All News & Insights

5G and Smart Manufacturing

As companies advance along their journey to smart manufacturing, 5G communication technology promises to overcome the current networking constraints hindering the creation of more effective real-time, high-volume, and secure data-driven applications.  By Dr. Jay Lee, Moslem Azamfar, and Marcella Miller

 

 

Remarkable technological advancements in wireless communication since the late 1900s have changed the way people and the environment interact with each other. As a result, global mass communication, smart cities, the Internet of Things (IoT), and enormous unforeseen applications have been envisioned. The impact of communication technology is pervasive and permeates into the core of our daily lives by connecting smart devices over the internet which in turn is transforming how we live and work.

It is expected that the number of globally connected devices will increase to more than 20.4 billion by 2020 which emphasizes the need for better coverage, high-efficient connectivity, and reliable networking. Smart devices operating with long-lasting batteries and equipped with wireless communication modules provide enormous data for a variety of intelligent-data-driven applications such as automation and control, job scheduling and production, predictive maintenance and near-zero downtime operation.

5G networking now promises to provide manufacturing companies with the communications capabilities to help realize this kind of data driven smart manufacturing. Although these applications serve different purposes, they all have one common characteristic: dependency on reliable and strong connectivity.

Evolution of Disruptive Communications 

The evolution journey of wireless communication technologies embraces decades of research and advancement in infrastructure design, IT engineering, and business intelligence. Figure 1 shows a summary of the evolution of seven disruptive wireless technologies, from the early days of the 1980’s and into the future.

For manufacturers, the problem is that current 4G wireless technology is not able to support reliable, high speed, and high coverage communication which is the core enabling technology requirement for smart manufacturing. On the other hand, fifth generation (5G) wireless technology, provides key advancements which are expected to address these concerns and realize the seamless connectivity and mobility of wireless networks.

End-to-end latency in a 5G network, for example, may reduce to less than 1ms and would support massive coverage with enhanced bandwidth and ultra-reliable connectivity. 5G promises a maximum speed of 20 Gbps compared to the maximum theoretical speed of 1 Gbps for 4G.

Looking further ahead, seventh-generation wireless technology (7G) is expected to integrate these 6G wireless networks with the satellite functionality to provide global network coverage with ultra-high-speed internet access. 7G is likely to appear sometime around the year 2040 and is expected to address the potential roaming issues of 6G networks. It will also introduce the concept of satellite connectivity for global mobile communication and include certain protocols and standards for mobile to satellite and satellite to satellite communications.

Today’s Industrial 5G 

The adaptation of 5G for industrial applications, known as Industrial 5G, can help support advanced Industry 4.0 strategies by bringing ubiquitous, high speed, reliable, high coverage connectivity to industrial environments and systems.

Industrial 5G has abundant untapped potential for new and innovative use cases in different industries. In Nov. 2019, Qualcomm Technologies, Inc., announced the findings of a study1 that predicted 5G will generate $13.2 trillion in sales enablement by 2035, by which time 5G global value will also support some 22.3 million jobs.

As one of the main targets of Industrial 5G, smart manufacturing would significantly benefit from this by supporting advancements in the adoption of cyber physical systems, digital twins, edge computing, and the implementation of industrial Artificial Intelligence (AI) in different operational or production systems.

“5G can help support advanced Industry 4.0 strategies by bringing ubiquitous, high speed, reliable, high coverage connectivity to industrial environments and systems.”

 

Among different industrial 5G leaders, for example, Nokia has utilized the low-latency and high-bandwidth capability of 5G networks alongside machine learning to ameliorate production issues in manufacturing environments and in particular, to monitor and analyze assembly line processes. Ericsson, meanwhile, in collaboration with the Fraunhofer Institute for Production Technology, is using 5G technology to improve jet engine manufacturing processes where high speed data collection is used for real-time monitoring in different operational conditions to update digital twin operational systems.

While significant benefits can be realized by utilizing 5G technology for existing applications, the best opportunities for 5G lie in capturing value from new use cases such as Industrial Internet of Things applications (IIoT), and real-time AI analytics. Once more companies begin to acknowledge the necessity of 5G and look for new potential applications, a more widespread rollout of 5G will become inevitable as manufacturers work to upgrade their systems to catch up to the emerging 5G leaders.

Current Constraints  

Current manufacturing systems are experiencing significant shifts in production strategies, customers’ requirements, and IT and infrastructure needs which demand more advanced tools capable of supporting the increase in automation and transparency across the manufacturing shop floor. Today, multiple devices are connected to the internet and provide information from different aspects of production to support a variety of applications, such as real-time control, predictive maintenance, and production scheduling. That requires key characteristics in the development of new communication technologies to support both the current and future requirements of smart manufacturing systems. Current communication technologies, however, are not capable of satisfying the advanced requirements of these future manufacturing systems.

