DEMAND

FORECASTING
Opening case
 In recent years, unilever china in its business
strategy responded very closely to the complex
needs of consumer in china. The Chinese consumers
have typical preferences with typical taste and local
needs. Modern buyers have also become quality
conscious .
 By studying the demand pattern in Chinese economy,
unilever develop a portfolio of brands of its product-
considering traditional Chinese outlook and trend of
technological improvements in the country.
Opening case
 With its joint venture and large scale consolidation
in 1999, the company brand folio contained a mix
of global and local brands.
 Dove, lux, ponds are the global brands of unilever
promoted and managed in china. Unilever’s
expertise and resources are extended to the
development of local brands
 In short, unilever china sought to balance the
global and local market segments in satisfying the
distinct demand composition of the consumers.
Introduction
 Forecasting Planning is the most important
function of managing.
 Planning is thinking before doing. It is done to
minimize the risks arising out of an uncertain
future.
 The risks associated with uncertain future can
be negated if one tries to make reasonable
assumptions about the course that the future is
likely to take.
 Such an estimation of the future situation is
known as forecasting.
Demand Forecasting
 Demand forecasting essentially involves
ascertaining the expected level of
demand during the period under
consideration.
 reasonable judgment of future based on
scientific background
 Cant be 100 % precise
 Reliable approximation regarding the
possible outcome.
Level of Demand Forecasting
 Micro level :
 forecasting by individual business firm
 Industry Level:
 demand estimate for the product of the
industry , related to market demand
 Macro level :
 Aggregate demand for the industrial
output by the nation
Significance of Demand Forecasting
 Product planning
 Sales forecast
 Control of business
 Inventory control
 Growth and long term investment
programmes
 Stability
 Economic planning and policy making
Short term and long term Forecasting
 Short term demand forecasting
 Evolving a sales policy
 Determining price policy
 Purchase policy
 Fixation of sales target
 Short term financial planning
Long term
 Business planning
 Manpower Planning
 Long term financial planning
Stages in Forecasting Demand
 Specification of objective(s)
 Selection of appropriate technique
 Collection of appropriate data
 Estimation and interpretation of results
 Evaluation of the forecasts
FORECASTING TECHNIQUES

 Qualitative and quantitative Techniques

 Accuracy
 Least possible cost
 Minimum use of other resources
 Choice of technique depends on-
Urgency and Availability of data
classification of Forecasting Techniques

 Qualitative techniques
 Obtain information about the likes and
dislikes of consumers.
 Suited for short term demand
forecasting.
 Demand forecasts for new products can
be made only by qualitative technique.
Quantitative Techniques
 Forecast future demand by using quantitative
data from the past and extrapolating it to make
forecasts of future levels.
 This is suited for long term demand forecasting.
 Forecasts for new products, for which no past
data is available, cannot be made by
quantitative techniques as extrapolation is not
possible.
 Demand for existing products can be forecast
by using this method also.
Sources of Data Collection
 Primary data :
 Original in character which are collected
for the first time for the purpose of
analysis
 Secondary data:
 Obtained from someone else record.
Already in existence in recorded or
published form
Secondary Data Sources
 Official publication from central and state
government
 – annual survey of industries, economic
survey, census on India
 Economic journals
 – stock exchange directory, economic and
political weekly
 Official publication from reserve bank of India
 Annual report on currency and finance
DIFFERENT TYPES OF QUALITATIVE TECHNIQUES

 Expert opinion method :

 Views of experts on the likely level of

demand in the future are sought.
 Advantages of expert opinion method:
 Simple to conduct
 Can be used where quantitative data is
not available
 The forecast is reliable as it is based on
the opinion of people who know the
product very well
 It is inexpensive
 Takes less time
 Disadvantages of using expert opinion
method:
 Results are based on mere hunch of
one or more persons and not on
scientific analysis.
 The experts may be biased
 This method is subjective and the
forecast could be unfavorably influenced
by persons with vested interests
 Panel consensus:

