

Mean ( X ) chart

Control Chart
 58

Sigma level 3
Since all the points lie within control limits therefore, the process is statistically under
control. The following data relates to the life (in hours) of 20 samples of 4 electric bulbs each
drawn at intervals of one hour from a production process.
Sample No. Life in Hours
1 47 32 44 35
2 33 33 34 34
3 34 34 31 34
4 12 21 24 47
5 35 23 38 40
6 19 37 31 27
7 23 45 26 37
8 33 12 29 43
9 25 22 37 33
10 29 32 30 13
11 40 18 30 11
12 21 18 36 34
13 26 35 31 29
14 52 29 21 18
15 26 20 30 20
16 19 1 30 30
17 28 34 39 17
18 29 25 24 30
19 21 37 32 25
20 24 22 16 35
 59

Draw the X - S control charts and comment

Solution:
First we construct the S-chart in order to see whether the process variations are in
control or not.

For given data n = 4, m = 20, s = 8.2594, x =288875, First find sample Standard
deviations Si for all i = 20
Si = 7.141428, 0.57735, 1.5, 14.89966, 7.615773, 7.549834, 10.14479, 12.91962, 6.946222,
8.75595, 12.84199, 9.069179, 3.774917, 15.38397, 4.898979, 13.68698, 9.469248,
2.94392, 7.135592, 7.932003
The control limits for s-chart are :

LCL  BA s  0

CL  s  8.2594
LCL  BA s  18.7161
Control chart for S

Control Chart

Sigma level: 3
Since, all points (Si) lie within limits therefore the variability is in control.

Now we construct ( x  S ) chart.

The control limits are: LCL  x  A2 s  15.4404 ; CL  x  28.8875

UCL  x  A2 s  42.3346
 60

Mean Chart ( X )

Control Chart

Sigma level: 3
Since all the points lie within control limits therefore, the process is statistically under
control.
Note: 1. If any point goes out of control one should get the revised control chart by
discarding out of control points and getting revised control limits.
2. Generally samples of size 4 or 5 are taken more frequently than larger samples as the
cost involved is less and the chance of a change occurring during the taking of the sample is
also small.
5.7 Process Capability
So far we have discussed ways of monitoring the production process to ensure that it
is in a state of control and that there are no assignable causes of variation. A critical aspect of
statistical quality control is process capability. To be capable of producing an acceptable
product, the process variation cannot exceed the preset specifications. Process capability
involves evaluating process variability relative to preset product specifications order to
determine whether the process is capable of producing an acceptable product.
Process capability is measured by the process capability index,O Cp, which is
computed as the ratio of the specification width to the width of the process variability:
USL  LSL Specification width
C pcs  
6 Pr ocess width
Where the specification width is the difference between the upper specification limit
(USL) and the lower specification limit (LSL) of the process.
There are three possible ranges of values for Cp that also help us interpret its value;
 61

Cp =1: A value of Cp equal to 1 means that the process variability just meets specifications.
We would then say that the process is minimally capable.
Cp < 1: A value of Cp below 1 means that the process variability is outside the range of
specification. This means that the process is not capable of producing within specification and
the process must be improved.
Cp > 1; A value of Cp above 1 means that the process variability is tighter than specifications
and the process exceeds minima! capability.

The Cp value greater than one.

A Cp value of 1 means that 99.74 percent of the products produced will fall within the
specification limits. This also means that .26 percent (100% - 99.74%) of the products will
not be acceptable. Although this percentage sounds very small, when we think of it in terms
of parts per million (ppm) we can see that it can still result in a lot of defects. The
number .26 percent corresponds to 2600 parts per million (ppm) defective (0.0026 x
1,000,000). That number can seem very high if we think of it in terms of 2600 wrong
prescriptions out of a million, or 2600 incorrect medical procedures out of a million, or even
2600 malfunctioning aircraft out of a million. We can see that this number of defects is still
high. The way to reduce the ppm defective is to increase process capability.
5.8 Summary
Control charts are the principal tools used for process control. These charts are used
in situations where the quality characteristic which is to be controlled can be measured. So
in this lesson we have discussed the different control charts for variables i.e. X -chart, R -
chart,  -chart for controlling the process.
5.9 Further Readings
1. John E. Freund and Frank, I Williams: Modern Business Statistics.
2. Montgomery, D.C. (2001) : Introduction to Statistical Quality Control; Wiley.
5.10 Model Question
A quality control inspector at the Cocoa Fizz soft drink company has taken twenty-
five samples with four observations each of the volume of bottles filled. The data and the
computed means are shown in the table. If the standard deviation of the bottling operation is
 62

0.14 ounces, use this information to develop control limits of three standard deviations for
the bottling operation.
Observations
Sample (bottle volume In ounces) Average Range
Number 1 2 3 4
1 15.85 16.02 15.83 15.93 15.91. 0.19
2 16.12 16.00 15.85 16.01 15.99. 0.27
3 16.00 15.91 15.94 15.83 15.92 0.17
4 16.20 15.85 15.74 15.93 15.93 0.46
5 15.74 15.86 16.21 16.10 15.98 0.47
6 15.94 16.01 16.14 16.03 16.03 0.20
7 15.75 16.21 16.01 15.86 15.96 0.46
8 15.82 15.94 16.02 15.94 15.93 0.20
9 16.04 15.98 15.83 15.98 15.96 0.21
10 15.64 15.86 15.94 15.89 15.83 0.30
11 16.11 16.00 16.01 15.82 15.99 0.29
12 15.72 15:85 16.12 16.15 15.96 0.43
13 15.85 15.76 15.74 15.98 15.83 0.24
14 15.73 15.84 15.96 16.10 15.91 0.37
15 16.20 16.01 16.10 15.89 16.05 0.31
16 16.12 16.08 15.83 15.94 15.99 0.29
17 16.01 15.93 15.81 15.68 15.86 0.33
18 15.78 16.04 16.11 16.12 16.01 0.34
19 15.84 15.92 16.05 16.12 15.98 0.28
20 15.92 16.09 16.12 15.93 16.02 0.20
21 16.11 16.02 16.00 15.88 16.00 0.23
22 15.98 15.82 15.89 15.89 15.90 0.16
23 16.05 15.73 15.73 15.93 1 586 0.32
24 16.01 16.01 15.89 15.86 15.94 0.15
25 16.08 15.78 15.92 15.98 15.94 0.30
Total 398.75 7.17
Suggested Answers to CYP
1. Control Charts are divided into how many Types; Control Charts with stand
Control Charts with standards are Not Given'.
2. There is no relationship between Control Limits and Specifications limits.
 63

Lesson-6

CONTROL CHARTS FOR ATTRIBUTES

Structure
6.0 Objectives
6.1 Introduction
6.2 Control Chart for Number of Non-Conforming (Defectives)
6.3 Control Chart for Fraction Defective
6.3.1 Example
6.4 Control Charts for Number of Defects (c-chart)
6.4.1 Example
6.5 Control Charts for Number of Defects per Units (u-chart)
6.5.1 Example
6.6 Summary
6.7 Further Readings
6.8 Model Questions
6.0 Objectives
After going through the lesson you will be able to:
 List at least three limitations in the control charts for mean, range and S.D.
 List at least situations where control charts for attributes could be used.
 Discuss at least two types of Control charts for attributes.
 Construct Control Chart for Number of Non-Conforming.
 Construct Control Chart for Fraction Defective.
 Construct Control charts for number of defects.
 Construct Control charts for number of defects per units.
6.1 Introduction
In the last lesson we have discussed control charts for mean, range and S.D. These
charts have the following limitations.

The control chart for variables (i.e. x & R ) can be used for quality characteristics
which can be measured and expressed in numbers. But if sometimes these characteristics
are too large (say 10,000), then it is not possible to have 10,000 control charts. Then we have
control chart for attributes which says each quality characteristic is an attribute such that
each item is recorded as either defective (non-confirming) or non-defective. To see whether
the process is in control we have to see whether the population fraction defective P is same
 64

for all sub-groups. For this we can have judgements based control charts for fraction
d
defective P  in the sample or the control chart for number of defectives say d, in the
n
sample, where n, as before denotes the number of items inspected per sub-group.
Control Chart for Attributes
An alterative to x -charts or R-charts, we can have control charts for attributes. These
control charts have to be used
(1) When a quality or characteristic is not capable of quantitative measurement.
(2) When a quantity or characteristic can be measured quantitatively but this is
not done and the data are studied on in basis of totality of the attributes.
Control charts for attributes are.
(1) Control charts for number of defectives (np-charts)
(2) Control charts for fraction defectives (p-charts)
(3) Control charts for number of defects (c-chart)
(4) Control charts for number of defects per units (u-chart)
6.2 Control Chart for Number of Non-Conforming (Defectives) d or np CHART
Assume that each random sample is taken with replacements or, even without
replacement, is taken from a practically infinite population; we may suppose that d=np
(number of non-confirming units) is distributed in the binomial form with
E (np)  nP
and np  nP(1- P)
 P(d  x)  (nx ) p x (1- p)n-x x  0,1, 2 n
 
 E (d )  np and V (d )  np(1- p) 
p being the same for all sub-groups if the process is in control
Standards are given:
Hence, if p' be the specified standard value of P. the control chart will be

LCL  np ' 3 np '(1  p ')

CL  np '
UCL  np ' 3 np '(1  p ')
Standards are not given:
Estimate P from the samples themselves, the appropriate estimate of P is the mean fraction
defective i.e. P is given by
m m
pi d
P  i
i 1 m i 1 mn
 65

di
Where pi  : i  1,..., m
n
and
m = no, of subgroups (samples) each of size n.
The lines on the control chart will then be
  
LCL  n P  3 n P(1  P)

CL  n P
  
UCL  n P  3 n P(1  P)
Note : np can never be negative so, if LCL comes out to be negative it is to be taken as zero.
6.3 Control Chart for Fraction Defective (P-Chart)
If we construct a control chart for p instead of np, here we have
E (p) =P

p(1  p)
and  P 
n
Standards are given :
if p’ be the specified standard value of P, the control chart for P will be

P '(1  P ')
LCP  p ' 3  P ' A P '(1  P ')
n
CL  P '
P '(1  P ')
UCL  P ' 3  P ' A P '(1  P ')
n
3
Where is A  tabulated for different values of n.
n
Standards are not given :
 np P1 np D1
Estimate P by P where in. P   .Then
m mn
LCL  P  3 P (1  P )
CL  P
UCL  p  3 p(1  p)  P  A np(1  P )
Note: Since p can never be negative so; if LCL comes out to be negative, its to be taken as
zero.
 66

6.3.1 Example
1. The data that follows give the number of nonconforming bearing and seal assemblies
in samples of size 100. Construct no of nonconforming and fraction nonconforming
control charts for these data.
Sample Number of Non-Conforming Asses
1 7
2 4
3 1
4 3
5 6
6 8
7 10
8 5
9 2
10 7
11 6
12 15
13 0
14 9
15 5
16 1
17 4
18 5
19 7
20 12
Solution
np-chart
For given data n = 100, m 20, P = 0,0537 where Pi are first calculated and shown below.
p 0.07 0.04 0.01 0.03 0.06 0.08 0.10 0.05 0.02 0.87 0.06 0.15 0 0.09 0.05 0.01 0.04 0.05 0.07 0.12

The lines on the control chart will then be

 
LCL  n p  3 n p1  p   0 (Since LCL, cannot be negative)

LCL  n p  5.37
 67

 
 
UCL  n p 3 n p1  p   12.13
 

np chart

Control chart

Since all the points lie within limits the process is under statistical control
P-chart
The lines on the control chart will then be

LCL  p  A np(1  p )  0 (Since LCL cannot be negative)

CL = p = 0.0537

UCL  p  A np(1  p)  0.1213

 68

Sigma level 3
Since, all the points lie within limits the process is under statistical control
P chart in case of varying sample size
In cane the sample size is constant, it is immaterial whether one uses the np-chart or
the p-chart If, however, the sample size varies, in the np-chart all three lines will vary with n
and the resulting chart will be highly confusing whereas in the p-chart the central line will be
invariant It is, therefore simpler and preferable to use the p-chart (or 100p-chart ) in case the
sample size varies.
Instead of computing control limits for each sample size separately, two sets of limits
may be computed based on the minimum and the maximum sample sizes. Action need not
be, taken for points lying within the inner set of limits, while action must be taken for points
lying beyond the outer limits. For other points, action should be based on exact control limits.
The confusion in a p-chart (or np-chart or 100p-chart) with varying control limits can
be avoided with some additional computation. For that, instead of plotting p in the control
chart, one should plot the standardized values, viz.
 69

p  p' p p
x or
p '(1  p ') ___________

n p(1  p)
n
According as the standard value for p is specified or not, P being the weighted mean
of sample proportions with the sample sizes as weights. The central line as well as the control
limits becomes invariant with n, since obviously here
LCL = -3; CL = o and UCL = 3
Self Assessment Questions
Please use the space given below to answer the question. And after giving your answers
match your answers with the suggested answers given at the end of the lesson.
1. Which Control Chart could be used when quality is not capable of quantitative
measurement?

