运营管理第14版蔡斯IChap018课件

1、FORECASTINGChapter EighteenCopyright 2014 by The McGraw-Hill Companies, Inc. All rights reserved.McGraw-Hill/Irwin运营管理第14版蔡斯IChap018Learning ObjectivesLO181: Understand how forecasting is essential to supply chain planning.LO182: Evaluate demand using quantitative forecasting models.LO183: Apply qua

2、litative techniques to forecast demand.LO184: Apply collaborative techniques to forecast demand.18-2运营管理第14版蔡斯IChap018The Role of ForecastingForecasting is a vital function and affects every significant management decision.Finance and accounting use forecasts as the basis for budgeting and cost cont

3、rol.Marketing relies on forecasts to make key decisions such as new product planning and personnel compensation.Production uses forecasts to select suppliers; determine capacity requirements; and drive decisions about purchasing, staffing, and inventory.Different roles require different forecasting

4、approaches.Decisions about overall directions require strategic forecasts.Tactical forecasts are used to guide day-to-day decisions.18-3运营管理第14版蔡斯IChap018Forecasting and Decoupling Point18-4Decoupling point: Point at which inventory is stored, which allows SC to operate independentlyThe choice of th

5、e decoupling point in a SC is strategic.Forecasting helps determine the level of inventory needed at the decoupling points.The decision will be affected by the error produced in the forecast and the type of product (easily inventoried or easily perishable).运营管理第14版蔡斯IChap018Types of ForecastingThere

6、 are four basic types of forecasts.1.Qualitative2.Time series analysis (primary focus of this chapter)3.Causal relationships4.SimulationTime series analysis is based on the idea that data relating to past demand can be used to predict future demand.18-5运营管理第14版蔡斯IChap018Components of DemandAverage d

7、emand for a period of timeTrendSeasonal elementCyclical elementsRandom variationAutocorrelationExcel: Components of Demand18-6运营管理第14版蔡斯IChap018TrendsIdentification of trend lines is a common starting point when developing a forecast.Common trend types include linear, S-curve, asymptotic, and expone

8、ntial.18-7运营管理第14版蔡斯IChap018Time Series AnalysisUsing the past to predict the future Used mainly for tactical decisionsShort term forecasting less than 3 months Used to develop a strategy that will be implemented over the next 6 to 18 months (e.g., meeting demand)Medium term forecasting 3 months to

9、2 years Useful for detecting general trends and identifying major turning pointsLong term forecasting greater than 2 years18-8运营管理第14版蔡斯IChap018Model SelectionChoosing an appropriate forecasting model depends upon1.Time horizon to be forecast2.Data availability3.Accuracy required4.Size of forecastin

10、g budget5.Availability of qualified personnel18-9运营管理第14版蔡斯IChap018Forecasting Method Selection GuideForecasting MethodAmount of Historical DataData PatternForecast HorizonSimple moving average6 to 12 months; weekly data are often usedStationary (i.e., no trend or seasonality)ShortWeighted moving av

11、erage and simple exponential smoothing5 to 10 observations needed to startStationaryShortExponential smoothing with trend5 to 10 observations needed to startStationary and trendShortLinear regression10 to 20 observationsStationary, trend, and seasonalityShort to medium18-10运营管理第14版蔡斯IChap018Simple M

12、oving AverageForecast is the average of a fixed number of past periods.Useful when demand is not growing or declining rapidly and no seasonality is present.Removes some of the random fluctuation from the data.Selecting the period length is important.Longer periods provide more smoothing.Shorter peri

13、ods react to trends more quickly.18-11运营管理第14版蔡斯IChap018Simple Moving Average Formula18-12运营管理第14版蔡斯IChap018Simple Moving Average Example18-13运营管理第14版蔡斯IChap018Weighted Moving AverageThe simple moving average formula implies equal weighting for all periods.A weighted moving average allows unequal we

14、ighting of prior time periods.The sum of the weights must be equal to one.Often, more recent periods are given higher weights than periods farther in the past.18-14运营管理第14版蔡斯IChap018Selecting WeightsExperience and/or trial-and-error are the simplest approaches.The recent past is often the best indic

15、ator of the future, so weights are generally higher for more recent data.If the data are seasonal, weights should reflect this appropriately.18-15运营管理第14版蔡斯IChap018Exponential SmoothingA weighted average method that includes all past data in the forecasting calculationMore recent results weighted mo

16、re heavilyThe most used of all forecasting techniquesAn integral part of computerized forecastingWell accepted for six reasons1.Exponential models are surprisingly accurate.2.Formulating an exponential model is relatively easy.3.The user can understand how the model works.4.Little computation is req

17、uired to use the model.5.Computer storage requirements are small.6.Tests for accuracy are easy to compute.18-16运营管理第14版蔡斯IChap01818-17Exponential Smoothing ModelExponential Smoothing Model运营管理第14版蔡斯IChap018Exponential Smoothing ExampleWeekDemandForecas8203680 8114655 7855750 7596802 757

