In our earlier blog, we discussed about how to leverage machine learning with the ISLM (Intelligent Scenario Lifecycle Management) framework. Now that we see the ISLM framework is available with improved features, more innovation is happening around the framework with more integrations into the other ML frameworks. While with the earlier PAi (Predictive Analytics Integrator) framework, you could only leverage the SAP HANA PAL and APL libraries for machine learning, with the ISLM framework you could also leverage the machine learning algorithms from Tensor Flow, Python, the open-source R etc. Initially the integration started with helping to build the side-by-side ML models with the SAP Data Intelligence platform and later on it is now getting extended to the SAP AI Business services using the AI foundation. There are more updates happening on the integration front with the SAP AI business services and hence we shall come back to this topic when we discuss the side-by-side ML scenarios with ISLM.
Let us now get back our focus on the embedded ML in SAP S/4HANA and the migration of PAi to the ISLM framework. I will not go into the details of the ISLM framework or the architecture concepts since we did discuss them in the earlier blogs. Hence the focus here is mainly on the embedded ML use cases that are migrated from the PAi framework to the ISLM framework.
As of May 2021 (the 2105 release), you will still have both the PAi framework and the ISLM framework available in the SAP S/4HANA system as shown in the picture above. The customers and partners can use either of the "Predictive models" Fiori app(from the PAi) or the "Intelligent Scenario Management" Fiori app(from the ISLM) to do training of their models. But the "Predictive Scenarios" Fiori app (from PAi) is read-only and cannot edit/publish the draft scenarios. Hence you will need to use the "Intelligent Scenarios" Fiori app (from the ISLM) to create/edit/publish scenarios.
Following are the embedded ML scenarios in SAP S/4HANA that have been successfully migrated from PAi to ISLM. There is not much difference on how you would train a model, activate a model and run the model which is similar to what you have done in the past. But the corresponding scope items and the documentation has been updated on the SAP Help portal etc.
The above are the ML scenarios migrated from PAi to ISLM that are embedded in SAP S/4HANA. We will provide detailed updates on the embedded ML scenarios leveraging ISLM in a best practices package as well that would be focused on the naunces of how-to-use these scenarios out of the box and also enhance them further. Please stay tuned for more updates and happy predicting the future with ML technologies!
Here are some quick links to the blogs in this series to give you a complete understanding of how Predictive Intelligence is infused into SAP S/4HANA.
Blog series: Approaches of doing Machine Learning with SAP S/4HANA
Blog series: Machine Learning with SAP S/4HANA using ISLM
Happy predicting the future!!
Let us now get back our focus on the embedded ML in SAP S/4HANA and the migration of PAi to the ISLM framework. I will not go into the details of the ISLM framework or the architecture concepts since we did discuss them in the earlier blogs. Hence the focus here is mainly on the embedded ML use cases that are migrated from the PAi framework to the ISLM framework.
As of May 2021 (the 2105 release), you will still have both the PAi framework and the ISLM framework available in the SAP S/4HANA system as shown in the picture above. The customers and partners can use either of the "Predictive models" Fiori app(from the PAi) or the "Intelligent Scenario Management" Fiori app(from the ISLM) to do training of their models. But the "Predictive Scenarios" Fiori app (from PAi) is read-only and cannot edit/publish the draft scenarios. Hence you will need to use the "Intelligent Scenarios" Fiori app (from the ISLM) to create/edit/publish scenarios.
Following are the embedded ML scenarios in SAP S/4HANA that have been successfully migrated from PAi to ISLM. There is not much difference on how you would train a model, activate a model and run the model which is similar to what you have done in the past. But the corresponding scope items and the documentation has been updated on the SAP Help portal etc.
| LoB/Industry | Use case | Technical Scenario | Algorithm | Component | Scope Item |
| Procurement | Supplier Delivery Prediction | SUPLRDELIVPREDICT | REGRESSION | MM-PUR-PO | 3FY |
| Procurement | Quantity Contract Consumption | QTY_CONTRACT_CNSMPN | REGRESSION | MM-FIO-PUR-ANA | 1QR |
| Sales | S/4HANA Sales: Predicted Delivery Creation Delay | PRDTDDELIVCRTNDELAY | REGRESSION | SD-ANA | 2YJ |
| Sales | S/4HANA Sales: Predicted Delivery Processing Delay | PRDTDDELIVPROCGDELAY | REGRESSION | SD-ANA | 2YJ |
| Sales | Sales Quotation Conversion Rate | SLSQTANPREDICTION | REGRESSION | SD-ANA | 2YJ |
| Sales | Sales Performance - Prediction | SALESVOLUME001 | REGRESSION | SD-ANA | 2YJ |
| Produce - Inventory Management | Stock in Transit material overdue | MATERIAL_OVERDUE_SIT | REGRESSION | MM-FIO-IM-ANA | 20N |
| Produce - Inventory Management | Consumption Data for Slow or Non-Moving Materials | MMSLO_CONSUMPTION_02 | REGRESSION | MM-FIO-IM-ANA | 20N |
| Produce - Quality Management | QM – Defect Code Proposal | QM_DEF_CODE_PRPSL | TEXT_ANALYSIS | VDM_QM_NOTIFICATION | 20N |
| Produce | Demand-Driven Replenishment: | INDIV_LEADTIME_STO2 | REGRESSION | PP-DD-BLM-BAS | 20N |
| Research & Development | Project Cost Forecast | PROJCOSTFRCAST01 | REGRESSION | PS-FIO | 2Y7 |
| Finance | Financial Statement Insights (sunset since Q1 2020 and not to be used) | FIN_STMNT_PRED_AMT | REGRESSION | LOD-MAP-FSI | |
| Finance | Check Assigned Liquidity Items version1 (also called Detect abnormal liquidity items) | FCLM_RDT_CALI_V1 | CLASSIFICATION | FIN-FSCM-CLM | 30K |
The above are the ML scenarios migrated from PAi to ISLM that are embedded in SAP S/4HANA. We will provide detailed updates on the embedded ML scenarios leveraging ISLM in a best practices package as well that would be focused on the naunces of how-to-use these scenarios out of the box and also enhance them further. Please stay tuned for more updates and happy predicting the future with ML technologies!
Here are some quick links to the blogs in this series to give you a complete understanding of how Predictive Intelligence is infused into SAP S/4HANA.
Blog series: Approaches of doing Machine Learning with SAP S/4HANA
- Part 1 - Leveraging Predictive Intelligence with SAP S/4HANA
- Part 2 - Architecture and deep-dive of the different approaches around Predictive Intelligence
- Part 3 - Process flow leveraging Machine Learning and Predictive Analytics
- Part 4 - Scope and functionality in the context of an end-to-end process leveraging ML
- Part 5 - Activating machine learning functionality for SAP S/4HANA
- Part 6 - Building ML into the digital core of SAP S/4HANA - Embedded ML
- Part 7 - Enhancing the digital core with ML Services (Side-by-Side ML)
- Part 8 - Extending the digital core by leveraging ML with SAP Analytics Cloud
- Part 9 - ML Extensions to SAP S/4HANA processes
Blog series: Machine Learning with SAP S/4HANA using ISLM
- Leveraging Machine Learning with the ISLM Framework
- Migration of PAi to ISLM with the embedded ML (this blog)
Happy predicting the future!!
- SAP Managed Tags:
2 Comments