2. What is P-Chart?

6.4 Control Chart for Number of Defects (for Non-Conformities)

C-Chart
A defective is an item that fails to fulfill one or more of the given specifications; a
defect is any instance of the item’s lack of conformity to specifications. Every defective item
thus contains one or more defects. E.g. weak spots in a given length of fibre or wire, some
defects in a piece of cloth, loose screws in a machine, surface defects in a piece of paper, no.
of broken rivets in an aircraft wing or so on.
In many manufactured items, the opportunities for defects to occur are numerous,
even though the possibilities for defect to occur in a particular spot are negligible. So, if n is
large, then p is finite, np is finite. Hence, the number of defects (c) may in most cases be
supposed to be distributed in the Poisson form, with parameter  .
A control chart for c will then detect any difference that may exist among .the Poisson
distributions for different sub-groups or in other words among the  - values for the sub-groups.

e   x
P( x)  ; x  0.1, 
x1
where x = no of nonconformities and)  > 0 is the parameter for Poisson distribution
Standards are given :
We know for a Poisson variable c with parameter 

E(c)  ; a,  
If C be the specified standard value of  then the control chart for c will be based on
 70

LCL  c  3 c '
CL  c '
UCL  c ' 3 c '
Standards are not given :
Estimate  from the observed c values. Let ci is the c - value for the sample taken from the
1th subgroup (i = 1, ..., m). Then appropriate value of  will be


nt
c
c iz1 i
m
Substituting c for c’ in the above.
We get the control lines for c-chart as:
 
LCL  c 3 c

CL  c
 
UCL  c 3 c
Note :
1. c cannot be negative, so if LCL comes out to be negative, it is to be taken as
zero.
2. The above formulae relate to the sample of constant size from all sub-groups.
Example 6.4.1 The following table gives the results of inspection of 100 yards pieces of
wooden goods.
Piece No.
No. of Defects
1 3
2 6
3 0
4 1
5 3
6 7
7 8
8 4
9 5
10 2
11 1
12 5
 71

13 2
14 3
15 0
. 16 1
17 4
18 3
19 1
Draw the appropriate control chart and comment
Solution
For given data the suitable control chart is c-chart
We get the control lines for c-chart as:

LCL  c  3 c  0 (since LCL cannot be negative)

CL  c  3.1053

UCL  c  3 c  8.3918

Sigma level 3
 72

Since, all the points (defects) lie within limits the process is under statistical control
6.5 Control Chart for Non-Conformities Per Unit
U-Chart
When the occurrence of defects may vary from sub-group to sub-group the
conventional c-chart showing only the number of defects is not applicable.
Then we have control chart for non-conformities per unit i.e. u-chart. For example,
inspection of rolls of cloth or paper leads to a situation when the size of the sample varies as
all rolls are not of same width.
c
u
n
E (c) 
E (u)    n
n n
V (c)  m
V (u)   2
n2 n n
n
u 
n
Standards are given:
Let  ' be given value of n
'
LCL   ' 3
nt
CL   '

'
UCL   ' 3
nt
Standards are NOT given:

Estimate u by u


mi
c Total defects found
Let u  ci i


mn
n Total units inspected
cni i

Then the u-chart is

u
LCL  u ' 3
ni
CL   '
 73

u
LCL  u ' 3
ni
6.5.1 Example It was found that when a manufacturing process is under control the
average number of defectives per sample batch of 10 is 12. What limits would you set in a
quality control chart based on the examination of defectives of sample batches of 10?
1.2
Solution. Preparation of defectives p  1
10
Standard error of defectives

 npq  10  .12  .88

 1.056  1.03
Usually 2 and 3 limits are marked on the control chart- So in the present case.
Outer control limits for number of
Defectives = Average No. of defectives ± 3 standard error.
= 1.2 ± 3 x 1.03
= 1.2 ± 3.09
= 4.29 and 0.
inner control limits
= average number of defective ± 2 x standard error
= 1.2 ± 2 x 1.03
= 1.2 ± 2.06
= 3.26 and 0.
6.6 Summary
The x , R and S charts, which we discussed in the least lesson, could be used only
when a quantitative expression for quality characteristic is done. And it is difficult to identify
the quality of a product by a single variable only. For example a steel pipes' quality can be
identified in terms of its thickness, length, breadth, diameter tensile strength etc. In such a
case, we may have to make a large number of control charts for each one of the variables.
Then it becomes, cumbersome time consuming and expensive. The control charts for
attributes could be used when a quality or characteristic is not capable of quantitative
measurement OR when a quantity or characteristic can be measured quantitatively but this
is not done and the data are studied on in basis of totality of the attributes. In this lesson we
have discussed the role and construction of some of the most prevalent Control charts for
attributes such as for fraction defectives (p-charts), for number of defectives (np-charts)
6.7 Further Readings
1. John E. Freund and Frank, I Williams: Modem Business Statistics,
2. Montgomery, D.C. ; Introduction to Statistical Quality Control; Wiley.
 74

6.8 Model Questions

Qn.1. In a factory producing spark plugs, the number of defectives found in inspection of 20
lots of 100 are given below. Construct an appropriate chart.
Lot No. No. of defective Fraction defective
1 5 0.05
2 10 0.10
3 12 0.12
4 8 0.08
5 6 0.06
6 4 0.04
7 6 0.06
8 3 0.03
9 5 0.05
10 3 0.03
11 4 0.04
12 7 0.07
13 8 0.08
14 2 0.02
15 3 0.03
16 4 0.04
17 5 0.05
18 8 0.08
19 6 0.06
20 10 0.10
Total 120

Qn.2. Twenty pieces of cloth out of different rolls contained respectively 1,4,3,2,5,4,6, 7,
2,3,2,5,7,6,4, 5,2, 1,3 and 8; thread counts. Ascertain if the process is in control.
(draw c- chart).
Qn.3. Average population of defectives in first 10 samples of size 150 each was observed to
the 0.04. What are the 1 in 1000 control limits? If later on it is noticed that the
machine is only producing 2% defective items, what are the revised control limits.
Qn.4. It was found that when a manufacturing process is under control the average number
of defectives per sample batch of 10 is 1.2. What limits would you set in a quality
control chart based on the examination of defectives of sample batches of 10?
 75

Qn.5. After the chart was set up the process was observed to be well under control for a long
period with an average of 4% and it was decided to reduce inspection for control. A
sample of size 25 was decided upon. Obtain the control limit for the revised chart. At a
later date, the control chart gave the following number of defectives per sample over
the last 0 samples, 0, 0, 2, 3, 0, 2, 1, 1, 0, 0, 1, 2, 1, 1, 1, 2, 4, 4, 3, 3. The quantity
control in charge decided on the basis of data that the process has to be checked. Is
the decision justified?
Suggested Answers to SAQ
1. Control Charts for Attributes
2. Control Charts for number of defects per unit.

------
 76

Lesson-7

PRODUCT CONTROL
Structure
7.0 Objectives
7.1 Introduction
7.2 Acceptance Sampling
7.3 AQL and LTPD
7.4 Producers Risk and Consumer Risk
7.5 Sampling Plans
7.6 OC Curves
7.7 Average outgoing Quality
7.8 Summary
7.9 Further Readings
7.10 Model Questions
7.0 Objectives
After going through the lesson you will be able to:
 Define acceptance sampling.
 How acceptance sampling is different from statistical process control.
 Define Acceptable quality level (AQL).
 Distinguish between AQL and Lot tolerance percent defective.
 Define Producers Risk and Consumer Risk.
 Evaluate how well sampling plans discriminate between good and bad lots.
7.1 Introduction
In the last lesson we had discussed the construction of Control Chart for Number of
Non Conforming, construction of Control Chart for Fraction Defective, and construction of
Control charts for number of defects. The object of product control is to decide whether to
accept or reject a lot produced for marketing purpose. The customer who makes use of the
products of any industry is faced with the problem of ascertaining whether the product
confirms to specifications. For this purpose some form of inspection is necessary. Earlier
data was collected as process was going on. We analysed and interpreted this data in order to
decide either to continue production as the process was running or to stop production and
take corrective action. In this lesson, we shall focus our attention on making an ‘after the
fact’ decision on materials submitted for acceptance. This type of judgment is necessary
when materials are received from some preceding Sub process or from an outside vendor as
well as when we are deciding whether to ship the product to customer or to scrap it. The
purpose of inspection is to distinguish good material from the bad. This is done, by a
procedure, known as acceptance sampling.
 77

7.2 Acceptance Sampling

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.
Acceptance sampling before the process involves sampling materials received from a supplier,
such as randomly inspecting crates of fruit that will be used in a restaurant, boxes of glass
dishes that will be sold in a department store, or metal castings that will be used in a
machine shop. Sampling after the process involves sampling finished items that are to be
shipped either to a customer or to a distribution center. 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. You may be
wondering why we would only inspect items in the lot and not the entire lot. 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. Another example of when acceptance sampling would be used is in
destructive testing, such as testing eggs for salmonella or vehicles for crash testing.
Obviously, in these cases it would not be helpful to test every item! However, 100 percent
inspection does make sense if the cost of inspecting an item is less than the cost of passing
on a defective item.
Acceptance sampling can be used for both attribute and variable measures, though it
is most commonly used for attributes.
The acceptance sampling has become an integral part of the product control. The
objective of product control may also be achieved by inspecting each and every item in a lot
and rejecting those items which do not conform to set standards. This type of 100 percent
inspection is neither profitable nor practicable. Incase testing is destructive, 100 percent
inspection is feasible. Sometime 100 per cent inspection does not always mean 100 per cent
assurance because neither men nor machines are infallible. This method is expensive and
time consuming. Therefore it has become a necessity to employ acceptance sampling to
decide whether to accept or reject a lot on the basis of evidence, provided the samples are
drawn at random.
Sampling inspection has a number of psychological advantages over cent per cent
inspection. It is a common knowledge that sometime, even 100 per cent inspection will not
eliminate all of the defective products from a stream of product a portion of which is
defective. The best protection against the acceptance of defective product is, of course, having
the product made right in the first place. Good sampling acceptance procedures may often
contribute to this objective through more effective pressure for quality improvement than can
be exerted with 100 per cent inspection.
Some sampling schemes also provide a better basis for diagnosis of quality troubles
than is common with 100 per cent inspection. It may be pointed out that whenever a portion
of the stream of products submitted for acceptance is defective, some defective items are
likely to be passed by any sampling acceptance scheme. The statistical approach to
acceptance sampling frankly faces this fact. It attempts to evaluate the risks assumed with
alternative sampling procedures and to make a decision as to the degree of protection needed
 78

in any instance. It is then possible to choose a sampling acceptance scheme that gives a
desired degree of protection with due consideration for the various costs involved.
Inspection may be visual (e.g. for surface defects, or cracks or be below holes in casting)
or by means of a guage (e.g. diameter of a metal rod) or by means of measurements e.g.
percentage of carbon in steel, length of life of an electric bulb). When the result of inspection
is one or more measurements, we have ‘inspection of variable’ and when the result of
inspection is merely a classification of the product into good or bad (acceptable or not
acceptable we have inspection of attributes’.
Again NOTE: Acceptance sampling can be used for both attribute and variable
measures, though it is most commonly used for attributes
Any quantify of manufactured product or other material presented for inspection for
conformance to given specifications will be called a ‘LOT’. It is customary to make pieces of
lot as homogeneous as possible by having separate lots for the products of different days
different raw materials, different machines etc. When we speak of sampling from a lot, it will
be understood what a random sample is meant.
Acceptance sampling is done with the help of sampling plans.
A ‘sampling plan’ is a procedure for taking a sample from a given lot and of disposing
off the lot on the basis of the results or inspection of the pieces in the sample. It usually
states one or more sample sizes together with corresponding acceptance numbers.
The plans for selecting samples depend on the following four factors—
(1) Producers risk (chance of rejection of a good lot) is small.
(2) The consumers risk (chance or accepting a bad lot) is small.
(3) Average quality of goods sent out of the factory may not be worse than some
specification
(4) The amount of Inspection is minimum.
7.3 Acceptable Quality Level (AQL) and Lot Tolerance Percent Defective (LTPD)
There are 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. The AQL of
a sampling plan is a level of quality routinely accepted by the sampling plan. AQL is generally
defined as that level of quality (percent defective, defects per hundred units, etc.) that the
sampling plan will accept 95% of the time. This means lots at or better than the AQL are
accepted at least 95% of the time.
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. The LTPD of a
sampling plan is the level of quality routinely rejected by the sampling plan. It is generally
defined as the percent defective (number of defectives per hundred units X 100%) that the
sampling plan will reject 90% of the time. This is also the percent defective that will be
accepted by the sampling plan at most 10% of the time. In other words it means that lots at
or worse than the LTPD are rejected at least 90% of the time and accepted at most 10% of the
time.
 79

