Demand Forecasting Demand Forecasting

Accurate demand forecasting is essentiaI for a firm to Accurate demand forecasting is essentiaI for a firm to
enabIe it to produce the required quantities at the right enabIe it to produce the required quantities at the right
time and arrange weII in advance for the various factors time and arrange weII in advance for the various factors
of production, viz., raw materiaIs, equipment, machine of production, viz., raw materiaIs, equipment, machine
accessories, Iabour, buiIdings, etc. accessories, Iabour, buiIdings, etc.
Factors invoIved in Demand Forecasting Factors invoIved in Demand Forecasting
1. 1. How far ahead? How far ahead?
a. Long term a. Long term - - eg., petroIeum, paper, shipping. TacticaI eg., petroIeum, paper, shipping. TacticaI
decisions. Within the Iimits of resources aIready avaiIabIe. decisions. Within the Iimits of resources aIready avaiIabIe.
b. Short b. Short- -term term - - eg., cIothes. Strategic decisions. Extending or eg., cIothes. Strategic decisions. Extending or
reducing the Iimits of resources. reducing the Iimits of resources.
Factors invoIved in Demand Forecasting Factors invoIved in Demand Forecasting
2. 2. Undertaken at three IeveIs: Undertaken at three IeveIs:
a. a. Macro Macro- -IeveI IeveI
b. b. Industry IeveI eg., trade associations Industry IeveI eg., trade associations
c. c. Firm IeveI Firm IeveI
3. ShouId the forecast be generaI or specific (product 3. ShouId the forecast be generaI or specific (product- -wise)? wise)?
4. ProbIems or methods of forecasting for "new" vis 4. ProbIems or methods of forecasting for "new" vis- - - -vis vis
"weII estabIished" products. "weII estabIished" products.
5. CIassification of products 5. CIassification of products - - producer goods, consumer producer goods, consumer
durabIes, consumer goods, services. durabIes, consumer goods, services.
6. SpeciaI factors pecuIiar to the product and the market 6. SpeciaI factors pecuIiar to the product and the market - - risk risk
and uncertainty. (eg., Iadies' dresses) and uncertainty. (eg., Iadies' dresses)
Purposes of forecasting Purposes of forecasting
Purposes of short Purposes of short- -term forecasting term forecasting
a. a. Appropriate production scheduling. Appropriate production scheduling.
b. b. Reducing costs of purchasing raw materials. Reducing costs of purchasing raw materials.
c. c. Determining appropriate price policy Determining appropriate price policy
d. d. Setting sales targets and establishing controls and incentives. Setting sales targets and establishing controls and incentives.
e. e. Evolving a suitable advertising and promotional campaign. Evolving a suitable advertising and promotional campaign.
f. f. Forecasting short term financial requirements. Forecasting short term financial requirements.
Purposes of long Purposes of long- -term forecasting term forecasting
a. a. Planning of a new unit or expansion of an existing unit. Planning of a new unit or expansion of an existing unit.
b. b. Planning long term financial requirements. Planning long term financial requirements.
c. c. Planning man Planning man- -power requirements. power requirements.
Demand forecasts of particular products form guidelines for related Demand forecasts of particular products form guidelines for related
industries (eg., cotton and textiles). Also helpful at the macro industries (eg., cotton and textiles). Also helpful at the macro
level. level.
riteria of a good forecasting method riteria of a good forecasting method
1. 1. Accuracy Accuracy -- measured by (a) degree of measured by (a) degree of
deviations between forecasts and deviations between forecasts and
actuals, and (b) the extent of success in actuals, and (b) the extent of success in
forecasting directional changes. forecasting directional changes.
2. 2. Simplicity and ease of comprehension. Simplicity and ease of comprehension.
3. 3. Economy. Economy.
4. 4. Availability. Availability.
5. 5. Maintenance of timeliness. Maintenance of timeliness.
#oIe of Macro #oIe of Macro- -IeveI forecasting in IeveI forecasting in
demand forecasts demand forecasts
'arious macro parameters found useful for demand 'arious macro parameters found useful for demand
forecasting: forecasting:
1. 1. National income and per capita income. National income and per capita income.
2. 2. Savings. Savings.
3. 3. Investment. Investment.
4. 4. Population growth. Population growth.
5. 5. Government expenditure. Government expenditure.
6. 6. Taxation. Taxation.
7. 7. redit policy. redit policy.
