The NBN is using machine learning to reduce customer frustrations

Ubtech’s Alpha 2 humanoid robot. (Source: Ubtech)

The National Broadband Network has launched a new big data project that will analyse the fault repair experiences for retail customers.

The new Tech Lab will collect information via voluntary surveys from customers that have had faults serviced. The data will then have machine learning and trend analysis applied to allow technicians, for example, to immediately determine if a problem can be fixed remotely or if a site visit is required.

“Once the investigation and implementation of the Tech Lab research is complete we could, for example, easily identify trends that occur in a failed activation in order to pre-empt problems before arriving at a house,” said NBN chief systems engineering officer John McInerney.

“Faults are an inevitable part of any technology network but minimising the disruption is key to improving the experience. We expect to see significant improvements as a result of early detection and quick resolution.”

There’s now an average of 45,000 premises connecting each week to the NBN, and the organisation says that makes it more imperative that sophisticated tools like Tech Lab are used for “seamless” experiences for customers.

The news follows reports earlier this week that the NBN was trialling diagnostic tools and “limited” free repairs to faults in wiring within the home, which are technically not within its jurisdiction but is suspected to be a culprit for many Australians suffering from poor performance.

The Tech Lab team will use open source tools like Apache SPARK, Kafka, Flume, Cassandra and JanusGraph, in conjunction with commercial products like Amazon Web Services, RStudio, H2O.ai and ArangoDB.

The NBN last month became available to six million premises, which is over the halfway point of the national rollout.

NOW WATCH: Tech Insider videos

Business Insider Emails & Alerts

Site highlights each day to your inbox.

Follow Business Insider Australia on Facebook, Twitter, LinkedIn, and Instagram.