Accessing IntentAI and Managing Intents

You can access IntentAI to manage AI-driven feature intents to optimize network performance.

Accessing IntentAI from the Navigation Bar

To access IntentAI from the navigation bar, complete the following steps:

  1. From the navigation bar, click AI Assurance > AI Analytics > IntentAI.
    The IntentAI page is displayed without any pre-selected filters, listing all available intents in the table. You can apply custom filters as needed to view specific intents.
    IntentAI Page

    The IntentAI page provides a detailed view of all available network intents in a table format, allowing users to manage and monitor intents efficiently. The page includes several filters and attributes that enable you to narrow down the search and focus on specific intents based on various criteria. The Network Hierarchy filter located at the top right of the UI helps to navigate through the organization’s network structure, filtering intents based on specific network segments or hierarchy levels.

    You can also narrow down the list of required intents by AI Feature, Intent, Category, Venue, and Status. When the list of intents spans multiple pages, you can choose how many intents to display per page using the drop-down menu at the bottom of the page.

    You can remove filters individually or clear them all at once. To remove a specific filter, click the 'x' button in the filter box. To clear all filters simultaneously, use the Clear Filters button.

    The table on the IntentAI page consists of the following columns that summarize each intent:
    • AI Feature: Displays the AI-driven feature associated with the intent (for example AI Operations, AI-Driven RRM). This helps you to identify the AI functionality leveraged to generate the intent.
    • Intent: Describes the network objective or optimization the intent focuses on. This provides a clear overview of the goals of each intent.
    • Category: Specifies the type of network optimization associated with the intent, such as Wi-Fi Experience or Infrastructure.
    • Venue: Displays the venue where the intent is available.
    • Status: Displays the current state of the intent.

      The intent progresses through various statuses during its workflow, depending on the actions taken. The status is set to New when an intent is available but has not yet been applied. Once an intent is scheduled for application, the status changes to Scheduled. After the scheduled intent is successfully applied, the status updates to Active. If there is a failure during application, the status changes to Failed. If an applied intent is later reversed, its status becomes Reverted. The actions available for an intent vary depending on its current status. For more information on the full list of statuses and corresponding actions, refer to Table 1.

      You can configure email notifications to receive updates whenever the status of an intent changes. Once configured, each status change automatically triggers an email notification to the network administrator, providing detailed information about the update. If you had previously configured email notifications for recommendations, the settings may have been reset. To continue receiving notifications for IntentAI status changes, re-enable the email notification configuration.

      For more information on configuring notifications, refer to the Configuring Global Notifications section.

    • Last Update: Displays the latest timestamp when the intent was updated.
    Table 1. Intent Workflow Status and Available Actions
    Status Description Available Actions
    New IntentAI has analyzed the data and generated a change recommendation that is awaiting your approval.

    To review the details, specify intent priority, and apply the recommendations, use the Optimize option. Alternatively, use the 1-Click Optimize option to instantly apply the changes with default priority.

    		Optimize, 1-Click Optimize, or Pause
    Scheduled The change recommendation has been scheduled either automatically by IntentAI or through user actions such as Optimize or 1-Click Optimize. Edit, Revert, or Cancel

    The Cancel option is not applicable for intents automatically scheduled by the system.
    Apply in Progress

    IntentAI-recommended changes are currently being applied to the venue.

    		 Not Applicable
    Active

    Intent is active and has successfully applied the changes to the venue.

    		 Edit, Revert, or Pause
    Paused, Applied Failed

    The recommended changes from IntentAI failed to apply to the venue, and the intent is now paused. To process new data and generate updated recommendations using ML algorithms, select the Resume action.

    		 Revert or Resume
    Revert Scheduled A reversion of the IntentAI-recommended changes has been scheduled. Revert or Cancel
    Revert in Progress The recommended changes from IntentAI are being reverted to the previous configuration. Not Applicable
    Paused, Revert Failed The reversion of the IntentAI-recommended changes has failed, and the intent is currently paused. To process new data and generate updated recommendations using ML algorithms, select the Resume action. Revert or Resume
    Paused, Revert Success The reversion was successful, and the intent is currently paused. To process new data and generate updated recommendations using ML algorithms, select the Resume action. Resume
    Verified

    IntentAI has validated the venue configurations. No new changes have been recommended. It will continue to monitor the venues and raise a recommendation if an improvement is required.

    		 Pause
    Paused The Intent is paused by the user action Pause. A paused Intent will not execute any tasks, including KPI measurement, ML model generations, recommendation generation, or configuration changes. Resume
    No Recommendation, No APs

    No recommendation was generated because IntentAI did not find any APs in the venue.

    		 Pause
    No Recommendation, Not Enough Data

    No recommendation was generated due to insufficient data in the venue. IntentAI requires at least 4 days of data to train its AI models.

