1. T18-03 - 1 T18-03 Exponential Smoothing Forecast Purpose Allows the analyst to create and analyze the "Exponential Smoothing Average" forecast. The MAD and MSE for the forecast are calculated, and a graphical representation of history and forecast are shown. Inputs Historical Period Demand Smoothing Constant (  ) Outputs Exponential Smoothing Forecast Forecast Error MAD & MSE Graph showing Historical Demand and Exponential Smoothing Forecast Limitations 60 Time Series Observations

2. T18-03 - 2 Quantitative Methodologies Quantitative methods model historical demand variation patterns (random, trend, seasonal or cyclical). Once past history has been explained by a model, extrapolations can be made about the future. Some simplistic techniques are Averaging/Smoothing Models. Naïve. Moving average. Weighted moving average. Exponential smoothing These techniques are best used when demand is stable with no trend or seasonal pattern Stable

3. T18-03 - 3 Averaging/Smoothing Models Exponential Smoothing – next period forecast is another weighted average method based on the premise that the most recent period has the highest predictive value. Horizon: Short range Method: Strength: Same as moving average, smoothing constant can balance the benefits of smoothing with the benefits of responding to real changes Weakness: Not very accurate when a longer forecasting horizon is necessary, lags actual demand

4. T18-03 - 4 Given that a forecast is rarely correct, the methodology you choose should be the one which provides the least error from the actual historical demand. Forecast error is defined as the difference between actual historical demand and the forecast. Forecast Accuracy

5. T18-03 - 5 Forecast Error

6. T18-03 - 6 There are two measures used to monitor the accuracy of a forecast. The Mean Absolute Deviation (MAD) and the Mean Squared Error (MSE). The MAD is the average of the absolute value of the forecast errors. The MSE is the average of the squared forecast errors. Note : The formula for the MSE shown above may vary slightly. Some textbooks divide the sum of the squared errors by n-1 rather than n. Monitoring the Forecast

7. T18-03 - 7 Exponential Smoothing Example Prepare an exponential smoothing forecast with a smoothing constant of.1 Calculate the exponential smoothing forecast, MAD and MSE.

8. T18-03 - 8 Input the History Values, and smoothing constant in the light green cells. The Exponential Smoothing Forecast, Error, MAD, and MSE are automatically calculated.

9. T18-03 - 9 A graph showing the History Values and Exponential Smoothing Forecast is automatically produced.