Demand Planning and Forecasting Session 3

1
Demand Planning and ForecastingSession 3

• Demand Forecasting Methods-1
• By
• K. Sashi Rao
• Management Teacher and Trainer

2
Forecasting in Business Planning
Inputs Market Conditions Competitor
Action Consumer Tastes Products Life
Cycle Season Customers plans Economic
Outlook Business Cycle Status Leading
Indicators-Stock Prices, Bond Yields, Material
Prices, Business Failures, money Supply,
Unemployment Other Factors Legal, Political,
Sociological, Cultural
Forecasting Method(s) Or Model(s)
Outputs Estimated Demands for each Product in
each Time Period Other Outputs
Management Team
Production Capacity Available Resources Risk
Aversion Experience Personal Values and
Motives Social and Cultural Values Other
Factors
Processor
Forecast Errors
Sales Forecast Forecast and Demand for Each
Product In Each Time Period
Feedback
3
Forecasting Methods
Forecasting
Qualitative Or Judgmental
Quantitative Or Statistical
Projective
Causal
4
Forecasting Basics

• Types
• Qualitative --- based on experience, judgment,
  knowledge
• Quantitative --- based on data, statistics
• Methods
• Naive Methods --- using ball-park numbers or
  assuming future demand same as before
• Formal Methods --- systematic methods thereby
  reduce forecasting errors using
• time series models (e.g. moving averages and
  exponential smoothing)
• causal models (e.g. regression)

5
Forecasting Approaches(1)

• JUDGEMENTAL APPROACHES The essence of the
  judgmental approach is to address the forecasting
  issue by assuming that someone else knows and can
  tell you the right answer. They could be experts
  or opinion leaders.
• EXPERIMENTAL APPROACHES When an item is "new"
  and when there is no other information upon which
  to base a forecast, is to conduct a demand
  experiment on a small group of customers and
  extrapolated to the wider population. Test
  marketing is an example of this approach.
• RELATIONAL/CAUSAL APPROACHES There is a
  reason why people buy our product. If we can
  understand what that reason (or set of reasons)
  is, we can use that understanding to develop a
  demand forecast. They seek to establish product
  -demand relationships to relevant factors and/or
  variables e.g. hot weather to cold drinks
  consumption.
• TIME SERIES APPROACHES A time series is a
  collection of observations of well-defined data
  items obtained through repeated measurements over
  time.

6
Forecasting Approaches(2)

• In general, judgment and experimental approaches
  tend be more qualitative
• While relationship/causal and time series
  approaches tend be more quantitative
• Still, these qualitative methods are also
  scientifically done with results that are
  expressed in indicative numbers and broad trends
• Time series/causal methods are completely based
  on statistical methods and principles

7
Qualitative Approach

• Qualitative Approach
• Usually based on judgments about causal
  factors that underlie the demand of particular
  products or servicesDo not require a demand
  history for the product or service, therefore are
  useful for new products/servicesApproaches vary
  in sophistication from scientifically conducted
  surveys to intuitive hunches about future events.
  The approach/method that is appropriate depends
  on a products life cycle stage
• Qualitative Methods
• Educated guessExecutive committee
  consensusDelphi methodSurvey of sales
  forceSurvey of customersHistorical
  analogyMarket research

8
Forecasting Methods-judgmental approach(a)

• Surveys - this involves a bottom up method
  where each individual/respondent contributes to
  the overall result this could be for product
  demand or sales forecasting also for opinion
  surveys amongst employees, citizen groups or
  voter groups for election polls
• Sales Force Composites- where the similar bottom
  up approach is used for building up sales
  forecasts on any criteria like region-wise or
  product wise sales territory groupings from sales
  force personnel
• Consensus of Executive Opinion -normally used in
  strategy formulation by sought opinions from key
  organizational stakeholders- managers, suppliers,
  customers, bankers and shareholders
• Historical analogy- used for forecasting new
  product demand as similar to the previously
  introduced new product benefiting from its
  immediacy that same demand influencing factors
  will apply

