How NTT DATA and Ericsson Scale Private 5G Edge AI

Real-time processing is moving to the centre of enterprise infrastructure strategy as organisations push decision-making closer to where data is generated.

NTT DATA is positioning itself within this shift by combining private 5G, edge AI and managed services, linking connectivity with compute and operational control.

Its latest partnership with Ericsson reflects this direction, focusing on deploying private 5G networks alongside embedded edge AI capabilities so enterprises can act on data at source rather than relying on centralised systems.

Shahid Ahmed, Global Head of Edge Services at NTT DATA, states: “As enterprises adopt AI at the edge, they need partners who can bring connectivity, intelligence and security together in a way that actually works in production.

“Together with Ericsson, we can deploy these solutions faster, operate them at scale and deliver outcomes. Private 5G gives enterprises the foundation they need to achieve real, measurable impact with edge AI and physical AI deployments.”

Private 5G as a processing layer

Within the partnership, private 5G acts as an enabling layer for real-time processing, supporting consistent data transfer between devices, sensors and edge compute systems. Delivered as a managed service, the model includes defined architecture and operational controls.

This allows enterprises to standardise deployments across locations while maintaining control over data flows and security. It also supports workloads requiring continuous data ingestion, including monitoring systems and automated operations.

Åsa Tamsons, Senior Vice President and Head of Business Area Enterprise Wireless Solutions at Ericsson, says: “Ericsson has been advancing enterprise connectivity for over a decade. This extends that capability to support edge AI and physical AI at scale across industries.

“By combining our global platforms with NTT DATA’s engineering and managed services, industry expertise and AI-driven operations, enterprises can move from experimentation to always-on, production-grade operations.”

Reliability underpins this model. In transport, energy and manufacturing sectors, uninterrupted data processing is essential where system responses directly affect physical processes.

Embedding AI into connectivity

A core part of NTT DATA’s strategy is embedding AI directly within network environments. Edge AI agents operate where data is generated, allowing systems to interpret and respond without routing information to central cloud environments.

This architecture supports use cases that require immediate action. In manufacturing, systems analyse sensor and vision data to detect faults.

In transport and logistics, vehicle and asset data informs routing and safety systems.

Meanwhile, energy and mining operations rely on continuous monitoring and automated responses.

Rather than separating data collection, transmission and processing, these functions are integrated into a unified environment. This reduces latency and simplifies system design.

“Private 5G is the backbone for scaling AI in production, where autonomous systems must operate reliably and at scale, but integration complexity often remains the final hurdle,” says Alejandro Cadenas, Associate Vice President of Worldwide Telco Research & Consulting, 5G, IoT and Mobility at IDC.

“The combined expertise of NTT DATA and Ericsson seamlessly integrates edge AI and physical AI with enhanced connectivity, overcoming operational, scalability and accountability challenges and accelerating the deployment of AI with confidence.”

Real-time processing in data centres

NTT DATA is also applying real-time processing within its own data centre operations. At its Rhine-Ruhr 1 facility in Bonn, Germany, the company has deployed an AI-driven system to manage cooling infrastructure.

Developed with software company etalytics, the platform uses a digital twin to simulate and optimise performance based on live operational data. Inputs such as ambient temperature, system load and flow rates are processed continuously, allowing the system to adjust control parameters in real time.

Processing takes place on-site within the data centre, ensuring decisions are made without dependency on external systems. This reduces latency and maintains operational continuity.

The deployment recorded a 19.1% reduction in energy consumption for cooling over the observed period. Safeguards are built in to prevent actions outside defined parameters, with operators able to revert to manual control if required.

Linking edge and core infrastructure

NTT DATA connects edge processing with core infrastructure through managed services. Data centres are part of the architecture, supporting workloads that require aggregation, storage or further analysis, while time-sensitive processes shift to the edge.

This balance allows enterprises to distribute workloads according to requirements. Integration between edge and core depends on consistent data models and orchestration, supported by managed services overseeing deployment and operation.

Scaling real-time systems

Scaling real-time systems can be a challenge as organisations move from initial deployments to broader operations. NTT DATA addresses this through repeatable solutions and standardised architectures within its private 5G and edge AI offering.

While environments vary for these solutions vary, the underlying architecture remains consistent, supporting faster deployment and reduced integration effort.

Managed services also provide a framework for maintaining performance across multiple sites, handling variations in infrastructure, data and regulatory requirements.

Operational control and safety

Real-time processing introduces requirements around operational control and safety, particularly where systems act on live data in physical environments.

NTT DATA incorporates safeguards within its deployments to ensure systems operate within defined parameters. In the Rhine-Ruhr 1 project, optimisation actions are restricted if conditions fall outside set ranges or if communication issues occur. Operators retain oversight and can intervene when necessary.

This balance between responsiveness and reliability ensures that real-time systems remain predictable while adapting to changing conditions.