Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

This document describes how the decision graph is stored in such a way that it follows the XSD of the Decision Model and Notation (DMN) specification.

The goal of storing a decision is that an organization can rebuild a model of the decisions that has have been made by the runtime. By rebuilding the decisions with the actual values, organization have insight how decisions are made. The meta model for the exchange of decisions according to the DMN specifications prescribes that each decision, each business knowledge model and each input variable is a separate DMN element. We will show how this works in practice with an example.

More information on the specification of the meta model and the XSD can be found here: http://www.omg.org/spec/DMN/1.0/Beta1/ 

Image Added

Image Removed

Description of all the parts of a Decision Rule Graph (DRG) at runtime

In the above example decision requirements graph (DRG) shown above, the decision was made on how high the amount of a monthly installment should be. When storing this decision in the trace engine, an XML representation of that decision is generated. The XML document contains the entire decision requirements graph ( DRG ) from the image shown above. All the XML components for a single DRG are placed between the definitions start and end tag.

...

XML representation of the decision requirements graph shown above. 

Decision

Between the decision tags, the DRG components are placed to identify and determine the decision. The attribute name attribute of the decision refers to an [entity.attributewhere ] that was subject to a decision had to be made for. The id attribute uniquely identifies the decision in the entire diagram. Inside the decision other default and custom XML components are included. The custom components can be identified by the bq prefix. The value element contains the value after determination of the decision. To determine a decision, information is needed. To accomplish this, informationRequirement tags are included inside a decision. The informationRequirement These tags point on their way forth to input data (inputData) or other decisions (decision).

For determining a decision, logic is needed. This logic can come from different sources, for example a decision table, business rule, default rule etc. To indicate which logic is used to determine the decision, a knowledgeRequirement tag is placed inside the decision. Take for example the attribute RequiredMonthlyInstallment where a DRG is made for, it uses a default rule.

When an attribute contains a justification text, it will also be displayed inside the DRG graph and the XML variant. In XML this text is displayed between the JustificationText start and end tags inside a decision.
BusinessKnowledeModel

Business knowledge model

 Inside the name attribute of the BusinessKnowledeModel the name of a logic component, such as a decision table, default rule, business rule is placed.InputData

Input data  

The data that is set by the user is placed inside the InputData element tags.