  • High data rate: Developments in IoT devices and edge and fog computing, along with the advent of cloud computing, virtual reality systems, and digital twins, have led to a huge amount of data which require uninterrupted high data rates of more than 25 Mbps and cannot be satisfied with the current communication technologies.
  • High coverage – due to the complexity in manufacturing production environments, such as the heavy use of metals and electromagnetic interference, ubiquitous connectivity cannot be satisfied.
  • Low latency: Rapidly developing manufacturing applications, such as real-time control, device to device communication, and autonomous shop floor operations, usually require low communication latency.
  • High reliability: In critical applications, such as real-time control and automation, it is necessary to have a reliable data transmission with the packet loss less than 1×10-12.
  • High security: Data leakage may lead to malicious attacks on manufacturing processes and equipment with the aim of paralyzing the production system and illegally extracting critical corporate information.
  • Scalability of the network: Wired communication is traditionally used in manufacturing environments due to its better performance compared to wireless technologies. However, as the number of connected devises increases, the complexity of scaling up the network becomes more and more challenging. In addition, updating the wired network is more complicated and costly compared to wireless networks.

“Current communication technologies are not capable of satisfying the advanced requirements of future manufacturing systems.”

5G & Smart Manufacturing 

5G on the other hand, utilizes advanced technologies such as Millimeter Wave and terahertz band, Network Function Virtualization (NFV), Wireless Software Defined Network (WSDN), Cloud Radio Access Network (CRAN), and Massive MIMO to provide low latency, high reliability, high transmission rate, high coverage, high security, and scalable networking which can better support the communication demands of future smart manufacturing (See Fig. 2: 5G Enablers & Impacts).

5G-Enabled Cyber-Physical Systems 

Real-time data flow from manufacturing assets with high reliability, speed, and data rates is an essential element in the realization of integrated cyber-physical systems to leverage intelligent applications such as autonomous Prognostic and Health Management (PHM), real-time monitoring and control, digital twin implementation, and many more. Although wireless technologies may have advantages compared to wired communication technologies, their application in manufacturing has been relatively limited due to low coverage, low on-site penetration, and an inability to meet the exponentially rising data demands, speed, and spectral efficiency requirements in fast-growing industrial networks. 5G, on the other hand, provides inherent advantages that appear as a real game-changer.

“High data rates could significantly impact the development and implementation of AI tools that involve real-time signal, image, and video processing.”

Typically, the established 5-level Cyber Physical Systems architecture (5C-CPS) is utilized to successfully bridge the gap between manufacturing physical systems and cyber computational space. This architecture introduces a five-layer systematic approach consisting of: 1/ a connection layer for smart sensing and enhancing self-awareness; 2/ a conversion layer for utilizing artificial intelligence to improve transparency; 3/ a cyber layer for implementation of digital twin, big data analytics, and data storage; 4/ a cognition layer for assisting the decision-makers; and 5/ a configuration layer for implementing data-driven insight to the physical systems in order to realize self-configure and self-optimization. Depending on the application, the use cases of 5G in each layer of 5C-CPS can be summarized into three main categories – URLLC (Ultra-Reliable and Low Latency Communication), mMTC (massive Machine Type Communications), and eMBB (enhanced Mobile Broadband). (See Fig. 3: 5G Application Scenarios)

Enhanced Mobile Broadband  

In the eMBB scenario, applications that require high data rates, such as augmented reality, real-time asset monitoring, and intensive collaborative tasks such as 3D design and digital twins, could significantly benefit.

The enhanced broadband features of 5G can help revolutionize the way industries train high skilled workers for sensitive applications. An operator can be trained to work with the cyber twin of a machine in a virtual environment prior to working with the real machine and hence the operational safety and performance would be significantly improved.

For instance, oil and gas company Total is pioneering the use of Siemens’ virtual reality software in training staff for offshore installation in a virtual 3D environment. In a similar way, Ford uses virtual reality in optimizing the automotive assembly process and has successfully reduced the accidental injury rate during production by 70%.

It is also expected that eMBB will help in the generation of new services and business models that could potentially create significant value. For instance, digital twin models and data could be shared as services (digital twin as a service) across factories and between different partners. This could generate new streams of revenue for the service providers and facilitate customer access to digital models.

Access to a high data rate is also one of the biggest concerns in cloud-based applications and significantly impacts customers’ experience and network operational performance. Using eMBB, could help unleash the full potential of cloud manufacturing where data, intelligence, infrastructure, and software are offered as on-demand services to the end-users. For instance, monitoring vibration signals on a cutting tool on a shop floor may provide significant information that companies and other developers could use to help design better analytics tools, minimize unplanned downtime, or enhance machine and component design.

High data rates could also significantly impact the development and implementation of AI tools that involve real-time signal, image and video processing. New infographic tools based on interactive image and video processing could also be created to support a better representation of the information to aid decision-makers.

