Forecasting of demand is the art of predicting demand
for a product or a service at some future date on the
basis of certain present and past behavior patterns of
some related events.
Forecasting is used in process design, capacity and
facilities planning, aggregate planning scheduling
inventory management etc
Types of forecasts
There are long-term forecasts as well as short-term
Operations managers need long-range forecasts to
make strategic decisions about products, processes
and facilities. Long tem forecasts are used to make
location, layout,and capacity decisions.
They also need short-term forecasts to assist them
in making decisions about production issues that
span only the next few weeks.
Since forecasting forms an integral part of
planning and decision-making , production
managers must be clear about the horizon of
forecasts-month or year, for example, additionally
they must also be clear about the method of
forecasting and unit of forecasting
Importance of Demand forecasting
Determination of sales territory.
To decide to enter a new market or not.
To determine how much production capacity to be
Helpful in deciding the number of salesman
required to achieve the sales objective.
To prepare standard against to which measure
To assess the effect of a proposed marketing
Helpful in the product mix decisions.
To decide the promotional mix.
To assess the effect of a proposed marketing
In deciding the channels of distribution and
physical distribution decision.
Criteria of good forecasting
Simplicity and ease of comprehension.
Methods of demand forecasting
I. Opinion polling.
II. Statistical method.
Opinion polling method:
1. Consumer survey method.
2. Sales force opinion method.
3. Delphi method.
Consumer survey method:
Sample survey test.
End use method.
Time series analysis.
Simultaneous equation method.
Time series analysis:
Chronologically arranged continuous past data.
Trend analysis method.
Least square method.
Moving average method
Exponential smoothing.:α Dt-1+(1- α)Ft-1
α = Exponential smoothing constant (0 to1).
Ft= forecasting period.
Dt-1= Actual demand for periodt-1.
Eg: F July= α DJune + (1- α)F June
Strategies for developing aggregate plans:
The aggregate plan is developed after careful
consideration of the different Variables which
influence the production plan.
Similarly the aggregate plan also influenced by no. of
Trend projection method:
These are generally based on analysis of past sales
Least squares method: certain statistical formulae
are here to find the trend line which best fits the
The trend line is the basis to extrapolarate the line for
future demand for the given product or service on
MOVING AVERAGE METHOD:
This method is based on the assumption that the
future is the average of past achievements.
Hence based on past achievement, future is
When the demand is stable this method can
provide good forecasts.
The main issue in moving averages is determining
the ideal number of periods to include the average.
Customer needs demand forecasts competition.
Financial conditions of the firm.
Labour training capacity.
New products product design changes Machines.
Suppliers capability storage capacity material
Machine capacity, workforce capabilities.
Under the barometric technique one set of data is used
to predict another set.
In other words, to forecast demand for a particular
product or service, use some other relevant indicator
of future demand.
Ex: the demand for cable TV may be linked to the
number of new houses occupied in a given area.
Simultaneous equation method
In this method, all variables are simultaneously
considered with the conviction that every variable
influences the other variables in an economic
It is a system of ‘n’ unknowns. It can be solved, the
moment the model is specified because it covers all
the unknown variable. It is also called complete
systems approach to demand forecasting.
Correlation and regression
Correlation and regression methods are statistical
Correlation describes the degree of association
between two variables such as sales and advertisement
expenditure. When two variables are tend to change
together, then they are said be correlated.
The extent to which they are correlated is measured by
Of these two variables, one is a dependent variable and
the other is independent variable.
An equation is estimated which best fits in the sets of
observations of dependent variables and independent
The best estimate which best fits in the sets of
observation of dependent variables and independent
the best estimate of the underlying relationship
between these variables is thus generated.
The dependent variables is then forecast based on this
estimated equation for a given value of the