    		 Pause
    No Recommendation No recommendation is currently available. The system is awaiting data processing and recommendation generation by the ML algorithms. Pause
    No Recommendation, Conflicting Configuration

    No recommendation was generated because IntentAI detected conflicting configurations in the venue.

    This status is displayed in the following scenarios:
    • When more than 20% of the APs in the venue have configurations with AP overrides.
    • When the venue has active mesh APs.
    		 Pause
    Note: If a venue is deleted, IntentAI may show the deleted venue with a status of either 'Paused' or 'No Recommendation, No APs'.

Applying an Intent Using the 1-Click Optimize Option

The 1-Click Optimize option provides a fast, automated method for applying the intent and optimizing the network with minimal manual input. This hands-off approach allows you to instantly apply the AI-recommended intent with the default configuration and priority setting.

The following task example shows how to apply an AI-Driven RRM intent. Complete the following steps to apply the intent using the 1-Click Optimize option.

  1. From the navigation bar, click AI Assurance > AI Analytics > IntentAI. The IntentAI page is displayed.
  2. In the IntentAI table, select the intent with the status labeled New.
  3. Click 1-Click Optimize at the top of the IntentAI table. A confirmation dialog box is displayed.
    1-Click Optimize Confirmation Dialog Box
  4. Optionally, click the Change time option if you want to schedule a different date and time for applying the intent, other than the system’s default. The system retains this scheduled time for future use, which IntentAI leverages daily to implement any necessary adjustments. If any subsequent network configuration changes are detected, the intent will be automatically applied at the same time as the originally scheduled time each day, ensuring consistent optimization based on the latest network conditions.
    Choose off-peak hours, when the number of Wi-Fi clients are at a minimum, to apply the changes. The time you select is based on your browser's time zone. If the APs are located in a different time zone, adjust the scheduled time accordingly to ensure changes occur during off-peak hours.
  5. Click Yes, Optimize to enable the intent to be applied at the scheduled time.

    Additionally, this feature supports bulk optimization, allowing you to select and apply multiple intents simultaneously to optimize the network efficiently.

Applying an Intent Using the Optimize Option

The Optimize option launches a self-explanatory workflow wizard that guides you through a detailed analysis of the intent, allowing you to review key performance indicators (KPIs), visual charts, trade-offs, and configuration options. The system presents two configuration choices based on network priorities that define the intent, and while IntentAI suggests an optimal solution, you can override it to select an alternative configuration aligned with your network goals.

The following task example shows how to apply an AI-Driven RRM intent. The workflow wizard remains consistent across other AI-feature intents, but there will be specific variations in the intent details, KPIs, intent priority, and summary information based on the AI-feature being applied.

Complete the following steps to apply an intent using the Optimize option.

  1. From the navigation bar, click AI Assurance > AI Analytics > IntentAI. The IntentAI page is displayed.
  2. In the IntentAI table, select the intent with the status labeled New.
  3. Click Optimize at the top of the IntentAI table. A workflow wizard is displayed to guide you through the configuration.
    AI-Driven RRM Optimization Workflow - Introduction
    The Introduction page of the workflow wizard provides a snapshot of the intent details such as the venue, status, the date on which the intent was last updated, brief description of the network intents, and the benefit of applying the recommended change.
    The interactive dot graph at the bottom of the page provides a side-by-side visual representation of both the current state and the expected changes once the recommendation is applied. You can drag the slider to the left or right to view the comparison graph.
  4. Click Next to move to the Intent Priority page of the workflow wizard.
    AI-Driven RRM Optimization Workflow - Intent Priority
    The system typically presents two configuration options based on specific network priorities that define the user intent. While IntentAI recommends the best course of action by default, you can override this recommendation and select the alternative intent based on your desired goal for network optimization. For more information about feature-specific intents and priority details, refer to AI Feature-specific Intents, Network Priorities, and Scope.

    The page also displays a Potential trade-off box that outlines the key considerations you should be aware of before applying the recommendation. It describes the scope of venue-level configuration changes that will be applied and the potential impact on the network after applying the recommendation.

  5. Click the radio button to select the intent priority of your choice.
  6. Click Next to move to the Settings page of the workflow wizard.
    AI-Driven RRM Optimization Workflow - Settings
  7. Select the date and time in the calendar to schedule the time at which you want to apply the recommendation.
    The system retains this scheduled time for future use, which IntentAI leverages daily to implement any necessary adjustments. If any subsequent network configuration changes are detected, the intent will be automatically applied at the same time as the originally scheduled time each day, ensuring consistent optimization based on the latest network conditions.
  8. Click Next to move to the Summary page of the workflow.
    AI-Driven RRM Optimization Workflow - Summary
    The Summary page provides a quick summary of the configuration recommendation, schedule of the intent activation, and a comparison graph that shows a side-by-side visual representation of both the current state and the expected changes once the recommendation is applied.
  9. Click View More to view the detailed key performance indications report.
    This report displays the current configuration, including the number of interfering links discovered in a venue and the expected value (when the recommended AI-Driven RRM is applied) in a graph data structure. The legend on the right provides information on channel bandwidth sizing as well as transmit power changes. Pausing the pointer over any dot will bring up a side-by-side example that further identifies specific APs as well as the proposed changes.
  10. Click Apply to enable the recommendation to be applied at the scheduled time.
    Intents that are applied successfully are also reflected on the Config Change page. For more information, refer to the Configuration Change section.

