Process for Escalating Incidents

Here are the main steps necessary to create and implement rules that escalate incidents. The process is the same for escalating answers, opportunities, and tasks.

1. Outline the chain of events for responding to an incident. For example, be sure you understand how much time an agent has to respond to an incident before the agent’s supervisor should be notified. Continue this process through all levels of escalation.
2. Add an escalation function to the incident rule base.
3. Add the first rule to the escalation function.
   This rule creates the escalation level and defines the conditions that cause the rules engine to set an escalation level. For example, the condition might be a previous escalation level or an Unresolved incident status.
4. Add the second rule to the escalation function.
   This rule specifies the action to take when the incident reaches the defined escalation level. For example, you might add an action to notify a supervisor.
5. Repeat steps 3 and 4 for every escalation level.
6. Add a rule that calls the escalation function. Create a rule with no conditions and an action that calls the escalation function you created in step 2, then add this rule to the initial state as the second-to-last rule.
   Note: Normally, you want the other rules in the state to be applied before escalation, and the last rule in a state is usually an action to transition to the next state.
7. Repeat step 6 for every state in the incident rule base.

Results:

What happens now when an incident is added or updated?

• The rules engine applies all rules in the initial state to the incident, including the rule that calls the escalation function. When it calls the escalation function, the rules engine begins processing the rules in the function, comparing the incident to the conditions of each rule. If the incident matches the conditions of a rule, the rules engine schedules the escalation and defines when it should escalate to the specified escalation level.
• Then, during a scheduled run, the Dbstatus utility compares the incident to the escalation schedule and, if the time indicated in the escalation action has passed, it sets the incident escalation level to the one defined in the rule.
• Because the incident has just been modified by having an escalation level set, the rules engine reprocesses the incident. When it applies the rules in the escalation function to the updated incident, the escalation level now matches the conditions of the second rule in the escalation function. Consequently, the escalation action (such as notifying a supervisor) occurs. Rules can escalate an incident multiple times before it is solved.