Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Forecasting Techniques


Published on

Interventions required to meet business objectives - from Forecasting Methods,
Forecast Accuracy / Error Reduction,
Integrate – Sales Forecast / Production to undertaking a CPFR

Published in: Business, Technology
  • Login to see the comments

Forecasting Techniques

  1. 1. Forecasting Techniques Interventions required to meet business objectives Anand Subramaniam
  2. 2. <ul><li>“ An ardent supporter of the hometown team should go to a game prepared to take offense, no matter what happens.” </li></ul><ul><ul><li>- Robert Benchley </li></ul></ul>
  3. 3. Highlights <ul><li>Forecasting Methods </li></ul><ul><li>Quantitative Methods – Examples </li></ul><ul><li>Forecast Accuracy / Error Reduction </li></ul><ul><li>Integrate – Sales Forecast / Production </li></ul><ul><li>CPFR - Collaborative Planning, Forecasting and Replenishment </li></ul>
  4. 4. Forecasting Methods
  5. 5. Planning Levels
  6. 6. Forecast Horizon <ul><li>Trend Exploration </li></ul><ul><li>Graphical Methods </li></ul><ul><li>Exponential Smoothing </li></ul><ul><li>Purchasing </li></ul><ul><li>Detailed Job Scheduling </li></ul>1 day ~ I year Short <ul><li>Time Series </li></ul><ul><li>Regression </li></ul><ul><li>Staffing Plans </li></ul><ul><li>Aggregate Production Plan </li></ul>1 season ~ 2 years Intermediate <ul><li>Economic </li></ul><ul><li>Demographic </li></ul><ul><li>Market Information </li></ul><ul><li>Technology </li></ul><ul><li>Facility Planning </li></ul><ul><li>Capacity Planning </li></ul><ul><li>Product Planning </li></ul>> 5 years Long Methods Applications Horizon Range
  7. 7. Major Areas of Forecasting Economic Forecasting Predicts what the general business conditions will be in the future (Eg. Inflation rates, Gross National Product, Tax, Level of employment) Technology Forecasting Predicts the probability and / or possible future developments in technology (Eg. Competitive advantage or firm’s competitors incorporate into their products and processes) Demand Forecasting Predicts the quantity and timing of demand for a firm’s products
  8. 8. Forecasting Methods Subjective Approach (Qualitative in nature and usually based on the opinions of people) Objective Approach (Quantitative / Mathematical formulations - statistical forecasting)
  9. 9. Qualitative Methods <ul><li>Executive Committee Consensus </li></ul><ul><li>Develop long ~ medium forecast by asking a group of knowledgeable Executives their opinions with regard to future values of the items being forecasted </li></ul><ul><li>Presence of a powerful member in the group may prevent reaching consensus </li></ul><ul><li>Delphi Method </li></ul><ul><li>Involves a group of Experts who eventually develop a consensus </li></ul><ul><li>They usually make long range forecasts for future technologies or future sales of a new product </li></ul><ul><li>Sales Force Composite </li></ul><ul><li>Sales people are a good source of information with regard to customers’ future intentions to buy a product </li></ul><ul><li>Customer Surveys </li></ul><ul><li>By using a customer survey, a Firm can base its demand forecast on the customers’ purchasing plans </li></ul>
  10. 10. Quantitative Methods Time Series Models (Only independent variable is the time used to analyse 1) Trends, or 2) Seasonal, or 3) Cyclical Factors that influence the demand data) Casual Models (Employ some factors other than Time, when predicting forecast values)
