Docstoc

Forecasting - PowerPoint Presentation.ppt

Document Sample
Forecasting - PowerPoint Presentation.ppt Powered By Docstoc
					   Fundamentals of
Operations Management
       BUS 3 – 140

     Forecasting
       Feb 5, 2008
Forecasting

     – A statement about the future value of a variable of interest
         • Future Sales
         • Weather
         • Stock Prices
         • Other Short term and Long term estimates

     – Several Methods
         • Quantitative
             • History and Patterns
             • Leading Indicators / Associations (Housing Starts &
               Furniture)
         • Qualitative
             • Judgment
             • Consensus

 Used for making informed Decisions and taking Actions based on those decisions

                                                            Page 2                2
Forecasting


   Forecasts make a MAJOR IMPACT (Positive or Negative) on:
     • Revenue
     • Market Share
     • Cost
     • Inventory
     • Profit




                                                 Page 3       3
Features Common to all Forecasts

      – Generally assumes that what drove past performance and
        behavior will drive future performance and behavior

         • Credit Rating
         • Insurance Rates
         • Other


      – More accurate for groups vs. individuals



      – Accuracy decreases as time horizon increases


    Forecasts WILL be wrong – the goal is to predict as closely as possible


                                                           Page 4             4
Three Major Types of Forecasts

   Judgmental
       – Uses subjective, qualitative “judgment” (opinions, surveys,
         experts, managers, others). Most useful when there is limited
         data and with New Product Introductions

   Time series
        – Observes what has occurred over previous time periods and
          assumes that future patterns will follow historical patterns



   Associative Models
       – Establishes cause and effect relationships between
          independent and dependent variables (rainy days and umbrella
          sales, pricing and sales volume, attendance at sporting events
          and food sold, others)


                                                        Page 5             5
Forecasting techniques (Table 3.6)
          Approach                   Technique                                      Brief Description

                            Consumer surveys               Questioning consumers on future plans


                            Direct-contact composites      Joint estimates obtained from sales or customer service
       Judgment /
                                                           Finance, marketing, and manufacturing managers join to prepare
        opinion:            Executive opinion
                                                           forecast
      QUALITATIVE                                          Series of questionnaires answered anonymously by knowledgeable
                            Delphi technique               people; successive questionnaires are based on information obtained
                                                           from previous surveys

                            Outside opinion                Consultants or other outside experts prepare the forecast


                                                           Next value in a series will equal the previous value in a comparable
                            Time series: Naïve
                                                           period

                            Time series: Moving Averages Forecast is based on an average of recent values

       Statistical:         Time series: Exponential
                                                           Sophisticated form of weighted moving average
     QUANTITATIVE           Smoothing

                            Associative Models: Simple     Values of one variable are used to predict values of a dependent
                            Regression                     variable

                            Associative Models: Multiple   Two or more variables are used to predict values of a dependent
                            Regression                     variable


* From Stevenson, Operations Management, Ninth Edition, McGraw Hill Irwin                          Page 6                         6
Elements of any Good Forecast




                                                 Timely



                                    Reliable                Accurate



                                                       Written



* From Stevenson, Operations Management, Ninth Edition, McGraw Hill Irwin


                                                                            Page 7   7
Steps in the Forecasting Process




                                                         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


* From Stevenson, Operations Management, Ninth Edition, McGraw Hill Irwin


                                                                                Page 8   8
Forecast Factors (Table 3.5)

                    Factor                         Short Range          Intermediate Range               Long Range

   Frequency                                           Often                    Occasional                Infrequent


                                                                                                     Total Output, Type of
   Level of Aggregation                                Item                   Product Family
                                                                                                       product / service


                                              Smoothing, Projection,        Projection, Seasonal,
   Type of Model                                                                                     Managerial Judgment
                                                  Regression                     Regression


   Degree of Management Involvement                     Low                      Moderate                    High



   Cost per Forecast                                    Low                      Moderate                    High




                     Forecasts are established with two (2) Units of Measure:
                                     1. Units
                                     2. Dollars
                           Both have significance to the Enterprise

* From Stevenson, Operations Management, Ninth Edition, McGraw Hill Irwin                       Page 9                       9
Start with what you KNOW
     – How many people will attend the next Giants game?
        •   Tickets already sold
        •   Patterns of walk up sales
        •   Visiting team
        •   Weather
        •   School day
        •   Other

     – How many Sewing Machines will Singer sell this week?
        • Orders in Backlog
        • Inventory in Stores
        • Production capacity

     – Household Budget
        • Rent
        • Car Payment
        • Bills
        • Rest of money

                                                              Page 10   10
A Demand Forecast serves many Purposes


                           WHAT is done and WHY?




                Product
Region                      Channel          Features       Product           Customer
                 Line

 Revenue Planning         Estimating TAM and Share      Scheduling Factory Volumes
 Revenue Scenarios        Pricing Targets               Materials Planning
 Allocation Criteria      Programs & Promotions         Balancing Factory Capacity
 Commissions &            Margins @ Mixes               Assessing Direct Cost @ Mixes
     Quotas               Message to Analysts           Analyzing Absorption implications


                             Business Need / Benefit



                                                                  Page 11                   11
How different Functions use Forecast information


       ORGANIZATION              KEY VALUE OF A FORECAST
   Sales & Marketing      Pricing, Promotions, Quotas, Commissions
   Operations             Schedules, Capacity, Capital
   Materials              Continuous supply, Inventory
   Logistics              Transportation Planning
   Finance & Accounting   Cash flow, cost, profits, PE estimates
   HR                     Hiring, recruiting, training
   MIS                    Hardware, connectivity, support
   Design                 New products and services




