1. © Wiley 20101 Chapter 8 - Forecasting Operations Management by R. Dan Reid & Nada R. Sanders 4th Edition © Wiley 2010

2. © Wiley 20102 Learning Objectives Identify Principles of Forecasting Explain the steps in the forecasting process Identify types of forecasting methods and their characteristics Describe time series and causal models Generate forecasts for data with different patterns: level, trend, seasonality, and cyclical Using linear regression to account for trend Method to forecast demand with seasonality and trend Compute forecast accuracy Explain how forecasting models should be selected

3. © Wiley 20103 Principles of Forecasting 1. Many types of forecasting models differ in complexity and amount of data & way they generate forecasts 2. Forecasts are rarely perfect 3. Forecasts are more accurate for grouped data than for individual items 4. Forecast are more accurate for shorter than longer time periods

4. © Wiley 20104 Forecasting Steps Decide what needs to be forecast Level of detail, units of analysis & time horizon required Evaluate and analyze appropriate data Identify needed data & whether it’s available Select and test the forecasting model Cost, ease of use & accuracy Generate the forecast Monitor forecast accuracy over time

5. © Wiley 20105 Types of Forecasting Methods Forecasting methods are classified into two groups:

6. © Wiley 20106 Types of Forecasting Models Qualitative methods – judgmental methods Forecasts generated subjectively by the forecaster Educated guesses Quantitative methods – based on mathematical modeling: Forecasts generated through mathematical modeling

7. © Wiley 20107 Qualitative Methods

8. © Wiley 20108 Quantitative Methods Time Series Models: Assumes information needed to generate a forecast is contained in a time series of data Assumes the future will follow same patterns as the past Causal Models or Associative Models Explores cause-and-effect relationships Uses leading indicators to predict the future Housing starts and appliance sales

9. © Wiley 20109 Time Series Models Forecaster looks for data patterns as Data = historic pattern + random variation Historic pattern to be forecasted: Level (long-term average) – data fluctuates around a constant mean Trend – data exhibits an increasing or decreasing pattern Seasonality – any pattern that regularly repeats itself and is of a constant length Cycle – patterns created by economic fluctuations Random Variation cannot be predicted

10. © Wiley 201010 Time Series Patterns

11. © Wiley 201011 Time Series Models Naive: G ood for level patterns where F t+1 = Forecast for period t + 1, A t = Actual value for period t. Moving Average: where n = number of periods The average value over a set time period (e.g.: the last four weeks) Each new forecast drops the oldest data point & adds a new observation More responsive to a trend but still lags behind actual data

12. © Wiley 201012 Time Series Models (cont.) Weighted Moving Average: w here C t = weight used for period t (C t < 1) All weights must add to 100% or 1.00 e.g. C t.5, C t-1.3, C t-2.2 (weights add to 1.0) Allows emphasizing one period over others; above indicates more weight on recent data (C t =.5) Differs from the simple moving average that weighs all periods equally - more responsive to trends

13. © Wiley 201013 Time Series Models (con’t.) Exponential Smoothing: Most frequently used time series method because of ease of use and minimal amount of data needed Need just three pieces of data to start: Last period’s forecast (F t ) Last periods actual value (A t ) Select value of smoothing coefficient,,between 0 and 1.0 If no last period forecast is available, average the last few periods or use naive method Higher values (e.g..7 or.8) may place too much weight on last period’s random variation

14. © Wiley 201014 Time Series Problem Determine forecast for periods 7 & 8 2-period moving average 4-period moving average 2-period weighted moving average with t-1 weighted 0.6 and t-2 weighted 0.4 Exponential smoothing with alpha=0.2 and the period 6 forecast being 375 PeriodActual 1300 2315 3290 4345 5320 6360 7375 8

15. © Wiley 201015 Time Series Problem Solution PeriodActual2-Period4-Period2-Per.Wgted.Expon. Smooth. 1300 2315 3290 4345 5320 6360 7375340.0328.8344.0372.0 8 367.5350.0369.0372.6

16. © Wiley 201016 Time Series: Linear Regression A time series technique that computes a forecast with trend by drawing a straight line through a set of data using this formula: Y = a + bx where Y = forecast for period X X = the number of time periods from X = 0 a = value of y at X = 0 (Y intercept) b = slope of the line

17. © Wiley 201017 Linear Regression Identify dependent (y) and independent (x) variables Solve for the slope of the line Solve for the y intercept Develop your equation for the trend line Y=a + bX

18. © Wiley 201018 Linear Regression Problem: A maker of golf shirts has been tracking the relationship between sales and advertising dollars. Use linear regression to find out what sales might be if the company invested $53,000 in advertising next year. Sales $ (Y) Adv.$ (X) XYX^2Y^2 1130324160230416,900 2151527852270422,801 3150507500250022,500 4158558690302524964 5153.8553 Tot58918928202925387165 Avg147.2547.25

19. © Wiley 201019 Correlation Coefficient How Good is the Fit? Correlation coefficient (r) measures the direction and strength of the linear relationship between two variables. The closer the r value is to 1.0 the better the regression line fits the data points. Coefficient of determination ( ) measures the amount of variation in the dependent variable about its mean that is explained by the regression line. Values of ( ) close to 1.0 are desirable.

