Digital Transformation in the PLM domain - distrib.pdf
Demand forecasting
1.
2. Demand forecasting is an estimation of sales in
money or physical units for a specified future
period under a proposed marketing plan.
We can thus define demand forecasting as the
scientific and analytical estimation of demand for
a product(good or service) for a particular period
of time.
3. It is the basis of planning production program.
It is an estimate or a forecast of sales in
future.
It depends on market planning.
It tries to find out lines or profitable
investment.
It is done for a particular period.
It tries to arrange appropriate promotional
efforts, advertisement, sales etc…
4. To produce required quantity.
To access probable demand.
Sales forecasting.
Control of business.
Inventory control.
To plan investment and employment.
To help Govt. to import a export policies.
Man power planning.
To call for team work
5. Appropriate production scheduling.
Helping the firm in reducing costs of purchase of raw
materials.
Determine appropriate price policy to maintain
consistent sales.
Forecasting short term financial requirements.
6. Planning of new unit or expansion of an existing unit.
Planning long term financial requirements.
Planning man-power require.
Planning a suitable statuary to produce goods in
accordance with the changing needs of society.
7. There are two methods of demand forecasting.
Subjective method
consumer’s opinion method-in this method buyers are asked
about their future buying intentions of products.
sales force method-in this method salespersons are asked about
their estimated sales targets in their respective sales territories
in a given period of time.
Expert opinion method (Delphi method)-in this method a
group of experts come to a consensus on the demand of a
particular good(generally a new one).It is less expensive.
Market simulation- Firms may create artificial market where
consumers are instructed to shop with some money.
Test marketing- in this market product is actually sold in certain
segments of the market, regarded as test market.
8. QUANTITATIVE METHOD
trend projection method
a. Secular trend-change occurring consistently over a long
time and is relatively smooth in its path.
b. Seasonal trend-seasonal variations of data within a year.
e.g. demand for woolen, ice cream.
c. Cyclic trend- cyclic movement in demand for a product
that may have a tendency to recur in a few year
d. Random Events-these are natural calamities, social unrest
etc.
Different Methods of trend projection-
a. Graphical method
b. Least square method
9. GRAPHICAL METHOD
This is the simplest technique to determine the trend. All values of
output or sells for different years are plotted on a graph.
Year Sales
1990 30
1991 40
1992 35
1993 50
1994 45
60
50
40
30
20
10
0
90 91 92 93 94
10. LEAST SQUARE METHOD
We can find out the trend values for each of the 5 years and also for the
subsequent years making use of a statistical equation, the method of Least
Squares.
In a time series, x denotes time and y denotes variable. With the passage of
time, we need to find out the value of the variable.
To calculate the trend values i.e., Yc, the regression equation used is-
Yc = a+ bx.
As the values of ‘a’ and ‘b’ are unknown, we can solve the following two
normal equations
simultaneously.
(i) ∑ Y = Na + b∑x
(ii) ∑XY = a∑x + b∑ x2
Where,
∑Y = Total of the original value of sales ( y)
N = Number of years,
∑X = total of the deviations of the years taken from a central period.
∑XY = total of the products of the deviations of years and corresponding sales
(y)
∑x2 = total of the squared deviations of X values .
When the total values of X. i.e., ∑X = 0
11. EXAMPLE
Year = n Sales in Deviation Square of Product Computed
Rs Lakhs from Deviation sales trend
Y assumed X2 and time values Yc
year X Deviation
XY
1990 30 -2 4 -60 32
1991 40 -1 1 -40 36
1992 35 0 0 0 40
1993 50 1 1 50 44
1994 45 2 4 90 48
N=5 ∑Y=200 ∑X=0 ∑x2=10 ∑XY=40
CALCULATION
Regression equation = Yc = a + bx
To find the value of a = ∑Y/N = 200/5 = 40
To find out the value of b = XY/ ∑x2 = 40/10 = 4
12. For 1990
Y = 40+(4*-2)
Y = 40-8= 32
For 1991
Y = 40+(4*-1)
Y = 40-4= 36
For 1992
Y = 40+(4*0)
Y = 40+0 = 40
For 1993
Y = 40+(4*1)
Y = 40+4 = 44
For 1994
Y = 40+(4*2)
Y = 40+8 = 48
For the next two years, the estimated sales would be:
For 1995
Y = 40+(4*3)
Y = 40+12 = 52
For 1996
Y = 40+(4*4)
Y = 40+16 = 56
13. Barometric techniques
In barometric forecasting we construct an index of a relevant
economic indicator and forecast future trends on the basis of
these indicators.
Regression analysis
Regression analysis relates a dependent variable to one or
more independent variable in the form of linear equation.
Y = a+bx
Where , y indicates future demand
a indicates fixed demand
b indicates rate of change of demand
x indicates value of related variables like price,income
of consumer,price of related commodity etc.
14. Change in fashion- it is an inevitable consequence of
advancement of civilisation.Results of demand forecasting
have short lasting impacts especially in a dynamic business
environment
Consumers’ psychology-results of forecasting depend
largely on consumers’ psychology, understanding which itself
is difficult.
Uneconomical-forecasting requires collection of data in
huge volumes and their analysis, which may be too expensive
for small firms to efforts.
Lack of experts-accurate forecasting necessitates
experienced experts who may not be easily available.
Lack of past data-demand forecasting requires past sales
data, which may not be correctly available.