When estimating the short term requirements, the forecasting module is accessed via the Forecasts button, either on the item Sales forecast window, or on the Forecasts table window. In this case, the forecasting module generates weekly forecasts and the historical data are weekly data (typically weekly sales). The first time a given item is accessed, the associate table includes four lines corresponding to the 52 weeks of the current yearly period.
The historical data are entered in the cells of the tables (one can use the arrow keys to go from one week to another). The corresponding graph is immediately updated. It should be noted that each line of the table corresponds to a quarter.
In order to enter data for the previous year, a set of four lines with empty cells can be created by clicking on the button Previous Year. Similarly, a set of four lines with empty cells can be created by clicking on the button Next Year, in order to enter data for the next yearly period.
The historical data related to a given year can be cancelled by clicking on the first or the last line of the corresponding table and then by clicking on the Delete button.
When all the historical data have been entered, the process is validated by clicking on the button OK and the maintenance window is closed. All the linked numerical results are updated.
Nota : for a family item type, it is also possible to define weekly historical data (although for such item types, the data are monthly data that are exploited in the Sales and Operations Plans, typically analyzed and optimized on a monthly basis). From such weekly data, the Forecasting module estimates the weekly family sales forecasts for the future periods and transfers these results into the Sales forecast window. Via the Commercial bill of materials, these aggregate weekly forecasts can then be separated into weekly forecasts for the normal items corresponding to the disaggregated family item. This corresponds to the implementation of a process of disaggregation of a family-aggregated forecast into simple-items forecasts. This forecast process is called a top-down process.