Where is Your CSP on the Closed Loop Automation Journey?
Automation in telecom is not new. But it has evolved.
Initially, it was about being more efficient at solving well-defined problems – the known knowns. Solutions like Root Cause Analysis that reduced tens of thousands of NOC alarms to a single “parent alarm”, which highlighted the underlying problem.
From there automation progressed to addressing known problems that are characterized by more dynamic scenarios – the known unknowns. With known unknowns you can’t know ahead of time the full spectrum of scenarios and parameters that you might confront – but you know the broad characteristics and can code for them ahead of time. An example of this is self-optimizing mobile networks. You don’t know where or when coverage holes will occur, but you can detect them and up-tilt surrounding antennas to mitigate the problem.
Today, we are moving into the era of the unknown unknows where software can identify problems that we can’t necessarily code for in advance. It’s like detecting and mitigating new cyber threats that were never encountered before. Machine Learning and Artificial Intelligence are the ‘secret sauce’, that, for example, could enable software to progressively learn traffic patterns, detect anomalous deviations from the norm and initiate relevant mitigation.