dpf ksrao ppt 3

Upload: khushnoodahmedkhan

Post on 06-Apr-2018

224 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/2/2019 Dpf Ksrao Ppt 3

    1/42

    Demand Planning and Forecasting

    Session 3

    Demand Forecasting Methods-1

    By

    K. Sashi Rao

    Management Teacher and Trainer

  • 8/2/2019 Dpf Ksrao Ppt 3

    2/42

    Forecasting in Business PlanningInputsMarket Conditions

    Competitor Action

    Consumer TastesProducts Life Cycle

    Season

    Customers plans

    Economic OutlookBusiness 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

    Sales Forecast

    Forecast and Demand

    for Each Product

    In Each Time Period

    Processor

    Production Capacity

    Available Resources

    Risk Aversion

    ExperiencePersonal Values and

    Motives

    Social and Cultural

    Values

    Other Factors

    Management Team

    Forecast

    Errors

    Feedback

  • 8/2/2019 Dpf Ksrao Ppt 3

    3/42

    Forecasting Methods

    Forecasting

    Qualitative

    Or

    Judgmental

    Quantitative

    Or

    Statistical

    Projective Causal

  • 8/2/2019 Dpf Ksrao Ppt 3

    4/42

    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)

  • 8/2/2019 Dpf Ksrao Ppt 3

    5/42

    Forecasting Approaches(1)

    JUDGEMENTAL APPROACHES: The essence of the judgmental approach is toaddress the forecasting issue by assuming that someone else knows and can tellyou the right answer. They could be experts or opinion leaders.

    EXPERIMENTAL APPROACHES: When an item is "new" and when there is noother information upon which to base a forecast, is to conduct a demandexperiment on a small group of customers and extrapolated to the widerpopulation. Test marketing is an example of this approach.

    RELATIONAL/CAUSAL APPROACHES: There is a reason why people buy ourproduct. If we can understand what that reason (or set of reasons) is, we can usethat 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.

  • 8/2/2019 Dpf Ksrao Ppt 3

    6/42

    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

  • 8/2/2019 Dpf Ksrao Ppt 3

    7/42

    Qualitative Approach

    Qualitative ApproachUsually based on judgments about causal factors that underlie thedemand of particular products or servicesDo not require a demand history for the product or service, therefore areuseful for new products/services

    Approaches vary in sophistication from scientifically conducted surveys tointuitive hunches about future events. The approach/method that isappropriate depends on a products life cycle stage

    Qualitative MethodsEducated guessExecutive committee consensus

    Delphi methodSurvey of sales forceSurvey of customersHistorical analogyMarket research

  • 8/2/2019 Dpf Ksrao Ppt 3

    8/42

    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 usedfor 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

  • 8/2/2019 Dpf Ksrao Ppt 3

    9/42

    Forecasting Methods-judgmental approach(b)

    Consensus thro Delphi method especially for new productdevelopments and technology trends forecasting

    It is the most formal judgmental method and has a well definedprocess and overcomes most of the problems of earlier consensusby executive opinion

    This involves sending out questionnaires to a panel of expertsregarding a forecast subject. Their replies are analyzed,summarized, processed and redistributed to the panel for revisionsin light of others arguments and viewpoints. By going thro such aniterative process say 3-4 times, the final panel forecast is consideredas fairly accurate and authentic

    Yet, difficulties do exist in planning, administering and integratingmember views into a meaningful whole

    Course Booklet has a separate chapter on the Delphi method( page107 onwards)

  • 8/2/2019 Dpf Ksrao Ppt 3

    10/42

    Forecasting Methods-judgmental approach(c)

    Method Short termaccuracy

    Medium

    term

    accuracy

    Long term

    accuracy

    Cost

    Personalinsights

    POOR POOR VERY POOR VERY LOW

    Panel

    consensus

    POOR TO FAIR POOR TO FAIR POOR LOW

    Market survey VERY GOOD GOOD FAIR VERY HIGH

    Historical

    analogy

    POOR FAIR TO GOOD FAIR TO GOOD MEDIUM

    Delphi method FAIR TO GOOD FAIR TO GOOD FAIR TO GOOD HIGH

  • 8/2/2019 Dpf Ksrao Ppt 3

    11/42

    Forecasting Methods

    - experimental approach

    Customer surveys- thro extensive formal market research using personalor mail interviews, and newly thro internet modes; also build demandmodels for a new product by an aggregated approach

    Consumer panels- particularly used in initial stages of productdevelopment 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 understandany problems or issues to fine-tune marketing plans and avoid costlymistakes before going in a big way

