PERTEMUAN 14 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-1 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Forecasting To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-2 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Learning Objectives Students will be able to: 1. Understand and know when to use various families of forecasting models 2. Compare moving averages, exponential smoothing, and trend time-series models 3. Seasonally adjust data. 4. Understand Delphi and other qualitative decision-making approaches To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-3 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Learning Objectives continued Students will be able to: 5. Identify independent and dependent variables and use them in a linear regression model. 6. Compute a variety of error measures. To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-4 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Chapter Outline 5.1 Introduction 5.2 Types of Forecasts 5.3 Scatter Diagrams 5.4 Measures of Forecast Accuracy 5.5 Time-Series Forecasting Models 5.6 Causal Forecasting Models 5.7 Monitoring and Controlling Forecasts 5.8 Using the Computer to Forecast To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-5 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Introduction Eight steps to forecasting: 1. Determine the use of the forecast 2. Select the items or quantities to be forecasted 3. Determine the time horizon of the forecast 4. Select the forecasting model or models 5. Gather the data needed to make the forecast 6. Validate the forecasting model 7. Make the forecast 8. Implement the results To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-6 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Forecasting Models Fig. 5.1 Forecasting Techniques Qualitative Models Time Series Methods Causal Methods Delphi Methods Moving Average Regression Analysis Jury of Executive Opinion Exponential Smoothing Multiple Regression Sales Force Composite Consumer Market Survey To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna Trend Projections Decomposition 5-7 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Annua l Sa le s Scatter Diagram for Sales Fig. 5.2 450 400 350 300 250 Televisions 200 150 100 50 0 0 2 4 6 8 10 Time (Years) To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-8 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 12 Decomposition of Time Series Time series can be decomposed into: • Trend (T): gradual up or down movement over time • Seasonality (S): pattern of fluctuations above or below trend line that occurs every year • Cycles(C): patterns in data that occur every several years • Random variations (R): “blips”in the data caused by chance and unusual situations To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-9 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Decomposition of Time Series Two Models Multiplicative model: demand = T * S * C * R Additive model: demand = T + S + C + R To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-10 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Product Demand Showing Components 650 Actual Data Demand 550 Trend 450 350 250 150 Cyclic 50 -50 Random -150 0 1 2 3 4 Time (Years To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-11 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 5 Moving Averages Moving average: demand in previous n periods n To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-12 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Calculation of ThreeMonth Moving Average Month Actual Shed Sales Three-Month Moving Average January 10 February 12 March 13 April 16 (10+12+13)/3 = 11 2/3 May 19 (12+13+16)/3 = 13 2/3 June 23 (13+16+19)/3 = 16 July 26 (16+19+23)/3 = 19 1/3 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-13 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Table 5.2 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-14 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Weighted Moving Averages Weighted moving average = Σ(weight for period n) (demand in period n) Σweights To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-15 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Calculating Weighted Moving Averages Weights Applied Period Last month 3 Two months ago 2 Three months ago 1 3*Sales last month + 2*Sales two months ago + 1*Sales three months ago Sum of weights 6 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-16 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Calculation of ThreeMonth Moving Average Month Actual Three-Month Moving Average Shed Sales January 10 February 12 March 13 April 16 [3*13+2*12+1*10]/6 = 12 1/6 May 19 [3*16+2*13+1*12]/6 =14 1/3 June 23 [3*19+2*16+1*13]/6 = 17 July 26 [3*23+2*19+1*16]/6 = 20 1/2 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-17 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Table 5.3 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-18 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Exponential Smoothing New forecast = previous forecast + (previous actual - previous) or: where Ft = Ft-1 + (At-1 - Ft-1) Ft = new forecast Ft-1 = previous forecast = smoothing constant At-1 = previous period actual To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-19 