You use this report to identify the impact of different events on the historical demand. If you want to consider the effects of events during forecasting, you use this report to identify events. The system searches for outliers in the historical demand.
To use this function, you must activate the business function SCM-APO-FCS, Events and Outliers, Causals, ABC/XYZ Classification
(SCM_APO_FORECASTING_1
). For more information, see SCM-APO-FCS,
Events and Outliers, Causals, ABC/XYZ Classification.
After you have run the report, you can use the events the system has found during forecasting to estimate the impact of future events. Additionally, you can clean the history from the impacts of events during forecasting.
The system uses the events created by the Automatic Outlier Detection and Event Generation
report as follows:
On the SAP Easy Access
screen, under you can see the event labels and their impact, if you configured the planning book to display the events and their impact.
For more information about interactive demand planning, see Interactive Demand Planning.
In forecasting, you can specify that you want the system to consider the events created by the Automatic Outlier Detection and Event Generation
report. This means that the system removes the effects of the events from the history that serves as the basis
of forecasting. On the SAP Easy Access
screen, under , on the Univariate Profile
tab page, in the Event Types
screen area, select the Clean History
checkbox.
You can also configure forecasting so that the system estimates the
future occurrences of event types based on the average value of past events of the same type. On the Univariate Profile
tab page, in the Event Types
screen area, select the Estimate Automatically
checkbox.
You can run this report if the following applies:
You have assigned a key figure to save the impact of events and maintained the promotion level for your planning area.
On the SAP Easy Access
screen, choose . Since the effects of events can be either negative or positive
relative to the actual sales history, you have allowed the system to store negative values in this key figure.
You have defined a planning level in Customizing, under
.Make sure that this planning level is the same as the one you use for forecasting and XYZ classification.
Note
When you create a planning level, make sure that the sequence of the characteristics reflects a hierarchical relationship, from general to specific. Additionally, the promotion planning level must be the last level in the planning level definition.
On the SAP Easy Access
screen, under , you have created the following objects:
An event catalog
Generic event types
Generic event types are, for example, STOCKOUT
, LOW_PROMO
, MEDIUM_PROMO
, and HIGH_PROMO
.
The following table provides an example of the values you can assign to generic event types in Customizing for Advanced Planning and Optimization
, under :
Lower Limit in % |
Upper Limit in % |
Event Type |
---|---|---|
30 |
50 |
LOW_PROMO |
50 |
70 |
MEDIUM_PROMO |
70 |
N/A |
HIGH_PROMO |
N/A |
-40 |
STOCKOUT |
Note
You have to assign different event types for outliers with positive and negative impact. The outlier detection limits must be continuous, there must not be a gap in the range for upper and lower limit.
Note
In display mode no entry is displayed as a zero (0
). This means that there is no limit (or infinite limit).
You have assigned these event types to one of the event catalogs you specified in Customizing for Advanced Planning and Optimization
, under , in the Event Catalogs Used for Planning Areas
view.
Other event types
Other event types, are, for example, Christmas, and other holidays, as well as other reoccurring or unsystematic event types such as periodic or one-time promotions, political events, weather-related events. You have assigned these event types
to one of the event catalogs you specified in Customizing for Advanced Planning and Optimization
, under , in the Event Catalogs Used for Planning Areas
view.
In Customizing, under
, you have performed the following activities:You have created outlier detection profiles.
You have specified the planning level that the system uses for different planning areas during outlier detection.
You have assigned event catalogs to planning areas.
For more information, see the Customizing activity documentation.
You have created a selection on which you want the system to perform automatic outlier detection.
You have included event labels in the planning book definition by adding an auxiliary key figure in the Key Fig. Attributes
(Key Figure Attributes) tab page, and have selected the Show Event Types
checkbox. You have configured the
event impact key figure on the SAP Easy Access
screen, under . Here, you assign a key figure to the planning area as an event key figure. Make sure you add this key figure to your planning book.
If you want the system to detect price change events, the following must apply:
You must have created a price key figure (for example, 9APRICEFC
)
You must have created a price change event type
The price change event type is a non-reoccurring event type that you can create on the SAP Easy Access
screen, under . When creating price change event types, you must make sure that the periodicity you specify for the price change event type matches the periodicity
you specified for the Automatic Outlier Detection and Event Generation
report in the Periods
field.
The report reads the historical demand of the specified selection for the specified time period, and based on the settings you specify in Customizing, under Configure Outlier Detection and Event Catalog Assignments
. The system generates events for the identified event types to store the impact of the events.