 If the forecasting is based on the

opinion of several experts, then it is
known as panel consensus. This kind of
forecasting minimizes individual
deviations and personal biases.
 Delphi Method:
 Olaf Helmer originated Delphi Method in late 1940’s
 Under this a group of experts are repeatedly questioned
for their opinion or comments on some issues and their
agreements and disagreements are identified.
 The responses received are analyzed by an independent
body.
 This method thus takes care of the disadvantages of
panel consensus where some powerful individual could
have influenced the consensus.
 Forecasting is done by analyzing the responses of
experts.
 Consumers complete enumeration
survey

 This method is based on a complete

survey of all the consumers for the
commodity under consideration

 Interviews or questionnaires are used

 Advantages

 Quite accurate
 Simple to use
 Not affected by personal biases
 Disadvantages
 Costly
 Time consuming
 Difficult and impossible to survey all the
consumers
 Open to faulty recording and wrong
interpretation
 Can be used only for products with
limited consumers
 Consumers Sample Survey
 Miniature form of complete enumeration
method

 Only few customers selected and their

views collected
 Advantages

 An important tool especially for short

term projections
 Simple; does not cost much
 Works quickly
 Gives excellent results, if used carefully
 Disadvantages

 Conclusions are based on the view of

only a few consumers and not all of
them
 Sample may not be a true
representation of the entire population
 Sales Force Opinion Survey

 Employees of the Company who are a

part of the sales and marketing teams
are asked to predict future levels of
demand.
 Advantages

 Simplest of the forecasting methods

 Less costly
 Easy to collect data
 Disadvantages:

 Consumers’ tastes and preferences keep

changing.

 Past trend may not continue in the future.

Opinion of sales force may thus be erroneous.
 Sales force may give biased views as the
projected demand affects their future job
prospects.
QUANTITATIVE TECHNIQUES
 These are forecasting techniques that
make use of historical quantitative data.
 A statistical concept is applied to the
existing data in order to generate the
predicted demand in the forecast period.
 These quantitative techniques are also
known as statistical methods.
1. Trend Projection Method:

 A time series analysis of sales data over

a period of time
 Assumes that past trend will continue in
future.
 Method used:
 Method of moving averages
 The least square method
Trend Projection Method

 Time series analysis Sales of a

commodity over the past 15 to 20 years
are plotted on a graph.

 The year to year oscillations are

smoothened and a trend line is fitted
using a statistical method.
Advantages

 Very simple method

 Provides reasonably accurate forecasts
 Quick and inexpensive
 Disadvantages
 Can be used only if past data is
available
 It is not necessary that past trend may
continue to hold good in future as well
2. Barometric Forecasting

 At times, a business concern may

assign the task of demand forecasting to
some expert agency
 It would attempt to forecast on the basis
of signals received from the policies
adopted or the events that had taken
place within the country or in other
countries.
  Advantages
 Simple method
 Predicts directional changes quite
accurately
 Disadvantages
 Does not predict the magnitude of
changes very well
 The method can be used for short term
forecasts only
Econometric Techniques

 It is assumed that demand is determined

by one or more variables e.g. income,
population, exports, etc.
 Demand is forecast on the basis of
systematic analysis of economic
relations by combining economic theory
with mathematical and statistical tools.
  Advantages
 Produces reliable and accurate results
 Forecasts not only the direction but also
the magnitude of the change
 Disadvantages
 Uses complex calculations
 Costly and time consuming
Criteria of a good demand forecasting
 Accuracy : can be measured by the degree of
deviations between actual and predicted b) the
extent of success in forecasting directional
changes .
 Economy : the results achieved by forecasting
must be weighted against the cost of the
method
 Availability : results must be readily available
and well understood, forecast must fully
understand the techniques, assumptions,
constraints and probability
 Timeliness: gap between occurrence of
the event and its forecast , known as
Lead time
 Lead time should be such that there is a
possibility to make adjustments with
upcoming market trends
 Simplicity : mathematical techniques
used must be accurate and simple to
understand by masses
SIGNIFICANCE OF DEMANDFORECASTING

 Useful to following decision makers:

 Producers
 To allocate various factors of production
for maximization of profit
 To plan for expansion of scale of
operations
 Policy makers and planners
 To formulate economic policies through
planning commission for allocating
resources
 To ensure adequate supply of inputs for
achieving the objectives of industrial policy,
import export policies, credit policy, public
distribution system and other related
policies Other groups of the society