7.4 Producers Risk and Consumer Risk

For an acceptance procedure there are always two parties. These parties are producers
and consumers or sellers and buyers respectively. The interest of two parties are conflicting.
The producers want protection against the rejection of a Good lot and a buyers want
protection against acceptance of a bad lot. As such sampling acceptance plan has a great
importance for the both. The sampling plan which is selected must be such that both the
producer’s as well as consumer’s interests are protected.
Producer’s Risk constitutes the rejection of a lot which is of acceptable quality and
the consumers risk is of accepting a lot which is of sub-standard quality. The producers risk
is denoted by  and consumers risk by . Thus the probability of rejecting a lot of good
quality is  and the probability of accepting a lot of bad quality is .
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 a alpha.
7.5 Sampling plans
We will look at the different types of sampling plans and at the ways to evaluate how
well sampling plans discriminate between good and bad lots
A common procedure in acceptance sampling is to consider each submitted lot of a
product separately and to base the decision on acceptance or rejection of lot on the evidence
of one or more samples chosen at random from the lot. If we make the decision on the
evidence of only one sample or two samples or three samples etc. the sampling plans are
called single, double or triple, as the case may be.
7.5.1 Single sampling plan
In a single sampling plan, a sample of size n is taken from the lot presented for
inspection and the number of detectives in the sample is found out by inspection of all pieces.
If this number happens to be equal to or less than the specified acceptance number c, then
the lot is accepted. Otherwise, the lot is rejected. Rejected lots have generally to be
completely accepted by replacing all the defective pieces by non-defective pieces.
N is number of items in lot.
n is size of sample drawn.
c is acceptance number. It is the maximum allowable number of defective pieces
in the sample. More than this will reject the lot.
i.e., if N =200. n = 20, c = 1.
We say, of a sample of 20 out of 200 contains more than I defective piece, reject the
lot, otherwise accept it.
 80

Flow chart for Single sampling scheme

Inspect a Sample of size n

If the number of defectives

Does not exceed c exceeds c

Accept the lot, replace the Reject the lot, inspect 100%,
defectives in the sample by replace the defectives by
non-defectives non-defectives

7.5.2 Double Sampling Plan

In a double sampling plan, we have two sample sizes n 1 and n2, and two
corresponding acceptance numbers c1 and c2. To begin with a sample of size n1 is taken from
the lot. This sample is inspected and the number of defectives in it is found. If this number is
less than or equal to acceptance number c1 then the lot is accepted. If on the other hand, the
number of detectives in the first sample is more than the acceptance number c2, then the lot
is rejected and/or completely inspected. If it happens that the number of defectives in the
first sample is more than c1 but less than or equal to c2, then a second sample of size n2 is
taken from the same lot. All the pieces in this second sample are then inspected and the total
number of defectives in the first and second samples taken together is found out. If this
number is less than or equal to the acceptance number c 2, that the lot is accepted, if not, the
lot is rejected.
We put it in notational form and explain with the help of an example.
n1 = first sample size
n2, =second sample size.
c1 =acceptance number in the first sample.
c2 = acceptance number for n1 + n2
We Interpret it as follows.
(i) If c1 is not exceeded in the first sample, accept the lot.
(ii) If c2 is exceeded in the first sample, reject this lot.
(iii) If c1 is exceeded in the first sample but c2 is not, a second sample, n2 is taken.
 81

(iv) The lot is accepted if c2 is not exceeded in the combined sample n1 + n2 and
rejected if c2 is exceeded.
The flow chart for double sampling plan is given below.
Flow chart
Inspect a sample of size n1

If the number of defectives in first sample

----------------------------------------------------------------------------------------

Does not exceed c1 Exceeds c1 but not c2 exceeds c2

Inspect a second sample of size n2

If the number of defectives in the combined samples

-----------------------------------------------------------------------

Does not exceed c2 exceeds c2

Accept the lot, replace the defectives found Reject the lot, inspect 100% and replace
in the sample by non-defectives defectives by non-defectives

Example : Suppose a lot contains 600 items and is ready for shipment. But the
producer wants to confirm of the lot. For this he adopts the following plan.
n1 = 50; c1 = 2
n2 = 75, c2 = 5
On the basis of this plan; the decision to accept or reject the lot for shipment depend
upon:
(i) If the first sample contains 2 or less defectives, the lot would be accepted.
(ii) If the first sample contains more than 5 defective items, the lot would be rejected.
(iii) If the first sample contains more than 2 but less than 5 defective items then second
sample would be tested.
(iv) The lot is accepted if the number of defective items is less than 5 and rejected if it is
more than 5 when the two samples are combined.
 82

The second sample is most frequently required when we deal both lots of marginal
quality. The main advantage, of double sampling plans is their strong Psychological
advantage in giving lots or marginal quality a second chance. This feature has a strong
appeal to partical minded production men.
7.5.3 Multiple / sequential sampling plan
In this manner, we can have sampling in three stages or any number of stages. Such
sampling plans are known as 'Sequential Sampling plans' or multiple sampling plans. In
these plans, we arrive at a decision to accept or reject after inspecting more than two
samples. We can continue sampling unless we reach at final decision, in this plan, units are
inspected one at a time and after examining each unit, we take, one or the following two
decisions.
(a) Accept or reject the lot.
(b) Continue sampling by taking another sample from the lot.
It can be shown that the inspecting a large number of lots, the total amount of
inspection required is generally less in double sampling than in single sampling. This is the
reason for the use of double sampling plans, even though they lead to complications in
organisation and supervision. Double 'sampling plans have a psychological advantage
because we are giving, so to, say a second chance to lots which fail to be accepted on the
basis of first sample. Sampling plans with three or more stages are seldom used in practice
except in situations where inspection is a very costly process, and hence even small
economies in the total number sampled become important.
A multiple sampling plan is illustrated below.
Sample Sample Size Combined Sample
Number
Size Acceptance Rejection
Number Number
1 30 30 1 3
2 30 60 3 6
3 30 90 5 7
4 30 120 6 8
5 30 150 7 9
6 30 180 8 9

From the above table, it is evident that the first five samples fall in the doubtful
category and the final decision can be made only on the basis of the last sample, i.e. for the
first four samples combined the acceptance number is 6. Whereas the rejection number is 8.
But it is possible that the number of defective articles may be 7, When this happens we can
not decide about the lot. Therefore there arises a need of selecting one more sample at
random and find decision would be made on the basis of results of all the 6 samples
combined. Now the lot would he accepted if the number of defective item is '8 or less whereas
the lot would be rejected if the number of defective items exceeds 9 or is 9.
 83

Self Assessment Questions

Please use the space given below to answer the question. And after giving your answers
match your answers with the suggested answers given at the end of the lesson.
1. What is Time Reversal Test?
-------------------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------
2. Besides formula error, which other errors could creep in the Index numbers?
-------------------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------

7.6 The Operating Characteristic (O.C.) Curve

Corresponding to any given sampling plan, there is an operating characteristic curve
which shows for each value of the process average the probability of accepting the lots on the
basis of given sampling plan. 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.
Example of an Operating Characteristic (O.C.) Curve
 84

The above 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.” They axis shows the probability or
chance of accepting a lot.
Regardless of which sampling plan we have selected, the plan is not perfect. That is,
there is still a chance of a Iots that are “bad” and rejecting “good” lots. The steeper the OC
curve, the better our sampling plan is for discriminating between “good” and “bad.” Figure
below shows three different OC curves, A, B, and C. Curve A is the most discriminating and
curve C the least. You 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.

Shape of an ideal OC. Curve

Ideal curve would be perfectly perpendicular from 0 to 100% for a given fraction
defective.
 85

The Ideal OC curves would be one for which all good lots are accepted and all bad lots
are rejected. No sampling plane can have an OC curve of this type. The degree to which an
actual OC curve approximates the ideal curve depends upon it and e, where n represent
sample size and a, the acceptance number i.e. the number of
 We know from own knowledge that Binomial Distribution is
P( X  X )  Ncx p x qn x defectives in the sample which is not to be exceeded.
Two Ways of Calculating OC Curves
— Binomial Distribution
— Poisson formula
Binomial Distribution Method
We take the following example.
A buyer and seller agree to use a sampling plant with sample size n =0 and acceptance
number c = 1. What Is the probability that a buyer would accept a lot having a fraction
defective
(a) p = 0.1 (b) p = 0.3 (c) p = 0.5 (d) p =
0 (e) p = 1.
So construct the OC curve for this plan.
Solution. First, we compute the probabilities of the sampling plan with the help of
Binomial distribution.
P( X  X )  Ncx p x q N  x
P (Accept/ P = .1) = 0.919, P (Accept / P = .3) = 0.528
P (Accept/ P= .5) = 0.188, P (Accept/P = 0) = 1, P (Accept/ P= 1) = 0
Plot of these probabilities against values of p gives the oc curve.
 86

OC – Curve

Poisson Distribution. Can be used in situations where the fraction motion defective (P)
is less than 0.1 and in lot size is at least 10 times the sample size. In most industrial
situations, the Poisson Distribution can be used.
 As n is larger and p is smaller for small sample sizes ñ>20 and p <= 0.05 Poisson can
be used.
 This would make calculation fairly easy however a summation of defects from A=0 to
the number of defects in the sample size is needed to get the probability of acceptance.
 Using Poisson equation makes calculating OC curves very difficult and repetitive. If
one uses a Poisson table we can make these curves much easier.
Calculation of OC Curve
 Find your sample size, n
 Find your fraction defect p
 Multiply n”p
 A=d
 From a Poisson table find your PA
 N = 1000
 n = 60
 p = .01
 A=3
Find PA for p = .01, .02, .05, .07, .1, and .12?
Probabilities are shown below.
n * p = 60 * .01 = .6 n * p = 60 * .02 = 1.2
A=d=3 A=d=3
 87

PA = 99.8% PA = 87.9
n * p = 60 * .05 = 3 n * p = 60 * .07 = 4.2
A=d=3 A=d=3
PA = 64.7 PA = 39.5
n * p = 60 * .1 = 6
A=d=3
PA = 15.1
n * p = 60 * .12 = 7.2
A=d=3
PA = 7.2
Plot of PA on Y- axis and p on x-axis gives the O.C curve
7.7 Average Outgoing Quality
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 approximately computed as follows:
AOQ (Pac)p,
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
Illustration
p .05 .10 .15 .20 .25 .30 .35 .40 .45 .50
Pac .9974 .9185 .8352 .7373 .6328 .5282 .4284 .3370 .2562 .1875
AOQ .0499 .0919 .1253 .1475 .1582 .1585 .1499 .1348 .1153 .0938
The following curve shows the shape of an AOQ curve.
 88