  11. 11. Time Series Models <ul><li>Trends </li></ul><ul><li>Gradual upward or downward movement of data over time </li></ul><ul><li>Trends reflect changes in population levels, technology, and living standards </li></ul><ul><li>Long term movement </li></ul><ul><li>Seasonality </li></ul><ul><li>Variation that repeats itself at fixed intervals </li></ul><ul><li>It can be as long as a Year, or as short as a few hours </li></ul><ul><li>Can correspond to the Seasons of the Year, Holidays, or other special periods </li></ul><ul><li>Short-term regular and repetitive variations in data </li></ul><ul><li>Cyclical </li></ul><ul><li>Has a duration of at least one year. The duration varies from cycle to cycle </li></ul><ul><li>Long(er) term, requires many years of data to determine its repetitiveness or unusual circumstances (Eg. ups and downs of general business economy, war) </li></ul><ul><li>Random </li></ul><ul><li>Variations in demand that cannot be explained by Trends, Seasonality, or Cyclicality </li></ul><ul><li>Caused by chance </li></ul>
  12. 12. Time Series Models <ul><li>Smoothing Models </li></ul><ul><li>Moving Average (Simple & Weighted) </li></ul><ul><li>Single Exponential Smoothing </li></ul><ul><li>Double Exponential Smoothing </li></ul><ul><li>Decomposition Models </li></ul><ul><li>Additive Models </li></ul><ul><li>Multiplicative Models </li></ul>
  13. 13. Quantitative Methods - Examples
  14. 14. Simple Moving Average F 4 =(650+678+720)/3 =682.67 F 7 =(650+678+720 +785+859+920)/6 =768.67
  15. 15. Simple Moving Average
  16. 16. Exponential Smoothing <ul><li>Premise — determine how much weight to put on recent information versus older information </li></ul><ul><li>0 < a < 1 </li></ul><ul><li>High a such as .7 puts weight on recent demand </li></ul><ul><li>Low a such as .2 puts weight on many previous periods </li></ul>F t+1 =  D t + (1-  )F t (  is the smoothing parameter)
  17. 17. Exponential Smoothing F 1 =820+(0.5)(820-820)=820 F 3 =820+(0.5)(775-820)=797.75
  18. 18. Effect of  on Forecast
  19. 19. Simple Linear Regression Model
  20. 20. Simple Linear Regression Model (Contd)
  21. 21. Simple Linear Regression Model (Contd) Y t = 143.5 + 6.3x 135 140 145 150 155 160 165 170 175 180 1 2 3 4 5 Period Sales Sales Forecast
  22. 22. Simple Linear Regression Model (Contd) Actual observation (y value) Least squares method minimises the sum of the squared errors (deviations) Time period Values of Dependent Variable Deviation 1 (error) Deviation 5 Deviation 7 Deviation 2 Deviation 6 Deviation 4 Deviation 3 Trend line, y = a + bx ^
  23. 23. Forecast Accuracy / Error Reduction
  24. 24. Forecast Accuracy <ul><li>Forecast bias </li></ul><ul><ul><li>persistent tendency for forecasts to be greater or less than the actual values of a time series </li></ul></ul><ul><li>Forecast error </li></ul><ul><ul><li>difference between the actual value and the value that was predicted for a given period </li></ul></ul>
  25. 25. Forecast Accuracy (Contd.) <ul><ul><li>where </li></ul></ul><ul><ul><ul><li>e t = forecast error for Period t </li></ul></ul></ul><ul><ul><ul><li>A t = actual demand for Period t </li></ul></ul></ul><ul><ul><ul><li>F t = forecast for Period t </li></ul></ul></ul>
  26. 26. Forecast Error Measures <ul><li>Bias </li></ul><ul><ul><li>indicates on an average basis, whether the forecast is too high (negative bias indicates over forecast) or too low (positive bias indicates under forecast) </li></ul></ul><ul><li>Mean Absolute Deviation (MAD) </li></ul><ul><ul><li>indicates on an average basis, how many units the forecast is off from the actual data </li></ul></ul><ul><li>Mean Absolute Percent Error (MAPE) </li></ul><ul><ul><li>indicates on an average basis, how many percent the forecast is off from the actual data </li></ul></ul><ul><li>Mean Squared Error (MSE) </li></ul><ul><ul><li>a forecast error measure that penalises large errors proportionally more than small errors </li></ul></ul>