Ultra-Reliable and Low Latency Communication  

In the URLLC scenario, the emphasis is on latency and reliability in data transmission. The objective is to transfer a small payload of data with ultra-low latency and high reliability. Possible applications include autonomous shop floor operations, remote control, and mission-critical applications. With everything connected, near real-time communication, and the use of Artificial Intelligence (AI) techniques, almost every object can become a smart, data-driven agent enabling more reliable and autonomous functionality. For instance, a CNC machine that is equipped with several smart sensors that interact with each other in a near real-time fashion can help monitor the health condition of the spindle, enhance the machine’s operation, extend the system’s effective lifetime, and finally create an efficient and worry-free production environment. In particular, the integration of 5G with mobile edge computing technology could facilitate immense data collection across all objects which further accelerates automation and the evolution of AI tools. With on-demand intelligence access and the use of ultra-high-speed and reliable communication, new applications such as autonomous Prognostic and Health Management (PHM) and real-time mass optimization can be developed.

Latency is one of the dominant factors contributing to low performance and financial losses in the control and automation industry. In general, latency occurs in both the communication and computation stages of systems’ operation. Enhancing microprocessor capacity, using efficient computational algorithms, and the distribution of computational tasks between multiple resources can significantly reduce computational latency. The implementation of URLLC could drastically reduce the communication latency between two devices. Edge and fog computing can play a more effective role when integrated with 5G technology to bring seamless computation and networking power to the system.

Massive Machine Type Communications 

The mMTC scenario focuses on providing wireless connectivity to a massive number of devices that are often sporadically active, have limited power capacity, and send small data payloads. This approach can improve interconnectivity, scalability, coverage, and indoor penetration and therefore provide ubiquitous connectivity for IoT devices, facilitating data collection across all devices on the manufacturing shop floor. In order to create an interconnected network, 5G utilizes and facilitates Device-to-Device (D2D) communication which enhances link reliability, spectral efficiency, and fulfills data transmission and network capacity requirements.

As smart manufacturing functionalities require enhanced connectivity to achieve their full potential, the development of 5G comes at just the right time.”

In a 5G network, D2D communication can be established between two devices directly without connection with the base station or the core network. An mMTC approach can help focus on enhanced coverage with extended battery life for IoT devices of the order of decade, and lower network complexity and coordination. These features help resource-restricted IoT devices such as sensors, actuators, and PLC modules to be accessible in data-driven applications. The adopted architecture can significantly help in local traffic offloading and backhaul control. Data gathered across systems could then be used for production optimization, predictive maintenance, and distributed decision making. In a full implementation of mMTC, lots of the resource restricted IoT devices that are currently used for simple local operational tasks could be empowered to play a role in more global decision making, which could help prevent catastrophic incidents and increase manufacturing productivity and operational efficiency.

Currently, mobile IoT technologies such as Narrow Band IoT (NB-IoT) and Long-Term Evolution machine-type communications (LTE-M) provide large scale, low cost, and secure connectivity for heterogeneous IoT devices. The integration of these technologies with 5G would facilitate low-power connectivity and high coverage for reliable data transmission between infrequently available IoT devices.

5G as a Catalyst for Smart Manufacturing 

Recent development trends in cyber physical production systems and the adaptation of advanced technologies such as the industrial internet, deep learning, digital twins, blockchain, and others have increased industrial demands for faster, more reliable, and higher throughput network connectivity. As smart manufacturing functionalities require this enhanced connectivity to achieve their full potential, the development of 5G comes at just the right time. Full scale implementation of 5G technology will address the reliability, coverage, and latency challenges associated with previous network generations and will facilitate the envisioning of new application areas, services, and use cases never dreamed of before.

A 5G-enabled cyber physical system will create a loop between manufacturing physical systems and cyber physical computational resources. In this loop, massive amounts of data across different applications are constantly pushed to cyberspace for conversion into valuable information, dynamic simulation, and predictions for future actions. Insights from cyberspace will feed-back to the physical systems for self-configuration and real-time execution. Company-wide implementation of 5G technology can significantly increase data availability and promote smart applications such as online monitoring and health management. Due to 5G’s massive data access and high-speed data transmission capabilities, distributed computing can be well integrated within different applications that are running on resource-restricted equipment such as sensors and actuators. Throughout this process, different elements of the network are empowered to make advanced decisions regardless of their built-in computational resources.

However, although the 5G network has the potential to revolutionize smart manufacturing and to enable new possibilities in terms of automation, productivity, and flexibility, practical uses of the technology are not yet well explored and there are limitations that must be considered by early adopters of the technology. A 5G network will reach its full sustainable potential once a 5G device is used on the both sides of a connection, the suitable use cases of the technology are correctly identified, the cost-benefit analysis supports the heavy initial investment, and the 5G network is well integrated with other components of the system. Companies should consider short- and long-term investment strategies with measurable risk success factors to guarantee a safe and successful transition to the 5G network. M

Footnote:
1 https://www.qualcomm.com/invention/5g/economy

View More