AOQ curve shows the relationship between Incoming lot quality p and AOQ. It is a
function of the incoming lot quality ‘p’. The maximum value of AOQ subject to variations in
‘p’ is known as average outgoing qualify limit (AOQL).
ASN (Average Sample Number) One of the most important considerations in choosing
among these sampling plans is number of observations required for a decision. For single
sampling plan, the size of sample is fixed. In other sampling plans, however, there is
variation in the number of items needed to make a decision. Average sample number (ASN)
will depend on the quality of material submitted for inspection. If the lot is of a high quality,
it will on the average be accepted early and ASN will be small.
If the lot is of poor quality, it will on the average be rejected early and ASN will again
be small. The ASN will be larger for lots of intermediate quality, where the appropriate
decision is less obvious.
ATI (Average Total Inspection) curve. It depicts graphically the relationship between
the incoming lot quality 'p' and average total number of unit inspected per lot including
sampling and sorting. For a single sampling plan with rectification of rejected lot let the plan
be (N, n, C). Since the probability of acceptance is Pa ATI is defined as
= Pa x n + (1 - pa) N
Example. N = 4000, n = 100, c = 2
P = 0.1, the probability of acceptance from OC curve Pg = .92
Hence
ATI =.92 x 100 + (1 - .92) 4000
= 412 Answer
Similarly the values or ATI for various values of p (Incoming lot quality) can be
calculated and ATI curve can be drawn.
Dodge and Roming sampling plans
Dodge and Romings acceptance sampling plans provide ready made plans which
provide adequate protection to both producers and consumers most economically and are
essentially based on AOQL,
 89

7.8 Summary
Having introduced elements of SQL in the previous lessons, in this lesson we focused
on reaching at a point where we could decide about accepting or repeated a lot produced for
marketing. In ' this context due regard is given to the interest of the producer and, the
consumer. In SQL the terms used in this regard are Producers' Risk and Consumers' Risk.
Where a producer's Risk constitutes the rejection of a lot which is of acceptable quality and
the consumers risk is of accepting; lot which is of sub-standard quality and a Consumer's
risk is the probability that a lot will be accepted that contains a greater number of defects
than the LTPD limit. The other terms which have been used are AQL, LTPD, Sampling plans
and O.C curves. It is the OC curve that is of ultimate importance for achieving at the crucial
decision. Note that the AQL of a sampling plan is a level of quality routineiy accepted by the
sampling plan. AQL is generally defined as that level of quality (percent defective, defects per
hundred units, etc.) that the sampling plan will accept 95% of the time. The LTPD of a
sampling plan is the level of quality routinely rejected by the sampling plan. Lots at or worse
than the LTPD are rejected at least 90% of the time and accepted at most 10% of the time.
7.9 Further Readings
1. John E. Freund and Frank, I Williams Modern Business Statistics.
2. Montgomery, D.C. : Introduction to Statistical Quality Control, Wiley.
7.10 Model Questions
1. What is Acceptance Sampling? Discuss its role in Product Control.
2. Distinguish between producers risk and Consumers risk.
3. What is on OC curve? Discuss its importance and calculations.

——
 90

Lesson-8

QUALITY: CLASSIC QUALITY CONTROL

& QUALITY ASSURANCE
Structure
8.0 Objectives
8.1 Introduction
8.2 Quality as Meeting to Customer’s Requirement
8.3 Quality is Affordable
8.4 Why Persue Quality?
8.5 Classic Quality Control
8.6 Quality Assurance
8.7 Summary
8.8 Further Readings
8.9 Model Questions
8.0 Objectives
After going through the lesson, you should be able to :
– Define Quality as linked to excellence
– Explain the concept of Quality as Meeting Customer’s Requirement
– Distinguish between Quality Management and Total Quality Management.
8.1 Introduction
In common practice, the concept of quality is linked to excellence and achieving the
best. Consider the following statements :
We shall strive for quality
or
This product is pure quality.
In this sense the concept of ‘quality’ is an absolute one and it is either very difficult to
achieve be achieved by the selected few. The concept is an essential part of our culture and
will so. From practical point of view, this concept of best or excellence as quality is not
helpful to run business or to run a company better. Consider a very some common statement
such as :
We only wish we could concentrate on quality goods, but we cannot afford to do so. We
would ourselves out of the market.
This only Implies that quality is either not affordable or not needed. This emerges from
 91

the lad many a times customers will overlook money of the product characteristics which the
manufacturer believe makes his product so excellence, provided the price is low enough.
We operate in a commodity market. Everyone’s product is the same and quality just
does not into it.
This concept at quality based on excellence or something very superior is a subjective
concept. The story below illustrates the quality concept based on best excellence.
Example
A bright new shopping centre is struggling to establish itself. The anticipated store
traffic has yet materialised due to the general business downturn or for other reasons. But
there is one lion, a rejection pottery shop selling ‘seconds’ of a famous brand pots at half
price or less is making thriving business. The Interesting thing is that excellent brand pottery
is not making brisk but only that stock which consists of the rejects (or ‘second’s) is
flourishing. The Managing is bitter, “This company has been committed to the highest quality
for a hundred years but where is it leading us to? Bankruptcy. The only thing I can see is to
increase our reject rate and sell most of our pots as seconds”.
All this reflects some major business issues but the paradoxes are the result of a
failure in understanding. What is required is a more practical and effective concept of quality.
This concept of quality based on excellence is not very popular. It is not a measurable
concept either. What is excellent for one person may not be Good for other.
More Practical View of Quality
In this view 'Quality' is not projected as an 'unattainable excellence' but if focuses on
the needs of a group on which every business entirely depends - the customer so long as the
product meets the requirements of the customers - which may differ from customer to
customer - it is pertaining to quality. This concept of quality form the basis of quality system.
Let us study this concept of quality in details.
8.2 Quality as Meeting Customer's Requirement
Regarded as such quality is the goal of every business. The concept may be simple but
there are quite a few implications to consider :-
(1) This concept of quality focuses on customers, who are brought right where they belong -
to the heart of the business. In the long run no business can keep their customers if they do
not meet their requirements.
In any market, customers are not homogenous and may be segmented into smaller
groups with common requirements. Quality in the sense of meeting customer requirement is
to develop a product range for a range of customer groups. In many markets a business will
realise that it cannot cater to the requirements of all possible customers and so by choice will
choose some selected group of customers. All such companies should be no less committed
to quality (meeting requirements) because a surviving company has to be a quality company.
In the example cited above of the pottery shop and the second shops, the concept of
quality in sense of 'meeting requirements' can point all some solutions. The shop is catering
to the customers - a budget conscious group - who overlook, the defects in compromise with
a low price. The 'second shop' in succeeding in meeting the requirements of its customers.
What about the quality of pots? The need may be to realise the need to build functional range
of pots which can be sold at a budget price and to curtail the supply of expensively produced
rejects or seconds. In this case, the customer's requirements may be even better met with
 92

budget as well perfect pots. It may result in the increase of the sale of some of the 'first rate'
stores and decline. In the sale of 'second shop' because it is likely that customers will find
their requirements met in a different way and the innovatory retailing skills of the 'second
shops' will be successfully redirected. All this results from the new understanding of quality.
(2) The term "requirements" in the concept of quality meeting customer's requirements also
needs to be understood in its widest possible sense.
Requirements, are the totality of all the features of a product as service of significance
to a customer. Some of these may be implicitly implied rather than asked specifically by the
customer. Still, these are important.
For example, while booking a flight a customer need not specify that the plane should
be adequately serviced but it is an implicit requirement that the authorities concerned
should take precautionary steps so as to minimise the chances of wings falling off. Hence,
quality implies in their sense, providing products and services which are fit for purposes, as
a part of meeting customer's requirements.
Note
Whether the requirements are implicit or explicit, meeting them requires their
identification and understanding. This may be achieved through discussion and agreement of
the specific needs of the individual customer of is a continual feedback achieved through
customer satisfaction monitoring or market research programmes.
This approach of quality based as customer's expectations is gaining world wide
acceptance among product/service organizations. This definition is also considered largely
accepted as a management tool. Before we look into the customer's expectations, let us
understand the concept of internal and external customers.

Internal Customers External Customers

Customer that exist within the business and are not the ultimate user of the product
are internal customers. These internal customers are interacting in business in the form of a
chain to provide the ultimate product to external customer. So, in meeting the customer's
expectations the satisfaction of internal customers is must.
The expectations of the customers may look for other parameters in a product as given
below :
(i) Fitness for purpose
(ii) Conformance to standards
(iii) Safety
(iv) Reliability
(v) Durability
 93

The needs of the customers can be known through two ways.

i) By customer himself
ii) By Market Research
The need of the customer as ordered by him are stated needs. The stated needs alone
may not be sufficient for a business to satisfy the customer fully. The unstated customer's
expectations are known through market research and are called implied needs.

8.3 Quality is Affordable

Viewing quality as a means of satisfying customer's requirements requires that quality
must be affordable. If we cannot afford quality, it implies that we are not meeting customer
requirements. In that case either the production method is to be changed to afford 'quality' or
business is to be redirected to a new group of customers whose requirements can be met.
Meeting customers' requirements is absolutely necessary for commercial success. It is
therefore implicit that the requirements are met at least cost and maximum profit. It is also
implied by this definition of 'quality' that the persuit of quality requires continual design
improvement, production innovation and maximum efficiency. Least cost does not imply
curtailing quality but it implies delivery of quality in most efficient manner. This may be
achieved by "getting it right first time".
8.4 Why Persue Quality?
It is clear that in any business we have no option than to persue quality, otherwise it
will fail to meet requirements who will consequently may switch their business to other
competitors. Quality is, therefore, a vital competitive strategy and every business will aim to
meet customer's requirements better than their competitors.
In this sense 'quality' may be expressed by some as something which not only meets
the customer's requirements but also "delights customers". It involves meeting requirements
in a manner which delights the customers. Achieving this not only keep customer's business
but they may cease to even consider alternative suppliers.
Example
Going little extra distance to meet customer's requirements - delighting them - is in
many business an attention to smallest details.
Equipment control knobs just to fit comfortably the hand of operator.
 94

The pack can be opened without finding a pair of scissors.

8.5 How to Get Quality - Classic Quality Control
The traditional or classical approach to quality focuses on the output. This can be
described as the "police method" of quality assurance. Sometimes this 'quality Policeman' is
assisted by his aid. The 'fireman'.
The Police Method or the Quality Policeman
In this approach the output or the products are put to quality inspection before they
leave the factory. The following figure explains the detailed working of this principle. This
approach has following features :
(1) The quality policeman note down the products before they go out to the
customers.
(2) Product standards are specified to meet the customer requirements and the
products are tested to conform to these standards. Only those goods which
conform to these standards are sent to the customer and the rest go as scrap or
for reworking.
Remark What happens inside the black box of the factory i.e., during production is of no
concern to the quality Policeman. This method has two fundamental drawbacks :-
1. Waste is institutionalised because waste in the quality department's output and
their employment depends on a suitable level of waste. If the factory stops
sending out defective items then quality assurance is not required.
2. In some of the businesses the product cannot be effectively tested for
conformity to customer requirements after it is produced.
Modified Police Approach
So a modified police approach to quality is represented in the following fig.2. In this
approach the quality policeman does not rigidly stays out of the factory and quality
inspection is carried out of the product at various stages before it moves to next stage.
However, quality is still outside the production process.
However, in may businesses effective quality assurance through post-production.
The international standards known as quality system standards infact started as
British standards. These days these standards as quality systems are known as ISO 9000
standards and are followed world over. The focus of quality system is the production process
itself and not the output. If we get the recipe right and train the cook, we know that food will
be right. We may do some tasting of food before serving. This assurance about quality of food
will build up in time and we may stop taking food entirely tater on.
Testing is just not feasible :
Example
We go to a restaurant and our requirements are not just confined to receiving tasty
food. We are also concerned that the waiter deals with us politely. If the waiter mistreats us
the damage is done and even if a quality inspector (The head waiter) identifies the "non-
conformance", we are still dissatisfied.
95
96
 97

This example shows that conformity to requirements depends an carrying out works
which cannot be adequately or practically tested after the event.

