DEMAND forecasting
What is meant by Forecasting and Why?
• Forecasting is the process of estimating a variable,
such as the sale of the firm at some future date.
• Forecasting is important to business firm,
government, and non-profit organization as a
method of reducing the risk and uncertainty
inherent in most managerial decisions.
• A firm must decide how much of each product to
produce, what price to charge, and how much to
spend on advertising, and planning for the growth
of the firm.
The aim of forecasting
• The aim of forecasting is to reduce the risk or
uncertainty that the firm faces in its short-
term operational decision making and in
planning for its long term growth.
• Forecasting the demand and sales of the firm’s
product usually begins with macroeconomic
forecast of general level of economic activity
for the economy as a whole.
Contd...
• The firm uses the macro-forecasts of general economic
activity as inputs for their micro-forecasts of the
industry’s and firm’s demand and sales.
• The firm’s demand and sales are usually forecasted on
the basis of its historical market share and its planned
marketing strategy (i.e., forecasting by product line and
region).
• The firm uses long-term forecasts for the economy and
the industry to forecast expenditure on plant and
equipment to meet its long-term growth plan and
strategy.
Elements of a Good Forecast
• Timely
• Reliable • Accurate
e
f ul us
n g to
ni • Written y
s
ea Ea
M •
•
Steps in the Forecasting Process
“The forecast”
Step 6 Monitor the forecast
Step 5 Make the forecast
Step 4 Obtain, clean and analyze data
Step 3 Select a forecasting technique
Step 2 Establish a time horizon
Step 1 Determine purpose of forecast
6
NEED OF DEMAND FORECASTING
• New facility planning
• Production planning
• Workforce scheduling
• Financial planning
Forecasts by Time Horizon
• Short-range forecast
– Up to 1 year (usually less than 3 months)
– Job scheduling, worker assignments, plan for
purchasing
• Medium-range forecast
– 1 Year to 3 years
– Sales & production planning, budgeting
• Long-range forecast
– 3 years, or more
– New product planning, facility location
Methods of Demand Forecasting
Methods of Demand Forecasting
• Qualitative Methods
– Survey Method
– Historical Analogies
– Collective Opinion (sale force opinion Method)
– Delphi Technique
– Test marketing
• Expert Opinion
• The expert opinion method, also known as expert consensus method, is being
widely used for demand forecasting.
• This method utilizes the findings of market research and the opinions of
management executives, consultants, and trade association officials, trade journal
editors and sector analysts.
• When done by an expert, qualitative techniques provide reasonably good forecasts
for a short term because of the expert's familiarity with the issues and the
problems involved.
• There are various methods of confirming the opinion regarding future demand by
experts. One of these methods is the Delphi method.
• Delphi method: In the Delphi method, the opinion
of number of experts is gathered individually.
• An analyst combines these forecasts using same
weighting system and passes on the combined
forecast to the forecasters.
• The forecasters make a new round of forecasts
with this information. The process continues till an
overall consensus is arrived at from all the panel
members.
• The Delphi method is primarily used to forecast
the demand for new products. Though it provides
valuable insights, it is rather expensive as the
experts may charge a high fee for their opinion.
• Survey:
Information can also be collected through surveys. Surveys can be carried out through
mail, e-mail, telephone or by directly speaking to respondents.
• Survey bye-mail an phone survey are conducted for prospective customers who are not
using a particular product or service of a company.
• A firm can determine the demand for its products through a market survey. It may
launch new products, if the survey indicates that there is a demand for that particular
product in the market.
• For example, Coke in India expanded its product range beyond carbonated drinks, after
the company conducted a nationwide survey. The survey revealed that about 80% of
the youth preferred to drink tea or coffee rather than carbonated drinks at regular
intervals. The remaining 20% preferred to have milk products, while only 2% preferred
to drink carbonated drinks like Coke.
• The survey results helped Coke to expand its product range. Coke introduced coffee and
tea in the
Market Experiment
• One major problem with the survey method is
that people may not reveal their true likes and
dislikes while giving responses.
• Responses to direct questions may not always be
correct. Market experiments can help to
overcome these problems as they generate data
before introducing a product or implementing a
policy.
• Market experiments are of two types: 1. Test
Marketing , & 2. Controlled Experiments.
• Test marketing : In this case, a test area is selected, which
should be a representative of the whole market in which the
new product is to be launched.
• A test area may include several cities and towns, or a
particular region of a country or even a sample of consumers.
• By introducing the new product in the test area consumers'
response about the product can be judged. More than one
test area can be selected if the firm wants to assess the
effects on demand due to various alternative marketing mix
i.e. changes in price, advertising or packaging can be done in
various market areas.
• Then the demand for the product can be compared at
different levels of price and advertising expenditure. In this
way, consumer's response to change in price or advertising
can be judged.
• Some of the drawbacks of test experiments are that they
are very costly and much time consuming. If in a test
market prices are raised, consumers may switch to the
competitor's products. It may be difficult to regain lost
customers even if the price is reduced to the previous
level.
