Aravind Chennuru (Software Architect)
Diana Varga (Marketing Manager Europe)
Enghouse Networks showcases how to reduce network issues and interruptions for networks in digital transformation.
In today’s complex telecom arena where networks are governed by a mixture of legacy and virtualized (SDN/NFV/5G) infrastructure, service providers experience challenging margin pressures and growing expectations from their customers. Continuously improving the quality of service (QoS) requires the utilization of artificial intelligence (AI), predictive analysis, and knowledge-based self-learning for zero-touch, automated Service Assurance.
This was the focus of the TM Forum Proactive Service Assurance via Closed Loop Predictive AI/ML Catalyst project led by Enghouse Networks, as presented at the recent Digital Transformation World event in Nice.
Proactive Service Assurance for Networks in Transformation
Championed by leading telecom Service Providers Du and Sri Lanka Telecom, the project uses the Operators’ actual network data and highlights their commitment to continuously improving the quality of service (QoS) for an enhanced customer experience.
· 10M+ subscribers and 150+ different network device types generating millions of alarms per day
· Service Assurance in the Radio Access Network (RAN) domain is a challenge to proactively monitor and manage
· Limited ability to rapidly process and resolve issues, even with sophisticated filtering and correlation tools
· Slow reaction times impact customer satisfaction, brand, churn, and revenue
Architecture of the Catalyst
The AI system utilizes data from the Service Assurance platforms and identifies issues based on patterns for problem identification, network type, equipment type, and problem severity together with previous occurrences of similar problems.
Once the problem is identified, the AI system determines a solution based on data stored in the ticketing system and knowledge repository. Successfully identified solutions are applied automatically through the provisioning system.
Reduced service issues and network interruptions should reduce the customer churn rate; required truck rolls; mean time to repair (MTTR); network queries into the contact centre; and mean cost to repair (MCTR).
The solution can be fully automated with a pass-through to the provisioning system, depending on the envisioned confidence level. If the problem identification and solution is not suited for full automation, the system may present the identified solution(s) to the engineer assigned to the issue.
Further reading on TM Forum Pages
For information that dives into the used datasets, business benefits, and the automation solution examples, click here to read more by TM Forum.