Forecast Profile

In the forecast profile , you can define, check, and change the control parameters for forecasting. There you can, for example, assign a forecast strategy to each location product. You can create a forecast profile manually or the Planning Service Manager (PSM) can create one automatically. For more information, see Forecast Profile Inheritance .

Forecast Run

You can either schedule the forecast run as a regular PSM planning service or you can start it manually on the Interactive Forecasting screen. During the forecast run, the system calculates the forecasting results at location product level. If you are not satisfied with the results, you can make manual corrections or adjustments to the demand history, forecast profiles, and PSM service profiles, and then restart the forecast run.

Standard Deviation and Mean Absolute Deviation (MAD)

The system calculates the standard deviation and the MAD for each location product. This calculation is realized as a PSM planning service.

Automatic Model Evaluation

Automatic model evaluation consists of Trigg’s tracking signal and tripping. Trigg’s tracking signal checks whether the forecast model is still optimal or whether there is a systematic forecasting error. Tripping reinitializes basic values and the deviation of an existing forecast model and shortens the demand history used for forecast creation. You can schedule Trigg’s tracking signal and tripping separately in the forecast service profile as part of the planning service. However, both functions are also included in the combined forecast.

For more information about the composite forecast, see Planning Services for Forecasting

Automatic Model Selection

You can use automatic model selection to automatically determine the forecast model best suited to a specific location product. You can start automatic model selection manually or you can schedule it in the PSM. If you want to schedule automatic model selection in the PSM, you can schedule it separately in the forecast service profile as part of the forecast service. However, automatic model selection is also included in the combined forecast.

Rough-Tuning and Fine-Tuning

For the forecast models that consider smoothing factors during forecast creation, you can automatically optimize these smoothing factors by using rough-tuning and fine-tuning . You can schedule rough-tuning separately in the forecast service profile as part of the forecast service. However, it is also included in the combined forecast. You can only start fine-tuning manually.

Forecast Disaggregation

The system calculates the forecast on the basis of aggregated demand history at the level of the entry location. It then disaggregates the forecast result to the child locations of the entry locations. Forecast disaggregation is realized as a stand-alone PSM planning service. However, the system also performs forecast disaggregation when you manually save the forecast results on the Interactive Forecasting screen.

Forecast Approval

Forecast approval calculates the final forecast for each location product and checks which forecast results require manual approval according to rules that you define. The system automatically approves all other forecast results and provides these results to other SPP planning services. Forecast approval is realized as a PSM planning service. However, the system also performs forecast approval when you manually save the forecast results on the Interactive Forecasting screen.

Historical Forecasts

To ensure an exact planning, the system records all calculated forecast results as historical forecasts . You can also schedule the PSM planning service Recalculation of the Forecast in the Past . If the forecast model changes, the system triggers this planning service so that it creates historical forecasts for the new combination of location product and forecast model.

Display Forecast Results

You can display forecast results for individual locations products on the Interactive Forecasting screen.

Leading-Indicator-Based Forecast

In addition to the forecast that the system creates based on the past demand of a location product, you can also create a leading-indicator-based forecast . Leading indicators form the basis for this type of forecast. These leading indicators arise from your items of equipment for which your service parts are used. Possible leading indicators are, for example, your installed base (that is all of your items of equipment), the operating time of each individual item of equipment, or the number of implementations of each item of equipment.

Phase-In Planning

Phase-in planning uses values based on experience to calculate forecast values for new products that do not yet have any historical data. Phase-in planning is realized as a PSM planning service.

Phase-Out Planning

Phase-out planning calculates forecast values at entry location level for products in discontinuation. Phase-out planning is realized as a PSM planning service.