Rule sets are groups of rules that can be executed as a single rule by the rule engine. Rule sets allow logic that is of a similar nature to be treated as a single item in execution (with multiple results) and grouped and edited together.
Rule sets also provide a structure for rule selection, where rules can be included or excluded based on ‘effective dates’ or subsets based on other classifications.
Example: Credit scoring rules might change each year (or month) with some old rules being replaced by newer rules.
Rule sets also can be constructed in a hierarchy to allow common rules that can be reused and specialized; rules that only apply to a subset of business conditions.
Example: Insurance approval rules where there is a base set of rules then a set for each policy that uses the base rules plus has additional rules that apply only to that policy.
Interceptor rules are a rule set construction aimed at data cleansing and mapping. Interceptor rules run against a stream of data and allow individual data elements (messages) to be evaluated, changed, or even stopped for an analyst to evaluate. The analyst can either fix the message data, stop the data permanently, or even construct a new rule to change this message and others like it – adding it to the rule set.
Example: Evaluating messages coming into a medical record (EHR) system and stopping messages that are missing data or badly formatted – ensuring data in the system maintains integrity.
The ‘Decisions’ Rule Engine provides common management of rule sets – however, unlike rules that follow mostly standard patterns – the exact logic of how rule selection (which rules to run) and rule result evaluation can vary greatly based on business requirements. Customized rule execution harnesses are built as sequential rules and have great flexibility in how rule sets are managed and what various rule execution results mean within the context of a specific execution.