Intelligent Advisor Release 24B (12.2.35) is now available for public cloud and private cloud (on-premise) customers.
Be sure to also review the features added to 24A since our last post.
Separate Hub roles for Permissions and Connections administrators (Monthly Update 2)
The existing Hub administrator role has been split into two roles: a Connections administrator role and a Permissions administrator role. This allows for users who need a smaller set of permissions to be assigned only the connections or Hub permissions aspect of that role. There is no change to functionality otherwise. Both roles are auto-assigned to existing Hub administrators. When a new site is provisioned, the initial admin account gets the Permissions administrator role only.
Test Suites REST API (Monthly Update 2)
The new Test Suites REST API provides the ability to maintain and assess user-defined test suites, and review the results of the executed test cases. This enables Intelligent Advisor test case management to be integrated into custom continuous integration and continuous deployment (CI/CD) tools.
Flow project authoring and management enhancements (Monthly Update 2)
Several improvements have been made to authoring and managing flow projects in Intelligent Advisor Hub. These are:
- The Flows tab on the Projects page now shows which flows are using a scheme. You can also filter the list by flow scheme. This enables managers to understand the potential impact of scheme changes.
- The scheme used for a flow project is now shown in the project header. You can click on the scheme name to select a different scheme to use. In addition, you can open the flow scheme from the Actions menu while authoring a flow. This enables managers to quickly make changes to the scheme.
- The configuration of flow goals and global input data is now more obvious with both included at the top of the flow editor.
- The object scope of controls inside containers in the flow editor is now shown on the tab for each container. This helps authors to know what parts of the data model they can be using at that point.
- An error will be shown in the flow editor if you try to use a field in the control’s expression that is collected later in the flow. Similarly, if the field being used in an expression is collected before the control but is always hidden, then you will get an error.
Heading level styles in flows (Monthly Update 2)
You can now choose a heading level for a label in a flow scheme. This style will then be applied to that label control when added to a flow. This enables you to add visual structure to your flow pages.
Side by side containers in flows (Monthly Update 2)
You can now place control containers (record collects and control groups) side-by-side with other controls so that you can make efficient use of the screen real estate.
Default attribute values and relationships (Monthly Update 2)
Policy model authors can now define default values for attributes and default (non-inferred) relationships to have zero instances using the attribute/relationship editor, thus ensuring that these attributes and relationships are never unknown. This enables projects to be de-cluttered by reducing and removing use of the DefaultWithUnknown function in rules. Testing can be streamlined by pre-setting relationships to complete unless provided.
Relevance API for the Flows SDK (Monthly Update 2)
The Flow Engine API now provides a mechanism to see whether a field was relevant to determining any goal. This information can be used to drive relevance-based visibility and allows you to implement features such as decision reports.
Decision service input/output descriptions (Monthly Update 1)
Decision services provide an OpenAPI specification that among other things describes the inputs and outputs of the decision service. This feature adds the ability to include in the OpenAPI description a description for the decision service as a whole and for each input/output field as part of the authoring process. This will allow the meaning or intent of a given field to be conveyed for uptake by a large language model (LLM) so that in the future decision services can be used from within AI Agents.
Debug decision service references in decision service rules
Live view in a decision service project now shows the returned values from each referenced decision service.
This view can be used to see how data objects and values flow through the rule logic, with the ability to focus on particular records in the input data.
Live view support for decision service deployments
Live view is now available for deployed decision service projects. Testers can reduce the time spent troubleshooting in test and production environments by quickly reviewing rule logic in the exact decision service version that is deployed and currently in use.
Deployment name and version rule functions in Policy Modeling
This feature adds two new functions to Policy Modeling projects:
- DeploymentVersion() – Returns the version of the deployment the session was created from as a number value. This allows audibility of which version of the rules were used to make any decision or assessment.
- DeploymentName() – Returns the name of the deployment the session was created from as a text value. This allows audibility of which project was used to make a decision, and can also be used to drive dynamic behavior in the project based on the name used to deploy the project.
Reorder attributes and entities in the Temporal Visualization view in the Debugger
The lines of attributes and entities displayed in the Temporal Visualization view in the Debugger can now be reordered. This enables you to read the information in the order you expect it.
Use Intelligent Advisor to create Help Desk requests in Fusion Service
Help Desk users in the Redwood Experience can now use Intelligent Advisor as a way to process requests. The interview can provide step-by-step guidance which can help make certain that agent is following the defined process and is performing all necessary tasks or collecting all necessary information before taking the next action, such as closing the request or escalating to a Case.
Want to start using the 24B release?
Public cloud: You can upgrade your testing sites at any time via Oracle Cloud Portal Service Cloud Configuration Assistant. All sites will be automatically upgraded according to the published schedule.
Private cloud (on-premises): Licensed customers should download from Oracle® Software Delivery Cloud.