The image shows a simplified data flow from two MiQ data extraction instances: on-premises and on Oracle Cloud Infrastructure (OCI).

The OCI region includes:

• The SLiQ application which runs in Oracle Container Engine for Kubernetes and comprises a transform process, machine learning analysis, and web/mobile actions interface.
• A data store with Oracle Cloud Infrastructure Object Storage and Autonomous Database
• An instance of MiQ for data extraction

Retail management systems collect inventory management and monitoring data periodically and pass it on to MiQ. Data flows along two slightly different paths, depending on whether MiQ is deployed on-premises or in the cloud.

1. For an on-premises MiQ instance:
   1. An on-premises MiQ instance extracts the data from the retail management system (RMS).
   2. Using REST APIs, MiQ sends the data to the SLiQ application running on OCI.
   3. SLiQ transforms the data.
2. For an MiQ instance in the cloud:
   1. The data from the retail management system (RMS) undergoes ETL (extract, transform, load).
   2. The data is stored in Oracle Cloud Infrastructure Object Storage.
   3. An MiQ instance on OCI extracts the data.
   4. The data is sent to the SLiQ application on OCI where it is transformed.
3. SLiQ loads this data into Autonomous Database.
4. SLiQ queries the data in Autonomous Database.
5. Machine learning analysis uses the data in Autonomous Database.
6. Machine learning trains the model, predicts inventory events, and makes recommendations.
7. Results are sent back to the retail employees who access the results on mobile devices.
8. Retail employees are provided with actionable insights such as which shelves to stock and when to restock them.
9. Employees act on insights, minimizing the time products sit in the stockroom or warehouse.