LESSON10.

DEMAND FORECASTING

Today business enterprises are working under the conditions of uncertainties. Uncertainties can
be minimized through planning and forecasting. The success of a business firm depends upon
its ability to forecast future events.

10.1 Meaning of Demand Forecasting

Future is uncertain. There is great deal of uncertainty with regard to demand. Since the demand
is uncertain, production, cost, revenue, profit etc. are also uncertain. Through forecasting it is
possible to minimize the uncertainties.

Forecasting simply refers to estimating or anticipating future events. It is an attempt to foresee

the future by examining the past. Thus demand forecasting means estimating or anticipating
future demand on the basis of past data.

10.2 Objectives of Demand Forecasting

A. Short Term Objectives

1. To help in preparing suitable sales and production policies.
2. To help in ensuring a regular supply of raw materials.
3. To reduce the cost of purchase and avoid unnecessary purchase.
4. To ensure best utilization of machines.
5. To make arrangements for skilled and unskilled workers to maintain a suitable labor force
6. To help in the determination of a suitable price policy.
7. To determine financial requirements.
8. To determine separate sales targets for all the sales territories.
9. To eliminate the problem of under or over production.

B. Long term Objectives

1. To plan long term production.
2. To plan plant capacity.
3. To estimate the requirements of workers for long period and make arrangements.
4. To determine an appropriate dividend policy.
5. To help the proper capital budgeting.
6. To plan long term financial requirements.
7. To forecast the future problems of material supplies and energy crisis.

Dr. Ramesh Raj Ayer PhD, DMS, MBA, MA, PGDM, PGCP, BSc
Professor, Corporate Mentor & Consultant
Mo: +91 9900704081 Email/LinkedIn: [email protected]
10.3 Factors Affecting Demand Forecasting

For making a good forecast, it is essential to consider the various factors governing
demand forecasting. These factors are summarized as follows.

1. Prevailing business conditions​: While preparing demand forecasts it becomes necessary to

study the general economic conditions very carefully. These include the price level
changes,change in national income, per capita income, consumption pattern,savings and
investment habits, employment etc.

2. Conditions within the industry: ​Every business enterprise is only a unit of a particular
industry. Sales of that business enterprise are only a part of the total sales of that industry.
Therefore, while preparing demand forecasts for a particular business enterprise, it becomes
necessary to study the changes in the demand of the whole industry,number of units within the
industry, design and quality of product, price policy,competition within the industry etc.

3. Conditions within the firm​: Internal factors of the firm also affect the demand
forecast. These factors include plant capacity of the firm, quality of the product, price of
the product, advertising and distribution policies, production policies, financial policies
etc.

4. Factors affecting export trade​: If a firm is engaged in export trade also it should consider the
factors affecting the export trade. These factors include import and export control, terms and
conditions of export, Exim policy, export conditions, export finance etc.

5. Market behavior : ​While preparing demand forecasts, it is required to consider the market
behavior which brings about changes in demand.

6. Sociological conditions​: Sociological factors have their own impact on the demand forecast
of the company. These conditions relate to size of population, density, change in age groups,
size of family, family life cycle, level of education, family income, social awareness etc.

7. Psychological conditions: ​While estimating the demand for the product, it becomes
necessary to take into consideration such factors as changes in consumer tastes,
habits,fashions, likes and dislikes, attitudes, perception, life styles, cultural and religious beliefs
etc.

8. Competitive conditions: ​The competitive conditions within the industry may

change.Competitors may enter into market or go out of market. A demand forecast prepared
without considering the activities of competitors may not be correct.

Dr. Ramesh Raj Ayer PhD, DMS, MBA, MA, PGDM, PGCP, BSc
Professor, Corporate Mentor & Consultant
Mo: +91 9900704081 Email/LinkedIn: [email protected]
10.4 Process of Demand Forecasting/ Steps in Demand Forecasting

Demand forecasting involves the following steps:

1. Determine the purpose for which forecasts are used.

2. Subdivide the demand forecasting programme into small I parts on the basis of product or
sales territories or markets.

3. Determine the factors affecting the sale of each product and their relative importance.

4. Select the forecasting methods.

5. Study the activities of competitors.

6. Prepare preliminary sales estimates after, collecting necessary data.

7. Analyze advertisement policies, sales promotion plans, personal sales arrangements etc. and
ascertain how far these programmes have been successful in promoting the sales.

8. Evaluate the demand forecasts monthly, quarterly, half yearly or yearly and necessary
adjustments should be done.

9. Prepare the final demand forecast on the basis of preliminary forecasts and the results of
evaluation.

10.5 Methods of demand forecasting (for established Products)

There are several methods to predict the future demand. All methods can be broadly classified
into two.
(A) Survey methods,
(B) Statistical methods

(A) Survey methods

Under this method surveys are conducted to collect information about the future purchase
plans of potential consumers. Survey methods help in obtaining information about the desires,
likes and dislikes of consumers through collecting the opinion of experts or by interviewing the
consumers. Survey methods are used for short term forecasting.

