Business Rules

In a decision management system, decisions are represented by business rules, also called decision logic. To make a decision, the conditions of business rules are evaluated and according actions are taken based on pre-defined expressions. Rules and rule sets can be organized and displayed in many formats – decision tables, decision trees, scorecards, and decision graphs. Normally a rule engine performs the task of delivering the decision services by executing rules against input data. The methods of such rule execution can be one of any forms among forward/backward chaining inference, Boolean logic, or runtime code generation. A true modern, powerful, and smart digital decisioning system delivers optimized decisions through high-performing execution and sophisticated yet flexible decision modeling facilities that can easily and seamlessly operationalize advanced analytics and machine learning algorithms.

To make business rules work in an automated decisioning system, they need to be defined, authored, configured, stored, tested, deployed, executed, optimized, and managed on an on-going basis. A flexible rule authoring environment should provide options for both business-type decision experts as well as technical professionals. With such flexibility, business users can take full advantage of intuitive, drag-and-drop graphical user interface to view, edit, and create rules and model decision flows. Yet the power is not lost at technical user side since the system also provides rich grammar, user-defined vocabulary, and English-like syntax rules for sophisticated rule authoring and programming capability.

As an active, emerging area in enterprise software industry, digital decisioning and decision modeling specifically enjoys wide adoption and support from standards bodies such as OMG (Object Management Group). Open standard, XML-based DMN (Decision Model and Notation) becomes very popular for decision modeling, FEEL (Friendly Enough Expression Language) for rule expression, and PMML (Predictive Model Markup Language) from Data Mining Group for describing and exchanging predictive models produced by data mining and machine learning algorithms. Coming with the decisioning products, FICO also developed SRL (Structured Rule Language) for codifying business rules. It’s English-like and easy-to-understand.