“Some pilots are proving quite interesting in enhancing Network Operations. Especially when AI / ML is used in the form of Natural language processing – NOC operators are talking to a ML system in natural language and asking questions in context to an alarm about deep topics ( BGP, 4G eUTRAN, etc). Do more with L1 engineers.
– Use of familiar chat front ends like FB Messenger, Hangouts to have conversation with a OSSChatBot which learns using ML, continuously, the intents of the questions being asked and responds by bringing information back from OSS using APIs. (Performance of a cell site, or Fuel level on a tower site, Weather information in a cluster of sites etc). Ease of use, Open new channels to use OSS
– Training the AI system in TAC tickets, Vendor manuals, Historical tickets, – ability to search and index unstructured data and recommend solutions for problems when they occur or about to occur – informed via OSS. efficiency increase, Lower NOC costs.
– Network planning using ML apis. Ticket prioritization using ML analytics ( which ones to act upon first – certainly not only by severity)”
Steelysan in response to an earlier post entitled, “A thousand AI flowers to bloom for OSS.”
Steelysan has raised some brilliant use cases for AI, chatbots specifically, in the assurance domain (hat-tip Steelysan!). OSSChatBots are the logical progression of the knowledge bases of the distant past or the rules / filters / state-machines of the current generation tools, yet I hadn’t considered using chat bots in this manner.
I wonder whether there are similar use cases in the fulfillment domain? Perhaps we can look at it from two perspectives, the common case and the exception case.
If looking at the common case, there is still potential for improved efficiency in fulfillment processes, particularly in the management of orders through to activation. We already have work order management and workforce management tools, but both could be enhanced due to the number of human interactions (handoffs, approvals, task assignment, confirmed completions, etc). An OSSChatBot could handle many of the standard tasks currently handled by order managers.
If looking at the exceptional case, where there is a jeopardy or fall-out state reached, the AI behind an OSSChatBot could guide an inexperienced operator through remedial actions.
These types of tools have the potential to give more power to L1 NOC operators and free up higher levels to do higher-value activities.
Over the years, I’ve often referred to The Corkscrew Analogy or Momentum Spiral to describe a mindset of incremental improvement that’s needed to keep an
We’ve embarked on a journey of trying to plan for the future of OSS/BSS, not by predicting the future, but by seeking ways to be
As discussed in the previous article, “Future Scenario Planning: Have you given up on doing an OSS transformation? (part 1)” part of the fun of
Earlier in the week, a friend mentioned he was reading “Alchemy” by Rory Sutherland. It’s not an OSS book per se, but does provide some
Here at Passionate About OSS, we are exactly that – Passionate About OSS (Operational Support Systems). We’re also passionate about BSS, NMS, or any other names you call the tools that help to operationalise your network.
2 Responses
Hi Ryan
Kind of you to quote me. Thank you very much.
not really order fallouts specifically, but using AI for O2A process by Infosys’s AI offering is interesting. of course, the devil is in the details.
https://www.infosys.com/nia/