
We use cookies to improve your browsing experience. By continuing to use the website you allow Senza Fili to deploy cookies, as detailed in our privacy policy.
The days of vertically integrated, single-vendor solutions are long gone. Network complexity requires a robust, interconnected ecosystem of vendors and service providers that support operators. The Catalyst projects at DTW showcase ecosystems that help operators progress toward autonomous networks, one step at a time, until they reach the final Level 5. Each project starts with an innovative solution, but equally importantly shows how participants work together toward a common goal.
This was the case with the second phase of the “CX optimization via AI-driven SOC over Autonomous Networks” project led by VIAVI Solutions, with Claro, Entel and Safaricom as the operator champions, and with Iquall, NTT DATA and Persistent as ecosystem partners.
The challenge operators presented was finding an effective way to resolve network inefficiencies that had long degraded the user experience and proven difficult to address. Operators needed a new way to identify the root causes of user experience issues and resolve them quickly and effectively. Could AI optimization, agentic AI or digital twins help? And beyond improving the user experience, could they automate the optimization process?
Yes, they can help, but not in a single leap. Operators need a gradual path, and a sharp focus on the issues to solve and the outcome they want.
It all starts with Levels 1 and 2, where most operators are today. This is where they build the foundation, with inventory and topology auto-discovery, event correlation, intelligent incident management, automated root cause analysis and workflow automation, and where they start exploring closed-loop automation. This stage delivers deeper network visibility, faster identification of network issues, and more effective problem resolution, along with initial, measurable results.
Once they have mastered Level 2, they can move to Level 3 and beyond and gradually introduce agentic AI and digital twins at scale in commercial networks. Agents will evolve from tightly supervised task executors to intent-driven experts capable of learning, analyzing, making decisions and executing them. Digital twins will evaluate proposed actions or remedies to ensure they work as intended, improving network performance and user experience, while protecting the network.
VIAVI’s CX Optimization Catalyst project guides operators from Level 2 to Levels 3 and 4 as they choose the right architecture and develop a compelling strategy, with the support of complementary partners, each with a specific area of expertise.
At DTW, Jose Romero at Entel Chile discussed how agentic AI and digital twins give Entel a more efficient way to manage its network to improve the user experience. The initial goal is to reduce the interference that affects 20-30% of cell sites in Entel’s 4G network.
Today’s operating model is fragmented across teams, driven by equipment alarms rather than user impact, and relies on case-by-case manual intervention. The result? It takes too long to identify and remedy issues, and unnecessary field visits add to the engineers’ workloads, while customer satisfaction remains low.
With automation and AI, Entel Chile can shift to a predictive approach that prevents network issues from affecting the user experience by addressing them as soon as they emerge. Interference detection becomes a real-time activity continuously tied to user impact. Resolution becomes driven by increasingly automated policies.
At Levels 1 and 2, this new interference management approach remains largely manual as operators refine automation processes and assess their effectiveness. As they move to Levels 3 and 4, two AI agents, the RF interference agent and the execution agent, take over identifying and resolving interference issues, with the assistance of digital twins. Entel Chile expects to see an 80% reduction in human effort per affected cell, from 3.75 to 0.75 hours.
Discussing the Catalyst project, Jose Romero noted: “We are at the beginning of an exciting journey that will equip us with the tools we need to address persistent RF interference issues. We are confident that we can now address issues before they affect our customers and expect AI agents to provide the scalability we need to roll out this solution across our entire network.”
The third phase, next year, will be the real test. Agents and digital twins will have to prove themselves across a wide commercial network, with all the ecosystem partners working together to improve the user experience as operators move to higher automation levels.
“We are at the beginning of an exciting journey that will equip us with the tools we need to address persistent RF interference issues. We are confident that we can now address issues before they affect our customers and expect AI agents to provide the scalability we need to roll out this solution across our entire network”, Jose Romero, Entel Chile
Supported by
More from VIAVI Solutions:
CX optimization via AI-driven SOC over autonomous networks (TM FORUM Catalyst project)
NITRO AIOPs Redefining Network Operations
VIAVI NITRO AIOps, Pathway to NOC Automation
From Complexity to Clarity: AI-First Network Inventory for Telecom
We use cookies to improve your browsing experience. By continuing to use the website you allow Senza Fili to deploy cookies, as detailed in our privacy policy.