Configuring an Intelligent IVR Flow

Procedure

1. Sign in to the AICC as a system administrator and choose Configuration Center > Chatbot Management > Flow Configuration > System > Trustlist.
2. Click Create. On the page that is displayed, add the IP address and port number for invoking the SmartCare API to the whitelist of the corresponding tenant space.

   

3. Sign in to the AICC as a tenant administrator and choose Configuration Center > Chatbot Management > Flow Configuration.
4. Choose Resource > Business Interface and click Create. The Add Business Interface Configuration page is displayed.
   1. Configure the SmartCare API for querying users with common issues.
      1. Basic information: request URL: https://{IP}:{PORT}/apiaccess/CC-Management/v1/smartcare/queryfaultyuser

         

      2. Message header example:

         x-app-key:9******************************3  
         Authorization:Bearer e******************************e  
         Content-Type:application/json

         

      3. Message body example:

         {
         "msisdn": "${msisdn}",
         "faultTime": "${faultTime}",
         "complaintScene": "${complaintScene}",
         "backtraceDuration": "${backtraceDuration}",
         "sessionID": "${sessionID}",
         "queryFlag": "${queryFlag}"
         }

         

      4. Output parameter:

         

   2. Configure the SmartCare quick fault demarcation API.
      1. Basic information: request URL: https://{IP}:{PORT}/apiaccess/CC-Management/v1/smartcare/submitquickdelimit

         

      2. Message header: Authorization and x-api-key

         

      3. Message body example:

         {
         "msisdn": "${msisdn}",
         "startTime": "${startTime}",
         "endTime": "${endTime}",
         "locale": "${locale}",
         "scene": "${scene}",
         "longitude": "${longitude}",
         "latitude": "${latitude}",
         "radius": "${radius}"
         }

         

      4. Output parameter: taskId of the demarcation task

         

   3. Configure the SmartCare API for quick demarcation result query.
      1. Basic information: request URL: https://{IP}:{PORT}/apiaccess/CC-Management/v1/smartcare/queryquickresult

         

      2. Message header: Authorization and x-api-key

         

      3. Message body:

         {
         "taskId": "${taskId}"
         }

         

      4. Output parameter: rootcause returned by the API

         

5. Choose Configuration Center > Chatbot Management > Flow Configuration > Flow > Orchestration and click Create. The Create Flow page is displayed.
   1. Create a flow.

      

   2. Edit the IVR flow.
      1. IVR flow configuration overview:

         

      2. Configure the start diagram element, as shown in the following figure.

         

      3. Configure the token query API diagram element, as shown in the following figure.

         

      4. Configure the diagram element of the SmartCare API for querying users with common issues. (By default, the fault time is the current system timestamp, and the MSISDN is the calling number.)

         

      5. Configure the branch judgment diagram element, as shown in the following figure.

         

      6. Configure the robot reply diagram element, as shown in the following figure.

         

      7. As shown in the following figure, you can use the Variable assignment diagram element to set the fault start time for query. (The start time is the system time minus one day.)

         

      8. Configure the SmartCare quick fault demarcation API, as shown in the following figure.

         

      9. Configure the robot reply diagram element, as shown in the following figure.

         

      10. Configure the SmartCare API for quick demarcation result query, as shown in the following figure.

          

      11. Configure the branch judgment diagram element, as shown in the following figure.

          

      12. Configure the variable assignment diagram element, as shown in the following figure.

          

      13. Configure voice playback for troubleshooting suggestion result processing, as shown in the following figure.