    Customer buying data bases- based on selected and acceptedindividuals/families on their buying behavior , patterns and expenditurescaptured using electronic means direct from retailer sales data; givesextensive clues on buying factors, customer attitudes, brand loyalty andbrand switching and response to promotional offers

  • 8/2/2019 Dpf Ksrao Ppt 3

    12/42

    Forecasting Methods- relationship/causal approach(1)

    Its basic premise is that relationships exist betweenvarious independent demand variables( likepopulation, income, disposable incomes, age, sex etcto consumer needs/wants/expectations( dependentvariables)

    Before linking these, we need to find the nature andextent of these causes/relationships in mathematicalterms as regression( linear/multiple)equations

    Once done, they can be used to forecast thedependent variable for available independent variables

    Various types of causal methods follow in next slide

  • 8/2/2019 Dpf Ksrao Ppt 3

    13/42

    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

  • 8/2/2019 Dpf Ksrao Ppt 3

    14/42

    Forecasting Methods- time series approach(1)

    Fundamentally, uses historical demand/sales data todetermine 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( thoughin reality may not always be the case !)

    They involve statistical methods of understanding andexplaining patterns in time series data( like constant series

    e.g. annual rainfall; trends e.g. growing expenditure withincomes; seasonal series e.g. umbrella demand during rainyseason; and any random/unexplained noise where actualvalue= underlying pattern+ random noise)

  • 8/2/2019 Dpf Ksrao Ppt 3

    15/42

    7-15

    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)

  • 8/2/2019 Dpf Ksrao Ppt 3

    16/42

    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 dueto 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

  • 8/2/2019 Dpf Ksrao Ppt 3

    17/42

    Forecasting Methods- time series approach(3)

    Basic concepts involved are those ofmoving averages and exponentialsmoothing

    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 limiteduse

    A moving average takes the average over a fixed number( by choice) ofprevious periods ignoring older data periods giving a sense of immediacyto the data used e.g. taking only past 3 months data as relevant forforecasting for next quarter with same weightage; later improved byweighted moving averages with unequal weightage

    All moving averages suffer in that(a) all historically used data are givensame /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 anon-linear basis

  • 8/2/2019 Dpf Ksrao Ppt 3

    18/42

    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.

  • 8/2/2019 Dpf Ksrao Ppt 3

    19/42

    7-19

    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

  • 8/2/2019 Dpf Ksrao Ppt 3

    20/42

    7-20

    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

  • 8/2/2019 Dpf Ksrao Ppt 3

    21/42

    7-21

    Static Methods

    Assume a mixed model:

    Systematic component = (level + trend)(seasonal factor)

    Ft+l= [L + (t + l)T]St+l

    = forecast in period tfor demand in period t+ lL = 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

  • 8/2/2019 Dpf Ksrao Ppt 3

    22/42

    7-22

    Adaptive Forecasting

    The estimates of level, trend, and seasonalityare 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 exponentialsmoothing (Winters model)

  • 8/2/2019 Dpf Ksrao Ppt 3

    23/42

    Moving Averages(1)

    This is the simplest model of extrapolative forecasting

    Since demand varies over time, only a certain amount of historicaldata is relevant to the future, implying that we can ignore allobservations older than some specified age

    A moving average uses this approach by taking average demandover 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 (150+130+125)/3= 135. If actual demand in4th month is 135 as forecasted( their differences are forecasting errorswhichwill discuss later), then forecast for 5th month is(130+125+135)/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 importanceto most recent periods

  • 8/2/2019 Dpf Ksrao Ppt 3

    24/42

    7-24

    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 levelLt = (Dt + Dt-1 + + Dt-N+1) / N

    Ft+1 = Lt and Ft+n = LtAfter observing the demand for period t+1, revise the estimates asfollows:

    Lt+1 = (Dt+1 + Dt + + Dt-N+2) / NFt+2 = Lt+1

  • 8/2/2019 Dpf Ksrao Ppt 3

    25/42

    Moving Averages(3)

    Include n most recent observations

    Weight equally

    Ignore older observations

    weight

    today

    123...n

    1/n

  • 8/2/2019 Dpf Ksrao Ppt 3

    26/42

    Moving Averages(4) Forecast Ft is average ofn 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.

    !

    !

    !

    t

    nti

    it

    ntttt

    Dn

    F

    DDDn

    F

    1

    1

    111

    1

    )(1

    .

