What’s New in SAS Decision Manager 3.1

Decision Builder enables you to combine analytical models, rule flows, and conditional logic into decisions. You can investigate scenarios, test and refine the decision logic, and then publish the decisions. You can choose to publish a decision as a SAS Micro Analytic Service module for use in online transactions or as a Decision Logic metadata object for use in batch applications such as SAS Data Integration Studio jobs. For more information, see Using the Decision Builder and Generating Enterprise Decision Management Output in SAS Data Integration Studio: User’s Guide.

SAS Decision Manager 3.1 is integrated with SAS Relationship Service and SAS Lineage. This integration enables you to view the relationships between the data, rule flows, and models, and the processes that consume them. SAS Lineage displays network, governance, and impact analysis diagrams. The information in these diagrams is drawn from the Relationship database.For more information, see Viewing Lineage Information for a Rule Flow and Add Model Dependencies and View Lineage.

You can now add dependencies to models within the SAS Decision Manager model repository. You can add dependencies between models that are located within the same project or within another project. You can also add dependencies to generic models that are located within a folder. Then you can view the relationships by using SAS Lineage. For more information, see Add Model Dependencies and View Lineage.

In SAS Decision Manager 2.2, when a rule flow is deployed in a SAS Data Integration Studio job, the job is associated with a specific version of a rule flow. Beginning with SAS Decision Manager 3.1, you can specify that a SAS Data Integration Studio job use the latest compatible version of a rule flow. You can use the &DCM_USE_LATEST_VERSION and &DCM_DEPLOYED_RULEFLOW_NAME macro variables to tell SAS Data Integration Studio to always use the latest compatible version of the rule flow. For more information, see Running Rule Flows Dynamically in SAS Decision Manager: Administrator’s Guide.

The %BRM_GET_RULE_FLOW_CODE macro creates a DS2 package that contains the code for a specific rule flow. You can execute this package outside of the SAS Decision Manager interface by using the %BRM_RULE_FLOW macro. For more information, see %BRM_RULE_FLOW in SAS Decision Manager: Administrator’s Guide.

The Inventory subcategory is added under the Models category. You can use this subcategory to access all of the models in the model repository in one place, whether they are located in a folder, portfolio, or project. As a result, you see improved performance when you search the model inventory. You can also filter the search results by date modified, keywords, model properties, and user-defined properties. You can add user-defined properties as columns in the list in order to display the values for each model. For more information, see Overview of Model Inventory.

You can add keywords to a model and also edit existing keywords. As a result, the keywords can be used with other models. You can also use model keywords to filter the search results of the model inventory. For more information, see Add and Edit Model Keywords.

You can add new model versions and view existing versions of a model. A new version is created automatically when the model is set as champion or is published from the project level. A new version is also created when you update a model that is located within a folder. For more information, see Managing Model Versions.

You can now add multiple user-defined properties to a model at one time.

You can import generic models from the SAS Workspace Server directly into a folder in the Projects category view. You can then update the models from or export the models to the SAS Workspace Server. You can also duplicate these models or move them to another folder. However, you cannot move or copy them to a project version. For more information, see Importing Models into a Folder.

If you have licensed both SAS Model Manager and SAS Factory Miner, you can register projects from SAS Factory Miner to the SAS Model Manager model repository. The SAS Factory Miner projects that are registered include their project segments and models. The SAS Factory Miner projects are managed as portfolios in SAS Model Manager. For more information, see Overview of Portfolios.

You can create input and output variables for models from a SAS code file. This feature enables you to generate missing metadata for model variables. For more information, see Create Input and Output Variables from a SAS Code File.

Beginning with SAS Decision Manager 3.1, the rule set editor does not automatically add the term name in an expression that begins with the leading minus sign (-). For example, if you enter –10 for term, SAS Decision Manager interprets this as term = –10.

The behavior for positive values has not changed. The value +10 is still interpreted as term = term +10 in action expressions.