Integration with Decision Intelligence Client Managed Software
zRule Execution Server for z/OS can run decision services developed in Decision Intelligence, a component for modeling and running decisions. It supports collaboration by providing a development environment for both business users and developers.
Decision Intelligence provides decision modeling capabilities that help business experts capture and automate repeatable decisions.
Business experts develop and validate decision services by using Decision Designer, a graphical user interface that provides powerful capabilities along with intuitive menus and wizards. In Decision Designer, business experts can use three main decision artifacts to develop their decision services: data models, decision models, and task models. They can also use Decision Assistant in Decision Intelligence to define decision services.
You can integrate predictive models with decision models to gain insight into your data and make business decisions with greater accuracy. Using Machine Learning with decision modeling allows for better insights and predictions about business. Machine Learning and rule-based decision modeling are complementary approaches to making decisions. While rule-based decision modeling takes a deterministic approach to making decisions, Machine Learning takes a probabilistic approach.
Collaboration between business experts is essential to the success of decision services. In Decision Intelligence, users can work together on a decision service by sharing and managing decisions in the same repository. They work locally in the application and can decide when to share their changes to the decision automation to make them available to others.
For more information, see the IBM Decision Intelligence Client Managed Software
documentation.
Workflow
The following diagram shows the workflow for using Decision Intelligence in zRule Execution Server for z/OS:
Setup and usage
The following activities cover the setup and usage of Decision Intelligence with zRule Execution Server for z/OS. You must have access to Decision Intelligence and its supporting software, including Maven.
| Activity | Description | Information |
|---|---|---|
| Installing | In Eclipse, install a wizard for enabling Decision Intelligence projects to run in zRule Execution Server for z/OS. | Installing the Decision Intelligence wizard |
| Configuring | Start zRule Execution Server for z/OS in the HBRMODE=ADS mode so that COBOL applications can use the Decision Intelligence projects. | |
| Creating and sharing | Develop the decision services in Decision Intelligence. When a decision service is fully developed and validated, it is deployed to a GitHub repository. | Developing projects in Decision Intelligence |
| Converting and enabling | Pull a decision service from the GitHub repository, and use the Maven build command and Eclipse wizard to enable an archive of the project to run in zRule Execution Server for z/OS. | |
| Running | You must run an instance of zRule Execution Server for z/OS in ADS mode for Decision Intelligence projects. COBOL applications can then call the decision services. | Starting zRule Execution Server for z/OS to serve Decision Intelligence execution requests |
| Applying | Add ADS API calls to client COBOL applications. The API calls the Decision Intelligence projects in zRule Execution Server for z/OS. | Workflow for developing COBOL applications that call Decision Intelligence |
| Sample | Run a sample that uses a Decision Intelligence in zRule Execution Server for z/OS. | Automated Decision Services sample |