While forecasting is, by definition, an inexact science, its purpose is to improve the quality of predictions and thereby enable the fine-tuning of the balance between supply and demand. In forecasting environments where product ranges and configurations are extensive, bills of material are complex and raw material and component lead times are long, improved forecast accuracy is essential to minimizing inventory holding and obsolescence costs without compromising customer service and brand loyalty.
SYSPRO provides tools to enable you to track and evaluate your forecasts, as well as to identify the possible causes of forecast errors. These tools assist you in optimizing your forecasts to produce the best possible outcome with minimum forecast error. The Inventory Optimization suite of modules enables the entry of manual forecasts, as well as the automatic generation of forecasts via a variety of forecast algorithms, including those that compute for trends, seasonality and cyclical events. In addition, the module provides a competition forecasting method (also known as focus forecasting or the tournament method) which attempts to select the most suitable forecast algorithm, based on a selected error measurement and your SYSPRO sales history data. This enables you to compare results to determine which method is the most suitable for any particular item, whether you forecast products at code, revision, release, or warehouse level. A tracking signal is used to indicate when the validity of the forecast might be in doubt; those items with high forecast errors are highlighted and can be reviewed and adjusted as required.
Forecast accuracy depends not only on the regular evaluation of the forecast error, but also on the integrity and nature of the source data. In SYSPRO, forecast accuracy is enhanced through the ability to automatically filter and adjust outliers, as well as to manually adjust forecast base data for quantitative and qualitative factors. Lost sales can also be included in the demand profile.
Furthermore, SYSPRO's proxy functionality enables the forecasting of new products based on the sales history of similar superseded products which is useful in industries where product life cycles are short.