Methods of demand forecasting Methods of demand forecasting
1. 1. Survey of buyers` intentions Survey of buyers` intentions
2. 2. Delphi method Delphi method
3. 3. Expert opinion Expert opinion
4. 4. ollective opinion ollective opinion
5. 5. Trend projection method Trend projection method
6. 6. Smoothing techniques Smoothing techniques
a. a. Moving average Moving average
b. b. Exponential smoothing Exponential smoothing
1. 1. Analysis of time series and trend projections Analysis of time series and trend projections
2. 2. Use of economic indicators Use of economic indicators
3. 3. ontrolled experiments ontrolled experiments
4. 4. 1udgemental approach 1udgemental approach
Methods of demand forecasting Methods of demand forecasting
Though statistical techniques are essential in clarifying relationships Though statistical techniques are essential in clarifying relationships
and providing techniques of analysis, they are not substitutes for and providing techniques of analysis, they are not substitutes for
judgement. What is needed is some common sense mean between judgement. What is needed is some common sense mean between
pure guessing and too much mathematics. pure guessing and too much mathematics.
1. 1. Survey of buyers` intentions: Survey of buyers` intentions: also known as Opinion surveys. also known as Opinion surveys.
Useful when customers are industrial producers. (However, a Useful when customers are industrial producers. (However, a
number of biases may creep up). Not very useful for household number of biases may creep up). Not very useful for household
consumers. consumers.
Limitation: Limitation: passive and ~does not expose and measure the variables passive and ~does not expose and measure the variables
under management`s control under management`s control
2. 2. Delphi method Delphi method: it consists of an effort to arrive at a consensus in an : it consists of an effort to arrive at a consensus in an
uncertain area by questioning a group of experts repeatedly until uncertain area by questioning a group of experts repeatedly until
the results appear to converge along a single line of the issues the results appear to converge along a single line of the issues
causing disagreement are clearly defined. causing disagreement are clearly defined.
Developed by Rand orporation of the U.S.A in 1940s by Olaf Developed by Rand orporation of the U.S.A in 1940s by Olaf
Helmer, Dalkey and Gordon. Useful in technological forecasting Helmer, Dalkey and Gordon. Useful in technological forecasting
(non (non- -economic variables). economic variables).
DeIphi method DeIphi method
Advantages Advantages
1. 1. Facilitates the maintenance of anonymity of the respondent`s Facilitates the maintenance of anonymity of the respondent`s
identity throughout the course. identity throughout the course.
2. 2. Saves time and other resources in approaching a large number Saves time and other resources in approaching a large number
of experts for their views. of experts for their views.
Limitations/presumptions: Limitations/presumptions:
1. 1. Panelists must be rich in their expertise, possess wide knowledge Panelists must be rich in their expertise, possess wide knowledge
and experience of the subject and have an aptitude and earnest and experience of the subject and have an aptitude and earnest
disposition towards the participants. disposition towards the participants.
2. 2. Presupposes that its conductors are objective in their job, Presupposes that its conductors are objective in their job,
possess ample abilities to conceptualize the problems for possess ample abilities to conceptualize the problems for
discussion, generate considerable thinking, stimulate dialogue discussion, generate considerable thinking, stimulate dialogue
among panelists and make inferential analysis of the among panelists and make inferential analysis of the
multitudinal views of the participants. multitudinal views of the participants.
3. Expert opinion / "hunch" method 3. Expert opinion / "hunch" method
To ask ~experts in the field to provide estimates, eg., dealers, To ask ~experts in the field to provide estimates, eg., dealers,
industry analysts, specialist marketing consultants, etc. industry analysts, specialist marketing consultants, etc.
Advantages: Advantages:
1. 1. 'ery simple and quick method. 'ery simple and quick method.
2. 2. No danger of a ~group No danger of a ~group- -think mentality. think mentality.
4. ollective opinion method 4. ollective opinion method
Also called ~sales force polling, salesmen are required to estimate Also called ~sales force polling, salesmen are required to estimate
expected sales in their respective territories and sections. expected sales in their respective territories and sections.
Advantages: Advantages:
1. 1. Simple Simple -- no statistical techniques. no statistical techniques.
2. 2. Based on first hand knowledge. Based on first hand knowledge.
3. 3. Quite useful in forecasting sales of new products. Quite useful in forecasting sales of new products.
Disadvantages: Disadvantages:
1. 1. Almost completely subjective. Almost completely subjective.
2. 2. Usefulness restricted to short Usefulness restricted to short- -term forecasting. term forecasting.
3. 3. Salesmen may be unaware of broader economic changes. Salesmen may be unaware of broader economic changes.
5. Trend projection Method 5. Trend projection Method
Trend projection models are based exclusively on historical Trend projection models are based exclusively on historical
observation of sales (or other variables such as earnings, cash observation of sales (or other variables such as earnings, cash
flows, etc). They do not explain the underlying casual flows, etc). They do not explain the underlying casual
relationships which produces the variable being forecast. relationships which produces the variable being forecast.
Advantage: Advantage: Inexpensive to develop, store data and operate. Inexpensive to develop, store data and operate.
Disadvantage: Disadvantage: does not consider any possible causal relationships that does not consider any possible causal relationships that
underlie the forecasted variable. underlie the forecasted variable.
