Change In-app AI

The AI Copilot for Change Management acts as a smart assistant that helps analysts, approvers, and managers streamline the end-to-end change lifecycle. It leverages historical data, predictive analytics, and natural language interaction to guide decision-making, reduce risks, and improve efficiency.

Note

The In-App AI features here are predicted by AI for every page load. The Prediction result may vary every time when the page is loaded. To arrive at the decision based on the predicted result, you should repeat this and take an average of the predicted result.

As an Analyst you will be able to use the below Copilot features in a Change Record.

• Suggested Change Type

• Suggested Risk

• Approval

Suggested Change Type  

Copilot analyzes historical change records and patterns to determine whether a proposed change should be classified as a Standard Change or a Normal Change. It considers factors like the nature of the change, potential impact, and past outcomes to ensure reliable categorization.

The Change Types are configurable and can vary from customer to customer. Based on the Change Types that are trained on to the LLMs and the historical data patterns from the previous Change Records, as a combined result the Change Type is predicted.

Note

The values in the Change Type dropdown are configurable. For more information, refer to Change Type.

To view Change Type prediction, perform the following steps:

1. Log in to the Application.
   

2. Navigate to Change  > User > Change Record List.
   Change Record List page is displayed.
   

3. Select Change Record ID on the list page
   

4. Change Record ID details are displayed along with the predicted Change Type.
   

   Figure: Change Type Prediction
   

5. Hover over the Predicted Change Type for more information about how the prediction is suggested. The Type is determined based on the attributes such as historical data for Change Type selected for the change category, and other relevant attributes.
   

   Figure: Suggested Change Type
   
   As Change Type is configurable, training the LLM with the change types on the CRs will enable you to view the suggestion accordingly.

Suggested Risk

This feature predicts the risk level associated with implementing a change. It assesses potential threats and Predicts a Risk Value such as Low, Medium, and High enabling analysts to plan and mitigate accordingly. Prediction is depends on the historical data from the last 6 months.

Risk Prediction depends on the previous Change Records of a customer. It considers the Change Description from the current Change Record and all the past change records with the similar Description for the CI. Based on the historical CR data for the CI which is available in the AI Agent and the information that is available in the CR such as Description, Urgency, Impact, Priority, Category details etc. combined the AI predicts the Risk level.

To view change record risk prediction, perform the following steps:

1. Log in to the Application.
   

2. Navigate to Change  > User > Change Record List.
   Change Record List page is displayed.
   

3. Select Change Record ID on the list page.
   

4. Change Record ID details are displayed along with the suggested Change Risk.
   

   Figure: Suggested Risk

   
   Hover over the Suggested Change Risk for more information about how the prediction is suggested. The Risk is determined based on the historical data patterns for the CI, Change Record Description, Symptom and other SLA parameters.
   

   Figure: Predicted Risk Message
   

   Notes

   • In-app AI displays a risk score (low, medium and high) for each Change Request.

   • The risk score updates automatically when request details are modified by the user.

   • In-app AI response time is less than 3 seconds for a Change Request.

   • AI uses only authorized data and does not reveal sensitive information.

   Change Risk is configurable, based on the configured Risk the predictions are displayed.

Note

Here, the AI Copilot will only predict and provide the Risk value as a suggestion. It is up to the Analyst to change the Risk value from the dropdown. If the Analyst is in agreement with the AI prediction then the user can change the value accordingly.

Approval

Similarly, as an analyst you can view the prediction that if the CR is safe to approve or not. Based on the data gathered from the current CR’s Description and CI which then compares with the past CRs to determine the success rate of the CR.

If the percent is higher then it indicates greater probability for the CR to be successful.

To view the likelihood of Change Record success, perform the following:

1. Log in to the Application.
   

2. Navigate to Change > User > Change Record List.
   Change Record List page is displayed.
   

3. Select Change Record ID on the list page.
   

4. Select Approval in the CR’s Action panel to view the likelihood of the CR success.
   

   Figure: Likelihood of Change
   
   Following is the likelihood of change success prediction.

   Percentage level	Success Rate	Description
   0% - 20%	Fail	Lacks feasibility and proper planning.
   21% - 40%	Revise	Needs major risk mitigation and resource assurance.
   41% - 60%	Clarify	Requires detailed impact analysis and mitigation plans
   61% - 80%	Condition	Needs final validation or supporting evidence
   81% - 100%	Success	Meets all requirements; ready for approval