The system takes the demand history in the specified time horizon as the basis for outlier detection. If a certain combination does not have a history at the beginning or at the end of the horizon, the system adjusts the outlier detection horizon by removing the periods that do not have historical data from the beginning and from the end of the horizon.
The system calculates a moving average for the number of periods configured in Customizing, and checks if the period currently processed reaches the threshold you defined in the outlier detection profile. The parameters for moving average calculation are based on the settings you defined in
Customizing, under Outlier Detection Period Before and After Settings
view.
The system processes the outliers by order of magnitude, which means that it first corrects the history for the period that has the largest deviation. This iteration makes sure that the outliers that would otherwise distort the moving average are handled first.
The following section describes how the system handles events in different cases:
Event types found in event catalogs
The system considers the event type definition for creating the events. For example, if you have created an event type that has a duration of 2 months and has an effect 1 month before the actual period the system creates an event with a duration of 3 periods, and calculates an uplift for all 3 periods.
Events detected at the end of the history horizon
If the event that the system would create is at the end of the history horizon, or according to the event type definition the event has to be created for a duration that partly lies in the future (outside the detection horizon), the system creates an event with a flag that makes sure that such events are reestimated the next time the automatic outlier detection report is run on the same selection. The re-estimation is executed until the event periods are fully in the past (inside the detection horizon).
Events created by future event estimation
Events created by future event estimation are reestimated in case the event periods are partly or fully in the past (inside the detection horizon). These events are reestimated until they are fully in the past.
When the system creates multiple events that overlap one another, only one event gets the full uplift value and all the other events that overlap this event get 0 or a very small number for the uplift value. The event identified being the most significant gets the full uplift.
Note
There could be a slight difference in the uplift values that the system calculates automatically and the events that are manually created in interactive planning due to the evaluation sequence (during automatic processing, the system processes events in the order of magnitude) and the moving average calculation horizon.
The system checks for changes in the price key figure, and creates price change events for these changes. Price change events model the short term effects of price changes on the demand. For more information, see Detecting Price Change Events.
If you enter the parallel processing profile you created for Automatic Outlier Detection and Event Generation
in the Paral. Proc. Profile
(Name of Parallel Processing Profile) field, the system runs the report split into parallel
processes, which can improve system performance. You define parallel processing profiles in Customizing for Demand Planning
under .
To access this report, on the SAP Easy Access
screen, choose .
When you run the Automatic Outlier Detection and Event Generation
report, the system performs the following activities:
The system reads the historical demand for the selection and time period you specified.
As an optional step, the system deletes events that were created previously.
The system performs this preprocessing step only in case you select Delete Automatically Created Past Events
or Delete Manually and Automatically
Created Past Events
in the Del. Auto.Detected Past Events
(Delete Automatically Detected Past Events) field. The system deletes the events for all event types, regardless of the event catalog you specified.
The system removes the impact of previously created events from the history.
If you selected Delete Automatically Created Past Events
in the previous step, the system only cleans the history of the impact of manually detected past
events. If the event needs to be reestimated in the current run of the report, the system does not remove its effects, but recalculates them. Such a case can occur when the event could not be fully calculated in a previous run of the report, or was estimated by forecasting. For example, if forecasting
estimated an event in the future, and the event lies in the past when you run the report, the system recalculates the impact based on the actual history, instead of the forecasted estimate.
If you have selected Only Price Change Detection
, or Outlier and Price Change Detection
in the Detect Price Change Events
field, the system proceeds as follows:
The system detects changes in the price key figure that you specified. On the SAP Easy Access
screen, choose , open your planning area, and in the system menu, choose . In the Price
Settings
screen area, in the Price Key Figure
field, enter the key figure you want to use as the price key figure. The system considers the price change only in case its absolute percentage value is greater than the value you specified in the Price
Change Thresholds
field.
The system removes the impact of previously created price change events from the history.
If the event needs to be reestimated in the current run of the report, the system does not remove its effects, but recalculates them.
The system calculates the short-term effect of price changes, and creates price change events.
The system identifies outliers based on the settings found in the Customizing activity Configure Outlier Detection and Event Catalog Assignments
. For more information, see the documentation, in Customizing, under .
The system tries to match the identified outliers to event types found in the event catalog you specified.
If none of the event types defined in the event catalog matches the outlier period, the system assigns the corresponding generic event type you maintained
in the Customizing activity Configure Outlier Detection and Event Catalog Assignments
.
For the matched event types, the system creates events.