Phase-In and Phase-Out Profiles
Manufacturers introduce new products into their product portfolio regularly. Demand planning aims at making high-quality automated forecasts for the future demand of a product. This is done by sophisticated statistical algorithms that calculate a future time series based on the knowledge of historic sales values. However, for new products, historic values do not exist.
You therefore search for products that have a similar historical sales pattern and assign them as reference products to the new product. These reference products are then taken into account dynamically in the forecast engine at the time of the forecast run for the new product.
When you introduce a new product or phase out an existing product, you expect the demand to be noticeably different to that in the "mature" phase of the product’s life. Phase-in/out modeling lets you take this behavior into account.
To define reference products for a new product in SAP IBP for demand, your planning area must fulfill the following prerequisites:
You have defined which attribute of the planning area represents the product ID.
You have created the respective planning objects for your new products using key figure upload or the copy operator, or you have created the planning objects in Microsoft Excel.
Recommended: You have defined which attribute of the planning area represents the product description and the product group. If you do not do this, the Manage Product Lifecycle app cannot handle descriptions or product groups for the products and the system uses the product ID instead.
Recommended: You have set the business meaning Actual Sales for one key figure of the planning area.
You can plan the phase-in and phase-out process for a new product by specifying the start and end dates for the phase-in and phase-out and optionally assigning a curve.
The phase-in start date defines the point in time when a product is sold in the market. Forecasting is not generating results for periods with an end date that is before the phase-in start date.
The phase-out start date is the point in time when the product is gradually sold less than in the maturity phase. The phase-out end date is the point in time when the product is no longer sold in the market. The forecast engine does not generate results after this point of time.
The curves show the expected demand development of the product during phase-in and phase-out. Phase-in and phase-out curves represent percentage time series. The percentages are multiplied by the forecast result in the phase-in time horizon that is defined by the phase-in start and end dates.
Predefined Curves
The following curve types are available:
Linear:
The linear curve is calculated by a linear interpolation between 0.1 and 0.9.
Sublinear:
The sublinear curve applies the quadratic function to the time series that we obtain when you do a linear interpolation between 0.316 and 0.949. The boundaries are calculated by applying the reverse function of the quadratic function to the standard boundaries 0.1 and 0.9.
Superlinear:
The superlinear curve applies the square root function to the time series that we obtain when we do a linear interpolation between 0.01 and 0.81. The boundaries are calculated by applying the reverse function of square root to the standard boundaries 0.1 and 0.9.
Custom Curves
You can configure your own phase-in and phase-out curves. To do so, select an existing curve type and adapt the curve parameters to your needs. You can define the following parameters:
The number of time periods you want to display
The function used to calculate the curve is, for example, square root or quadratic
The start and end values of the curve in percent
After changing the parameters, you can simulate the resulting curve and, if you are happy with the result, save it for future use and assign it directly to the relevant launch dimension value. You can use the value label switch to display or hide the values for each point of the curve.
The phase-in start date defines the date from which the new product is sold on the market. The system does not generate any forecasts for time periods that lie before this date. The phase-in end date defines the point in time when the product has reached maturity. Once the phase-in end is reached in the historical horizon of the forecast, forecasting is based on the product's own sales history.
The phase-out start date defines the date from which the product will be gradually withdrawn from the market. The phase-out end marks the date at which the product is no longer being sold. The system does not make any forecasts after this date.
Note
You can generate the phase-out for a reference product based on the phase-in of a new product. The generated phase-out period of the reference product is identical to the phase-in period of the reference product. The generated phase-out curve mirrors the selected phase-in curve: The phase-out curve starts at the end value of the phase-in curve and ends at the start value of this curve. The generated curve name contains the date that the curve was created, for example, Generated 20180205 Superlinear Curve. If there is already a curve with matching parameters, the system uses this curve instead of generating a new one.
Usage Variants of Phase-in/Phase-out Function in SAP IBP
There are six different variants of usage for the phase-in/phase-out function in SAP IBP:
Phase-in without curve:
Weight of reference products can be set to use a % of actual delivery of reference products in the future. Phase-in start and end dates should be similar. In this case, 100% of the forecast for a new product is available starting from the selected date.
Phase-in with curve:
Weight of reference products can be set to use a % of actual delivery of reference products in the future. Phase-in start and end dates should be different. In this case, the forecast for new product is available starting from the selected date and increases in accordance with the selected curve.
Phase-out without curve:
In this usage, you do not need to add additional reference products because it is a phase-out. Set the weight of the product = 100%. Phase-out start and end dates should be similar. In this case, 0% of the forecast for an old product is available starting from the selected date.
Phase-out with curve:
- In this usage, you do not need to add additional reference products because it is a phase-out. Set the weight of the product = 100%. Phase-out start and end dates should be similar. In this case, the forecast for an old product becomes unavailable starting from the selected end date and decreases in accordance with the selected curve.
Replace without curve:
The weight of reference products can be set to use a % of the actual delivery of reference products in the future. Phase-in start and end dates should be similar. In this case, 100% of the forecast for a new product is available starting from the selected date. Choose the necessary curve with only one period inside.
At the same time, generate the phase-out profile for reference products. The start and end dates should be aligned with the phase-in dates of the new product.
Replace with curve:
The weight of reference products can be set to use a % of the actual delivery of reference products in the future. Phase-in start and end dates should be similar. In this case, 100% of the forecast for a new product is available starting from the selected date. Choose the necessary curve with more than one period inside.
At the same time, generate the phase-out profile for reference products. The start and end dates should be aligned with the phase-in dates of the new product in accordance with the selected curve.
Note
If you execute a statistical forecast at the Product level and, at the same time, a phase-in/out profile is generated at the Product-Customer level, the phase-in/out dates would not be used in the forecast, because they have different planning levels.
Historical data for a new product can be simulated in the Manage Product Lifecycle app to check all related information for the selected reference products, curves, periods, and dates. It can be helpful to generate your own curve with different reference products for different new products and take seasonality, trends, and so on, into account.