Reverting an Intent

If the AI-recommended configuration is not aligned with your desired network goal or if you do not want to apply it, the intent can be reverted. Reverting an intent restores the network to its original configuration prior to any changes made by IntentAI. However, if the network topology or dependent configurations have changed, the revert action may fail, causing the intent to enter a paused state. In such cases, you can resume the intent, and IntentAI will continue its analysis based on the currently applied configuration.

Complete the following steps to revert an intent.

  1. From the navigation bar, click AI Assurance > AI Analytics > IntentAI. The IntentAI page is displayed.
  2. In the IntentAI table, select the intent you want to revert.
    The Revert option is available based on the status of the intent. For more details on statuses and their corresponding actions, refer to Table 1.
  3. Click Revert at the top of the IntentAI table. A calendar prompt is displayed for scheduling.
  4. Select the date and time in the calendar to schedule the time at which you want to revert the intent.
  5. Click Apply to enable the recommendation to be reverted at the scheduled time.

    You can revert multiple intents simultaneously. However, only intents with the same status (that support the Revert action) can be reverted in bulk.

Pausing an Intent

Pausing an intent halts any further configuration changes, leaving the network in its current state. While the intent is paused, IntentAI will not make new recommendations or apply any adjustments until the intent is resumed. The network configuration remains as-is, preserving the last applied settings.

Complete the following steps to pause an intent.

  1. From the navigation bar, click AI Assurance > AI Analytics > IntentAI. The IntentAI page is displayed.
  2. In the IntentAI table, select the intent you want to pause.
    The Pause option is available based on the status of the intent. For more details on statuses and their corresponding actions, refer to Table 1.
  3. Click Pause at the top of the IntentAI table. The selected intent will remain in its current state until resumed.

    You can pause multiple intents simultaneously. To pause intents in bulk, select multiple intents (with the statuses that support the Pause action) and click Pause.

Resuming an Intent

Resuming an intent restarts the IntentAI assessment and learning process from the current network configuration. Once resumed, IntentAI will analyze the most recent data and begin generating new recommendations.

The intent's workflow for the venue will be triggered at the predefined time, allowing the system to continue optimizing network performance.

Complete the following steps to resume an intent.

  1. From the navigation bar, click AI Assurance > AI Analytics > IntentAI. The IntentAI page is displayed.
  2. In the IntentAI table, select the intent you want to resume.
    The Resume option is available based on the status of the intent. For more details on statuses and their corresponding actions, refer to Table 1.
  3. Click Resume at the top of the IntentAI table. The selected intent will restart and resume automated assessment and recommendation generation at the predefined time.

    You can resume multiple intents simultaneously. However, only intents with the same status (that support the Resume action) can be resumed in bulk.

Editing an Intent

Intents can be adjusted at any time while in a Scheduled or Active state, allowing you to make real-time modifications to intent parameters such as priority or scheduled time. This flexibility provides better control and visibility over AI-recommended changes.

Complete the following steps to edit an intent.

  1. From the navigation bar, click AI Assurance > AI Analytics > IntentAI. The IntentAI page is displayed.
  2. In the IntentAI table, select the intent you want to edit.
    The Edit option is available based on the status of the intent. For more details on statuses and their corresponding actions, refer to Table 1.
  3. Click Edit at the top of the IntentAI table. The Intent workflow wizard page opens, allowing you to modify the intent's priority and scheduled time.
  4. Make the necessary changes and click Apply.

    The Edit option is applicable only to individual intents; multiple intents cannot be edited simultaneously.

Canceling an Intent

An intent can be canceled if it is in a Scheduled or Revert Scheduled state before the changes are applied. This allows you to stop the execution of a scheduled intent. However, canceling is only possible for user-scheduled intents and does not apply to those automatically scheduled by the system.

Complete the following steps to cancel an intent.

  1. From the navigation bar, click AI Assurance > AI Analytics > IntentAI. The IntentAI page is displayed.
  2. In the IntentAI table, select the intent you want to cancel.
    The Cancel option is available based on the status of the intent. For more details on statuses and their corresponding actions, refer to Table 1.
  3. Click Cancel at the top of the IntentAI table to cancel the intent.

    You can cancel multiple intents simultaneously. However, only intents with the same status (that support the Cancel action) can be canceled in bulk.