Documentation Index

Fetch the complete documentation index at: https://help.symphonysummitai.com/llms.txt

Use this file to discover all available pages before exploring further.

Incident In-app AI

  • 5 minute(s) read
Prev Next

Incident Details for an Analyst is packed with Generative AI / Predictive AI capabilities, where the user can use the AI features to deliver the best possible service to a customer.

Following are the Generative AI / Predictive AI Capabilities provided to Analyst on the Incident Details page.

Generative AI

Predictive AI

Note

There is no Admin UI available to configure In-app AI.

The following table provides the various states of an Incident with the AI functionality.

New

Assigned

In-progress

Pending

Resolved

Pending Approval

Closed

Cancelled

Generate Insights

✅

✅

✅

✅

✅

✅

✅

✅

Generate Major Incident Summary

✅

Generate Resolution Summary

✅

✅

Risk of Escalation and SLA Management

✅

✅

✅

✅

✅

Note

User cannot generate Resolution Summary, when the Incident is in Pending Approval stage.

How Generate Action works within the Application?

Based on the incident metadata (considers details from the current incident) such as Symptom, Description, Priority, KB articles etc, the Incident summaries are generated. LLM (Large Language Models) uses these details to generate details such as  Incident Summary, Investigation/Interaction, Sentiment, Suggested Knowledge Articles and AI suggestions.

This logic is applicable for all the below Generate actions:

  • Major Incident Summary

  • Generate Resolution Summary

The following is applicable for Generate Insights:

Considers the history of the single incident itself (including communication logs) rather than the history of all incidents.

Generative AI

Generate Insights

Utilizes the Generative AI capability by generating the Incident summary for the Analyst, that include Incident Summary, Investigations and Interactions performed, Sentiment, Suggested Knowledge Articles and AI Suggestions.

Click Generate Insights to view the Incident Summary.
Figure: Generate Insights

Generate MI Summary

Generates Insights for Major Incidents. These recommendations are based on the analysis of the root cause and the resolution process, ensuring continuous improvement in incident management practices.

Note

You can Generate MI Summary after entering details in the Solution field.

Click Generate MI Summary to view and copy the content to a clipboard or print.

Figure: Major Incident Summary

Note

If the Solution field is empty, then you cannot generate MI Summary, a message is displayed as below in the following image.

Figure: Major Incident Summary warning message

Generate Resolution Summary

This functionality ensures that analysts can quickly generate detailed and accurate resolution notes without spending excessive time on writing and editing.

Click Generate Resolution Summary.
The summary is generated with the details below.

Resolution Summary: This section contains Symptom and Description.
Investigations and Interactions Performed: Consists of the investigation on the issue and workaround  performed to fix the issue.


Figure: Generate Resolution Summary

User is prompted with Override the Solution and Append to the Solution options. When user selects Override the Solution, previously added content will be removed and replaced with the generated content. If user selects Append to the Solution, the generated content is integrated to the existing content.

To add to the resolution summary, click Append to the Solution.
Previously, the field did not contain any content text hence Append action will add the resolution summary to the Solution field.

Figure: Generate Resolution Summary

Predicitive AI

Risk of Escalation and SLA Management

Is the Predictive AI feature provided to represent the probability that an incident will need to be escalated to meet SLA deadlines. Based on current and historical data patterns.

Risk of Escalation  

This metric predicts the probability that the incident may be escalated  by the end user.

AI evaluates the following:

  • Past incidents with similar symptoms, categories, or CI (Configuration Items)

  • Incident priority and impact patterns.

A high score—such as 67%—indicates that the incident is likely to be escalated based on the historical ticket escalation of this type. By acting on the prediction early, Analysts can significantly reduce delays and improve the user’s overall wait time.

Example:

A mobile app crash incident historically gets routed to the “Mobile Engineering” group even if analysts initially send it to “Application Support.” Predictive AI will flag a high Risk of Escalation.

Risk of Response SLA  

This metric predicts the likelihood that the Response SLA will be missed. Evaluates the probability that the incident will not receive an initial response within the agreed timeframe.

AI evaluates the following:

  • Current Response SLA deadline vs. typical handling speed.

  • Past performance of the assigned group on similar tickets.

A high percentage—such as the 88% shown—indicates that the assigned team may not acknowledge the incident in time, putting the Response SLA at risk even before any work begins. This early warning helps Analysts take proactive action by immediately prioritizing the ticket, notifying the responsible team, or reassigning the incident to a group that can respond faster.

Example:
If an incident is logged during peak hours or when the team already has a queue of P1/P2 tickets, AI predicts a high Risk of Response SLA.

Risk of Resolution SLA  

This metric predicts the probability that the Resolution SLA may  be breached for this particular incident, meaning the incident may not be resolved within the time defined for its priority. Assesses the likelihood that the incident will not be fully resolved before the resolution deadline.

What the AI evaluates:

  • Historical resolution times for similar issues

  • Known backlog or resource constraints

  • Complexity of the incident based on description/logs

  • Typical resolution patterns for the assigned team

A low score—such as 1%—indicates that the incident is likely to be resolved comfortably within the SLA window, while a high score suggests potential challenges such as increased complexity or the possibility that the incident is assigned to the wrong team. These insights help Analysts act early by escalating the issue when appropriate, involving subject matter experts to accelerate troubleshooting, or breaking the incident into  tasks or child incidents to manage it more efficiently.

Example:
A complex network issue typically takes 5–6 hours to fix, but the SLA window is 4 hours > high Risk of Resolution SLA.

Service Desk Intelligence

Based on the historical data and previous records similarity, AI automatically predicts values for fields such as Classification, Urgency Priority, Category, Workgroup and Analysts when an Incident is viewed by an Analyst. Configure Service Desk Intelligence in AI Configuration.

To view the suggestion, perform the following steps:

  1. Log in to the Application as an Analyst.

  2. Navigate to Incident > User > Manage Incidents > Incident List.

  3. Select the required Incident ID.

  4. AI Suggestion is generated based on the historical data.

  5. Select the values on the Apply checkbox and click Submit.
    Figure: AI Suggestion

  6. The applied suggestion is displayed on the record.
    Figure: Applied AI Suggestions