Setting Up Seasonality Profile Groups

This chapter provides an overview of seasonality profile groups, list prerequisites, and discusses how to create and maintain seasonality groups.

Understanding Seasonality Profile Groups

Seasonality is a repetitive pattern of demand for an item during certain periods of a forecast. A seasonality group is a flexible, user-defined grouping of forecast items that helps you analyze and understand patterns of demand over time. For example, you might base groupings on some combination of product, product group, market, and geographic area.

The system automatically maintains seasonality profiles for each user-defined seasonality group. If an item is seasonal but doesn’t have sufficient history to determine its own profile, then you can assign a seasonality group profile to the item to provide a profile or pattern of demand. You assign seasonality groups unique seasonality group codes. You can define items as users only, as contributors only, or as both a user and contributor of a seasonality group.

Note. In PeopleSoft Demand Planning, the terms seasonality group and seasonality profile are almost interchangeable. Seasonality groups refer to a set of related forecast items. Seasonality profiles refer to factors and processing controls that you establish for seasonality groups.

You use contributor items that are representative of the group and that exhibit stability. PeopleSoft recommends at least two years of demand history for items. The system uses the demand history of contributors to determine a group seasonality profile. To do this, the system aggregates the demand for these items and calculates the seasonality profile (indices) associated with the aggregate.

You can manually adjust the seasonality profile, if required. It is then used in the forecast calculation as the seasonality profile for all the items in the group in place of any individual profiles. The seasonality group profile is more stable than individual profiles of the contributors because the aggregation process smooths out random errors.

A key benefit of the seasonality group is that it enables seasonal items to be forecast by using seasonal models with items that have fewer than 12 periods of demand history.

Prerequisites

Before the system can begin using a seasonality group, its seasonality profile must be established. To establish a new seasonality profile:

Classify forecast items.

Forecast items with the same or similar seasonality should be grouped together. When reviewing forecast items, consider seasonality differences that may be attributed to such things as seasonal popularity and regional preferences.

If necessary, assign separate seasonality codes to different forecast item aggregates. You must decide which level in the structure can support the seasonality structure that you want. This is the level that determines the seasonality group profile.

Decide on a seasonality group code and use the Seasonality Profiles page to add it.
Associate forecast items with the seasonality group.

You must select a subset of forecast items in the group to be the contributors of the seasonality group. Contributors should be representative of the group and have a stable demand pattern, the recommended two years of historical demand, and little or no promotional activity.

After you determine the contributors, maintain all forecast items in the group. You must associate each item with a seasonality group code and indicate whether they are users, contributors, or both.
Generate the seasonality profile.

Use the Calc (calculate) Seasonality Profiles process (DP_CALCSEAS) to calculate the profile for the new seasonality group. The system uses the demand history for all forecast items that are contributors to the group during this process.
Calculate forecasts.

For the profile to have an impact on the forecast, you must recalculate the forecast for all of the forecast items associated with the seasonality group. You do this by using the Calculations feature.

PeopleSoft recommends that you use the Simulation option for a subset of the forecast items to confirm that the seasonality grouping is valid. Resulting forecasts from this test simulation should be stable and accurate before you proceed with a full forecast recalculation.

Creating and Maintaining Seasonality Groups

Seasonality groups aren’t required, and you can add new ones at any time. However, after you complete the tasks for adding a new seasonality group, you must recalculate the forecast for it to take effect.

If you change an existing seasonality group, the system uses the new parameters in calculations for the associated items the next time that a forecast run or simulation is performed.

This section discusses how to:

Create seasonality profile groups.
Update controls for seasonality profile groups.
View items associated with a seasonality profile.
Generate seasonality profiles.