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Selecting the Smoothing Constant () Select to minimize: Mean Absolute Deviation = MAD Σ | forecast errors | n Mean Square Error = MSE Σ(forecast errors) 2 n Mean Absolute Percent Error = MAPE 1 forecast error Σ n actual Bias = forecast errors To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-20 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Table 5.4 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-21 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Table 5.5 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-22 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Exponential Smoothing with Trend Adjustment Forecast including trend (FITt+1) = new forecast (Ft) + trend correction(Tt) Tt = (1 - )Tt-1 + (Ft – Ft-1) where Ti = smoothed trend for period 1 Ti-1 = smoothed trend for the preceding period = trend smoothing constant Ft = simple exponential smoothed forecast for period t Ft-1 = forecast for period t-1 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-23 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Exponential Smoothing with Trend Adjustment • Simple exponential smoothing first-order smoothing • Trend adjusted smoothing second-order smoothing • Low gives less weight to more recent trends, while high gives higher weight to more recent trends To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-24 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Trend Projection General regression equation: Ŷ a bX where Ŷ computed value of the variable to be predicted (dependent variable) a Y - axis intercept XY - nXY b 2 2 X nX To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-25 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Midwestern Manufacturing’s Demand 160 150 140 130 120 110 100 90 80 70 60 Trend Line Forecast points Actual demand line 1993 1994 1995 1996 1997 1998 1999 2000 2001 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-26 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Seasonal Variations Month Sales Demand Average Two-Year Demand Average Monthly Demand Seasonal Index Year Year 2 1 Jan 80 100 90 94 0.957 Feb 75 85 80 94 0.851 Mar 80 90 85 94 0.904 Apr 90 110 100 94 1.064 May 115 131 123 94 1.309 … … … … … … Total Average Demand 1,128 Seasonal Index: = Average 2 -year demand/Average monthly demand To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-27 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Using Regression Analysis to Forecast Y X Triple A' Sales Local Payroll ($100,000's) ($100,000,000) 2.0 1 3.0 3 2.5 4 2.0 2 2.0 1 3.5 7 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-28 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Using Regression Analysis to Forecast - continued Sales, Y Payroll, X X2 XY 2.0 1 1 2.0 3.0 3 9 9.0 2.5 4 16 10.0 2.0 2 4 4.0 2.0 1 1 2.0 3.5 7 49 24.5 Y = 15 X = 18 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna X2 = 80 XY = 51.5 5-29 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Using Regression Analysis to Forecast - continued Calculating the required parameters: X ΣX 18 3 6 3 Y ΣY 15 2.5 6 6 b ΣXY nX Y ΣX 2 nX 2 51.5 6 * 3 * 2.5 0.25 2 80 6 * 3 a Y - bX 2.5 - 0.25 * 3 1.75 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-30 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Tr iple A's Sale s ($100,000) Standard Error of the Estimate 4 3 2 1 0 0 1 2 3 4 5 6 7 Area Payroll ($100,000,000) To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-31 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 8 Standard Error of the Estimate - continued S Y ,X (Y Y )2 c n2 where Y Y value of each data point Y value of the dependent variable c computed from the regression equation n number of data points or: S Y,X To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna Y 2 aY b XY n 2 5-32 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Triple A’s Calculations To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-33 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Triple A’s Calculations - continued SY ,X SY ,X Y 2 a Y b XY n 2 39.5 (1.75 )(15.0 ) (0.25 )(51.5 ) 6 2 0.09375 0.306 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-34 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Correlation Coefficient r nX nX ΣXΣY 2 (X) 2 nY 2 (Y) 2 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-35 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Triple A’s Calculations - continued r nΣΣ nΣΣ ΣXΣY 2 (ΣΣX nΣΣ (ΣΣY 2 2 6 * 51.5 - 18 * 15.0 (6 * 80 - 18 2 ) (6 * 39.5 - 15.0 2 ) 309 - 270 156 * 12 39 1872 39 43.3 0.901 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-36 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 2 Correlation Coefficient - Four Values - Fig. 5.7 To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-37 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Monitoring/Controlling Forecasts The Tracking Signal RSFE TrackingSi gnal MAD Σ(actual demand in period i forecast demand in period i) MAD where Σ | forecast errors | MAD n To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-38 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458 Monitoring/Controlling Forecasts The Tracking Signal To accompany Quantitative Analysis for Management, 8e by Render/Stair/Hanna 5-39 © 2003 by Prentice Hall, Inc. Upper Saddle River, NJ 07458