  27. 27. Forecast Error Measures <ul><li>Bias = </li></ul><ul><li>MAD = </li></ul><ul><li>MSE = </li></ul><ul><li>MAPE = </li></ul><ul><li>Standard Deviation (σ) = </li></ul>
  28. 28. Mean absolute deviation (MAD) <ul><li>the average absolute forecast error </li></ul><ul><ul><li>where </li></ul></ul><ul><ul><ul><li>| e t |= absolute value of the forecast error for Period t </li></ul></ul></ul><ul><ul><ul><li>n = number of periods of evaluation </li></ul></ul></ul>
  29. 29. Mean Absolute Percentage Deviation (MAPE) <ul><li>the average absolute percent error </li></ul><ul><ul><li>where </li></ul></ul><ul><ul><ul><li>et = forecast error for Period t </li></ul></ul></ul><ul><ul><ul><li>n = number of periods of evaluation </li></ul></ul></ul><ul><ul><ul><li>A t = actual demand for Period t </li></ul></ul></ul>
  30. 30. Running Sum of Forecast Errors (RSFE) <ul><li>provides a measure of forecast bias </li></ul><ul><ul><li>where </li></ul></ul><ul><ul><li>e t = forecast error for Period t </li></ul></ul>
  31. 31. Tracking Signal <ul><li>The ratio of cumulative forecast error to the corresponding value of MAD </li></ul><ul><li>Used to monitor a forecast </li></ul>
  32. 32. Mean Absolute Deviation Month Sales Forecast Abs Error 1 220 n/a 2 250 255 5 3 210 205 5 4 300 320 20 5 325 315 10 40 Note that by itself, the MAD only lets us know the mean error in a set of forecasts.
  33. 33. Forecast Error Measures Period Sales (A) Forecast E |E| E 2 |E|/A 1 1600 1650 -50 50 2500 0.0313 2 2200 2010 190 190 36100 0.0864 3 2000 2200 -200 200 40000 0.1000 4 1600 1580 20 20 400 0.0125 5 2500 2480 20 20 400 0.0080 6 3500 3520 -20 20 400 0.0057 7 3300 3310 -10 10 100 0.0030 8 3200 3200 0 0 0 0.0000 9 3900 3850 50 50 2500 0.0128 10 4700 4720 -20 20 400 0.0043 10     -20 580 82800 0.2639 Bias = -2 low/High MAD = 58 MSE = 8280 MAPE= 2.64%
  34. 34. Integrate – Sales Forecast / Production
  35. 35. Forecasting Process Services Collect Data Select Model Plot Data Develop Forecast Check Accuracy Forecast Adjust Forecast Monitor Forecast Sales and Operations Planning Master Scheduling Customer Scheduling Materials Planning Workforce Scheduling Order Scheduling Manufacturing Forecasting
  36. 36. Integrate - Sales Forecast & Production
  37. 37. CPFR - Collaborative Planning, Forecasting and Replenishment
  38. 38. CPFR - Overview <ul><li>Developed by Wal-Mart and Warner-Lambert in 1995 </li></ul><ul><li>Recognised as a breakthrough business model for planning, forecasting, and replenishment which goes beyond the traditional practice </li></ul><ul><li>Uses Internet-based technologies to collaborate from planning to execution </li></ul><ul><li>Creates a direct link between the consumer and the supply chain </li></ul><ul><li>Improves the quality of the demand signal for the entire supply chain through a constant exchange of information from one end to the other </li></ul><ul><li>Focuses on information sharing among supply chain trading partners for purposes of planning, forecasting, and inventory replenishment </li></ul>
  39. 39. CPFR Model
  40. 40. CPFR - Process <ul><li>The plan and the forecast are entered by suppliers and buyers into an Internet accessible system </li></ul><ul><li>Within established parameters, any of the participating partners is empowered to change the forecast </li></ul>
  41. 41. <ul><li>“ You may have to fight a battle more than once to win it.” </li></ul><ul><ul><li>- Margaret Thatcher </li></ul></ul>
  42. 42. <ul><li>Good Luck </li></ul><ul><li> </li></ul>