Access Official Technical PDF

Share Summary:LinkedInEmail

5G in smart manufacturing is not simply a wireless replacement for factory cabling. It can be part of a broader shift toward connected production systems, where robots, sensors, mobile equipment and software services exchange information across the factory. The potential applications range from mobile robot control and machine vision to worker interfaces, process monitoring and links between industrial networks.

The 5G-SMART project examined 13 forward-looking manufacturing use cases. Seven were selected for trials at three European sites, using 3GPP Release 15 technology as the project baseline. The work offers a useful way for automation engineers and plant decision-makers to assess where 5G may fit—and what must be considered before relying on it in production.

Why manufacturers are considering 5G

Manufacturing is moving toward more flexible production and more autonomous operations. Product customization and changing production needs place pressure on factories to adapt equipment and workflows without sacrificing process quality or safety. At the same time, real-time monitoring can help production systems make better use of operational information.

This transformation brings information technology (IT) and operational technology (OT) closer together. Some computing functions that traditionally ran on dedicated devices or controllers may be placed in an edge or cloud environment. A wireless network can support this architecture by connecting equipment that moves, adding sensors where cabling is difficult, and making it easier to reconfigure parts of a factory.

That does not mean every wired industrial network should be replaced. Existing networks represent significant investments, and future architectures may need to integrate legacy systems and newer technologies such as Time-Sensitive Networking (TSN). For many plants, the practical question is how wireless connectivity can complement the existing automation infrastructure.

5G in smart manufacturing: seven trial use cases

The 5G-SMART trials covered different automation settings across Ericsson’s smart factory in Kista, the Fraunhofer IPT shop floor in Aachen, and the Bosch semiconductor factory in Reutlingen. Together, the selected use cases examined factory automation, robotics, production IT and monitoring.

  • Robot collaboration and human-robot interaction: At the Kista trial, stationary robots and a mobile robot were connected to a 5G test network. The scenario involved transferring an object between robots while a worker moved through the area. A vision system supplied information about object and robot positions, supporting motion-planning functions.
  • Factory-floor visualization: Another Kista use case considered presenting information from stationary and mobile robots through an augmented-reality headset. Information collection and pre-processing could be handled at an edge platform, while a worker viewed relevant factory-floor data.
  • Wireless acoustic workpiece monitoring: The Aachen trial addressed monitoring workpieces using acoustic information, a case where wireless sensing can be relevant to production processes.
  • Multi-sensor platform for a digital twin: Also in Aachen, a versatile sensor platform was considered in connection with digital-twin applications.
  • Cloud-based mobile robotics: The Reutlingen trial examined mobile robotics with computing functions supported by a cloud-based approach.
  • TSN and industrial LAN over 5G: This use case focused on connecting industrial networking concepts with 5G, reflecting the need to consider wireless and wired infrastructure together.

The trials were designed to investigate 5G capabilities in practical manufacturing scenarios, not to establish that one network approach suits every application. The sites brought different perspectives, from robotics testing to network integration and evaluation in an operating factory environment.

Matching communication to the application

Industrial applications place different demands on a network. Some need to move substantial volumes of sensor or video data; others depend on timely, dependable delivery of relatively small control messages. The project considered requirements in terms of measures such as latency, availability, data rate, device mobility and service area.

In the robot collaboration scenario, motion-planning messages had application-level timing requirements. The report specified a target communication service availability of at least 99.99% for these streams, with end-to-end latency below the message transfer interval. The stated transfer intervals ranged from 5 to 40 milliseconds for stationary robot motion planning and 10 to 50 milliseconds for the mobile robot. These are requirements for the described use case, not universal targets for all industrial 5G deployments.

The distinction between network performance and application performance matters. A robot-control loop includes communication as well as processing in the robot and edge environment. The system must be designed and assessed end to end, with application timing, network behavior and equipment responsibilities considered together.

Edge computing, safety and deployment challenges

Placing selected functions in an edge cloud can support local processing close to factory equipment. In the Kista scenarios, the edge environment was intended to host functions such as task planning, robot motion planning and localization-related processing. This approach could reduce the amount of specialized computing hardware required on robot units, while shared infrastructure such as cameras could serve multiple applications.

However, moving functions off equipment does not remove the need for local safeguards. The described robot use cases retained functions such as safety management and status reporting on the robot. The report also identified challenges in supporting several low-latency, deterministic communication streams on shared infrastructure and in allocating edge resources to meet end-to-end needs.

Operational considerations include network configuration, fault and alarm management, mobility support, security, and the ability to monitor service performance. Integrators also need to consider how application settings and network resources interact. A design that meets one stream’s requirements may create trade-offs when multiple applications share the same system.

Beyond the initial trials

Six additional use cases explored future directions beyond the seven trial scenarios: remote expert support; cross-domain and inter-company collaboration; automated guided vehicles with AI-assisted trajectory adaptation; metrology and maintenance across machine and factory boundaries; seamless device plug and play; and AI-assisted production quality management. These topics illustrate the range of manufacturing workflows that could benefit from improved connectivity, while also pointing to requirements that may need further development in 5G standards and systems.

Practical implications for industrial teams

For plant managers and engineering teams, the value of 5G depends on identifying a specific operational problem and translating it into measurable communication and application requirements. Mobility, difficult-to-cable sensor locations, robot coordination and the integration of distributed computing are relevant areas to evaluate. So are network availability, latency, security, existing infrastructure and the consequences of a connectivity interruption.

The 5G-SMART use cases show that industrial 5G is best assessed as part of a complete automation architecture—not as a standalone radio upgrade. Trials can help teams test how network performance, edge computing, application logic and safety arrangements work together. That disciplined approach can clarify where wireless connectivity complements wired control systems and where further technical validation is needed before deployment.

Access Official Technical PDF

About The Author