Decision Tables

Decision tables let you encode business rules as a grid of inputs and outputs — without writing code. Each row is a rule; each column is an input test or output expression. The engine evaluates every rule and applies the configured hit policy to pick the final output.

Conduit uses DMN 1.5 for decision tables and FEEL for expressions.

Quick Start

1. Design your table in the Conduit UI (Decision Table editor)
2. Deploy it via POST /api/v1/decisions with the DMN XML body
3. Reference it from a Business Rule Task in your BPMN process using the decision’s id

Deploying a Decision

POST /api/v1/decisions
Content-Type: application/xml

<?xml version="1.0" encoding="UTF-8"?>
<definitions xmlns="https://www.omg.org/spec/DMN/20191111/MODEL/"
             id="shipping-decisions" name="Shipping Decisions"
             namespace="http://camunda.org/schema/1.0/dmn">
  <decision id="shipping-cost" name="Shipping Cost">
    <decisionTable hitPolicy="FIRST">
      <input id="weight_in" label="Weight (kg)">
        <inputExpression typeRef="number"><text>weight</text></inputExpression>
      </input>
      <input id="zone_in" label="Zone">
        <inputExpression typeRef="string"><text>zone</text></inputExpression>
      </input>
      <output id="cost_out" label="Cost" name="shippingCost" typeRef="number"/>
      <rule id="r1">
        <inputEntry><text>&lt;= 1</text></inputEntry>
        <inputEntry><text>"domestic"</text></inputEntry>
        <outputEntry><text>5.00</text></outputEntry>
      </rule>
      <rule id="r2">
        <inputEntry><text></text></inputEntry>
        <inputEntry><text></text></inputEntry>
        <outputEntry><text>weight * 2.5</text></outputEntry>
      </rule>
    </decisionTable>
  </decision>
</definitions>

The decision id (shipping-cost) is what you set as the Decision Ref in the Business Rule Task.

Data Wiring

Input columns

Each input column’s header expression is a process variable name. The engine looks up that variable in the current BPMN execution context and passes it as the input value for that column.

Example: a column with expression weight reads the process variable weight.

Input entry cells — unary tests (mini-FEEL)

Input cells contain unary tests — they are tested against the column’s input value:

See the FEEL Reference for the full input syntax.

Output entry cells — full FEEL

Output cells contain full FEEL expressions evaluated when a rule matches. They can reference process variables, call functions, or use conditionals:

weight * 2.5
upper case(status)
if amount > 1000 then "premium" else "standard"

Result variables

Each output column’s name attribute is written back as a process variable after the decision runs. In the example above, the output column named shippingCost creates (or overwrites) the process variable shippingCost, available to all subsequent BPMN elements.

Hit Policies

The hit policy controls what happens when multiple rules match. See Hit Policies for the full reference.