3 models 3 models
1. To use actual sales of the current period as the forecast for the next 1. To use actual sales of the current period as the forecast for the next
period; then, period; then, Y Y
t+1 t+1
Y Y
t t
2. If we consider trends, then, 2. If we consider trends, then, Y Y
t+1 t+1
Y Y
t t
+ (Y + (Y
t t
-- Y Y
t t- -11
) )
3. If we want to incorporate the rate of change, rather than the 3. If we want to incorporate the rate of change, rather than the
absolute amount; then, absolute amount; then,
Y Y
t+1 t+1
Y Y
t t
(Y (Y
t t
/ Y / Y
t t- -11
) )
6. Smoothing techniques 6. Smoothing techniques
A. A. Moving average: Moving average: are averages that are updated as new information are averages that are updated as new information
is received. With the moving average a manager simply is received. With the moving average a manager simply
employs, the most recent observations, drops the oldest employs, the most recent observations, drops the oldest
observation, in the earlier calculation and calculates an average observation, in the earlier calculation and calculates an average
which is used as the forecast for the next period. which is used as the forecast for the next period.
Limitations: Limitations:
One has to retain a great deal of data. One has to retain a great deal of data.
All data in the sample are weighed equally. All data in the sample are weighed equally.
B. B. Exponential smoothing: Exponential smoothing: uses uses weighted average weighted average of past data as the of past data as the
basis for a forecast. basis for a forecast.
Y Y
t+1 t+1
aY aY
t t
+ (1 + (1- -a) Y a) Y
t t
or Y new a Y old + (1 or Y new a Y old + (1- -a) Y` old, where, a) Y` old, where,
Y new exponentially smoothed average to be used as the forecast Y new exponentially smoothed average to be used as the forecast
Y old most recent actual data Y old most recent actual data
Y`old most recent smoothed forecast Y`old most recent smoothed forecast
a smoothing constant a smoothing constant
Smoothing constant (or weight) has a value between 0 and 1 inclusive. Smoothing constant (or weight) has a value between 0 and 1 inclusive.
ExponentiaI smoothing ExponentiaI smoothing
The following rules of thumb may be given : The following rules of thumb may be given :
1. 1. When the magnitude of the random variations is large, give a When the magnitude of the random variations is large, give a
lower value to ~a so as to average out the effects of the random lower value to ~a so as to average out the effects of the random
variation quickly. variation quickly.
2. 2. When the magnitude of the random variation is moderate, a When the magnitude of the random variation is moderate, a
large value can be assigned to the smoothing constant ~a. large value can be assigned to the smoothing constant ~a.
3. 3. It has been found appropriate to have ~a between 0.1 and 0.2 It has been found appropriate to have ~a between 0.1 and 0.2
in many systems. in many systems.
Advantages: Advantages:
Exponential smoothing is a forecasting method easy to use and Exponential smoothing is a forecasting method easy to use and
efficiently handled by computers. Although a type of moving efficiently handled by computers. Although a type of moving
average technique, it requires very little record keeping of past average technique, it requires very little record keeping of past
data. This method has been successfully applied by banks, data. This method has been successfully applied by banks,
manufacturing companies, wholesalers and other organizations. manufacturing companies, wholesalers and other organizations.
. AnaIysis of time series . AnaIysis of time series
The time series relating to sales represent the past pattern of The time series relating to sales represent the past pattern of
effective demand for a particular product. Such data can be effective demand for a particular product. Such data can be
presented either in a tabular form or graphically for further presented either in a tabular form or graphically for further
analysis. The most popular method of analysis of the time series is analysis. The most popular method of analysis of the time series is
to to 5roject the trend 5roject the trend of the time series.a trend line can be fitted of the time series.a trend line can be fitted
through a series either visually or by means of statistical through a series either visually or by means of statistical
techniques. The analyst chooses a plausible algebraic relation techniques. The analyst chooses a plausible algebraic relation
(linear, quadratic, logarithmic, etc.) between (linear, quadratic, logarithmic, etc.) between sales sales and the and the
independent variable, independent variable, time. time. The trend line is then projected into The trend line is then projected into
the future by extrapolation. the future by extrapolation.
Popular because Popular because: simple, inexpensive, time series data often : simple, inexpensive, time series data often
exhibit a persistent growth trend. exhibit a persistent growth trend.
Disadvantage Disadvantage: this technique yields acceptable results so long as : this technique yields acceptable results so long as
the time series shows a the time series shows a 5ersistent tendency to move in the same 5ersistent tendency to move in the same
direction. direction. Whenever a Whenever a turning 5oint occurs, turning 5oint occurs, however, the however, the trend trend
5rojection breaks down. 5rojection breaks down.
The real challenge of forecasting is in the prediction of turning points The real challenge of forecasting is in the prediction of turning points
rather than in the projection of trends. rather than in the projection of trends.