9
Forecasting Methods-judgmental approach(b)

• Consensus thro Delphi method especially for
  new product developments and technology trends
  forecasting
• It is the most formal judgmental method and has a
  well defined process and overcomes most of the
  problems of earlier consensus by executive
  opinion
• This involves sending out questionnaires to a
  panel of experts regarding a forecast subject.
  Their replies are analyzed, summarized, processed
  and redistributed to the panel for revisions in
  light of others arguments and viewpoints. By
  going thro such an iterative process say 3-4
  times, the final panel forecast is considered as
  fairly accurate and authentic
• Yet, difficulties do exist in planning,
  administering and integrating member views into
  a meaningful whole
• Course Booklet has a separate chapter on the
  Delphi method( page 107 onwards)

10
Forecasting Methods-judgmental approach(c)
11
Forecasting Methods- experimental approach

• Customer surveys- thro extensive formal market
  research using personal or mail interviews, and
  newly thro internet modes also build demand
  models for a new product by an aggregated
  approach
• Consumer panels- particularly used in initial
  stages of product development and design to match
  product attributes to customer expectations
• Test marketing- often used after product
  development but before national launches by
  starting in a selected target market/geography to
  understand any problems or issues to fine-tune
  marketing plans and avoid costly mistakes before
  going in a big way
• Customer buying data bases- based on selected and
  accepted individuals/families on their buying
  behavior , patterns and expenditures captured
  using electronic means direct from retailer sales
  data gives extensive clues on buying factors,
  customer attitudes, brand loyalty and brand
  switching and response to promotional offers

12
Forecasting Methods- relationship/causal
approach(1)

• Its basic premise is that relationships exist
  between various independent demand variables(
  like population, income, disposable incomes, age,
  sex etc to consumer needs/wants/expectations(
  dependent variables)
• Before linking these, we need to find the nature
  and extent of these causes/relationships in
  mathematical terms as regression(
  linear/multiple)equations
• Once done, they can be used to forecast the
  dependent variable for available independent
  variables
• Various types of causal methods follow in next
  slide

13
Forecasting Methods- relationship/causal
approach(2)

• Econometric models like discrete choice and
  multiple regression models used in large-scale or
  macro-level economic forecasting
• Input-output models used to estimate the flow of
  goods between markets and industries, again in
  macro-economic situations
• Simulation models used to establish raw materials
  and components demand based on MRP schedules ,
  driven by keyed-in product sales forecasts to
  reflect market realities and imitate customer
  choices
• Life-cycle models which recognize product demand
  changes during its various stages(i.e.
  introduction/growth/maturity/decline)
  particularly in short life cycle sectors like
  fashion and technology

14
Forecasting Methods- time series approach(1)

• Fundamentally, uses historical demand/sales data
  to determine future demand
• Basic assumptions are that
• Past data/information is available
• This data/information can be quantified
• Past patterns will continue into the future and
  projections made( though in reality may not
  always be the case !)
• They involve statistical methods of understanding
  and explaining patterns in time series data( like
  constant series e.g. annual rainfall trends e.g.
  growing expenditure with incomes seasonal series
  e.g. umbrella demand during rainy season and any
  random/unexplained noise where actual value
  underlying pattern random noise)

15
Forecasting Methods-time series approach(2)

• Static elements
• Trend
• Seasonal
• Cyclical
• Random
• Adaptive elements
• Moving average
• Simple exponential smoothing
• Exponential smoothing (with trend)
• Exponential smoothing (with trend and seasonality)

16
Time Series-static elements

• Trend component- persistent overall downward or
  upward pattern due to population, technology or
  long term movement
• Seasonal component- regular up and down
  fluctuations due to weather and/or seasons whose
  pattern repeats every year
• Cyclical component- repeated up and down
  movements due to economic or business cycles
  lasting beyond one year but say every 5-6 years
• Random component- erratic, unsystematic, residual
  fluctuations due to random events or occurrences
  like one time drought or flood events

