SAP S/4HANA Cloud offers enticing Machine learning – Embedded Predictive Procurement Innovation Solution – “Supplier Delivery Prediction”.
Intelligent enterprises effectively use the procurement innovation to achieve their desired outcomes faster – and with less risk. Responding to individual customer needs, engaging suppliers in new ways, and creating disruptive business models are critical business imperatives. By becoming an Intelligent Enterprise, you can achieve these goals – and more.
Here’s everything you need to know about deploying machine learning to build an intelligent enterprise, and the key steps you should avoid if you don’t want to stumble when trying to gain your footing around new technology.
Let’s take deeper look at the business-driven innovation with Procurement Predictive Embedded Analytics with SAP S/4HANA Cloud.
Predictive capabilities empower users making better decisions for procurement activities with a self-learning analytics solution.
Supplier Delivery Prediction avoid delayed raw material availability and production rescheduling ( such as Impact on on-time production at plant and rescheduling assembly lines which is very costly).
Materials required for production are supplied by multiple suppliers. A delay in delivery can impact on-time production at the plant and cause rescheduling of assembly lines, which is very costly. Indirect material delay also causes interruption in supporting employees with required products or services. The Machine Learning algorithms identify the supplier delays based on the multiple situation and predicts the chances of delay.
During creation of purchase orders or purchase requisitions, the lead time from the material master sometimes doesn’t consider the processing and approval time. For More Information visit Best Practices Explorer – Supplier Delivery Prediction (3FY).
Predictive Embedded Analytics App is Accessible via the SAP Fiori launchpad.
The Machine Learning algorithms identify the supplier delays based on the multiple situation and predicts the chances of delay in supply of materials
Recommendations are enriched with insights gathered by the machine learning algorithm to Predict more reliable delivery lead time when creating PR / PO and Predict the arrival date of a shipment and classify the status into different classes
The results based upon transaction data predictions can later be reused at triggering sources such as Material Requirement Planning (MRP) or Source of Supply(SoS) to update delivery lead time or related parameters to avoid delays, thereby achieve delivery performance optimization.
Business Conditions should be met as Prerequisite
- Scope item Procurement of Direct Materials (J45) should be installed and activated
- Role SAP_BR_PURCHASER
- The application Monitor Purchase Order Item shall be enhanced with new fields such as Forecast Delivery Date and Prediction Date.
- A set of historical Purchase Orders are collected to understand and train the system with the Supplier’s delivery behavior. The Purchase Orders contains items representing Materials, Schedule line(s) where a Purchaser shall mention the Expected Delivery Date(s) for each Material. The Actual Delivery Date on which the Supplier has delivered the material is derived from Goods Receipt document.
- A comparison on the Expected Delivery Date and the Actual Delivery Date with the Purchase Order Creation Date shall be used to identify the nature of Supplier Delivery. Based on this information, the data is classified as Delay, On Time and Early Delivery.
Innovation with Application Monitor Purchase Order Item
A CDS view shall be created to extract the data from Schedule Line(s) (I_PurchaseOrderScheduleLine) and Goods Receipts corresponding to the PO Item from Purchase Order History (I_PurchaseOrderHistory).
The Following Artifacts shall be required for the development and delivery of ML using Predictive Analytics Integrator (PAI):
- Training CDS View
- Predictive Model
- Predictive Scenario
- CDS Table Function (to wrap generated API)
- Prediction CDS View
Model Status – Set Active/Retrain and Delete. Predictive model can have several versions and each version can have a different status. “Ready” status indicate that model can be set to active or retrained if necessary.
Predictive power KI and the prediction confidence KR indicators are used to evaluate the performance of a model version
For more information on SAP S/4HANA Cloud, check out the following links:
- SAP S/4HANA Cloud release info: http://www.sap.com/s4-cloudrelease
- Sven Denecken’s 1811 blog on SAP News here
- Best practices for SAP S/4HANA Cloud here
- SAP S/4HANA Cloud User Community: register here
- Feature Scope Description here
- What´s new Viewer here
- What’s New here
- Help Portal Product Page here
Stay tuned for more updates about SAP S/4HANA Cloud Innovations! in next quarter .
Source: https://blogs.sap.com/2018/11/15/intelligent-erp-update-sap-s4hana-cloud-1811-release-innovation-deep-dive-for-procurement-solution/
Nenhum comentário:
Postar um comentário