AnaIysis of time series and trend AnaIysis of time series and trend
projections projections
Four sets of factors Four sets of factors: secular trend (T), seasonal variation (S), : secular trend (T), seasonal variation (S),
cyclical fluctuations ( ), irregular or random forces (I). cyclical fluctuations ( ), irregular or random forces (I).
O (observations) TSI O (observations) TSI
Assumptions: Assumptions:
1. 1. The analysis of movements would be in the order of trend, The analysis of movements would be in the order of trend,
seasonal variations and cyclical changes. seasonal variations and cyclical changes.
2. 2. Effects of each component are independent of each other. Effects of each component are independent of each other.
. Use of economic indicators . Use of economic indicators
The use of this approach bases demand forecasting on certain The use of this approach bases demand forecasting on certain
economic indicators, eg., economic indicators, eg.,
1. 1. onstruction contracts sanctioned for the demand of building onstruction contracts sanctioned for the demand of building
materials, say, cement; materials, say, cement;
2. 2. Personal income for the demand of consumer goods; Personal income for the demand of consumer goods;
3. 3. Agricultural income for the demand of agricultural inputs, Agricultural income for the demand of agricultural inputs,
implements, fertilizers, etc,; and implements, fertilizers, etc,; and
4. 4. Automobile registration for the demand of car accessories, Automobile registration for the demand of car accessories,
petrol, etc. petrol, etc.
Steps for economic indicators: Steps for economic indicators:
1. 1. See whether a relationship exists between the demand for the See whether a relationship exists between the demand for the
product and certain economic indicators. product and certain economic indicators.
2. 2. Establish the relationship through the method of least squares Establish the relationship through the method of least squares
and derive the regression equation. (Y a + bx) and derive the regression equation. (Y a + bx)
3. 3. Once regression equation is derived, the value of Y (demand) Once regression equation is derived, the value of Y (demand)
can be estimated for any given value of x. can be estimated for any given value of x.
4. 4. Past relationships may not recur. Hence, need for value Past relationships may not recur. Hence, need for value
judgement. judgement.
Use of economic indicators Use of economic indicators
Limitations: Limitations:
1. 1. Finding an appropriate economic indicator may be difficult. Finding an appropriate economic indicator may be difficult.
2. 2. For new products For new products -- no past data exists. no past data exists.
3. 3. Works best when the relationship of demand with a particular Works best when the relationship of demand with a particular
indicator is characterized by a time lag. Eg., construction indicator is characterized by a time lag. Eg., construction
contracts will result in a demand for building materials but with contracts will result in a demand for building materials but with
a certain amount of time lag. a certain amount of time lag.
. ControIIed experiments . ControIIed experiments
Under this method, an effort is made to vary separately certain Under this method, an effort is made to vary separately certain
determinants of demand which can be manipulated, e.g., price, determinants of demand which can be manipulated, e.g., price,
advertising, etc., and conduct the experiments assuming that the advertising, etc., and conduct the experiments assuming that the
other factors remain constant. other factors remain constant.
Example Example -- Parker Pen o. Parker Pen o.
Still relatively new and untried Still relatively new and untried: :
1. 1. Experiments are expensive as well as time consuming. Experiments are expensive as well as time consuming.
2. 2. Risky Risky -- may lead to unfavourable reaction on dealers, may lead to unfavourable reaction on dealers,
consumers, competitors, etc. consumers, competitors, etc.
3. 3. Great difficulty in planning the study.difficult to satisfy the Great difficulty in planning the study.difficult to satisfy the
condition of homogeneity of markets. condition of homogeneity of markets.
10. JudgementaI approach 10. JudgementaI approach
Required when: Required when:
1. 1. Analysis of time series and trend projections is not feasible Analysis of time series and trend projections is not feasible
because of wide fluctuations in sales or because of anticipated because of wide fluctuations in sales or because of anticipated
changes in trends; and changes in trends; and
2. 2. Use of regression method is not possible because of lack of Use of regression method is not possible because of lack of
historical data or because of management`s inability to predict historical data or because of management`s inability to predict
or even identify causal factors. or even identify causal factors.
Even statistical methods require supplementation of judgement Even statistical methods require supplementation of judgement: :
1. 1. Even the most sophisticated statistical methods cannot Even the most sophisticated statistical methods cannot
incorporate all the potential factors, e.g., a major technological incorporate all the potential factors, e.g., a major technological
breakthrough in product or process design. breakthrough in product or process design.
2. 2. For industrial products For industrial products -- if the management anticipates loss or if the management anticipates loss or
addition of few large buyers, it could be taken into account only addition of few large buyers, it could be taken into account only
through judgement approach. through judgement approach.
3. 3. Statistical forecasts are more reliable for larger levels of Statistical forecasts are more reliable for larger levels of
aggregations. aggregations.