17
Forecasting Methods- time series approach(3)

• Basic concepts involved are those of moving
  averages and exponential smoothing
• A simple average forecast method is usable if
  past pattern is very stable, but very few time
  series are stable over long periods, hence are of
  limited use
• A moving average takes the average over a fixed
  number( by choice) of previous periods ignoring
  older data periods giving a sense of immediacy to
  the data used e.g. taking only past 3 months data
  as relevant for forecasting for next quarter with
  same weightage later improved by weighted
  moving averages with unequal weightage
• All moving averages suffer in that(a) all
  historically used data are given same /unequal
  weight and (b) works well only when demand is
  relatively constant. Its handicaps are overcome
  by exponential smoothing
• Exponential smoothing is based on idea that as
  data gets older it becomes less relevant and
  should be given a progressively lower weightage
  on a non-linear basis

18
Forecasting Examples

• Examples from Projects
• Demand for tellers in a bank
• Traffic flow at a major junction
• Pre-poll opinion survey amongst voters
• Demand for automobiles or consumer durables
• Segmented demand for varying food types in a
  restaurant
• Area demand for frozen foods within a locality
• Example from Retail Industry American Hospital
  Supply Corp.
• 70,000 items
• 25 stocking locations
• Store 3 years of data (63 million data points)
• Update forecasts monthly
• 21 million forecast updates per year.

19
Components of an Observation

• Observed demand (O)
• Systematic component (S) Random component (R)

Level (current deseasonalized demand)
Trend (growth or decline in demand)
Seasonality (predictable seasonal fluctuation)

• Systematic component Expected value of demand
• Random component The part of the forecast that
  deviates from the systematic component
• Forecast error difference between forecast and
  actual demand

20
Time Series Forecasting Methods

• Goal is to predict systematic component of demand
• Multiplicative (level)(trend)(seasonal factor)
• Additive level trend seasonal factor
• Mixed (level trend)(seasonal factor)
• Static methods
• Adaptive forecasting

21
Static Methods

• Assume a mixed model
• Systematic component (level trend)(seasonal
  factor)
• Ftl L (t l)TStl
• forecast in period t for demand in period t l
• L estimate of level for period 0
• T estimate of trend
• St estimate of seasonal factor for period t
• Dt actual demand in period t
• Ft forecast of demand in period t

22
Adaptive Forecasting

• The estimates of level, trend, and seasonality
  are adjusted after each demand observation
• General steps in adaptive forecasting
• Moving average
• Simple exponential smoothing
• Trend-corrected exponential smoothing (Holts
  model)
• Trend- and seasonality-corrected exponential
  smoothing (Winters model)

23
Moving Averages(1)

• This is the simplest model of extrapolative
  forecasting
• Since demand varies over time, only a certain
  amount of historical data is relevant to the
  future, implying that we can ignore all
  observations older than some specified age
• A moving average uses this approach by taking
  average demand over a fixed number of previous
  periods( say 3 as in below example)
• Example If product demand is 150, 130 and 125
  over the last 3 months, then forecast for 4th
  month is (150130125)/3 135. If actual demand
  in 4th month is 135 as forecasted( their
  differences are forecasting errors which will
  discuss later), then forecast for 5th month is
  (130125135)/3 130 and this process is
  repeated for subsequent periods
• In above example, all past periods were given
  equal weightage which can then be differentially
  weighted to give more importance to most recent
  periods

24
Moving Averages(2)

• Used when demand has no observable trend or
  seasonality
• Systematic component of demand level
• The level in period t is the average demand over
  the last N periods (the N-period moving average)
• Current forecast for all future periods is the
  same and is based on the current estimate of the
  level
• Lt (Dt Dt-1 Dt-N1) / N
• Ft1 Lt and Ftn Lt
• After observing the demand for period t1,
  revise the estimates as follows
• Lt1 (Dt1 Dt Dt-N2) / N
• Ft2 Lt1

25
Moving Averages(3)

• Include n most recent observations
• Weight equally
• Ignore older observations

weight
1/n
...
1
2
n
3
today
26
Moving Averages(4)

• Forecast Ft is average of n previous
  observations or actual Dt
  
• Note that the n past observations are equally
  weighted.
• Issues with moving average forecasts
• All n past observations treated equally
• Observations older than n are not included at
  all
• Requires that n past observations be retained
• Problem when 1000's of items are being forecast.

