Define a decision logic

Define a decision table

Decision logic can be presented as a table. A business rule that exists in your company is represented by a decision rule in the decision table. All cells in a table row form a decision rule.

Table inputs are specified in columns. Each input has a specific data type. The output column shows the result of the applied business rule. Like the inputs, each output also has a data type.

The decision rule then compares input data with the values in the cells.

Add rules to a decision table

To create decision rules, you must configure input data in the Inputs column and map them to output data in the Output column.

Any input data modeled on the canvas is already available when you open decision table. You can add missing input data by defining input columns in the table.

All input data defined via the table is added as input data elements to the canvas when the table is saved. You also need to define decision rules and map the outputs.

To define input data and add decision rules, follow these steps:

1. Open the decision table. To do so, select the decision to which you want to add rules and click .

   The decision logic configuration dialog opens.

   You can also open this dialog via the Decision logic attribute in the attribute panel.

2. Configure the input data. To do so, follow these steps:

   1. Double-click New Input in the column header, and enter a name for the input data.

      If a dictionary entry exists, you can select it.

   2. Below the input data name, double-click Text, select the type of the input data, and define its options.

      If you have chosen a dictionary entry as data input, it depends on the entry's configuration whether you need to specify the data type.

      The following data types are available:

      • Enumeration

      • Text

      • Number

      • Boolean

      • Hierarchy

      • Date

3. In the same way, configure the output data in the Outputs column.

4. In the same way, you can configure the Annotations column to add information to the decision rules.

5. Define the first decision rule:

   1. Double-click the left area of the first table cell. A list opens.

   2. Select an operator to define the relation between the input and the output.

      Read more on operators in section Decision table operators.

   3. Double-click the right area of the same cell. Again, a list opens.

   4. Select or enter an input value.

   5. In the Outputs column, double-click the cell and enter an output value.

6. With the buttons at the top of the table, you can add more rules, input data, output data, and annotations.

7. To define more rules, you have the following options:

   • You can continue defining rules cell by cell as described above.

   • With Add rule > Duplicate rule, you can copy and paste existing rules.

   • With Import/Export > Text Import, you can add text-based decision rules instead of defining the rules cell by cell. Read more in section Import decision rules.

8. Select a hit policy for the decision table and specify the completeness requirement.

   In case of a multi hit policy, you can also select an aggregation. Read more in section Define a hit policy and the completeness requirement.

9. With Verify, you can check the logic for completeness and consistency. Below the table, all combinations of input values are displayed that aren't covered by a decision rule.

10. Confirm with Apply.

    Your decision logic is set up.

Decision table operators

Import decision rules to the decision table

Instead of defining the decision rules cell by cell, you can import text-based decision rules into the table. The columns still need to be configured as described above. The individual rules can be in the form of a list, separated by the delimiters tab, semicolon, or comma.

The import function doesn't overwrite or delete existing rules. Imported rules are always added.

Text-based rules must have the following structure:

• Relational operators like = and <= must be part of the field they relate to.

• For each rule, you need to start a new line.

• The import only supports the following literal expressions:

  • not(value)

  • not(value1, ..., valueN)

  • !=value

To add text-based decision rules, follow these steps:

1. Open the decision table and click Import/Export > Text Import in the top-right corner.

   The text import dialog opens.

2. Select a delimiter.

3. Either enter the decision rules or copy them, for example from a text file or spreadsheet, to the editor field.

4. Click Import.

   The rules are added to the table. You can add more by opening the text import dialog again.

Define a literal expression

As an alternative to the logic in the decision table, you can define logic via a literal expression.

Literal expressions represent predefined logical algorithms or rules that can be used to automatically create output results for decisions, often but not necessarily in a formal expression language. You can use literal expressions to add logic that can't be expressed in the decision table, for example, function calls, interim values, or list manipulation.

Literal expressions are written in FEEL, the 'friendly enough expression language' specified as part of the DMN standard. You can download the specification along with examples for literal expression use cases at https://www.omg.org/spec/DMN/1.1/.

To define a literal expression, follow these steps:

1. Select a decision and click .

   The decision logic editor opens.

2. Select the Expression tab on the left side and add the literal expression.

3. Confirm with Apply.

Read more about literal expressions in section Using advanced literal expressions (functions in DMN decision elements).

Define an informal expression

If decision logic can't be expressed formally, you can describe it with natural language. The description doesn't have to be formal or executable.

To define an informal expression, follow these steps:

1. Select a decision and click .

   The decision logic editor opens.

2. Select the Informal expression tab on the left side and add the description of the decision.

3. Select the data type of the decision logic via the Value domain list in the upper right corner.

4. Confirm with Apply.

Next steps

Link the logic of another decision

Reuse decision logic via business knowledge models

Create dictionary entries for input or output data in decisions