Self Assessment Questions

Please use the space given below to answer the question. And after giving your answers
match your answers with the suggested answers given at the end of the lesson.
1. Towards Quality, besides focusing on External Customers, another area of focus
is on:
-----------------------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------------------
2. List two approaches in the 'Classic Quality Control'.
-----------------------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------------------
8.6 Quality Assurance and Meeting Requirements
The preceding paragraph have dealt with the various approaches to quality. As we
have seen. The approach of quality based on customers expectations have become the most
significant and central aspect. The International standards on Quality Management too, talk
about the customers expectations from business or service sector.
Quality :- Totality of characteristics of an entity that bear on its ability to satisfy stated and
implied needs.
Quality Control :- Operational techniques and activities that are used to fulfil requirements
For quality.
Quality Management :- All activities of the overall management function that determine the
quality policy, objectives and responsibilities and implement them by means such as quality
planning, quality control, quality assurance and quality improvement within the quality
system.
Total Quality Management :- Management approach of an organization, centered on quality,
based or. The participation at all its members and aiming at long-term success through
customer satisfaction and benefits to all members of the organization and to society.
Quality Assurance :- All the planned and systematic activities implemented within the
quality system, and demonstrated as needed, to provide adequate confidence that an entity
will fulfil requirements for quality.
Quality System :- Organizational structure, procedures, processes and resources needed to
implement quality management.
8.7 Summary
In this lesson we have introduces the concept of quality from the perceptive of
excellence, in Meeting customer's requirement and as is affordable. The lesson has also
discussed basic concepts which answer questions such as How to achieve quality control in a
classic way, what is quality assurance, what is TQM?
 98

8.8 Further Readings

1. John E. Freund and Frank, I. Williams: Modem Business Statistics,
2. Montgomery, D.C. : Introduction to Statistical Quality Control; Wiley.
8.9 Model Questions
1. What is Quality? Discuss its concept in the light of customer's requirement and
affordability.
2. Distinguish between quality management and Total Quality Manpower.
3. Discuss two methods of achieving quality.
Suggested Answers to SAQ
1. Internal Customers
2. The Police Method, and the Modified Police Approach

-------
 99

Lesson-9

ISO 9000 SERIES OF STANDARDS,

EVALUATION MODELS

Structure
9.0 Objectives
9.1 Introduction
9.2 Development of Quality System Standards
9.3 Quality Assurances - Three Models
9.4 Elements of Quality Systems
9.5 Quality Manual
9.6 Summary
9.7 Further Readings
9.8 Model Questions
9.0 Objectives
After going through the lesson you will be able to:
• List at least five stages in the development of Quality System Standards
• List three models of Quality Assurances
• Discuss the Elements of Quality Systems
• Distinguish between Design Control and Document Control
9.1 Introduction
In today's technology riddled world, the pursuant for quality constantly presenting
fresh. challenges and only those firms that are able to meet those challenges will be able to
survive. To achieve consistent high quality, an industry cannot continue receiving supplies of
uncertain quality from its sub-contractors and subjecting them to inspection and testing for
segregating goods lots to feed the production line'. This not only increases cost of production
but also does not ensure consistent quality. For consistency in the quality, it is of utmost
importance to ensue consistent, quality. For consistency in the quality, it is of utmost
Importance to ensure consistency in the quality of in coming material, It is, therefore,
necessary that suppliers (manufacturer) and sub-contractors follow documented procedures,
methods and controls, applicable to their operations. Such procedures, methods and controls
constitute their individual quality Management System.
 100

9.2 Development of Quality System Standards

9.2.1 The first attempt to standardise quaFy system was in America wherein MIL-1-
9858 and MIL-1-45208 were used as Quality Systems Specification and Inspection standards
respectively in the defence contracts.
9.2.2 These two specifications formed the basis for a series of 3 standards designed
for NATO use. These were called Allied Quality Assurance Publications (AQAP).
9.2.3 Despite its membership of NATO, the United Kingdom did not accept the AQAP
standards and
introduced a series of 3 specifications which wore nailed defence standards or
DEF.STANS. In the DEF.STANS, the United Kingdom introduced some
requirements of design to the quality systems. Later on, the AQAP standards
were aligned with DEF.STANS which are now obsolete and UK depends on the
AQAP standards for their defence needs.
9.2.4 A need was felt to adopt the quality systems standards in industry. The solution
for this came forth in 1970 with the publication BS 5750. The Committee of the
International Standards Organisation (ISO) under the Chairmanship of Canada,
later worked to produce a series of international quality systems standards in
1987 popularly known as ISO 9000 series of standards. Like 70 other
countries, India too has adopted- these quality systems standards as Indian
standards. These are popularly known as IS/ISO 9000 series of standards.
These are identical (word by word) to ISO 9000 series except national foreword.
Also provides guidance on the technical, demonstrative and human factors
affecting the quality of
9.2.5 After every five years, international standards are subject to review for
incorporation of updations taking place around the world. The ISQ 9000 series
of standards were also revised in 1994. The revision has not incorporated any
structural changes but language has been simplied in many clauses to make
the standards more user friendly.
9.2.6 Principal Considerations
An organisaion should seek to accomplish the following three objectives with regard to
quality :
a) Meeting customer requirement consistently : The organisation should
achieve and sustain the quality of the product so as to meet continually the
customers stated or implied need.
b) Confidence to self : The organisation should provide confidence to its own
management that the intended quality is being achieved and sustained.
c) Confidence to customers : The organisation should provide confidence to the
customer that the intended quality is being or will be, achieved in the delivered
product or service provided. When contractually required, this provision of
confidence may involve agreed demonstration requirements.
The concepts included in IS/IS 9000 series are applicable to all industry, be it
manufacturing or service based. The emphasis of the standards is on quality assurance
which is far broader than the traditional approach to controlling quality that has been in
existence for many years. It is concerned with far more than just inspection and corrective
 101

action after production. It is an all embracing system for effective quality management based
on the Principle that 'Prevention is better than cure' and as such involves a conscious
analysis of the management system to correct the cause of reject, not just treat symptom.
It is a common fallacy that Quality System related to production alone. All areas of a
company are involved in the provision of a product that is fit for the purpose intended and
supplied to a customer when he wants it and at an economic price. As such, the
requirements of the standards can be applied throughout a company's organisation.
9.2.7 Standards
1. The IS/ISO 9000 series of standards describe the requirements by which quality
systems can be installed operated and evaluated the three quality assurance models
representing three distinct forms of functional or organisation capability suitable for two
different situations contractual and non-contractual.
2. Guidelines
(a) IS/ISO 9000 : This standard provides the guidelines for the selection and use
of family of International Standards on quality systems that can be used for
internal quality management (ISO 9004) as well as external quality assurance
purposes (IS 9001), (ISO 9002) and (ISO 9003). It clarifies the relationship
among various quality concepts and specifies fuels for using the three models
given in IS/ISO 9001, IS/ISO 9002 and IS/ISO 9003. IS/ISO 9000 introduces
the notion of degree of demonstration concerning the adequacy of the quality
systems and the conformity of the product with the specified requirements.
(b) ISO 9004 (In parts) : This standard, together with ISO 9000, provides guidance
to all organizations on quality management. Each of the quality system
elements listed in ISO 9000 has been explained in ISO 9004 (in parts). The
standards helps in developing and implementing a quality systems as also
determining the extent to which each quality element is applicable. It is
products or services at all states - from detection of customer's needs to their
satisfaction. Throughout this standard, emphasis is placed on the satisfaction
of the customer's needs and establishment of functional responsibilities. The
object is to minimize the cost of the quality project while maximizing the
benefits.
9.3 Quality Assurance - Three Models
9.3.1 ISO 9001 "Quality Systems : Model for quality assurance in
design/development, production, installation and servicing" - This standard is
for use when conformance to specified requirements is to be assured by the
supplier during several stages which may include design/development,
production, installation and servicing. It is used when the contract specifically
requires design effort and the product requirements are stated (or need to be
stated) principally in performance terms.
9.3.2 ISO 9002 "Quality System : Model for quality assurance in production,
installation and servicing" - This standard is for use when the specified
requirements for products are stated in terms of an already established design
or specification. The supplier's capabilities in production, installation and
servicing only are to be demonstrated. The quality system elements required to
implement this standard are same as listed in IS/ISO 9001 except 'design
 102

control'.
9.3.3 ISO 9003 "Quality System - Model for quality assurance in final inspection
and testing": This standard applies to situations where the supplier's
capabilities for inspection and tests (conducted on the product as supplied only
are required to be demonstrated).
It is intended that these international standards will normally be adopted in their
present form, but on occasions they may need to be tailored for specific contractual
situations. ISO 9000 provides guidance on such tailoring as well as slection of the
appropriate quality assurance models, viz ISO 9001, ISO 9002 or ISO 9003.
It is intended that these international standards will normally be adopted in their
present form, but on occasions they may need to be tailored for specific contractual
situations. ISO 9000 provides guidance on such tailoring as well as slection of the
appropriate quality assurance models, viz ISO 9001, ISO 9002 or ISO 9003.

Self Assessment Questions

Please use the space given below to answer the question. And after giving your answers
match your answers with the suggested answers given at the end of the lesson.
I. Which country first introduced the Quality Standards in the modern age?
---------------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------------
2. What are basic three models of Quality Assurance?
---------------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------------
9.4 Elements of Quality Systems
Quality Policy Management Responsibility : The Top Management of the supplier
(organization producing /processing / manufacturing a product or (Quality Policy) affording
a service) should make its policy transparent through a statement to this effect. This
statement should be easily understandable to those responsible to implement the policy. The
Management should also evolve suitable procedure to monitor and ensure that the contents
of the policy statement are put to practice by all at appropriate levels and there are no lapses
in activities for continually achieving the objectives of the policy.
Organisation Responsibility and Authority : In the organizational set up for works
affecting Quality there are workmen who actually carryout specific jobs, those who manage
such workmen and those who verify performance of such groups. Responsibility attached to
each grade/cadre amongst them should be clearly laid down and communicated. In regard to
performance of functions, each person should know his delegated authority and as to whom
he is responsible for what. Knowledge of Do's and Don'ts, how to act and co-ordinate upto
what extent, is vital for generating accountability and avoiding confusion. Well defined
freedom and Authority to act particularly relevant to the following functionaries :
a) Those who organise activities that prevent creation of non-conforming Products.
These may be persons engaged in design control, material inspection, production
control, callbration, in-process inspection etc.
b) Those who observe problems relating to product quality and report or record such
 103

problems. These may be supervisors/inspectors engaged in stage/patrol/finished

goods inspection and tests.
c) Those who organise, propose and carryout corrective actions with the concurrence
of specified authorities. These may be QA Manager, R& D Scientists and Engineers
Technologists.
d) Those who ensure that corrective actions have been actually carried out. These
may be Factory Manager or production Manager or Quality Auditor.
e) Those authorized to suspend further production, sale/installation of defective
goods until non-occurrence of deficiency is ensured. These may be any of the
designated persons in the top management level such as Director, President or
General Manager.
Verification Resource and Personnel
Another important part of the Organisation is the internal set up for verification.
Verification includes audit/inspection/test and monitoring of materials and activities at all
stages. Design production, installation and final products are to be systematically reviewed.
These verifications should be carried out by persons independent of those who perform the
work/function, especially in critical areas like Quality System Audit. Design review and audit
of Processes/Products.
The organisation set up should therefore include trained manpower for these verification
functions. The management shall also allocate adequate resources to support the verification
activities.
Management Representative : ISO 9001 stipulates that a particular person within
the organisation should be designated as Management Representative for implementing the
system. It does not however state that such a designated person should be full-time engaged
on this assignment. It only states that the designated representative should be vested with
defined responsibility and delegated authority to ensure his function and also to monitor his
accountability.
The person to be designated to this post should be chosen carefully. Ability to
communicate and convince others, ability to plain and organise, bend of mind of system
orientation, self-discipline not to compromise on quality requirements are certain-qualities to
be looked for while selecting the person. It is possible that the person currently holding the
post of QA/QC Manager may not be suitable to this position. Under such circumstances, it is
always advisable to identify another person either from other sections within or by induction
from outside.
Initially, during the stage of organizing for ISO 9000 the job will involve lot of imaginative
and creative work. Once the system is installed, the representative has only to make sure that
the ISO 9001 requirements are actually being complied on a continuing basis.
With the passage of time, a quality system can turn out to be ineffective or even
obsolete. It (Management) may also be that certain conceptual changes or new (Review)
technologies require upgrading the system. With (4.1.3) these reasons in view the ISO 9001
stipulates that the management of the supplier should heave a set up within the organisation
to review the quality system at given intervals. Ideally the job may be entrusted to a
committee of members drawn from departments directly responsible to implement the
system. Internal audit reports could be normally evaluated during reviews. Items on the
Agenda and decisions thereof should be documented and maintained.
104
105
 106