27
Moving Averages(5)
n 3
28
Simple Moving Averages(6) example
29
Weighted Moving Averages(1)

• This is to overcome the lacuna of ALL past
  periods being given SAME importance
• Here, different past periods are given different
  weightage
• In same earlier example, let us take past periods
  weightage as 0.60, 0.30 and 0.10( totaling 1 or
  100) then forecast for 4th month is (
  125x0.60 130x0.30 150x0.10) 753915 129 and
  further forecast for 5th month as
  (129x0.60125x0.30130x0.10) 127.9 and so
  on..
• Idea is to give more importance to most recent
  observations
• But problems relate to the logic of deciding the
  number of past periods and the given differential
  weightage
• Generally, if the demand is stable, then larger n
  values are chosen if not, then a smaller n and
  using weightage factors is better

30
Weighted Moving Averages(2)-example
31
Moving Averages- closing remarks

• All moving average methods( besides exponential
  smoothing to be taken up later) focus on short
  term forecasting and provide such capability
  without consideration of any time series patterns
• But when medium term( say 1 year) or long term( 5
  years or more) forecasting needed, then time
  series data patterns need looking into
• These data patterns relate to trend, cyclical,
  seasonal and random forms( as introduced earlier)
• Once these patterns are extracted from a given
  time series data , they can be used for
  forecasting

32
Time Series Patterns(1)
33
Time Series Patterns(2)
34
Time Series Patterns(3)
35
Time Series Patterns(4)
36
Causal Forecasting(1)

• Basic idea is to use a cause or a relationship
  between and amongst variables as a forecasting
  method e.g. product sales is dependent on its
  price
• Need to identify the independent and dependent
  variables
• Causal forecasting is illustrated by linear
  regression

37
Linear Regression

• It looks for a relationship of the form
• Dependent variable(P) q r multiplied by
  independent variable (S) or P q r S where
• q intercept and r gradient of the line

Dependent Variable P
.

Gradient r ( gt0)
r(lt0)
Intercept q
Independent variable S
38
Linear Regression - example

• A manufacturer of critical components for two
  wheelers is interested in forecasting the trend
  in demand during the next year as a key input to
  its annual planning exercise. Information on past
  demand is available for last three years( next
  slide). We need to develop a linear regression
  equation to extract the trend component of the
  time series and use it for predicting the future
  demand for components

39
Linear Regression example(contd.)
40
Linear Regression example(contd.)
41
Linear Regression example(contd.)

• Linear regression equation P q rS
• Using method of least squares, the regression
  coefficients are worked out as X 78/12 6.50 and
  Y 5379/12 448.25
• Then the gradient r 37193-(12x6.50x448.25)/650-
  (12x6.50x6.50) 2229.5/143 15.59
• The intercept q 448.25-15.59x6.50 346.91
• Final regression equation is P 346.91 15.59S
• Thus Forecast for Year 4 Q1 346.91 15.59x13
  550
• Forecast for Year 4 Q2 346.91
  15.59x14 565
• Forecast for Year 4 Q3 346.91
  15.59x15 581
• Forecast for Year 4 Q4 346.91
  15.59x16 596

42
Multiple Regression

• When there are many independent variables
  involved which influence a dependent variable
  then issues become complicated
• Then not only linear regression equations are
  required but also multiple regression analysis is
  involved where the interdependency of the various
  independent variables are taken into account
• These involve complex statistics beyond the scope
  of this course
• For their practical use, advanced techniques and
  tools are available thro MS Excel tools, SPSS and
  other software packages