Taking TNBA Further

Case study:

Automating AI Decisions with IPA

Delivered with:
Find out how

The ask

Last year, we created an Advanced Machine Learning model called “Ticket Next Best Action” to help MetTel’s trouble ticket system select its next best action. Once our AI was integrated into MetTel’s ticket resolution system, the telco asked if our software services department could automate the decisions the AI had selected.



What We Did

Not only does TNBA make each decision, it now executes the resulting action using Robotic Process Automation (RPA), or automation through software. 

The Process

The first step before making a decision is triage. Information must be gathered from internal tools and logs about the current and previous status of each affected system. The automation engine pulls the identification from the trouble ticket, connects to several different APIs to request the status of every related system, and prepares this information to assign the ticket to the best resolution team or to request a better decision from TNBA for achieving a resolution.

  1. System status is obtained accessing several APIs across MetTel infrastructure
  2. And internal queue system ensures both robustness and scalability required to meet SLAs
  3. Integration with several notification systems, dashboards and log/audit repositories.


TNBA resolves each ticket on behalf of MetTel’s operators, allowing them to work on more complex tasks, and removing human intervention entirely on many tickets. 



For a free assessment, answers to your questions, or to discover how software and AI can be integrated into your business, don’t hesitate to get in touch.