Introduction to Defining Decision Tables in RASON

Users of RASON have the ability to create and evaluate Decision Tables. A decision table contains a set of rules which specify actions to perform based on specific conditions. Decision tables are a good tool to use when there is a consistent number of rules, or conditions, to be evaluated followed by a specific set of actions to be performed once a rule, or condition, is met. For example, the simple decision table below returns an employee's pay based on the number of hours worked.

If the employee worked from 1 to 40 hours in a week, the pay is $25 * hours. If any overtime is worked, the employee gets paid $45 an hour for any hour above 40.

A RASON decision table has a name, input/output parameters and a body implementing the logic through a structured framework. In RASON, a decision table is created in the decisionTable section.

This same decision table written in the RASON modeling language is below.

{ 
 modelName: "DTIntro",
 modelType: "calculation",
 decisionTables: {
   "tblPayDay": {
	   inputs: ["hours"],
	   outputs: ["pay"],
	   refTypes: ["number", "number"],
     hitPolicy: "C+",
     rules: [
	     ["[1..40]", "25 * hours"],
	     [">40", "45*(hours – 40)"]
	   ]
   }
 },

Use the formulas section to calculate the decision table given the input parameter, hours = 45.

 formulas: {
  "paycheck": {formula: "tblPayDay(,,45)", comment: "calculate pay per employee", finalValue:[]}
 }
}

The modeling language inside of Rason's Decision Table functionality is S-FEEL extended to standard conversion functions in FEEL. For more information on Decision Tables, we invite you to reference the following: DMN Method and Style by Bruce Silver (Cody Cassidy Press, September 28, 2018 and DMN Cookbook by Bruce Silver & Edson Tirelli (Cody-Cassidy Press, April 4, 2018).