Outage Identification with Process AI

CHALLENGE

Telecommunication companies experience outages for a variety of reasons (weather, fiber cut, etc.). Some are major outages, impacting tens of thousands of customers, while others are much smaller and harder to detect quickly. Our client was receiving an increased call volume when these smaller network outages occurred, impacting Customer Service Rep availability, average speed to answer, call abandonment, etc. Their challenge was to come up with a solution that would allow them to detect these outages sooner and communicate directly with impacted customers.

SOLUTION

The solution we developed utilized Pega’s Process AI feature. We were able to take streaming data from the network and calls to the call center to predict where one of the smaller outages may have occurred. This allows them to notify customers proactively via SMS that there may be an outage in their area and keep them updated on repair status. Doing this eliminates calls to the center, freeing customer service representatives to answer calls quicker.

RESULTS

Through ProcessAI and the corresponding LLM, our client is now able to identify outages up to 30 minutes faster and at a higher accuracy rate than the manual analysis they were using previously. This has eliminated calls to the center that far exceeded the business case projections.