Pages Used to Create and Maintain Seasonality Groups

Page Name	Object Name	Navigation	Usage
Define Seasonality Profiles	DP_SEASONPROFILES	Demand Planning, Process Forecast, Seasonality, Define Profiles, Define Seasonality Profiles	Create seasonality profile groups by defining profile factors for individual periods in a forecast view.
Update Control	DP_SEASONPROFILES1	Demand Planning, Process Forecast, Seasonality, Define Profiles, Update Control	Update controls for seasonality profile groups by defining data that controls the calculation of the seasonality profile factors from the contributing forecast items.
Where Used	DP_SEASONPROFILES2	Demand Planning, Process Forecast, Seasonality, Define Profiles, Where Used	View the items associated with a seasonality profile.
Generate Seasonality Profiles	DP_CALCSEASPROFILE	Demand Planning, Process Forecast, Seasonality, Generate Profiles	Generate seasonality profiles. Define the seasonality groups to include in the generation of the profile, and whether to use view evaluation periods. You must first have defined seasonality groups.
Delete Forecast View	DP_DELETEVIEW	Demand Planning, Process Deletions, Delete Forecast Views	Delete all seasonality profiles associated with a forecast view. Select the Seasonality Profiles check box to remove a profile from the view.

Creating Seasonality Profile Groups

Access the Define Seasonality Profiles page.

Period	Indicates the period to which the system applies the seasonal profile factor that you enter.
Profile Factor	Enter a factor for the effect that you want each period to have in the seasonality profile. The system uses data from the contributing forecast items to calculate profile factors during the calculation process. Items that have the Manual Updates Only check box selected are not included in the calculation. If you are using data from an external source, manually enter it in the field. Values for profile factors are between 0 and 10,000. The sum of the values must equal 1,200. The system defaults are 100.000.
Normalize	Click to have the system calculate the weighted factor values automatically so that they total 1,200. If you save, and the total of the factors is greater than 1,200, an error message appears saying that you must normalize profile factors before saving. The system allocates to all indices the difference between the calculated sum and 1,200, based on the proportion on the index value.
Refresh	Click to update the chart based on the changes that you made to the profile factors.

Updating Controls for Seasonality Profile Groups

Access the Update Control page.

Manual Updates Only	Select if the seasonality group profile has been developed outside of the system by using data from an external source. If this is the case, you maintain profile factors by using the Define Seasonality Profiles page. If you select this option, the remaining fields on the page are unavailable for entry.
Determined By Level	Select a forecast view level of the actual demand used for calculating the seasonality profile. The default value for the field is 1.
Last Generation Date	Displays the last time this seasonality profile group was generated.
UOM Group (unit of measure group)	Select a UOM group to correlate different units of measure so that the system produces a single standard UOM output for a specified UOM group.
Group Key	Select a control group. The system uses the group processing options during the seasonality profile calculation step. If you do not select to update the seasonality profile manually, you are required to define a control group.
Periods of Demand to Use	Enter the number of periods to include in the profile calculation. PeopleSoft recommends that you use at least two years (24 months) of data. Valid values are between 0 and the number of historical periods defined on the view. The default value is calculated by multiplying the number of periods per year by two.
Elimination Demand Periods	Enter the minimum number of periods of historical data that an item must have to be included in the calculation of the seasonality profile factors. Valid values are between 0 and 999. The default value is 24.

Viewing Items Associated With a Seasonality Profile

Access the Where Used page.

Contributes to Season Profile	Select this check box to view items that contribute to the creation of this seasonality profile group. This means that the system accumulates these items' demand to determine the profile factors for the seasonality profile group.
Uses Seasonality Profile	Select this check box to view items that derive their seasonality profile factors from this seasonality profile group.
Refresh Grid	After you have selected one or both of the check boxes, click this button to view the items.

Generating Seasonality Profiles

Access the Generate Seasonality Profiles page.

Using this page, you run the Calc Seasonality Profiles process to recalculate the profile factors for the selected seasonality groups. You can recalculate profiles for one or more seasonality groups at the same time. The system recalculate each selected seasonality profile in sequence.

Select the Recalculate check box to include the corresponding seasonality group in the profile that you are calculating.

Note. If a profile is not produced after generating the seasonality profile, make sure that there is at least one contributor and that the seasonality group isn’t set for manual update only on the Update Control page. Also, verify that the Effective Demand Periods field on the Model Control page of the Maintain Forecast Item feature is less than or equal to the Periods of Demand to Use field on the Update Control page for the seasonality group.