 Useful to researchers, social workers
and others with futuristic approach

 To understand the gaps in supply and

their impact on prices or the economy
Demand Forecasting for Existing Product

 By taking past forecast and present

demand conditions
 Demand for immediate future is possible
by analyzing present condition, provided
if the demand remain more or less
stable
 By extending the trend fitted into the
data for the past years like moving
averages, least squares method
 Input- output tables used for
forecasting , based on techniques
coefficient which indicate the relations
between inputs and outputs
 Direct methods like opinion surveys of
the customers
 Use of indirect method like expert
opinion etc
Demand Forecasting for New Product
 Survey of buyer intention (or consumer
survey )

 Test marketing

 Life cycle segmentation analysis (introduction

, growth, maturity, saturation and decline )
 Bounding method (marker data share
method )
Market Experimentation
 Market experimentations may be conducted to
make certain observations.

 Experimentation in laboratory ( consumer clinic

method. Small laboratory is formed by creating
an artificial market situation to study consumer
reactions in controlled conditions. Consumer
are given money and ask to buy items.
 Expensive and time consuming
 Test marketing ( performed under actual
market conditions )
 Business firms conducts experiment in
market by changing one or two variables.
 Consumer are then observed and
recorded.
 Unlike consumer surveys, market
experimentation are estimated based on
actual purchase.
Methods of measuring Trend( statistical
method)
 1. FREE HAND CURVE
 simplest method of trend fitting.
 Original data of time series is plotted on
graph.
 Taking care of the fluctuations, a smooth
line curve is drawn which passes
through mid points of the fluctuations of
time series.
Fit the trend line
Year 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

Prod( 150 155 165 152 174 150 174 175 160 180
000).
2. Semi – average methods
 Time series is divided in two equal parts,
separate arithmetic mean is calculated
for each part.
 Values are plotted to obtain trend line
 Case 1 : when no is even

Year 2000 2001 2002 2003 2004 2005 2006 2007

Sales 412 438 444 454 470 482 490 500

Case 2 :
when no is odd
 Mid year figure will be dropped

Year 1981 1982 1983 1984 1985 1986 1987

Profit(0 20 22 27 26 30 29 40
00)
3. Moving average method
 Widely used method
 Case 1 : odd period moving average say
3 , 5 , 7 years
Year 1981 1982 1983 1984 1985 1986 1987

Product 412 438 446 454 470 483 490

ion
 Case 2 : even period moving average

Year 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979
Sale 7 8 9 11 10 12 8 6 5 10
s
(cror
e)
4. Least square method
 Best method of fitting trend
 Mathematical method obtained in such a
way the 2 conditions are fulfilled
 1. ∑(Y-Yc)=0, sum of deviations of
actual values of Y and computed trend
values is zero
 2. ∑(Y-Yc)²=least sum of square of
deviations is least.
Fitting straight line trend
 Y=a+bX,
 Y = trend values
 X= Unit of time
 a is Y intercept
 b is slope
 ∑Y=Na+b∑X
 ∑XY=a ∑X +b∑X²
Fit a straight line (using 1978
as year of origin), obtain trend
values
Year 1979 1980 1981 1982 1983 1984
Prod. 5 7 9 10 12 17
Consumption level Method
 Estimated on basis of coefficient of
income elasticity and price elasticity of
demand
 Income elasticity
 D*=D(1+M*em)
 D*- projected per capita demand
 D- per capita Demand
 M*- % change in per capita income
 Em- income elasticity of demand
 Suppose the income elasticity of
demand for chocolates is 3 . In the year
1995, per capita income is 500$ and per
capita annual demand for chocolates is
10 million in a city .
 It is expected that in the year 2000 per
capita income will increse by 20% .
Calculate the projected demand for
2000.
 For Price elasticity
 D*= D(1+P*.e)
 D*- projected demand
 D- present demand
 P*- % change in price
 E- price elasticity of demand
 Present market demand for a commodity X is
2000 kg at Rs. 10 per kg. its price elasticity is
2. suppose price declines to 5, then the
expected change in demand is :