What financial benefits can you unlock by moving to fully autonomous networks? A recent report by STL Partners concludes that an average-sized communications service provider (CSP) can expect cost savings and revenue uplift of around US$794 million per year, or about 5% of annual revenues.

The evolution towards autonomous networks will take time: learn how automation and AI applications such as GenAI, AIOps, etc. can accelerate your journey by enabling service intent resolution, feasibility checks, dynamic service orchestration and others.

Key telecom challenges

The telecom industry faces many issues, ranging from managing cost pressures, the growing complexity of networks and delivering a differentiated service experience for their customers. As a CSP, some of the challenges you need to deal with today include:

• Network complexity: 5G, IoT and edge computing have significantly increased network complexity, making management more demanding.

• Operational costs: Greater complexity drives up costs, requiring significant investment in resources to maintain both legacy and new technologies.

• Traffic growth: The surge in data from streaming, IoT and cloud services is overwhelming networks, making performance maintenance a constant challenge.

How can automation help to address these challenges?

How automation is crucial in telecom networks

Automation is essential to address business, operational and technical challenges in many dimensions such as:

• Accelerates issue resolution, enhancing reliability and reducing downtime.

• Reduces manual intervention, lowering workforce costs and time spent on resolving issues.

• Allows networks to scale without proportional increasing complexity or resources.

• Enables real-time network adjustments, optimizing bandwidth, energy use and load balancing based on traffic.

• Supports monetization efforts with better service intent resolution as shown in the diagram below

However, the introduction of automation can create new complexities. How do you properly introduce new workflows, business process model and notation (BPMNs), scripts, systems and interfaces? The solution lies not in replacing current systems, but in enhancing and evolving them to reach the next level. This gradual advancement unlocks hidden resources and amplifies performance resulting in significant improvements while leveraging the existing technology stack. This is something we are doing at Nokia across our portfolio, to power fully autonomous networks.

How GenAI improves automation and orchestration

In a previous blog, we discussed how GenAI powers the operation of autonomous networks and touched at a high level on its role in intent-based orchestration. GenAI takes telecom automation to a whole new level by introducing intelligent decision-making and adaptability into network and service orchestration systems, being fully integral and native to automation process. Let’s explore how it enhances various aspects of network automation:

User interaction with orchestration systems

Traditionally, using orchestration systems required extensive technical expertise. GenAI transforms this by integrating natural language processing, enabling engineers to interact through intuitive, conversational commands as shown in the diagram below.

With GenAI-driven interfaces, users can now execute complex workflows and easily request detailed metrics. This broadens accessibility and reduces the need for specialized knowledge.

This will lower time-to-market for new services by an estimated 50-70 percent. Creation of the initial workflows will be done in minutes instead of hours. And the time to design, test and deploy new workflows or use cases is expected to be reduced by 50 percent.

Enhancing workflow engines

Telecom networks rely on workflow engines to manage interconnected processes like provisioning, configuration and fault management. GenAI can advance these engines by learning from historical patterns, dynamically optimizing processes and anticipating future needs. It's essential in intent-driven orchestration to align the current state with the desired target state and find the most efficient path from A to B, using both historical data and trend analysis.

GenAI modules serve as continuous enhancements to the automation system’s "brain"—the workflow engine—by calculating optimal execution paths and assembling workflows from multiple sub-tasks before passing them to the workflow engine for execution. When combined with inventory knowledge graphs, this creates a cohesive link between network states and simulations, enabling faster execution with minimal manual intervention.

Expected benefits include faster order processing, improved intent negotiation and optimized resource allocation, with workflow engine and order management performance gains of up to 40 percent.

Automating troubleshooting

Detection and resolution of network issues are some of the most time-consuming operational tasks. While design-time automation and dynamically captured intent drive initial processes within order management systems, handling a wide range of exceptions remains a challenge. GenAI systems can analyze extensive data from network logs, swiftly identifying root causes and even predicting potential failures. Automated troubleshooting shortens mean time-to-resolve (MTTR), prevents escalation and enhances network resilience. By translating machine logs into readable formats, pinpointing root causes for order or task failures and suggesting the next best actions with linked troubleshooting guides, AI significantly aids operational teams.

MTTR and mean time to detect (MTTD) reductions of up to 85% are expected.

GenAI as an embedded part of autonomous networks applications

The future of telecom lies in the deep integration of GenAI into autonomous networks applications. This means AI is no longer just an add-on but an integral part of how networks operate. GenAI is enhancing how telecom networks are orchestrated, enabling better user interactions, streamlining workflows and automating complex troubleshooting processes.

Moreover, GenAI models are being trained continuously with new data, allowing them to improve over time. This self-learning capability ensures that GenAI-powered systems remain cutting-edge, adapting to new challenges and opportunities as they arise.

As an integrated and embedded part of telecom applications, GenAI is not just transforming current operations but also paving the way for the future of autonomous networks, enabling you to take advantage of the significant financial benefits along the way.

To learn more about how Nokia’s Digital Operations software can help you to accelerate your journey towards autonomous networks please visit our website and request a private demonstration.