                                                         Page 12     12
Forecast accuracy varies over time




       Over
        Expected Errors




                          0   +1     +2   +3   +4 …………………………………………                     +n
                                                    Time in Future (Weeks)



      Under




                                   The further into the future, the harder
                                     to predict details with accuracy

                                                                             Page 13        13
Detailed Product Forecast Accuracy will vary by Time Horizon


 Current Week should approach 100%
           5%   10%   15%   20%   25%   30%   35%   40%    45%    50%   55%   60%   65%   70%   75%    80%   85%   90%   95% 100%

   Week                                                   Known                                                    High Prob. TBD




 Current Month should be greater than 80%
           5%   10%   15%   20%   25%   30%   35%   40%    45%    50%   55%   60%   65%   70%   75%    80%   85%   90%   95% 100%

  Month                           Known                                       High Probability / Influence               TBD




 Quarter should be at least 70%

           5%   10%   15%   20%   25%   30%   35%   40%    45%    50%   55%   60%   65%   70%   75%    80%   85%   90%   95% 100%

 Quarter          Known                 High Probability and/or can Influence                         To Be Determined




                                                                                                 Page 14                        14
Tracking Forecast Accuracy

    – Level of Aggregation
       • Item (Mix of individual SKU’s)
       • Family
       • Product Line            Absolute values and square roots eliminate the
       • Channel                  possibility of positive and negative variances
                                 canceling each other out – key for Mix tracking;
       • Customers                      less critical for Revenue tracking

    – Quantity

    – Time Buckets

    – Final consumer sales

     Regular tracking and monitoring with enable Demand SENSING,
      as well as contribute to increased accuracy of future forecasts


                                                               Page 15              15
Relationship of Lead Time, Forecast, Inventory, and Cost



                               Inventory
                    Need to                Cost to         Risk of
                               Levels in
                    Forecast               Manage          Excess
                                Pipeline



 Long Lead Time      High       High       Higher          Higher



 Short Lead Time     Low        Low        Lower           Lower




                                                 Page 16             16
Time Series Forecasts (and Behaviors)


     Trend - long-term movement in data


     Seasonality - short-term regular variations in data


     Cycle – wavelike variations of more than one year’s duration


     Irregular variations - caused by unusual circumstances


     Random variations - caused by chance




                                                           Page 17   17
Graphs help interpret Time Series data (Figure 3.1)


                               Irregular
                               variatio
                               n
                                                       Trend



                                                                     Cycles

                                                                                        90
                                                                                        89
                                                                                        88
                                                  Seasonal variations




* From Stevenson, Operations Management, Ninth Edition, McGraw Hill Irwin     Page 18        18
Relevance of SUPPLY on Forecasts


     Historical Sales does not always equal historical Demand
       – Stockouts
       – Substitutions
       – Causal Factors may distort the analysis (pricing,
         promotions, competitor performance)


     Scarcity Behavior
      – Allocation
      – Advance buying
      – Hedging
      – Hording




                                                        Page 19   19
Guide to selecting Forecasting methods (Table 3.4)

    Forecasting           Amount of                                                                           Personnel
                                              Data Pattern      Forecast Horizon   Preparation Time
      Method            Historical Data                                                                      Background

                                             Data should be
   Moving Average      2 - 3 observations                             Short               Short           Little sophistication
                                               stationary


 Simple exponential                          Data should be
                    5 - 10 observations                               Short               Short           Little sophistication
    smoothing                                  stationary

   Trend-adjusted
                                                                                                              Moderate
     exponential      10 - 15 observations        Trend         Short to medium           Short
                                                                                                            sophistication
     smoothing

                         10 - 20; for
                                                                                                              Moderate
    Trend models      seasonality at least        Trend         Short to medium           Short
                                                                                                            sophistication
                        5 per season

                                             Handles cyclical
                       Enough to see 2
      Seasonal                                and seasonal      Short to medium    Short to moderate      Little sophistication
                      peaks and troughs
                                                patterns

                      10 observations per                                          Long development
  Causal regression                       Can handle complex Short, medium, or                              Considerable
                         independent                                               time, short time for
      models                                 data patterns          long                                    sophistication
                           variable                                                  implementation


* From Stevenson, Operations Management, Ninth Edition, McGraw Hill Irwin


                                                                                            Page 20                               20
Selecting the most useful Forecasting technique(s)


     No single technique works in every situation


     Two most important factors
      – Cost
      – Accuracy


     Other factors include the availability of:
      – Historical data
      – Computers
      – Time needed to gather and analyze the data
      – Forecast horizon




                                                     Page 21   21
Causal Factors

   External
        – Market conditions (e.g. paintings when the Painter passes
          away)
        – New competition
        – Competitors cannot supply

   Internal
        – Pricing
        – Promotions
        – Incentives




                                                        Page 22       22
Barry Bonds Home Run Totals


            750

            675                                                          ?????????
            600

            525
Home Runs




            450

            375

            300

            225

            150

            75

             0
                  22   23   24   25   26   27   28   29   30   31   32    33   34   35   36   37    38   39   40   41   42
                                                                    Age




                                                                                              Page 23                    23
Other Points to consider
  Do not “second guess” the forecast
   • significant judgment and even debate contribute to the final forecast.
     Once the forecast is finalized it then becomes the Demand Plan of
     Record for the enterprise

  …… and do not say, “If only we got a better forecast” ……
   • The forecast should be generated as a team and managed as a team

  It is helpful to provide a range of expected Demand
     • A useful application of Confidence Intervals from Statistics

  Product Transitions are very difficult to forecast, but require special
  attention and monitoring
    • New Product Introduction
    • End Of Life

  Peter Drucker: “The best way to predict the future is to CONTROL it”

                                                            Page 24           24

				
DOCUMENT INFO