Monitoring and Analytics for Task Intelligence for ITSM

  • Release version: Washingtondc
  • Updated February 1, 2024
  • 1 minute to read
  • You can view the impacts of your trained incident prediction models. Monitor model performance overtime, track business value, and view what predictions your agents did and didn't use.

    To access the Task Intelligence Analytics dashboard, navigate to All > Task Intelligence for ITSM > Monitoring.

    Use the Model drop-down list to select a model. Select Apply to open the model's configuration.

    The Analytics dashboard contains the following sections:
    • Get an overview
    • See how your trained model is doing

    UI of the Task Intelligence Analytics dashboard.

    Get an overview

    The Analytics dashboard uses visuals to represent the performance overview of the model.
    Number of predictions
    The line graph shows the number of fields that the model predicted over time. As the model continues to learn, it can increase the number of predictions.
    Incident Mean time to resolve (MTTR)
    The line graph shows the average amount of time that it takes to resolve incidents over time. As the model makes more predictions, the MTTR must decrease as the predictions help agents.

    See how your trained model is doing

    The Analytics dashboard uses visuals to track how the model used predictions over time.
    Predictions agents accepted
    The line graph shows the correct predictions that agents used over time during incident management. If this number is trending downwards, consider retraining your model. For more information on editing a trained model, see Edit an incident prediction model in Task Intelligence for ITSM.
    Predictions agents replaced
    The line graph shows the incorrect predictions that agents removed over time during incident management. If this number is trending upwards, consider retraining your model. For more information on editing a trained model, see Edit an incident prediction model in Task Intelligence for ITSM.