Define Rules to Trigger Incidents

Trigger an incident when a certain condition that causes a production issue evaluates to true. Maintenance technicians can immediately view the list of incidents from their mobile app to repair the affected machine preventing an outage or minimizing the outage time.

Rules can use conditions based on sensor attribute values, metric values, anomaly conditions, or prediction values.
  1. Click Menu Menu icon and then click Rules.
  2. Click Add Add icon.
  3. Enter a name to identify this rule.
  4. From the Apply To list, select one of the following options:
    • Select Specific Machines, and then click the field that appears next to it to select a machine.

    • Select All Machines Within Type, and then click the field that appears next to it to select a machine type.

      If you select a machine type, the rule applies to all machines of the selected machine type.

      Optionally select Use Global Metrics to use a previously defined global metric for the rule condition. Global metrics are calculated for a machine type as a whole, as opposed to metrics that are calculated per machine.

    • Select Specific Factories, and then click the field that appears next to it to select a factory.

    • Select All Machines if you wish to use an aggregate metric defined for all machines as your rule condition.

      For example, if you have defined an aggregate machine metric to calculate the current average production quantity, you can define a rule to trigger a warning if this number falls below a threshold value.

    • Select All Factories if you wish to use an aggregate metric defined for all factories as the rule condition.

      For example, if you have defined an aggregate factory metric to calculate the maximum number of machines that are down currently, you can define a rule to trigger a warning if this number exceeds a threshold value.

  5. In the Condition area select an attribute, a condition, and a value.
    You can apply multiple conditions to the same rule.
    You can also use anomaly-based conditions. So, for example, you can configure a condition to trigger the rule if an anomaly occurred in the last one hour.
    You can also use prediction value conditions. So, for example, you can configure a condition to trigger the rule if the predicted value for a machine attribute exceeds a threshold value.
  6. In the Fulfillment section, from Fulfill When select All Conditions Apply to trigger the rule if all conditions are true, or Any Conditions Apply to trigger the rule if any of the conditions is true.
  7. From Generate, select Incident.
  8. In the Incident Details section, enter a summary that describes this rule.
  9. From the Type list select a type of incident.
    Available values are: Outage, Maintenance, and Routine.
  10. From the Level list select a priority for this incident.
  11. Use the up and down arrows in the Suppression field to specify the number of minutes to wait before triggering this incident again after the incident is resolved.
  12. (Optional) Add one or more tags.
    You can access these tags using REST APIs or external systems.
  13. (Optional) Enter a description.
  14. Click Save.
You can disable or enable existing rules from the Rules page.
The Rules page shows you how many incidents were generated since you defined the rule.