9.4.2 Quality Systems

The Quality System should aim to establish also to provide evidences that product or
service when made will conform to the specifications. The task has two components. These
are :
a) Bringing out documents containing procedures and instructions reflecting ISO
9001 equirements.
b) Effectively implementing the procedures and instructions as documented.
The basic document that evidence Quality Systems also to the customer and external
Auditor can be in the form of Quality Manual (Please refer chapter on Quality Manual). Those
information for directions in detail enabling accurate and standardized guidance to workforce
can be brought out as "Procedures" "Work Instructions" and "Quality Plans". While Quality
Manual requires to be made available for external reference, other documents may be
restricted to internal purposes only.
In actually establishing and implementing the system, the following activities will have
to be attended to :
(i) Locating and acquiring production resource/skills, processes, Controls, fixture
and inspection equipment.
(ii) Adopting view or updated techniques for quality control, inspection and testing.
This may also involve developing new instrumentation.
(iii) Procure in time, the measuring instruments/equipments with better
capabilities (e.g. closer accuracy, faster reading, accessibility to different
features, newer technology etc.) that may be required to match the emerging
System /Product standards.
(iv) Clearly lay down criteria for judging conformance and non-conformance in
respect of all parameters. Such criteria should be laid down also in respect of
subjective parameters through it may be a more difficult task.
(v) Ensure that the design, Production process, test/inspection/installation
fucncitons, and documentation relating to all these aspects are harmonized and
or made compatible.
(vi) Decide as to what type of records are required to be made in respect of Quality
System Functions and intiate their preparation.
9.4.3 Contract Review
Entering into a contract without looking into the capability of the supplier; ambiguities
in stating requirements of the purchaser; the contract details varying from those originally
mentioned in tender are some of the contract-related aspects that would lead to problems,
ISO 9001 stipulates that the supplier should evolve and practice procedures for reviewing
each contract and ensure that such situations are either avoided or suitably resolved. It is
also stipulated that such review should be documented and records maintained. The review
matters including the internal discussion and inter-departmental communications within the
supplier's organisation should be co-ordinated with the purchaser's organisation as may be
necessary.
Contracts are normally handled by the Marketing (Sales) or Export department. Those
only with commercial background may not alone be able to critically look into the technical
 107

and Quality requirements of contracts. It is for this reason that inter-departmental co-
ordination is considered useful. In contract reviews. An identical set of activities would be
required amongst the purchase department and technical personnel of the organisation of
the Purchaser (Importer). The success of contract review and resolving difference would
therefore depend on effective co-ordination between the supplier's and purchaser's
organizations. A complete approach to the subject like this only would avoid post-transition
disputes/claims.
9.4.4 Design Control
This element in the requirements of ISO 9001 is the major factor that render the
standard superior to 9002 & 9..3. In the manufacturing/processing and service industries. It
is not that all organisation go in for own Design and Development activities. Those who do
not design by themselves rely on establied designs. Such organizations are eligible for
registration only under ISO 9002.
Certain set up of industrial environments and mandatory compulsions; continuing
needs for review and design changes in the contexts of market forces and newer customer
expectations necessitates to have in house Design and Development Activities. Very often the
activity is referred to as R&D. The unit having in house R & D are viewed as the elite group
and their capabilities are related high. They are also benefited by certain incentives and
special facilities offered by government departments.
What ISO 9001 stipulates under this head, as in the case of many other elements, are
to plan draw up procedures, assign responsibilities, control, verify, review and document
them in evidence of such activities.
The end use of a product has certain specified or (if not) implied requirements. The
design of the product when developed should satisfy such requirements. The basic approach
should be to lay down procedures to counter-check if the design satisfy this vital aspect,
continue to carryout such controls and verifications.
Design & Development Activity and the Activity of verifying the designs are to be
allocated only to technically qualified and competent personnel. They shall be provided
adequate resources to satisfactorily carryout their functions. There should also be planned
approach to assigning responsibilities on these activities and such plans be described,
indexed and updated.
Another aspect to be given special attention is avoidance of overlapping and
duplicating situations in respect to both organisation structure as well as technical inputs,
information on these should be documented, circulated and reviewed.
Design input : There are information, data and material inputs required for Design
function. Selection of such requirements should be closely examined to ensure their
adequacy and problems such as ambiguity, conflicting requirements etc. duly tackled co-
ordinating with the persons drawing up the requirements. All these activities should also be
properly documented.
Design output : The first and foremost expectation is that the output meets the input
requirements. It may be that the input information did not contain mandatory requirements
applicable to the product. Safety during operation and proper functioning depend on certain
characteristics considered to be crucial. Such requirements are to be identified and
incorporated. The criteria based on which the design is accepted should be either described
or relevant reference cited.
 108

The above mentioned aspects of design output are to be documented correlating

requirements, calculations and analysis.
Design Verification : Design control measure such recording of design reviews.
Carrying out providing test/demonstration, making alternative calculations and comparing
the new design with an already available proven design, are must to adjudge the output-
input harmony. The verification functions encompassing these activities should be
incorporated in the organization's plans, properly documented, assigned to competent
personnel.
Design Changes : Modification and change in design are to be considered as routine
requirements. Wherever such change/modification is made, there should be provision to
review and approve the same at an appropriate level. There should be established procedures
for such approvals and to identify & document the changes/modification.
9.4.5 Document Control
Date and documents are pre-requisites for operational functions. Updated Data and
documents should be available to persons whose works are critical to effectiveness of Quality
System. The word "Updated" should be given deserving importance. Occasions are not rare
when obsolete documents or data are still used by even those performing critical functions,
due to the only reason that a system is either not evolved or followed to remove the obsolete
ones.
Another very pertinent issue is the correctness or adequacy of document/data
generated competent persons should be designated with authority to examine and approve
documents/data before release and only such actually approved ones should be used.
The ISO 9001 stipulates that the supplier shall establish and maintain procedures to
control documents/data relating to the standard (ISO 9001).
The documents in use are occasionally, either changed or modified. The authority that
approved the original document itself or another authority if so delegated should approve
such modified/changed documents. Prior to such approval the particular authority should go
through the reasons for modifications/change and get convinced. Whenever a document is
changed, the new document itself or an attachment can carry a short write-up about the
cause for change and nature of revision.
How to identify and ensure what is the updated/current one ? This is very vital. One
method is to re-issue the entire document with date and render the earlier on obsolete.
Alternately the document user should check with the document control section which should
have established suitable procedure for the purpose.
9.4.6 Purchasing
Purchases of items that go into the making of product/service are critical functions
since the end product quality would depend on quality of purchased items. Either raw
materials or semi-finished or ready to use components/constituents are purchased ISO 9001
standard expects that the Quality System Controls in respect of purchases are effective. For
this purpose, all purchase orders should be complete with proper descriptions and data such
as reference to applicable specifications, drawings, process instructions, norms for
inspection, acceptance criteria etc. The description should also mention the class, grade,
type, style under the relevant design/specifications as may be applicable to the particular
purchase order. All these requirements should be ensured by reviewing the purchase
documents. It may be that compliance to ISO Quality System Standard such as ISO 9002 or
 109

9003 is also specified, particularly in respect of sub-contracted components.

Sub-contractors or Vendors are to be scientifically assessed for their capability before
selection and their performance monitored based on a system. There are vendor rating and
vendor quality rating in vogue for these purposes.
Notwithstanding the in-built safeguards as aforesaid, it is desirable to make provisions
in the contract for counter checks of purchased items. It may be carried out at source or on
receipt in the purchaser's unit. It the contract provides for, the sub-contractor should
facilities inspection at his premises. While such an act is not a guarantee against subsequent
rejection, the purchaser himself also should not use the acceptance after 'inspection at
source' as a sufficient enough evidence of sub-contractor's effective quality control.
The manufacturer (Supplier) - Sub-contractor relationship in terms of Quality system
requirements can be generally equated to the Purchaser (Customer) - Supplier relationship. It
thus emerges that Quality System standards would gradually be applied in all the three tiers,
viz.
- Export-import transactions
- Domestic final product transactions
- Sub-Contracted transaction for manufacture.
9.4.7 Purchaser Supplied Product
The Customer (purchaser) or a final product himself may supply certain items
(products) for incorporation into the mix or assembly by the manufacturer (Supplier) in such
contexts also, there shall be established procedures for checking their acceptability for
storage and maintenance. The manufacturer's verification (inspection) activity is not intended
to absolve the customer from his responsibility to supply acceptable components.
9.4.8 Product Identification and Traceability
There are specified requirement under certain law to enable traceability of a product to
the supplier, date of production, quality grade etc. In other cases also traceability procedures
are being established and adopted. Traceability doesn't confine to tracing of a product to the
supplier only. Within the supplier's activities for production, at each stage, the relevant
specifications/drawings process instructions or other documents should be identifiable and
performance of right or wrong action by each individual should be traceable by creating and
implementing suitable system for this purpose.
When the requirement of traceability is within a law, identification codes incorporating
Batch number, manufacturer's name, date of production, specification number etc, are
affixed on the product. In other situation also if such procedures are implemented the
supplier's quality system is strengthened and his commitment for quality and accountability
made transparent.
9.4.9 Process Control
Process control refers to carrying out production under controlled conditions. The
supplier should identify and plan both the processes and controls for processes.
The process control include proper working environment, use of suitable equipment,
compliance with standards/codes of practices/quality plans and work instructions. The work
instructions should define manner of production and installation in detail, using easy to refer
format/language and should highlight such aspects non-compliance of which would affect
 110

quality.
The process control should also cover monitoring and control of product
characteristics at different stages of production. Workmanship of a product should receive
greater attention while dealing on characteristics. In the event of difficulty to bring out
written down standards, representative samples can form the criteria for workmanship.
Specific procedures for approval of equipment and processes are to be incorporated in
the process control plans. Such procedures should among other aspects, specify the
authority delegated to approve and norms for approval.
Effect of certain process cannot be ascertained by inspection and testing of the
(special) product, but deficiencies in the processes will (Processes) surface when the product
is in use in regard to (4.9.2) such processes, strict procedures should be implemented to
ensure that specified requirements have been actually ment continuous monitoring and
documenting the observations are important techniques employed to control special
processes and the equipment/personnel used for the same.
9.4.10 Inspection & Testing
Inspection and Testing activities are broadly grouped into 3 categories and these are :
(i) Receiving inspection & testing.
(ii) In process inspection & testing.
(iii) Final inspection & testing.
The supplier has to make Quality Plans or Procedures to plan and carryout the above
functions as a part of the ISO 9001 standard requirements.
For any inspection and testing the basic requirements are the specifications against
which the inspections are to be carried out the methodology or procedures for testing. These
are to be identified/established as the First step and duly documented. While incoming
material are normally released for used in production only on ensuring conformity with
specifications, under the positive recall procedure, waving this step can be permitted when a
material is required for urgent production purposes. If so authorized by the competent
authority.
Any inspection has an element "sampling" sample quantity can be reduced if the
supply source is known for it's inhouse control or when a test certificate accompanies the lot.
If otherwise, a higher sampling may have to be resorted to.
In process inspection refers to not only checking of products under process but also
monitoring of processes and control methods.
Stage inspection and patrol inspection are the two approaches that could be employed
for inprocess product inspection. The objective inprocess product inspection would be to
identify non-conforming products. If any, at every stage of production process. When a non-
conformity is detected, it may be due to material defect, machine setting defect, improper
process control or techniques etc. and their timely correction would benefit not only the
quality assurance process but also the economy of manufacture.
Final inspection and testing are carried out to ensure the Final product's conformity to
the specifications. Sampling for final inspection should ideally be done when the product is
in a ready to dispatch status so that aspects such as labeling and packaging requirements
could also be checked during this inspection. Before taking up final inspection, the inspector
 111

should scrutinize data on receiving and inprocess inspections and ensure that specified
requirements were complied and conformance evidenced. Supplier's system shall provide for
this authority which would enhance system reliability itself.
Products accompanied by a "passed" status report duly authorized by competent
persons, only shall be allowed for despatch. Persons so authorizing should make sure that all
activities in the specified Quality Plans and procedures have been satisfactorily completed
and authorized documents and data to this affect are physically available.
9.4.11 Inspection, Measuring and Test Equipment
Measurement capability in terms of range and precision\accuracy are the factors to be
looked for while selecting an equipment in relation to a given requirement. The equipment
should also ensure consistency with the measurement capability.
Use of all the inspection, measuring and test equipment and devices such as gauges,
instruments, jigs, fixtures, templates etc. within the required accuracy/wear-limit is possible
only if their calibrations are systematically attended to.
The periodicity of calibration depends on the nature of the measuring instrument etc.
and the intensity of use (wear). For example a gauge used to check large number of
measurements every day be calibrated once in a week whereas an equipment used to test a
parameter occasionally may be calibrated only once in a year. The calibration interval
applicable to each item should be decided and indicated both on the item and in the relevant
records.
All items should be serially numbered, their location and calibration status monitored
centrally, procedures of calibration and the follow up action in the event of detecting
unsatisfactory status on checking are to be established and documented in respect of all
inspection, measuring and test equipment. This requirement may vary also depending on the
equipment location. A gauge used by operator during production process may have different
calibration requirement when compared with same type of gauge used in final product
inspection.
Maintaining records of calibration in the form of History cards or in any other form
helps to understand current status and to trace history, wear limit etc.
Masters/standard pieces, using which comparison/adjustment are made during
calibration process, may preferably conform to a National Standard. Otherwise the basis of
calibration should be freshly documented. All such items used for calibration should be
maintained and handled under suitable environmental condition (e.g. An Air-conditioned
Area) without affecting Accuracy and Fitness.
9.4.12 Inspection and Test Status
As direct result of inspection and Test carried out on a product at any stage, the
product is adjudged either as conforming or non-conforming. Both these status especially the
non-conforming ones should be indentifiable. The status can be indicated by a method
appropriate to the product such as marking, labeling etc. it is also advisable to physically
segregate non-conforming products away from the conforming product to make sure that a
product dispatched, used for installed had passed in the required inspection and test.
Records should also indicate the inspection authority responsible for releasing a
passed product for use.
 112

