Topics AI (Artificial Intelligence), Network as a Service (NaaS), Open Digital Architecture (ODA)
A new AI-driven cloud native stack for network assurance and service management
Project companies
GenAI driven end to end automated Network operations can deliver “Zero-X” experience for NetCo organization, delivering improvements in operational metrics:
* >80% improvement in MTTD,
* > 50% reduction in trouble ticketing,
* > 50% in MTTR and
* >80% reduction in customer complaints
Context:
Telco's across globe are witnessing a massive transformation of technology, explosion of services and redefining its interaction with the customers. They are experiencing severe operational challenges due to inorganically grown complex OSS / BSS tools, fragmented inventory and domain specific strategies. The challenge we are trying to address is the fragmented nature of service management layer, characterized by disparate tools, processes, and interfaces across operational groups. This fragmentation leads to inefficiencies, delays, and increased operational costs. This particularly got amplified during the COVID-19 pandemic, which underscored the limitations of our manual processes and culture.
For most CSPs, harnessing the scalability and flexibility of the cloud isn’t easy. That’s because CSPs’ existing IT systems tend to encompass many legacy platforms - which have grown organically over the years, or inorganically through acquisitions. Today, CSPs tend to be employ a mixture of physical network functions and virtual network functions, which coexist and need to be managed seamlessly.
The Catalyst aims to build a complete cloud native stack that will address key challenges in the network assurance and service management layer and enable ‘zero-X’ operations use cases.
To implement new innovations, while safeguarding existing investments in network infrastructure and operations support systems, Catalyst envisions the creation of a ‘lights out’ or ‘dark’ NOC (network operations center), where operations located outside the NOC will define policies and automation workflows.
Objective:
The project aims to align with industry-standard architecture, such as TM Forum ODA (open digital architecture) and TM Forum Open APIs, while harnessing the power of AI by employing large language models to reduce costs and scale the solution across other fixed and mobile domains. Once complete, the Catalyst should help CSPs to modernize their service management and build a foundation for future cloud native networks.
This Catalyst project can help CSPs to create lean NOCs for NetCo, improved operational efficiency and reduce Opex, leveraging the benefits of GenAI. GenAI acts as an accelerator for improving data inventory quality, detection of root cause and improving engineering activities for remediation.
Over 300-500K manpower hours can be saved/year for higher skill tasks in a typical 60 million subs network, as an example, while operational tasks are carried out through the Dark NOC framework offered by the catalyst.
The project uses different vendors to collect PM and FM insights from 5G Mobile packet core, SDIN Voice core network and the 5G RAN network. For this, Microsoft Azure Operator Insights, Infosys Live Operations, and MYCOM OSI EAA & GenAie are used. ServiceNow TSM, TSOM and OMT, Red Hat’s Ansible Automation Platform and Lightspeed, IBM - Watsonx Code Assistant offer the service level orchestration. Iquall Networks MAT offers the network discovery feed to complete the required ecosystem.
The platform will be able to address the challenges of acquiring data from multiple network domains and technologies, fragmented view of assets, data quality issues and siloed operations. The NetCo would be enabled to realize the intra and cross domain fault & performance management and deliver self-healing capabilities using cutting edge predictive and generative AI technologies, within a TMF ODA compliant architecture framework: DarkNOC
Solution:
Imagine arriving at your Network Operations Center everyday, and not having to worry about the performance and fault information to be processed, because the DarkNOC has been able to automatically track all network domains, push the network problems for prioritized processing, after checking its service impact, and fed the corrective information back to the network and reduced the impact of the problems. All automatically and with the use of AI, without the intervention of NOC engineers.
This is what this innovative catalyst has aimed for. Using network analytics, predictive and generative AI, we have designed a highly re-usable and automated NOC framework for CSPs that continuously adapts to a customer’s needs and offers unique recommendations, and solves network problems, zero-touch. The Dark NOC use case is aimed at saving annual manpower at the NOC, increasing efficiency, reducing error and dollar savings. By leveraging the latest in predictive AI and GenAI, and network insights improve NOC efficiency manifold and deliver proactive customer outreach, preventing customer churn.
Most large networks have complex NetCo landscape, comprising physical and virtual network functions across multiple domains, with poor intra and cross domain fault & performance management and low levels of self-healing capabilities with minimal or missing AI technologies. With the advent of AI, the catalyst is timely in its utilization of GenAI, to bring the required revolution in operating the NOC, for a Zero-X experience and high efficiency.
The automated DarkNOC approach breaks away from the traditional OSS with high Opex, brings multiple OSS stakeholders together using the ‘TMF API glue’ seamlessly, and automates the feedback mechanism using AI, and further enhances it using GenAI. Right from discovery of network assets for improved data quality, to anomaly detection, and cross domain (Core, Transport, RAN) data correlation completing the loop with auto-remediation, and GenAI verifications.
This catalyst helps CSPs fundamentally rethink their NOC model, automation of corrective actions and proposes a radical shift through an AI and GenAI driven approach to NOC automation, inspired by the TMF Dark NOC framework. A framework much needed for CSPs to realizing the long pending intelligent automation of their NOCs, and releasing their highly skilled engineers for high level skills such as policy definition and AI modelling.
Welcome to the GenAI empowered NetCo Dark NOC!
Come and see for yourself in Booth C16
Resources
1. Reference Architecture
DarkNOC : Reference Architecture
2. Business Benefits
DarkNOC : Business Benefits for CSP
3. Arena Presentation
DarkNoc : Arena Presentation
4. Catalyst Demo
DarkNOC : Demo of key use case
Contact team
Email the members of the Catalyst team to request more details.