Dr. Ramesh Raj Ayer PhD, DMS, MBA, MA, PGDM, PGCP, BSc
Professor, Corporate Mentor & Consultant
Mo: +91 9900704081 Email/LinkedIn: [email protected]
Important survey methods are (a) consumers interview method (b) collective opinion or sales
force opinion c) experts opinion method (d) consumer clinic and (f) end use method.

(a) Consumers' interview/survey method​:

Under this method, consumers are interviewed directly and asked the quantity they would like
to buy. After collecting the data, the total demand for the product is calculated. This is done by
adding up all individual demands. Under the consumer interview method, either all consumers
or selected few are interviewed. When all the consumers are interviewed, the method is known
as complete enumeration method. When only a selected group of consumers are interviewed,
it is known as sample survey method

Advantages
1. ​I​t is a simple method because it is not based on past records.
2. It is suitable for industrial products.
3. The results are likely to be more accurate.
4. This method can be used for forecasting the demand of a new product.

Disadvantages
1. It is expensive and time consuming.
2. Consumers may not give their secrets or buying plans.
3. This method is not suitable for long term forecasting.
4. It is not suitable when the number of consumers is large.

(b)Collective opinion method: ​Under this method the salesmen estimate the expected sales in
their respective territories on the basis of previous experience. Then demand is estimated after
combining the individual forecasts (sales estimates) of the salesmen.This method is also known
as sales force opinion method.

Advantages
This method is simple.
1. It is based on the firsthand knowledge of Salesmen.
2. This method is particularly useful for estimating demand of new products.
3. It utilizes the specialized knowledge of salesmen who are in close touch with the
prevailing market conditions.

Disadvantages
1. The forecasts may not be reliable if the salespeople are not trained.
2. It is not suitable for long period estimation.
3. It is not flexible.
4. Salesmen may give lower estimates to easily achieve sales quotas fixed.

Dr. Ramesh Raj Ayer PhD, DMS, MBA, MA, PGDM, PGCP, BSc
Professor, Corporate Mentor & Consultant
Mo: +91 9900704081 Email/LinkedIn: [email protected]
(c) Experts' opinion method​: This method was originally developed at Rand Corporation
in 1950 by Olaf Helmer, Dalkey and Gordon. Under this method, demand is estimated on the
basis of opinions of experts and distributors other than salesmen and ordinary consumers. This
method is also known as the Delphi method. Delphi is the ancient Greektemple where people
come and prey for information about their future.

Advantages
1. Forecast can be made quickly and economically
2. This is a reliable method because estimates are based on expert’s knowledge & experience
3. The firm need not spare its time on preparing estimates of demand.
4. This method is suitable for new products.

Disadvantages
1. This method is expensive.
2. This method sometimes lacks reliability

(d) Consumer clinics: ​In this method some selected buyers are given certain amounts of money
and asked to buy the products. Then the prices are changed and the consumers are asked to
make fresh purchases with the given money. In this way the consumers``responses to price
changes are observed. Thus the behavior of the consumers is studied.On this basis demand is
estimated. This method is an improvement over consumer's interview method.

Merits
1. It provides an opportunity to study the behavior of consumers directly.
2. It provides a reliable and realistic picture about future demand.
3. It gives useful information to aid in the decision making process.

Demerits
1. It is a time consuming method.
2. Selecting the participants is very difficult.
3. It is expensive.
4. Consumers may take it as a game. They may not reveal their preferences.

(e) End use method: ​This method is based on the fact that a product generally has different
uses. In the end use method, first a list of end users (final consumers, individual industries,
exporters etc.) is prepared. Then the future demand for the product is found either directly
from the end users or indirectly by estimating their future growth. Then the demand of all end
users of the product is added to get the total demand for the product.

Dr. Ramesh Raj Ayer PhD, DMS, MBA, MA, PGDM, PGCP, BSc
Professor, Corporate Mentor & Consultant
Mo: +91 9900704081 Email/LinkedIn: [email protected]
(B) Statistical Methods

Statistical methods use the past data as a guide for knowing the level of future demand.
Statistical methods are generally used for long run forecasting. These methods are used for
established products.
Statistical methods include: (i) Trend projection method,(ii) Regression and Correlation, (iii)
Extrapolation method, (iv) Simultaneousequation method, and (v) Barometric method.

(i) Trend projection method​: Future sales are based on the past sales, because the future is the
grand-child of the past and child of the present. Under the trend projection method demand is
estimated on the basis of analysis of past data. This method makes use of time series (data over
a period of time). We try to ascertain the trend in the time series. The Trend in the time series
can be estimated by using any one of the following four methods:
(a) Least-square method, (b) Free-hand method, (c) Moving average method and (d)
semi-average method.
The Least square method uses the concepts of Intercept & Slope given by the straight line
equation

Y = a + bt Where Y = Sales or Demand a = Constant, b = Slope and t = Time period

For projection over a time series data the following can be used:

∑ tY − ∑ t ∑ Y ∑Y ∑t
b= 2 and a= n –b{ n }
2
n ∑ t − (∑ t )

(ii) Regression and Correlation: ​These methods combine economic theory and statistical
technique of estimation. Under these methods the relationship between the sales (dependent
variable) and other variables (independent variables such as price of related goods,
income,advertisement etc.) is ascertained. Such relationships established on the basis of past
data may be used to analyze the future trend. The regression and correlation analysis is also
called the econometric model building.

