Student Learning Outcome: The student will apply a time series forecasting method to calculate forecasts, and use the Mean Absolute Deviation (MAD) and Tracking Signal to evaluate the accuracy of the forecasts.
Scenario: A manager must ensure that an adequate supply of copier paper is maintained in inventory. This item is bulky and takes up space, so the manager wants to avoid having excess inventory on hand. However, running out of paper will immediately cause problems in the daily functioning of the office, by preventing necessary activities from taking place. They operate seven days per week. Because demand for (usage of) the paper varies from day to day, the manager needs to forecast (estimate) how much will be used in the future. The following data is available on daily demand over the past two weeks. Past forecasts are also given in the table where available. Forecasts are rounded to the nearest tenth of a case for each day.
Day 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Demand (cases) 11 14 12 10 15 17 16 20 18 18 22 21 20 24 ?
Forecast 10.0 10.2 11.0 11.2 11.0 11.8 12.8 13.4 14.7 15.4
1. Using simple exponential smoothing with a smoothing constant of α = 0.2, and the demand and forecast data as given above, calculate forecasts for days 11 through 15. Put your answers in column 3 of the table below.
2. Starting with day 6, calculate the MAD and Tracking Signal (TS) for days 6 through 10. Put your answers in columns 6 and 7 of the table below. Based on the TS values, is the forecast in control during this period? If not, briefly describe the problem and suggest a change that could be made to improve future forecasts.
Day Demand Forecast Error RSFE MAD TS
6 17 11.8
7 16 12.8
8 20 13.4
9 18 14.7
10 18 15.4
11 22 XXXXX XXXXX XXXXX XXXXX
12 21 XXXXX XXXXX XXXXX XXXXX
13 20 XXXXX XXXXX XXXXX XXXXX
14 24 XXXXX XXXXX XXXXX XXXXX
15 ? XXXXX XXXXX XXXXX XXXXX