Forecasting Models—Exercises

IMPORTANT
It is extremely important that you set up and work within your own Server ID to maintain your data integrity. If you work within any other server ID, you will compromise your own exercise data as well as the data of other students. Predictable exercise results require that your data be isolated in your own server ID.

Basic Data Setup

Defining Schedule

Purpose: The purpose of this exercise is to show you how to set up a schedule for the steps in the forecast cycle.

Windows:
Demand Plan Server Setup/Scheduling
Forecast Server

  1. Set up a schedule for your server ID.
  2. Start the Demand Planning Server.

General exercise for Defining Schedule

Main Exercises

Purpose: The purpose of these exercises is to give you an overview of a few of the forecast models and smoothing parameters. Changes will be made using both the Centura client and the Demand Client. However, they would normally be made in the Demand Client.

Windows:
Forecast Client
Forecast Part
Forecast Server

Forecasting Models—Naive

  1. On the Windows taskbar, click the Start button, point to Programs, and then click Forecast Client. Log on to the Demand Plan Client using appowner as the user ID and password.
  2. In the Flow ID list, select one of your flows.
  3. In the Part No list, select a part.
  4. In the Forecast Model list, select the Naïve model. Note that the system forecast will now be at the same level as the last adjusted demand. The explanation forecast will also change and be equal to the adjusted forecast, except with one period offset.

Forecasting Models—Manual

  1. If the demand plan server is running , exit the server by double-clicking the Demand Plan Server icon in the lower right corner of your screen.
  2. On the File menu, click Exit, and then click OK.
  3. Open the Forecast Part window. Query for one of your flows and a part within the selected flow. 
  4. Click the Forecast Model tab.
  5. In the Manual Yearly Forecast field, enter a manual forecast quantity. In the Forecast Model list, click Manual to change the forecast model to manual.
  6. On the Windows taskbar, click the Start button, point to Programs, and then click Demand Plan Server to start the server. 
  7. On the File menu, click StopServer. On the System menu, click Create Forecast.
  8. To start the server again select StartServer on the File menu.
  9. On the Windows taskbar, click the Start button, point to Programs, and then click Demand Plan Client. Log on to the Demand Plan Client using appowner as the user ID and password.
  10. In the Flow ID list, select the flow that contains the part for which you entered a manual forecast. In the Part No list, click selected part.
  11. The yearly demand has now been split over all forecast periods, and the explanation forecast has been brought to the same level.
  12. In the Std  Manual Yearly field, change the yearly quantity. The values will be an updated online in the graph.  

Forecasting Models—Moving Average

  1. On the Windows taskbar, click the Start button, point to Programs, and then click Forecast Client. Log on to the Demand Plan Client using appowner as the user ID and password.
  2. In the Flow ID list, select one of your flows.
  3. In the Part No list, select a part.
  4. In the Forecast Model list, select the Moving Average model. The system forecast is now the average of the adjusted demand over the number of months specified in the Moving Average Periods field in the Forecast Part/Forecast Model tab. The number of moving average periods can also be viewed or updated in the Moving Average Periods field in the client.

Forecasting Models—Exponential Weighted Moving Average

  1. On the Windows taskbar, click the Start button, point to Programs, and then click Forecast Client. Log on to the Demand Plan Client using appowner as the user ID and password.
  2. In the Flow ID list, select one of your flows.
  3. In the Part No list, select a part.
  4. In the Forecast Model list, select the EWMA Level model. The system forecast is now a weighted moving average of the adjusted demand. All periods are weighted unless the alpha value is set to 1.
  5. In the Alpha field, change the alpha value to 1. All weight is now on the most recent adjusted demand and the system forecast is set to the same level.
  6. Change the alpha value to 0.15. Now, greater weight is given to demands further away, and the result is that the system forecast level increases or decreases depending on historically adjusted demand.

Forecasting Models—Exponential Weighted Moving Average with Level and Trend

  1. On the Windows taskbar, click the Start button, point to Programs, and then click Forecast Client. Log on to the Demand Plan Client using appowner as the user ID and password.
  2. In the Flow ID list, select one of your flows.
  3. In the Part No list, select a part.
  4. In the Forecast Model list, select the EWMA Level and Trend model. The system forecast is now a weighted moving average of adjusted demand with consideration taken to the trend of the adjusted demand. All periods are weighted except when the alpha value is set to 1.
  5. In the Alpha field, change the alpha value to 1. In both the Beta and the Rho fields, enter a value of 1. All weight is now on the most recent demand. Because a trend is involved, the system forecast curve points steeply in another direction depending on the nature of the most recent adj. forecast demand.
  6. To dampen the trend, change the value in the Rho field to 0.9.
  7. To decrease the impact of the trend, change the value in the Beta field to 0.15.
  8. Change the value in the Alpha field to 0.15 to put more weight on adjusted demand periods that are further away. 
  9. Change the parameters as desired, and analyze the effect of your changes.

Forecasting Models—Least Square Regression

  1. On the Windows taskbar, click the Start button, point to Programs, and then click Forecast Client. Log on to the Demand Plan Client using appowner as the user ID and password.
  2. In the Flow ID list, select one of your flows.
  3. In the Part No list, select a part.
  4. In the Forecast Model list, select the Least Squares model. 
  5. In the Rho field, change the value to 1. 
  6. Change the rho value again—this time to 0.1. As you can see, the least square regression method is not sensitive to short-term fluctuations. This makes it suitable for intermediate- to long-term forecasts.