• Controlled experiments: Controlled experiments are
conducted to test the demand for a new product
launched or to test the demands for various brands of a
product.
• In this method, a sample of consumers which are
representative of the target market are selected. They
are requested to visit the store of that firm where various
varieties or brands of the product are kept for sale.
• Their preferences are recorded. They are then provided advertising
materials for various brands.
• The selected consumers are given some fixed money and are
allowed to make purchases of different variety of products or
various brands of a product.
• The quantity of the product or particular brands of a product
purchased by them is recorded. They are also requested to fill a
questionnaire asking reasons for the choices they have made. The
price of the product or its variety may be changed and the
experiment is repeated.
• Controlled experiments provide more accurate results than
consumer surveys. This is because in this case, the consumers are
asked to make actual decisions regarding their purchases, while
the consumer surveys show their intention to buy.
• Some of the drawbacks of controlled experiments are that they
may be biased in the process of selection of a sample of consumers
on which experiments is to be performed.
Quantitative Methods
• Time Series
• Moving averages
• Leading Indicator method
• Correlation and regression Equations
• Extrapolation
Quantitative techniques
• Time Series Analysis:
• Here, the past sales and demand are taken into consideration. The
time series analysis is divided into four categories - trends, seasonal
variations, cyclical variations and random fluctuations.
• In trend analysis, past data is used to predict the future sales of a
firm. A trend is a long term increase or decrease in the variable.
• Seasonal variations take into account the variations in demand
during different seasons. For example, the sale of the cotton dresses
increases in summer, while the sale of woolen clothes increases in
winter.
• Cyclical variations are the variations in demand due to fluctuations
in the business cycle - boom, recession and depression.
• Random fluctuations may happen due to natural calamities like
flood, earthquake, etc. which cannot be predicted accurately.
• There are two types of time-series analysis - moving average and
exponential smoothing.
• The moving average is a series of arithmetic averages and can be
divided into simple moving average and weighted moving average.
For example, to predict sales for the next period using a simple
moving average.
• On the other hand, exponential smoothing works on a premise that
the most recent occurrences are more indicative of the future than
the past ones.
Commonly observed “normal” demand
patterns
Constant demand Linear trend
D D
t t
Cyclic trend Seasonal pattern
D D
t t
Moving average
• Short range time series type of forecasting
model that forecasts sales for the next time
period.
• In this model the arithmetic average of the
actual sales for a specific number of most
recent past time periods is the forecast for the
next time period.
Category of Forecasting Method
Moving Average
Month Demand 3 Months MA 6 Months MA
Jan 500
Feb 525
Mar 575 533.33
Apr 560 553.33
May 600 578.33
Jun 615 591.67 562.50
Jul 595 603.33 578.33
Aug 609 606.33 592.33
Sept 696 633.33 612.50
Oct 690 665.00 634.17
Nov 700 695.33 650.83
Dec 900 763.33 698.33
• Barometric Analysis: Barometric analysis or forecasting
can be defined as "the prediction of turning points in one
economic time series through the use of observations on
another time series called the barometer or the indicator."
• In barometric analysis, the economic time series are
divided into three groups - leading indicators, coincident
indicators and lagging indicators.
• Leading indicators contain data that move ahead of the
series in question. The growing number of senior citizens is
a leading series for the demand for home for the aged.
Leading indicators index includes such things as average
weekly hours worked and claims for unemployment
insurance, manufacturers' new orders, stock prices, orders
for plant and equipment, index of consumer expectations
etc.,
• Coincident Indicators use data that move up and down
corresponding to some other series - for example, the
relationship between national income and employment (in
the short term).
• Components of an Index of Coincident indicators are
employees on non agricultural payrolls, industrial
production, personal income minus transfer payments,
manufacturing and trade sales.
• Lagging indicators move behind the series in question - for
example, manufacturer's inventory is a lagging series for
sales.
• The lagging indicator composite includes changes in labor
costs per unit, ratio of inventory to sales, and figures on
instalment credit and loans, among other items.
Linear Regression
• A model uses least-square method to identify the relationship
between a dependent variable and one or more independent
variables that are present in a set of historical observations.
• In simple regression there is only one independent variable.
• In multiple regression there is more than one independent
variable.
• If the historical data set is a time series, the independent
variable is the time period and the dependent variable in
sales forecasting is sales.
• A regression model does not have to be based
on a time series.
• In that case, the knowledge of future values of
the independent variable known as causal
variables is used to predict future values of
the independent variable.
• Linear regression is ordinarily used in long
range forecasting.
Exponential smoothing
• Short range forecasting model that forecasts
sales for the next time period.
• Forecast sales for the last period is modified
by information about the forecast error of the
last period.
• This modification of the last period’s forecast
is the forecast for the next time period.
Exponential smoothing with trend
• The exponential smoothing model but
modified to accommodate data with a trend
pattern.
• Such patterns can be present in medium range
data.
• Also called double exponential smoothing