What Do You Mean by Demand Forecasting? What Are Its Various Types?
What Do You Mean by Demand Forecasting? What Are Its Various Types?
What Do You Mean by Demand Forecasting? What Are Its Various Types?
What do you mean by demand forecasting? What are its various types?
A forecast is an estimate of a future situation. Forecasting demand devotes an
estimation of the level of demand of the product at a future period under given circumstances.
It helps a firm to assess the possible demand for its products and plan its production
accordingly.
Types:
Short – term demand forecasting:
This is limited to short period not exceeding one year. It concern with policies relating
to sales, purchases, pricing and finance. This is useful in taking ad hoc decision concerning
the day to day working of the concern
Medium term demand forecasting:
Medium term forecasting is intermediate between the short term and the long term
situations. Its need is felt by a firm when the industry to which the firm belongs, is subjected
to the trade cycle of a medium term (varying between say; two to five years). Engineering
goods industries and government manufacturers often find such patter of demand behaviour in
the market.
Long – term forecasting:
It involves the assessment of long - term demand for the product. It involves the study
of technological developments, economic trends and consumer preferences and manpower
planning. It enables to take major strategic business decisions.
Delphi method:
Under this method, a panel is selected to give suggestions to solve the problems in hand.
Both internal and external experts can be the members of the panel. Panel members are kept
apart from each other and express their views in an anonymous manner. There is also a
coordinator who acts as an intermediary among the panelists. He prepares the questionnaire
and sends it to the panelists. At the end of each round, he prepares a summary report. On the
basis of the summary report the panel members have to give suggestions. This method has
been used in the area of technological forecasting. It has proved more popular in forecasting
non-economic rather than economic variables.
STATISTICAL METHOD:
In this method, some statistical or mathematical techniques are used to predict the
demand. This method is useful for long – run forecasting and for the products already in the
market.
Time series analysis or Trend projection method:
In the time series or trend projection method past data of sales are used to forecast
future. This method is used to forecast the demand for a product whose past sales records are
available for a number of years. These data will be analyses in order to establish the nature of
trend in sales over a period. So that the possible trend in the future can also be inferred.
In order to find out the fluctuations in sales, it has got four types of components, namely,
Secular trend, Seasonal variation, Cyclical variations and Random variations. For e.g., if a
manufacturer of T.V. decides to forecast the next year sales of his product by this method, he
collects data about his sales for the past five years.
If the sales are plotted in a graph it shows an upward trend during the whole period.
This line drawn is called the ‘trend line’. The trend line is fitted by developing an equation to
find out nature and magnitude of this trend. The common method to construct the line of best
fit is by the method of least squares.
The equation of this line is Y = a + bx;
Where,
‘a’ is the intercept and ’b’ shows the impact of the independent variable.
Y = na + bx
XY= ax + bx
Moving averages:
This method is based on the assumption that the future is the average of past
achievements. Hence based on past achievements future is predicted. When the demand is
stable this method can provide good forecasts.
Exponential smoothing:
In the technique of moving averages, all time period s are weighted equally. But
recent observations should be given more importance as they contain the most important
observations than older observations. This satisfied by exponential smoothing by giving
more weights to the recent observations and decreasing weights to the older values.
Regression and Correlation method:
Regression and correlation are used for forecasting demand. Based on past data the
future trend is forecasted. If the functional relationship is analyzed with one independent
variable it is simple correlation. When there are several independent variables it is multiple
correlation. This method is also known as Econometric Model Building. E.g., the relationship
between sales and other variables.
Barometric technique:
Under barometric method, present events are used to predict the direction of change in
future. This is done with the help of economic and statistical indicators.
Problems:
1. Calculate price elasticity of demand for the following data.
Q = 2,000 P = 10
Q1 = 2,500 P1 = 9
Where,
Q - refers to initial demand
Q1 – refers to changed demand
P – refers to initial price
P1 – refers to changed price
2. Suppose the price of a commodity decreases from Rs. 200 to Rs. 120, as a result, demand
increases from 10,000 units to 15,000 units. Calculate price elasticity of demand.
3. DD is the demand curve of a consumer for good X. At price Rs. 10, 4 units of goods X
are demanded. When price goes down to Rs. 8, quantity demanded increases to 6. Find price
elasticity of demand.
4. If a consumer’s demand for a commodity increases from 100 units per week to 200 units
per week, when his income rises from Rs. 2000 to Rs. 3000. Find his income elasticity of
demand.
5. Suppose a consumer’s income is Rs. 1,000 and he purchases 10 kilos of sugar. If this
income goes up to Rs. 1,100, he is prepared to buy 12 kilos of sugar. Calculate income
elasticity of demand.
6. The price of coffee increases from Rs. 50 per kg to Rs. 70 per kg, and as a result the
demand for tea increases from 5 kg to 10 kg. What is the cross elasticity of demand of tea for
coffee?
7. Based on the trend projection equation, find out the trend value and forecast the sales for
the year 1994 by using the data given below.
Sales of the firm
Year Sales in 000’s
1989 60
1990 67
1991 62
1992 73
1993 80
8. The following date refer to sales, in thousands of rupee (X) of a certain product during
five years.
Sale of the industry
Year Sales in 000’s
1993 605
1994 715
1995 830
1996 790
1997 835
Assuming the present trend continues, in which year will you expect 1994 sales to be
doubled.