  • 8/2/2019 Dpf Ksrao Ppt 3

    27/42

    Moving Averages(5)

    Internet Unicycle Sales

    0

    50

    100

    150

    200

    250

    300

    350

    400

    450

    Apr-01 Sep-02 Jan-04 May-05 Oct-06 Feb-08 Jul-09 Nov-10 Apr-12 Aug-13

    Month

    Units

    n = 3

  • 8/2/2019 Dpf Ksrao Ppt 3

    28/42

    Simple Moving Averages(6) example

    Month Actual Sales ForecastChosen 3 months moving

    average

    Jan 24500

    Feb 27000

    Mar 19950

    Apr 26000 23817

    May 21200 24317

    June 18900 22383July 17500 22033

    Aug 19000 19200

    Sep 18525 18467

  • 8/2/2019 Dpf Ksrao Ppt 3

    29/42

    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)= 75+39+15= 129; and further forecast for 5th month as

    (129x0.60+125x0.30+130x0.10)= 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

  • 8/2/2019 Dpf Ksrao Ppt 3

    30/42

    Weighted Moving Averages(2)-example

    Month Actual Sales Forecast

    Chosen 3 months moving

    average

    Weightage- immediate

    past as 0.45, then 0.30and then 0.25

    Jan 24500

    Feb 27000

    Mar 25500

    Apr 26000 25700

    May 21200 26100

    June 18900 23715

    July 17500 21365

    Aug 19000 18845

    Sep 18525

  • 8/2/2019 Dpf Ksrao Ppt 3

    31/42

    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 ormore) 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

  • 8/2/2019 Dpf Ksrao Ppt 3

    32/42

    7-32

    Time Series Patterns(1)

    0

    10,000

    20,000

    30,000

    40,000

    50,000

    97,2

    97,3

    97,4

    98,1

    98,2

    98,3

    98,4

    99,1

    99,2

    99,3

    99,4

    00,1

  • 8/2/2019 Dpf Ksrao Ppt 3

    33/42

    7-33

    Time Series Patterns(2)

    0

    10000

    20000

    30000

    40000

    50000

    1 2 3 4 5 6 7 8 9 10 11 12

    Period

    Demand

    Dt

    Dt-bar

  • 8/2/2019 Dpf Ksrao Ppt 3

    34/42

    Time Series Patterns(3)

  • 8/2/2019 Dpf Ksrao Ppt 3

    35/42

    Time Series Patterns(4)

  • 8/2/2019 Dpf Ksrao Ppt 3

    36/42

    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

  • 8/2/2019 Dpf Ksrao Ppt 3

    37/42

    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

    Independent variable S

    Dependent

    Variable P

    Intercept q

    .

    Gradient r ( >0)

    r(

  • 8/2/2019 Dpf Ksrao Ppt 3

    38/42

    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 pastdemand 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 forcomponents

  • 8/2/2019 Dpf Ksrao Ppt 3

    39/42

    Linear Regression example(contd.)ACTUAL DEMAND FOR LAST THREE

    YEARS( in 000 units)

    PERIOD Period Number(X) ACTUAL DEMAND(Y)

    Year 1- Q1 1 360

    Year 1- Q2 2 438

    Year 1- Q3 3 359

    Year 1- Q4 4 406

    Year 2- Q1 5 393

    Year 2 -Q2 6 465

    Year 2- Q3 7 387

    Year 2- Q4 8 464

    Year 3- Q1 9 505

    Year 3- Q2 10 618

    Year 3- Q3 11 443

    Year 3- Q4 12 540

  • 8/2/2019 Dpf Ksrao Ppt 3

    40/42

    Linear Regression example(contd.)

    Period X Y XY XX

    PERIOD PERIOD Number ACTUAL DEMAND(Y)

    Year 1- Q1 1 360 360 1

    Year 1- Q2 2 438 876 4

    Year 1- Q3 3 359 1078 9

    Year 1- Q4 4 406 1625 16

    Year 2- Q1 5 393 1965 25

    Year 2 -Q2 6 465 2790 36

    Year 2- Q3 7 387 2709 49

    Year 2- Q4 8 464 3712 64

    Year 3- Q1 9 505 4545 81

    Year 3- Q2 10 618 6180 100

    Year 3- Q3 11 443 4873 121

    Year 3- Q4 12 540 6480 144

    SUM 78 5379 37193 650

  • 8/2/2019 Dpf Ksrao Ppt 3

    41/42

    Linear Regression example(contd.)

    Linear regression equation P= q+ rS

    Using method of least squares, the regression coefficients areworked 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

  • 8/2/2019 Dpf Ksrao Ppt 3

    42/42

    Multiple Regression

    When there are many independent variables involvedwhich influence a dependent variable then issuesbecome complicated

    Then not only linear regression equations are required

    but also multiple regression analysis is involved wherethe interdependency of the various independentvariables are taken into account

    These involve complex statistics beyond the scope ofthis course

    For their practical use, advanced techniques and toolsare available thro MS Excel tools, SPSS and othersoftware packages