How to Use Forecasting Methods in SAP APO DP

ByShaun Snapp

Executive Summary

  • We cover the forecasting methods (aka methodologies) available in SAP DP.
  • To use the forecasting methods, it is necessary to use the Forecast Profiles, Master Profile, the Univariate tab, MLR tab, and the Composite tab.
  • Best fit requires setting up the Forecast Profile and the Forecast Strategies.

Introduction to Forecasting Methods and Parameters

Forecast parameters control the methods within statistical forecasting systems. A specific combination of parameters converts a forecast method (say Holt-Winters or a Constant) into a specific model. In this article, you will learn about the methods and parameters in SAP DP.

Our References for This Article

If you want to see our references for this article and other related Brightwork articles, see this link.

Notice of Lack of Financial Bias: We have no financial ties to SAP or any other entity mentioned in this article.

  • This is published by a research entity, not some lowbrow entity that is part of the SAP ecosystem. 
  • Second, no one paid for this article to be written, and it is not pretending to inform you while being rigged to sell you software or consulting services. Unlike nearly every other article you will find from Google on this topic, it has had no input from any company's marketing or sales department. As you are reading this article, consider how rare this is. The vast majority of information on the Internet on SAP is provided by SAP, which is filled with false claims and sleazy consulting companies and SAP consultants who will tell any lie for personal benefit. Furthermore, SAP pays off all IT analysts -- who have the same concern for accuracy as SAP. Not one of these entities will disclose their pro-SAP financial bias to their readers. 

Forecast Profile and Forecast Strategies with Forecast Methods

SAP always misnamed Forecast Strategies. The generally accepted nomenclature would be the “forecast model.” The top five Forecasting Strategies in DP are the following:

  • Constant Model
  • First Order Exponential Smoothing
  • Constant Model with Auto Alpha Adaptation
  • Moving Average
  • Weighted Moving Average

SAP DP ships with 30 standard forecasting methods; however, some are not really used very often, such as the linear regression models. A few models, such as Forecast with Automatic Model Selection, are not models but are rather triggers to employ SAP DP’s best-fit functionality.

The SAP DP Forecasting Methods That Ship with DP

  • First Order Exponential Smoothing
  • Constant Model with Auto Alpha Adaptation
  • Moving Average
  • Weighed Moving Average
  • Forecast with Trend Model
  • Holt’s Method
  • Second-Order Exponential Smoothing
  • Trend Model with Automatic Alpha
  • Forecast with Seasonal Model
  • Seasonal Model Based on Winter’s Model
  • Seasonal Linear Regression
  • Median Method
  • Forecast with Seasonal and Trend Model
  • Holt and Winters’ Exponential Smoothing
  • Forecast with Automatic Model Selection
  • Test for Trend
  • Test for Season
  • Test for Trend and Season
  • Seasonal Model and Test for Trend
  • Trend Model and Test for Seasonal Pattern
  • Model Selection Procedure 2
  • Historical Data Adopted
  • Manual Forecast
  • Croston’s Model
  • Linear Regression
  • No Forecast
  • External Forecast

Every forecast method can be adjusted by adjusting its parameters.

However, no matter how the best fit is determined, the assignment must be created between the product location and the method. There are two basic areas where this assignment is created — the first, which I call the “initial assignment,” is in the transaction:

  • /SAPAPO/MSDP_FCST2 – Assign Forecast Profiles to a Selection.
  • The second place is in the Planning Book. Typically /SAPAPO/MSDP_FCST2 is set up by those that configure the system, while planners create the Planning Book Assignment.

The Forecast Profile

The forecast profiles the settings that can be saved and applied to the demand history. The profile configuration is segmented into two forecasting categories.

  • Univariate Forecasting
  • Multiple Linear Regression Forecasting

The forecasting profile configuration also allows the third tab for composite forecasting, combining the two categories.

Master Profile

The profile begins with the Master Profile tab. This tab holds the forecasting method independent fields that select things like the forecast horizon.

These values are important because SAP DP does not ship standard with every model to tune DP’s forecast models to meet the product/location demand history requirements. For instance, in many cases, when I perform best-fit forecasting, I end up with a forecast model assignment called Level-Seasonal-Trend. That is, it has elements of each of these parameters in the demand history.

However, I can take an existing “Forecast Strategy” that accounts for Seasonality already but then increase the parameters for Level and Trend components.

Forecast Parameters for the Univariate Tab

It should be understood that just because the parameters are blank in the Forecast Strategy screen does not mean no parameters are applied. The value 0.2 for Alpha, 0.1 for Beta, 0.3 for Gamma, and Delta by default. Of course, the default can be overridden. The following is copied from a previous article I wrote on these values.

  • Alpha: This is also known as the base value. This value determines how much more recent or how many past periods should be weighed in the forecast calculation. A higher value means more recent data points are given more weight.
  • Beta: This is also known as the trend value. It determines the degree of the ascent or descent value that should be used to adjust the forecast. This tells the system how much to focus on recent changes in trends.
  • Gamma: This is the seasonal component of the forecast, and the higher the parameter, the more the recent seasonal component is weighed.

Alpha, Beta, and Gamma are a bit misleading because the higher the value, the more recent influences are considered. It is effortless to fall into the process that Alpha, Beta, and Gamma control recent history, trends, and seasonality full stop. For a refresher on forecast parameters, see the article How to Understand Alpha, Beta, and Gamma in Forecasting.

Go to top

Conclusion

Forecast profiles in DP are pretty flexible. However, this is all very standard but not particularly impressive.

Regarding statistical forecasting, best fit, and MLR (which is very infrequently used), DP has it covered, but nothing here is not in just about every other forecasting package.

Configuration in DP is relatively straightforward, though it could be far better designed.

The real trick is figuring out the right forecasting method for which products. It was thought that the best fit could always be used for the right selection. This is something promoted not only by SAP but by many software vendors. And it is entirely untrue. I have worked with several clients who enabled and then disabled best-fit forecasting in SAP DP. I cover this topic in this article. When I perform best-fit forecasting, I don’t use DP as DP’s best fit is not worth using. The only people who use the best fit in DP are those who don’t have another application and feel they have to use the best fit in DP to be “SAP compliant.”