Forecast Evaluation

This process is used to evaluate the quality of the forecasts made for planning and inventory control purposes. This evaluation considers both how well the forecasts correspond to actual demand and how the periodic forecasts changed from one period to the next. Evaluation can be done for separate items or for groups of items.

This process is performed every period, most suitably at the end, when demand history is complete.

Before you start

The process can be started when these prerequisites are met:

  • The item groups included in a forecast comparison are set.
  • The forecasts are created and saved for the items included in the evaluation.
  • Updated forecast comparisons from previous forecasting periods are saved.

Follow these steps

The process includes these activities:

  1. Create Forecast IDs

  2. Update Forecast Comparisons

  3. Print Basis for Evaluation

  4. Analysis and Evaluation

Create Forecast IDs

The process starts by assigning IDs to the forecasts from the latest period to be evaluated, as described in procedure Create Forecast Comparison.

Forecast comparisons are made of a set of forecasts for specific items, warehouses, and periods. The forecast IDs facilitate the process of comparing these forecasts. The ID shows both the forecast and the period for which the forecast was made. For example, forecast ID PrjfrA9604 shows forecast PrjfrA created for period 4 in year 1996.

Update Forecast Comparisons

This activity covers saving sets of forecast from the preceding period as described in procedure Update Forecast Comparison.

Print Basis for Evaluation

This activity involves obtaining two different printouts used for forecast evaluation. The first type is for forecast comparison and is used to determine the changes that take place in forecasting as more current forecasts are made over a specified period. For more information, see procedure Print Forecast Comparison.

This activity is most suitable for printing forecasts for the same groups of items over separate periods together. In this way corresponding forecasts from two or more periods can compared.

The second type of printout is for actual demand in relation to the latest forecast. This is used to determine the accuracy of the forecast in predicting actual events. For more information, see procedure Print Item Forecast.

Analysis and Evaluation

This activity involves analyzing the printed forecast comparisons to determine the quality of the forecasts. This evaluation is made to determine:

  • The degree to which the precision of the forecasting changes when shorter forecasting terms are used.
  • The accuracy of forecasting for the length of time corresponding to supply lead time.
  • The accuracy of the latest forecasts.
Related topics