SAP IBP for Demand

AI-assisted Forecast Result Analysis now explains the impact of independent variables and system generated features.

• Supported algorithms:
  • Gradient Boosting of Decision Trees
  • Multiple Linear Regression
  • Auto(S)ARIMAX
• Coming later:
  • Extreme Gradient Boosting

The impact of independent variables and system-generated features is available in the "Forecast Algorithms" section

AI-assisted explanation now includes information related to Product Lifecycle Management.

• The goal is to raise attention to important factors while keeping the extent manageable
• Additional explanation in 2602:
  • Product references were evaluated at a more granular level
  • In case of Aggregated Phase-in / Phase-out planning, the level where forecast dates were applied
  • Reference product, in case only one was used
  • Note when forecast was not generated prior to the Phase-in start date or after the Phase-out end date
  • Note if Manual forecasting was used, ignoring the rest of the forecast model settings
• Note: to analyze Product Lifecycle Management settings in detail, open the Product Lifecycle side-panel

Navigate from Statistical Forecast Details to various Fiori apps.

Open the Manage Forecast Models app and preload the applicable forecast model

Open the Application Logs app and preload the log of the relevant forecast job

Navigate to the involved Forecast Automation profile

Outlier correction with Smoothing method now consideres the level of intermittency.

• From 2602 on, outlier correction with Smoothing method ignores zeros, if more than two occur in the time series
• This way, the correction value will not inflate due to a nearby zero
• Impacted functions:
  • Outlier correction with smoothing

• Automatic Outlier Correction