Chapter 2. DMN support in Red Hat Decision Manager
Red Hat Decision Manager provides design and runtime support for DMN 1.2 models at conformance level 3, and runtime-only support for DMN 1.1 and 1.3 models at conformance level 3. You can integrate DMN models with your Red Hat Decision Manager decision services in several ways:
- Design your DMN models directly in Business Central using the DMN designer.
- Import DMN files into your project in Business Central (Menu → Design → Projects → Import Asset). Any DMN 1.1 models that you import into Business Central, open in the DMN designer, and save are converted to DMN 1.2 models.
- Package DMN files as part of your project knowledge JAR (KJAR) file without Business Central.
In addition to all DMN conformance level 3 requirements, Red Hat Decision Manager also includes enhancements and fixes to FEEL and DMN model components to optimize the experience of implementing DMN decision services with Red Hat Decision Manager. From a platform perspective, DMN models are like any other business asset in Red Hat Decision Manager, such as DRL files or spreadsheet decision tables, that you can include in your Red Hat Decision Manager project and deploy to KIE Server in order to start your DMN decision services.
For more information about including external DMN files with your Red Hat Decision Manager project packaging and deployment method, see Packaging and deploying a Red Hat Decision Manager project.
2.1. Configurable DMN properties in Red Hat Decision Manager
Red Hat Decision Manager provides the following DMN properties that you can configure when you execute your DMN models on KIE Server or on your client application:
When enabled, this property disables by default any extensions or profiles provided beyond the DMN standard, such as some helper functions or enhanced features of DMN 1.2 backported into DMN 1.1. You can use this property to configure the decision engine to support only pure DMN features, such as when running the DMN Technology Compatibility Kit (TCK).
When enabled, this property enables verification of actual values conforming to their declared types in the DMN model, as input or output of DRD elements. You can use this property to verify whether data supplied to the DMN model or produced by the DMN model is compliant with what is specified in the model.
By default, this property makes the result of a decision service defining a single output decision be the single value of the output decision value. When disabled, this property makes the result of a decision service defining a single output decision be a
contextwith the single entry for that decision. You can use this property to adjust your decision service outputs according to your project requirements.
When valorized with a Java fully qualified name, this property loads a DMN profile onto the decision engine at start time. You can use this property to implement a predefined DMN profile with supported features different from or beyond the DMN standard. For example, if you are creating DMN models using the Signavio DMN modeller, use this property to implement features from the Signavio DMN profile into your DMN decision service.
When valorized with a Java fully qualified name, this property loads and registers a DMN Runtime Listener onto the decision engine at start time. You can use this property to register a DMN listener in order to be notified of several events during DMN model evaluations.
You can also configure this property in the
kmodule.xmlfile in your project.
<kmodule xmlns="http://www.drools.org/xsd/kmodule"> <configuration> <property key="org.kie.dmn.runtime.listeners.mylistener" value="org.acme.MyDMNListener"/> </configuration> </kmodule>
When enabled, this property enables DMN decision table logic to be compiled into executable rule models during run time. You can use this property to evaluate DMN decision table logic more efficiently. This property is helpful when the executable model compilation was not originally performed during project compile time. Enabling this property may result in added compile time during the first evaluation by the decision engine, but subsequent compilations are more efficient.