Demand Planning Data

The figure above illustrates the overall conceptual flow of the demand plan module.  The server reads data from the IFS database (the demand and forecast model for each part). The demand is aggregated from transactions and collected into periods (typically monthly). All this data is stored in memory on the Demand Plan Server. The Demand Plan Client sends commands to the server; these commands are processed by the server and a response is sent back to the client. As a result of this demand planning client and server interaction, a new forecast is generated. A typical work schedule for developing new demand plans might be:

1. On the third of each month, the demand from the previous month is aggregated.
2. A new system forecast is created for this month.
3. A user adjusts the computer-generated forecasts.
4. On the eighth of each month, the Demand Plan Server is locked, i.e., no changes in forecasts are allowed.
5. On the eighth of each month, run the Writeback Forecasts to the database scheduled job.
6. On the ninth of each month, the new forecasts are available to the IFS Planning MRP and MASSCH components to import.

The server can be set up to perform these jobs automatically (except for 3). See Forecast Server Setup/Scheduling.

It is critical to take backups from the Demand Plan Server after you have made changes to the forecasts. This is because the server stores all forecast in memory; data is only written to the database or a file when the backup is executed. In case of a power failure, all adjustments done after the last backup are destroyed. Backups can be scheduled to run each day or each hour.