20. Forecasting Demand with Trend and Seasonal Characteristics Step 1: Calculate the seasonal index Step 2: Use linear regression to forecast the total demand. (In the following example, use the year as dependent variable, and yearly demand as independent variable) Step 3: Use the forecasted total demand (obtained in Step 2) and multiply by the seasonal index to determine the seasonal demand.

21. Seasonality: repetitive increase/ decrease in demand  Let i = season and j = year  Let  D ij  Let  j D ij = sum of demands for all the i seasons  Let  D ij  Let  i  j D ij = sum of all demands Season i index = S i =  D ij  j D ij  D  i  j D ij Step 1: Calculation of Seasonal Index

22. Step 1: Illustration 2007 12.68.66.317.545.0 2008 14.110.37.518.250.1 2009 15.310.68.119.653.6 Total 42.029.521.955.3148.7 DEMAND (1000’S PER QUARTER) YEAR1234Total S 1 = = = 0.28 D1D1DDD1D1DD 42.0148.7 S 2 = = = 0.20 D2D2DDD2D2DD 29.5148.7 S 4 = = = 0.37 D4D4DDD4D4DD 55.3148.7 S 3 = = = 0.15 D3D3DDD3D3DD 21.9148.7

23. Step 2: Use linear regression to forecast the total demand y= - 8584.83 + 4.30(2010) = 58.17 Using Linear regression, for 2010

24. Step 3: Calculation of seasonal demand Multiply the seasonal index by the forecasted total demand to determine the seasonal demand. SF 1 = (S 1 ) (F 2010 ) = (0.28)(58.17) = 16.28 SF 2 = (S 2 ) (F 2010 ) = (0.20)(58.17) = 11.63 SF 3 = (S 3 ) (F 2010 ) = (0.15)(58.17) = 8.73 SF 4 = (S 4 ) (F 2010 ) = (0.37)(58.17) = 21.53

25. © Wiley 201025 Measuring Forecast Error Forecasts are never perfect Need to know how much we should rely on our chosen forecasting method Measuring forecast error: Note that over-forecasts = negative errors and under-forecasts = positive errors

26. © Wiley 201026 Measuring Forecasting Accuracy Mean Absolute Deviation (MAD) measures the total error in a forecast without regard to sign Cumulative Forecast Error (CFE) Measures any bias in the forecast Mean Square Error (MSE) Penalizes larger errors Tracking Signal Measures if your model is working

27. © Wiley 201027 Accuracy & Tracking Signal Problem: A company is comparing the accuracy of two forecasting methods. Forecasts using both methods are shown below along with the actual values for January through May. The company also uses a tracking signal with ±4 limits to decide when a forecast should be reviewed. Which forecasting method is best? MonthActual sales Method AMethod B F’castErrorCum. Error Tracking Signal F’castErrorCum. Error Tracking Signal Jan.3028221 27221 Feb.2625132 132 March32 03329363 April29302227284 May3130133292105 MAD12 MSE1.44.4

28. © Wiley 201028 Selecting the Right Forecasting Model 1.The amount & type of available data  Some methods require more data than others 2.Degree of accuracy required  Increasing accuracy means more data 3.Length of forecast horizon  Different models for 3 month vs. 10 years 4.Presence of data patterns  Lagging will occur when a forecasting model meant for a level pattern is applied with a trend

29. © Wiley 201029 Forecasting Software Spreadsheets Microsoft Excel, Quattro Pro, Lotus 1-2-3 Limited statistical analysis of forecast data Statistical packages SPSS, SAS, NCSS, Minitab Forecasting plus statistical and graphics Specialty forecasting packages Forecast Master, Forecast Pro, Autobox, SCA

30. © Wiley 201030 Guidelines for Selecting Software Does the package have the features you want? What platform is the package available for? How easy is the package to learn and use? Is it possible to implement new methods? Do you require interactive or repetitive forecasting? Do you have any large data sets? Is there local support and training available? Does the package give the right answers?

31. © Wiley 201031 Other Forecasting Methods Focus Forecasting Developed by Bernie Smith Relies on the use of simple rules Test rules on past data and evaluate how they perform Combining Forecasts Combining two or more forecasting methods can improve accuracy

32. © Wiley 201032 Forecasting within OM: How it all fits together Forecasts impact all business decisions such as: product designs that are expected to sell (Ch 2), choice of competitive priorities, changes in processes, and large technology purchases (Ch 3), the quality of product to produce (Chs 5 and 6), capacity and location needs (Ch 9), future space requirements (Ch 10), the amount of labor needed (Ch 11), the amount of needed supplies and materials (Ch 12) tactical planning & developing worker schedules (Ch 15).

33. © Wiley 201033 Chapter 8 Highlights Forecasts are rarely perfect; they are more accurate for groups than individual items, and are more accurate in the shorter term than longer time horizons. The forecasting process involves five steps: decide what to forecast, evaluate and analyze appropriate data, select and test model, generate forecast, and monitor accuracy. Forecasting methods can be classified into two groups: qualitative and quantitative. There are four basic patterns of data: random variation, trend, seasonality, and cycles. Forecast models: naïve, simple mean, simple moving average, weighted moving average, and exponential smoothing. Linear regression model is used to forecast trends. Three useful measures of forecast error are mean absolute deviation (MAD), mean square error (MSE) and tracking signal. There are four factors to consider when selecting a model: amount and type of data available, degree of accuracy required, length of forecast horizon, and patterns present in the data.