9.4.13 Control of Non-Conforming Product

The very Quality System itself is intended to make sure that
(i) non-conforming products are not produced;
(ii) if at all produced they are not allowed to be sold or installed.
The above element deals with the procedure to be followed if non-conforming product
is detected to have been produced. The procedure should primarily ensure that such non-
conforming products in a unit should be identifiable, they are segregated and prevented from
being used or installed. After documenting detection and preventing use, there are certain
further steps that may be usefully employed i.e. review and resolve as to how best the non-
conforming products be disposed and to find out as to how the non-conformity was caused.
If the deviation from specified requirement is such that it can be corrected, the
product could be re-worked to meet the specified requirements. Before releasing for use re-
worked items should be reinspected.
In certain other situation, if deviation is minor and correction is not required/possible,
a delegated person may accept the product with concession.
In both the above situations, the customer should be informed and his concurrence
obtained.
In respect of certain products where grading is adopted, it is possible to regrade,
assign the appropriate grade and release the product. This is advised only if the defect
observed does not relate to safety requirements.
When none of the three options as above is possible, the product should be rejected
and scrapped.
All the above activities shall be performed as per documented procedures, by the
persons authorized to do so and their record maintained.
9.4.14 Corrective Action and Preventive Action
This element of ISO 9001 is quite important. It deals with investigating/analyzing the
non-conformity and implementing actions, controls and revisions not to recur such non-
conformity.
While investigating the causes, it may also become necessary to analyse process, work
operations, concessions, quality records, service reports and customer complaints. Such
elaborate approach is required especially to detect potential causes of non-conformity.
Preventive action identified as appropriate to tackle the causes(s) should be
commensurate with the level of risk that a particular non-conformity may create. Such
actions may mean revision of procedures or work instructions. Initially it becomes necessary
to monitor effectiveness of the changes introduced before stabilizing the same. There should
also be adequate controls to make sure that corrective action is implemented. Changes
adopted are to be incorporated in the relevant documents and all corrective action data
properly recorded.
9.4.15 Handling, Storage, Packaging and Delivery
The supplier (manufacturer) has to adopt documented procedures and ensure their
regular compliance in respect of all these activities which also quite significantly contribute
in assuring quality to the customer. Very often these aspects are treated with less
 113

importance. Damage arising from improper packaging or handling of a final product is more
serious, since such damage caused is on value added products on one hand, on the other the
non-delivery due to damage may upset the purchaser's work schedule/time bound usage
resulting in losses and bad reputation etc.
Non-delivery of a product as per schedule defeats the very purposes of all the previous
activities, systems and control.
ISO 9001 stipulates that supplier should provide proper methods and facilities for
handling storage areas suitable to adequately preserve products until dispatch, laydown
controls for packing/marking and ensure total protection after final inspection until the
suppliers responsibility ceases. If the contract provides for delivery at customer's premises
such activities for protection extends upto destination.
9.4.16 Quality Records
In regard to performance of each system activity already dealt with, preparing recourse
has been specified as a necessity. These records are useful to evidence the system operation
and to demonstrate quality achievement. The records should be fegible. The formals of all
records should provide for entries by which the corresponding product is traceable.
When large number of records on different quality functions are generated, how and
where these are to be maintained/stored and when to dispose are to obvious questions.
Depending on the organisation and activities involved, the supplier should have established
procedures to identity, store and dispose the records. The procedure or system for record
maintenance should facilitate easy retrieval of a particular record when required. The type of
storage used should be good enough not to cause deterioration, damage or loss of records. In
regard to disposal, the procedure adoped by most of the organizations is to specify retention
period for each type of record so that at the end of such specified periods authorized persons
can weed out them.
ISO 9001 also stipulates that, the Quality records shall be made available to
purchaser, if contract so provides.
9.4.17 Internal Quality Auditing
Internal Quality Audits are carried out for two broad purposes :
(i) To find out whether all quality related functionaries comply with the laid down
procedures and
(ii) To determine whether the implemented system as such is effective.
Ideally, the internal audit should be carried out by persons independent of quality
control and production activities. Periodicity, procedure to be adopted, documenting and
intimating results follow up action and such other aspects of internal audit activity shall be
planned depending on the activities and status of the unit concerned. Such plans and
procedures should be documented.
A report on internal audit should be sent to the department/section audited and those
delegated with the responsibility shall promptly act to set right the deficiencies, if any,
brought out in the report.
9.4.18 Training
Those operating Quality Systems can be rated a grade higher to those operating
production functions for the simple reason that former group is to the effectiveness or
 114

correctness of the activities of the later group. Hence it implies that those assigned to Quality
System responsibilities have appropriate educational and training/experience background.
At induction stage, orientation training is imparted. Further training for refreshing
and updating skills are to be given as an ongoing programme of the organisation. With these
in view, the training needs are to be identified, documented and implemented.
It is also necessary that the particulars of training received by each person goes into
records and such recourse are traceable when needed.
9.4.19 Servicing
After-sales servicing is an area still to be cared for adequately. The servicing activity if
made a part of the terms of supply, such supplier gets higher rating and most of the
suppliers are organizing for this function, ISO 9001 stipulates this activity, however, only if
the contract so provides. The supplier, in such situations should document procedures for
servicing and also to verify it's effectiveness.
9.4.20 Statistical Techniques
Statistical techniques are to be applied, to evaluate product and to assess process
capability. In many industries these techniques are provided to be helpful. ISO 9001 provides
for its application only if considered appropriate.
9.5 The Quality Manual - A Structured Approach
9.5.1 IS/ISO 9000 series specifically calls for documentation of a quality manual which may
include or refer to requirements of Quality System standard. The format and design of
a quality manual should be the sole province of the company and organisation that
has to use the document on a regular basis. It is possible to produce manual in a
number of tiers or levels of information.
9.5.2 The first level manual is generally called Apex Manual/Policy Manual/ Quality
Manual. It essentially deals with the policy and derives the way in which the
company system is structure to meet the requirements of the applicable quality
management standard. The lower tier deal with procedures which can be
developed on a departmental/section/activity basis and also work
instructions/test methods/basic data and forms at the lower tiers.
9.5.3 Many companies successfully operate a system where all these various tiers are
included in the one document, but these tend to be the smaller concerns where it is
beneficial to maintain one point of reference for all information. A company of bigger
size or complexity will benefit from breaking the quality system down into a number of
workable elements in a similar manner as explained in 5.2 above.
9.5.4 While developing the qualify manual, it should be remembered that primary purpose
of quality manual is to provide an adequate description of the quality management
system whilst serving as a permanent reference in the implementation and
maintenance of that system. It is strongly recommended that an efficient and effective
method of updating is provided to allow the system to change overtime.
9.5.5 The structure of the Quality manual can very from company to company, although
organizations currently develop the policy manual or first tier documentation in the
order of the requirements included in the applicable quality management standard. In
this way, the policy manual provided detailed index to the quality systems operated by
 115

the company. A manual presented in this way is somewhat easier to audit but that is
not the main criteria for structuring the document should be produced to meet the
needs of the users than the auditors.
9.5.6 Whatever be the approach of presenting the quality manual. It should reflect what a
company does rather than provides as indication of what is ought to do/like to do.
9.5.7 Companies, in general hold majority of the Information required by IS/ISO 9000 series
in order to be able to conduct their operation, service clients etc. However, much of it
is most probably in unstructured form and vary likely has out of date material in use-
By applying the disciplines of IS/ISO 9000 series ft provides a structure to the
company's existing operations and systems whilst at same time providing a vehicle for
their continued updating. Thus crisis management by finding a problem and then
writing a memo to be concerned to prevent its recurrence becomes a thing of past,
because the manual is the source of information regarding how the system functions
and, if there is a need for change, the change is incorporated in that one document.
The quality manual should reflect what a company does rather than provides
an indication of what it would like to do.
9.6 Summary
In the technology driven world, quality is posing recover challengers to its practioner
on how to achieve constant quality, in this lesson the focuss has been on development of
quality system standards, where in experience of other countries is referred to in the light of
institutions as also about defining standards. In this lesson as many as 20 elements of
quality systems are discussed in this lesson. Eventually the lesson discusses the structural
approach to Quality Mannual.
9.7 Further Readings
1. John E. Freund and Frank, I Williams: Modem Business Statistics.
2. Montgomery, D.C. : Introduction to Statistical Quality Control; Wiley.
9.8 Model Questions
1. What are the principal consideration that are organisation has to seek to
accomplish to ensure quality.
2. Discuss three Models of quality assurance.
3. Discuss structured approach to Quality Manual.
Suggested Answers to SAQ
1. USA by ML-1-9858.
2. ISO 9001, ISO 9002, and ISO 9003

------
 116

Lesson-10

APPLICATION OF ISO 9000 TO INDUSTRY

AND SERVICE SECTOR

Structure
10.0 Objectives
10.1 Introduction
10.2 Why ISO 9000?
10.3 Application of ISO 9000 in Food Industries
10.4 ISO 9000 in Service Sector
10.5 Summary
10.6 Further Readings
10.7 Model Questions
10.0 Objectives
After going through the lesson you should be able to :
– discuss the importance of ISO 9000 with special reference to EC.
– discuss the role of ISO 9000 in service sector.
10.1 Introduction
ISO 9000 series of Quality System Standards have come to stay Global recognition
that System Standards should complement Product Standards takes us in this present
scenario.
The concepts and guidelines contained in these Standards are highly useful to any
enterprise be it a manufacturing or a service industry. A few writings here and there, a few
seminars and announcement of Government policies about ISO 9000 have created a lot of
inquisitiveness is not only Quality related workers but also general Public and even Students.
All are not able either to attend expensive seminars or to have access to these standards and
activities around their application.
A handy take-home reading material for those who are inquisitive and a down to earth
guidance at one place to those Intending to involve in detail are the requirements that this
guide will serve.
There are large number of small scale industrial units engaged in production for
export to European Countries. They need to be awakened, motivated and assisted if they are
to comply with the ISO 9000 requirements. A very, highly skillful activity is still left
untouched by most of them. A long way to go ! Why not get to know and get, started?
Perhaps a Guide of this type would help the process.
Being an edition delivered with a very short gestation. Inadequacies and inaccuracies
 117

are possible. The publishers of the Guide look for suggestions and comments to make the
next edition more useful and comprehensive.
10.2 Why ISO 9000
The immediate compulsion for manufacturers to turn to ISO 9000 was the consensus
among the European Community countries to unify their markets by the end of 1092, during
which process the community anticipate also compliance to ISO 9000 by companies that
export to/or operate within Europe.
The EC countries alone are having 320 million consumers. The European Free Trade
Association countries with another 30 million consumers have bridging agreements with
EEC. A total 350 million consumers of Europe will thus become the largest Trade market in
the free-world.
Since ISO 9000 Registration is stated to be mandatory for exporting to Europe by the
end of 1992, there is naturally concern and panic on this score. Recently the Export-import
policy announced by the Indian Government has offered special facilities to those enjoying
ISO 9000 registration. Well established and large private and public sector companies have
already started demanding compliance to ISO 9000 by their sub-Contractors and Suppliers.
Apart from the immediate European compulsion, the emerging trend is that the compliance
to ISO 9000 will slowly become a requirement within the domestic markets too, before long.
By going through a product standard a customer understands the characteristics of
the product intended to be produced against the standard. However the customer has further
questions about the product such as :
- Whether the particular piece of the product customer buys will really conform to
the standard ?
- Whether the product will perform as expected during its life expectancy and
demonstrate its reliability ?
- Whether the next piece and more pieces of the same product of the same
manufacturer would demonstrate consistence in conformance to the Standard and
reliability ?
While answer to the first question can be found, by carrying out inspection and testing,
finding answers to the second and third questions is a more complex exercise. The only
solution for the manufacturer is to demonstrate his commitment, policies and systems which
ensures that non-conforming products are not produced and from piece to piece the
conformance to Standard will be consistent. This takes us to a requirement for Quality
System.
ISO 9000 series of Standards are Quality System Standards for the purpose of
adopting and demonstrating quality assurance system at all stages of manufacture and
service activities. Today, with the evolution of these concepts and their adoption, the
customer looks at the supplier's organisation first in terms of it's management commitment
to strategies that makes quality a concern of every one in the organisation.
Adoption of ISO 9000 standards, already recognized by over 90 countries in the world,
helps the manufacturer to have easy access to international markets. As the customer would
prefer an effective marketing tool. It also helps to avoid barriers on the grounds of technical
standards to minimize product liability disputes and claims and to enjoy special grants or
incentives offered by the Government.
 118