18、7798 7668689 7729775 75610 76018-18运营管理第14版蔡斯IChap018Exponential Smoothing Effect of Trends18-19运营管理第14版蔡斯IChap018Example Exponential Smoothing with Trend Adjustment18-20运营管理第14版蔡斯IChap018Choosing Alpha and Delta18-21运营管理第14版蔡斯IChap018Linear Regression AnalysisRegression is used to identify the func

19、tional relationship between two or more correlated variables, usually from observed data.One variable (the dependent variable) is predicted for given values of the other variable (the independent variable).Linear regression is a special case that assumes the relationship between the variables can be

20、 explained with a straight line.Y = a + bt18-22运营管理第14版蔡斯IChap018Example 18.2 Least Squares MethodQuarterSalesQuarterSales160072,60021,55082,90031,50093,80041,500104,50052,400114,00063,100124,900The least squares method determines the parameters a and b such that the sum of the squared errors is min

21、imized “least squares”18-23运营管理第14版蔡斯IChap018Example 18.2 Calculations16006001360,000801.321,5503,10042,402,5001,160.931,5004,50092,250,0001,520.541,5006,000162,250,0001,880.152,40012,000255,760,0002,239.763,10018,600369,610,0002,599.472,60018,200496,760,0002,959.082,90023,200648,410,0003,318.693,80

22、034,2008114,440,0003,678.2104,50045,00010020,250,0004,037.8114,00044,00012116,000,0004,397.4The forecast is extended to periods 13-1618-24运营管理第14版蔡斯IChap018Regression with ExcelMicrosoft Excel includes data analysis tools, which can perform least squares regression on a data set.18-25运营管理第14版蔡斯IChap

23、018Time Series DecompositionChronologically ordered data are referred to as a time series.A time series may contain one or many elements.Trend, seasonal, cyclical, autocorrelation, and randomIdentifying these elements and separating the time series data into these components is known as decompositio

24、n.18-26运营管理第14版蔡斯IChap018Seasonal VariationSeasonal variation may be either additive or multiplicative (shown here with a changing trend).18-27运营管理第14版蔡斯IChap018Determining Seasonal Factors : Simple Proportions Example 18.3The seasonal factor (or index) is the ratio of the amount sold during each se

25、ason divided by the average for all seasons.SeasonPast SalesAverage Sales for Each SeasonSeasonal FactorSpring200Summer350Fall300Winter150Total100018-28运营管理第14版蔡斯IChap018Example 18.3 Continued18-29ExpectedDemand forNext YearAverageSales forEach Season(1,100y4)SeasonalFactorNext YearsSeasonalForecast

26、Spring275X0.8=220Summer275X1.4=385Fall275X1.2=330Winter275X0.6=1651100运营管理第14版蔡斯IChap018Decomposition Using Least Squares Regression1.Decompose the time series into its components.a.Find seasonal component.b.Deseasonalize the demand.c.Find trend component.2.Forecast future values of each component.a

27、.Project trend component into the future.b.Multiply trend component by seasonal component.18-30运营管理第14版蔡斯IChap018Decomposition Steps 1 and 218-31Using the data for periods 1-12, apply time series analysis (decomposition, linear regression, trend estimate & seasonal indices) to forecast for perio

28、ds 13-16 运营管理第14版蔡斯IChap018Decomposition Steps 3 and 4Develop a least squares regression line for the deseasonalized data.Project the regression line through the period of the forecast.Regression Results: Y = 555.0 + 342.2tForecast for periods 13-1618-32运营管理第14版蔡斯IChap018Decompostion Step 5Create th

29、e final forecast by adjusting the regression line by the seasonal factor.PeriodQuarterY from RegressionSeasonal FactorForecast (F x Seasonal Factor13I5,003.50.824,102.8714II5,345.71.105,880.2715III5,687.90.975,517.2616IV6,030.11.126,753.7118-33运营管理第14版蔡斯IChap018Forecast ErrorsForecast error is the d

30、ifference between the forecast value and what actually occurred.All forecasts contain some level of error.Sources of errorBias when a consistent mistake is madeRandom errors that are not explained by the model being usedMeasures of errorMean absolute deviation (MAD)Mean absolute percent error (MAPE)

31、Tracking signal18-34运营管理第14版蔡斯IChap018Forecast Error MeasurementsIdeally, MAD will be zero (no forecasting error).Larger values of MAD indicate a less accurate model.MAPE scales the forecast error to the magnitude of demand.Tracking signal indicates whether forecast errors are accumulating over time

32、 (either positive or negative errors).18-35运营管理第14版蔡斯IChap018Computing Forecast Error18-36运营管理第14版蔡斯IChap018Causal Relationship ForecastingCausal relationship forecasting uses independent variables other than time to predict future demand.This independent variable must be a leading indicator.Many ap

33、parently causal relationships are actually just correlated events care must be taken when selecting causal variables.18-37运营管理第14版蔡斯IChap018Multiple Regression TechniquesOften, more than one independent variable may be a valid predictor of future demand.In this case, the forecast analyst may utilize multiple regression.Analogous to linear regression analysis, but with multiple independent variables.Multiple regression supported by statistical software packages.18-38运营管理第14版蔡斯IChap018Qualitative Forecasting TechniquesGenerally used to take advantage of expert kn