The Correlation coefficient ​τ​measures the degree of association between y and x.

The values range from -1 to +1

Dr. Ramesh Raj Ayer PhD, DMS, MBA, MA, PGDM, PGCP, BSc
Professor, Corporate Mentor & Consultant
Mo: +91 9900704081 Email/LinkedIn: [email protected]
 n ∑ xy− ∑ x ∑ y
τ= 2 2

√ [n ∑ x2 −(∑ x) ][n ∑ y 2 −(∑ y ) ]

If more than one independent variable is to be used then unlike the linear model seen in least
squares we use multiple regression,

The general form of multiple regressions is

y​ = β​0 + β​
.​ i​ ​ 1x1,i​ + β​2x2,i +
​ ……………………………..+β​k​x​k,i +
​ e​i

where y​i​ is the variable to be forecast and x​1,i ……………….., ​x​k,i ​ are the k predictor variables. Each of
the predictor variables must be numerical. The co-efficient β​1,………….​β​k measure ​ the effect of
each predictor after taking into account effect of all other predictors. Thus the coefficients
measure the marginal effects of the predictor variables.

As for simple linear regression we make the following assumption for the errors (e​1,……..​e​n​) i.e.

- The errors have mean zero

- The errors are uncorrelated with each other
- The errors are uncorrelated with the predictor x​j,i

It is also useful to have errors normally distributed with constant variance in order to produce
prediction intervals, however this is not required for forecasting.
-
(iii) Extrapolation: ​Under this statistical method, the future demand can be extrapolated by
applying Binomial expansion method. This method is used on the assumption that the rate of
charge in demand in the past has been uniform.

(iv) Simultaneous equation method.-​This involves the development of a complete econometric

model for simulation which can explain the behavior of all the variables which the company can
control. This method is however not very popular.

(v) Barometric technique​: This is an improvement over the trend projection method. According
to this technique the events of the present can be used to predict the direction of change in the
future. Here certain economic and statistical indicators from the selected time series are used
to predict variables. Personal income, non-agricultural placements,gross national income,
prices of industrial materials, wholesale commodity prices,industrial production, bank deposits
etc. are some of the most commonly used indicators.
Dr. Ramesh Raj Ayer PhD, DMS, MBA, MA, PGDM, PGCP, BSc
Professor, Corporate Mentor & Consultant
Mo: +91 9900704081 Email/LinkedIn: [email protected]
Advantages of Statistical Methods
1. The method of estimation is scientific
2. Estimation is based on the theoretical relationship between sales (dependent variable) and
price, advertising, income etc. (independent variables)
3. These are less expensive.
4. Results are relatively more reliable.

Disadvantages of Statistical Methods

1. These methods involve complicated calculations.
2. These do not rely much on personal skill and experience.
3. These methods require considerable technical skill and experience in order to be effective​.

Methods of Demand Forecasting for New Products

Demand forecasting of new products is more difficult than forecasting for existing products.
The reason is that the product is not available. Hence, no historical data are available.
Forecasting is to be done by taking into consideration the inclination and wishes of the
customers to purchase. For research there is one problem as it is difficult for a customer to say
anything without seeing and using the product. Prof. Joel Dean has suggested the following
methods for forecasting demand of new products:
1. Evolutionary approach​: This method is based on the assumption that the new productis the
improved evolution of the old product. The demand is forecasted on the basis of the demand of
the old product. For example, the demand for black and white TVshould be taken into
consideration while forecasting the demand for color TV sets.
2. Substitute approach: ​Here the new product is treated as a substitute of an existing product,
e.g. polythene bags for cloth bags. Thus the demand for a new product is analyzed as a
substitute for some existing goods or service.
3. Growth curve approach​: Under this method the demand of a new product is estimated on
the basis of the growth rate of demand of an existing product.Suppose a new cosmetic like
Pears soap is to be introduced in the market. The average sale of Pears soap will give an idea as
to how the new cosmetic will be accepted by the consumers.
4. Opinion poll approach: ​The demand for a new product can be estimated on the basis of
information collected by polling like/don’t like with consumers.
5. Sales Experience approach: ​Here, the new product is offered as a sample, i.e. by direct mail
or through multiple shops or departmental shops. From The response the total demand is
estimated.
6. Vicarious approach: ​This method consists of surveying consumers' reactions through the
specialized dealers who are in touch with consumers. On the basis of their reports demand can
be estimated.
It is desirable to use a combination of two or more methods in order to get better results.

Dr. Ramesh Raj Ayer PhD, DMS, MBA, MA, PGDM, PGCP, BSc
Professor, Corporate Mentor & Consultant
Mo: +91 9900704081 Email/LinkedIn: [email protected]