Thus it is emerging that adoption of quality system standards in regard to products

and services will be the least expectation of the suppliers and customers of the future,
though the starting point now may be the EEC factor. All those involving in
production/manufacture and services cannot afford to sit on the fence as onlookers any
longer. They should get started and march towards compliance to ISO 9000 and achieve the
Registration.
In this regard there are aspects to be understood properly to make the journey easier.
This publication in intended to serve as a Road map for the journey to the ISO 9000
Registration.
10.3 ISO 9000 in Food Industries
Food industries in India are on the uptrend because there is a growing swing towards
usage of processed foods by the otherwise orthodox populace. The second factor that catalyse
the growth is the potential for export of sea and Agrobased food products. The Union
Government has identified these sectors as highly promising to contribute to the country's
export growth and initiated a number of measures to encourage export oriented food
industries.
Food as a product has a very sensitive status since it's safety is the greatest concern of
the civilised society. A number of standards and systems are already in operation to regulate
food manufacture and sale. In most of the countries. WHO, FAO and Codex Alimentarices
commission have either separately or jointly, brought out standards and codes of practices
for guidance of the food industries. For these reasons it is incorrect to assume that system
standards are difficult to be implemented in the Food industries. On the other hand, these
industries do have a good base to start on, provided they receive adequate guidance for
orienting themselves specifically to ISO 9000 requirements.
The major Indian food products that enter export trade are Marine products, Fruits
and Vegetables products, Spice and Spice products, Cashew Kernel, Rice and a variety of
other processed foods. Tea and Coffee are also exported in large volumes. It is significant that
all of these items have good markets in Europe.
A good number of governmental and Quasi-government institutions do operate in the
country specifically for export promotion of these food products and to help development of
their industries.
Most of the Food units being in the small scale sector, systems are more informal and
less visible. It is said that "It is not enough only by being good, but one should also appear to
be good". Formalising what are informal and creating better visibility for quality, are the
initially possible actions to work towards ISO 9000 compliance. ISO 9000 only guides what
elements should compose the Quality System and does not dictate what the element-content
should be. It however expects that a company formulate its own system, document it, operate
it, maintain it through continual operation, control it, review it and revise it as may be
required; make these activities visible through updated documents and records.
Atleast a few of the food units have not yet made any efforts to know what these ISO
9000 standards are and others who got to know about it are still reluctant to get into it
(either due).
Use of Consultants
Use of external assistance may be required for most of the small scale units for
inherent reasons. Even if there is a designated Quality Manager or Supervisor, he is unable
 119

to draw attention of top management, get them committed and pan authorization from them
to work on the project. It is also possible that the Quality Manager or Supervisor himself has
not been able to comprehend the standard and the route to the destination in view of
complexities involved.
Use of the consultants either by individual companies or a group of companies having
a common forum may be required to get initiated. The advantages around using consultants
may be :
- A consultant can more freely access to the top management and get across the
message.
- The internal organisation hierarchy doesn't affect or inhibit his working.
- He is more competent to decide as to what re required, to design a system, prepare
Quality Manual, organise training, and carryout pre-registration checks.
- Management can avoid employing a highly paid expert on a regular basis for a one-
time job or formalizing a system.
An Analysis of ISO 9001 requirements
Let us now go a little in depth to relate the ISO requirements to the Food industries
and understand if there are requirements that would offer difficulties while implementing.
(Please refer also to the chapter on Elements of Quality System).
Elements Remarks
Management Responsibility Bringing out a Quality Policy statement, defining
responsibility and authority, appointing a
management representative, periodic system
review by management are easily implementable.
The world design, should be understood to mean
the food formula or product standard.
The world 'installation' and 'servicing' not being
normally relevant can be generally standard.
Quality System Preparation of Manuals instructions, Quality
plans and formats are more easier since there are
documents of the kind used in Govt. regulations,
available to emulate.
Control Review The only snag is that the contracts mostly
mention only a grade or a simple commercial
nomenclature to imply the requirements.
Across reference to the original document or one-
time preparation of written down specifications
against implied requirements can be done.
If there are no in house testing facilities for certain
difficult characteristics, the procedure can suggest
reliance on independent outside facilities and thus
establish capability to meet the contractual
requirements.
 120

Design Control "Design" should be interpreted to mean

"formulations". "Product mix" or ingredient input
etc. By this interpretation all the resources and
activity requirements can be met by food industry.
If the basic data are drawn or adapted from an
established source, the same can be stated and
provisions made to effect changes when originally
relied source carries out changes. With this
understanding the activities for planning,
verification, document control etc. can be
established and followed.
Purchasing Most of the raw material suppliers may not be
able to demonstrate their ability in terms of
systems and controls. The purchase specifications
of the processing unit (company) may also be
deficient as a technical document.
In these areas some spade work will be required.
Supplier may be helped by tailor made Quality
Plans. In respect of purchases of items like
additive, flavours, containers and other packing
materials sub-contractors can be scientifically
selected, rated and controlled.
The "drawings" may be relevant only to the
specifications for purchase of containers and
packing cases.
Purchaser supplied Product Such instances are very rare. In buy-back and
original brand packing arrangements, there may
be some items like labels, containers etc.
Procedure for control of such items can easily be
established.
Product Identification & traceability This is very important to food products. Applying
identification codes by embossing or labeling is
the procedure used. For in process identification,
travelling batch cards can be used. Finished
product may be stored codewise. Procedures as
above can be easily documented and used to meet
this requirement.
Process Control The process control forms the backbone of any
food processing industry. Implied and traditional
controls, if not documented controls, are normally
known. It is enough to formalize them through a
structured procedure and document.
Inspection & Testing Most of the food product industries under
statutory regulations are already implementing
receiving. In process and final inspection and test.
These also can be formalized conforming to the
 121

requirement of ISO 9001.

For abruptly introduced contractual test
characteristics and where no test facilities exist,
procedure can incorporate provision for such tests
to be performed by third party.
Inspection, Measuring and Test There are atleast a few who assume that this
Equipment element is unimportant for food industry. It
should not be taken lightly. There are weighing
balances, pressure and tremperature gauges etc.
used whose non-calibration will affect product
quality. Hence a limited in-house facility or third
party calibration procedure should be finalized
and the system documented.
Inspection and Test Status A procedure of batch numbering or lot coding and
relating non-conformity to such batches and codes
are ideal procedures for this purpose. Such
systems are in vogue which can be formally
documented and procedure controlled. Others can
easily develop an appropriate procedure.
Central on Non-conforming Product There are procedures of re-grading and
reprocessing of non-conforming products in vogue
in some of the food industries. Procedures not to
mix up such products are important. Being food
products concessions should be used only with
utmost care, under proper authorization. The
product declared as unfit for human consumption
due to serious deficiencies should be destroyed.
These are the aspects requiring special
procedures. Depending on the nature of the
product, an appropriate system on this element
also can thus easily be established.
' Corrective action Analysing reasons for non-conformity and revising
processes and procedures are important to avoid
loss, in food industries. No problem is envisaged in
documenting and implementing a procedure for this
purpose.
Handling, Storage. Packing and Preservation is the foremost activity in any good
Delivery industry. In most companies, there are well laid out
systems for handling & storage. Under Indian
conditions, packaging aspect may need better
attention,
Quality Records This element is equally important and
implementable.
Internal Quality Adults Most of the companies in the small scale sector
may not have organizational provision for this
function. Resources and procedures can be
 122

identified and documented.

Training Providing for his requirement in the case of food
industries should be considered as an investment
for the future. Training on Hygiene & sanitation
controls upto workers level assumes importance in
this industry.
Servicing This element has practically no relevance to this
industry,
Statistical Techniques Most of the companies do not empioy this
technique. If there is a contractual requirement the
technique can be easily employed for establishing
product & process acceptance.
Note : Examples of Quality Plan and Quality Policy Manual etc, appearing as Annexure,
relate in FOG Industries.

Self Assessment Questions

Please use the space given below to answer the question. And after giving your answers
match your answers with the suggested answers given at the end of the lesson.
1. When did ISO 9000 regulations became mandatory for exporting to Europe?
-------------------------------------------------------------------------------------------------------------------------------------
-------------------------------------------------------------------------------------------------------------------------------------
2. What is the limitation of ISO 9000?
-------------------------------------------------------------------------------------------------------------------------------------
-------------------------------------------------------------------------------------------------------------------------------------

10.4 ISO 9000 in Services

The notes below 'Definitions in ISO 9001, 9002 arid 9003 state that for the purposes
of these standards the term "product" is also used to denote "service" as appropriate. ISO
9004-1987 that set out Guidelines on Quality management and System elements have made
mention about "product and services". However for the purposes of applying these Guidelines
for "Services", ISO 9004-1987 was not adequate. ISO 9004- Part-2 of 1991 was therefore
prepared by the technical committee ISO/TO. 176 as guidelines for services only. As a result
ISO 9004-1987 is being revised as ISO 9004- Part I.
ISO 9004 - Part-2 helps successfully apply Quality Management to "Services". The
industries handling services are now therefore able to achieve compliance to ISO 9000 by
developing systems for quality management as per the guidelines of ISO 9004-2.
The application of ISO 9000 is not only intended to "Service" but also "Service delivery".
Handling both these together is considered a necessity for successful introduction of ISO
9000 in service industries. As such specification for the service to be provided and the
service.
Examples of services to which ISO 9004-2 may be applied are :
- Health services - Training & Education
- Transport-Railways - Maintenance
 123

- Telecommunications - Electricity & water supply

- Banking, Insurance - Trading
- Project Management - Scientific Research
Service requirements are to be defined by characteristics that may be quantitative or
qualitative. While certain characteristics may be observable and subject to customer
evaluation, there may be another set that are not observable by customers but directly affect
service performance. These aspects are to be properly understood while establishing the
requirements.
Examples of characteristics that may be specified in requirement documents are :
- Facilities, capacity, number of personnel and quantity of materials.
- Waiting time, delivery time and process time
- Hygiene, safety, reliability, and security.
- Responsiveness, accessibility, dependability, credibility etc.

10.5 Summary
Having introduced, 150 9001, ISO 9002 and ISO 9003, this lesson discusses ISO 9000
which provides a set of international standards on quality management and quality
assurance developed to help companies effectively document the quality system elements to
be implemented to maintain an efficient quality system. In this lesson a special focus has
been laid on the food industry and the service sectors.
 124

10.6 Further Readings

1. John E. Freund and Frank, 1 Williams: Modern Business Statistics.
2. Montgomery. D.C. : Introduction to Statistical Quality Control; Willey.
10.7 Model Questions
1. Discuss the importance of ISO 9000, with special reference to European
Community.
2. Write a note on the ISO 9000 requirements.
3. Discuss the role of ISO 9000 in Services.
Suggested Answers to SAQ
1. 1992
2. It guides what elements should compose the Quality system and not the element-
content.

--------

Typesetting by : Chanakya Graphics, Kesho Ram Complex, Sec. 45, CHD.

USOL (307) 18.01.2021