May I ask how the base line WMAPE got calculated in Demand sensing profile, as it's works as threshold to decide if the demand sensing will be triggered for respective planning objects.
The forecast/demand planning qty. is got calculated monthly and got disaggregated into tech week, why the demand sensing calculated in daily time profile.
The WMAPE consider how many periods, at which level please.
Thanks a lot.
In SAP Integrated Business Planning (IBP), the Weighted Mean Absolute Percentage Error (WMAPE) is used as a baseline threshold in the Demand Sensing profile to determine whether the demand sensing algorithm should be triggered for the respective planning objects.
The WMAPE is calculated using historical data and serves as a benchmark to assess the accuracy of the forecast. It measures the average absolute percentage error of the forecasted quantities compared to the actual demand. A lower WMAPE indicates higher forecast accuracy.
Regarding the time profile for calculating the demand sensing, SAP IBP calculates the WMAPE at a daily level, even if the forecast or demand planning quantities are calculated at a higher level (e.g., monthly or weekly). This allows for a more granular analysis of forecast accuracy and the detection of short-term fluctuations in demand.
The specific number of periods used in the WMAPE calculation and the level at which it is applied can be configured in the Demand Sensing profile in SAP IBP. The number of periods typically refers to a historical time range used to calculate the WMAPE, such as the last six months or twelve months. The level at which it is applied depends on how the data is structured in your planning model, whether it's at a daily, weekly, or monthly level.
The configuration of the WMAPE calculation and the corresponding time profile settings can be adjusted based on your specific business requirements and historical data availability. It's important to strike a balance between capturing short-term demand variations and ensuring statistical significance in the forecast accuracy assessment.
The calculation of the baseline WMAPE involves comparing the historical demand data (actuals) with the forecasted or planned quantities for the respective planning object. The WMAPE is calculated as the weighted average of the absolute percentage errors for each time period, where the weights represent the relative importance of each period.