In any type of manufacturing, the endless push for greater efficiencies and productivity often feels like an exercise in incrementalism. Any significant payoff from continuous improvement initiatives seems to come at a glacial pace.
But modern standards in wireless technologies, especially newer standards like Wi-Fi 6 and 5G wireless networks, are poised to fuel dramatic reductions in data latency along with tangible increases in reliability, and the trade-offs required for their higher performance are, for once, very palatable indeed.
Source: metamorworks/Adobe Stock
What is low latency manufacturing?
Low latency manufacturing relies heavily on real-time communications between machines, sensors (including wearables) and control systems that enable instant decision-making, rapid adaptation to changing priorities and procedures, and timely intervention with real-time troubleshooting assistance when challenges arise — all in service of operational continuity.
The network of connected workers and machines is the foundation of low-latency manufacturing — in the parlance of Industry 4.0, it’s known as the industrial internet of things (IIoT) — where ultra-reliable low-latency communication (URLLC) is at a premium.
Any tool, worker, process or machine can be a node in the network, a component connected through a mesh of wireless networks and routers, receiving and sending data in real time, in service of a limitless array of use cases.
With increasing reliance on artificial intelligence (AI) and machine learning on the backend and more workers on the front lines wearing sophisticated tech like augmented reality (AR) headsets, optimizing network performance is not optional. That means maximizing bandwidth and speed, as well as exploiting strategies that drive latency ever lower. Wi-Fi 7 promises even greater performance, but just adopting current standards can be an immediate game changer.
Latency versus speed
While terms like latency, bandwidth and speed are routinely conflated, critical distinctions are worth noting.
Speed is a rate, of course, and in networking it measures how much data the network can transmit every second, expressed as the number of bytes that can be sent or received per unit time. The typical units are MB or GB per second.
Speed is an important performance indicator for sure, given the increasing volume of real-time data flowing between the cloud and the edge. But there’s more to performance than speed in real-time applications.
Latency is a measure of time, usually milliseconds (ms). Unlike speed or bandwidth, it’s a duration, not a rate. Latency and speed each influence a user’s perception or the efficiency of processes in unique ways.
Consider wireless communication with an AR headset overlaying imagery that is visually aligned and dynamically updated with the current state of a machine on the factory floor. The purpose may be to guide a worker through dynamic assembly or troubleshooting procedures in real time.
In that kind of scenario, a deficiency in the network’s capacity for speed, aka "bandwidth," could mean frequent interruptions of the incoming video, a reduced frame rate, degradation of the visual quality or all of the above — think movie-streaming interrupted by an intermittent "loading" widget, or the image abruptly switching to a lower resolution while the incoming data "catches up." Those built-in compensations may be annoying in an action flick, but are a deal-breaker in mission-critical, AR-assisted tasks. That’s why lower-bandwidth standards like Bluetooth are insufficient for any media type more demanding than audio.
Assuming available bandwidth is sufficient, as it must be, it’s low latency that makes the crucial difference in the stability of the control loop — in regard to both successful outcomes and the quality of the user experience. Higher latency makes systems feel sluggish and unresponsive to human users when there’s a tangible delay between an input to a system and its reaction. Whether you’re a human or a machine, that delay can introduce insurmountable control challenges in any process. Is a worker making control inputs because they’re reacting to the state of the system 3 seconds ago, or 3 milliseconds ago? A 3 second delay in response to control input doesn’t work in most real-time manufacturing interactions any more than it would when driving a car.
Minimizing the time between input and response is clearly a critical design imperative, on the road and on the factory floor.
Importance of low latency in manufacturing
When sports car enthusiasts talk excitedly about “responsive” handling, they’re essentially saying that the driver’s steering inputs are met with an instant and proportional change of direction — the automotive version of low latency.
Minimizing latency is a key design goal in any control-feedback loop, even in purely mechanical systems, and certainly it’s most critical when human input is part of that loop.
In the context of Manufacturing 4.0, the IIoT is nothing if not an intricate mesh of control-feedback loops, so the importance of low latency in IIoT-based manufacturing can’t be overstated. It is the cornerstone upon which the responsiveness and reliability of real-time systems are built. Lower latencies only further extend core capabilities:
● Real-time monitoring and control: Immediate feedback and control over manufacturing processes reduce downtime and optimize operations.
● Predictive maintenance: Real-time data analysis allows for predictive maintenance, reducing unexpected equipment failures and extending the life of machinery.
● Enhanced safety and security: Immediate responses to safety and security issues can be addressed, protecting workers and assets.
● Flexible manufacturing: Quick adaptation to changes in production requirements or customization requests becomes feasible, enhancing competitiveness.
In turn, the implementation of low latency wireless technologies in manufacturing environments unlocks new possibilities.
Smart factories
Factories become smarter, more connected and more efficient, with machines and systems that can communicate, learn and make decisions independently. Low latency tightens the turnaround between control-sensor feedback, whether it’s humans doing the controlling or AI, or both. Low latency makes possible instantaneous delivery of data to and from the factory floor — response to interruptions in nearly any process can be met in near real time with contextually aware AI recommendations, for example.
Augmented reality (AR) and virtual reality (VR)
These technologies require both low latency and high bandwidth to provide immersive training, maintenance and remote assistance applications.
Automation and robotics
Enhanced wireless communication enables more sophisticated automation and robotics applications, with more compute available closer to real-time, enabling machines to perform more and more complex tasks with less and less human intervention.
Supply chain optimization
Real-time tracking and communication across the supply chain improve efficiency, reduce costs and enhance the ability to respond to changes in processes, fluctuations in resources and shifts market demands.
Wireless technologies in industrial environments
Wireless technologies play a pivotal role in this transformation, offering the flexibility, scalability and speed needed for highly competitive modern manufacturing.
The key wireless technologies at play are Wi-Fi 5/6, 5G, and yes, even Bluetooth. All can play a key role in lowering latency and bringing measurable productivity gains to the workshop or factory floor, where real-time communication is key.
Wi-Fi 5 and Wi-Fi 6
Wi-Fi technology has been a communication workhorse in industrial environments, with Wi-Fi 5 and the newer Wi-Fi 6 bringing significant improvements. Wi-Fi 5, with its support for the 5 GHz frequency band, offered increased data rates and reduced interference compared to its predecessors.
Wi-Fi 6 has extended these benefits, providing higher data throughput, improved energy efficiency and better performance in environments with many connected devices. Its OFDMA (Orthogonal Frequency Division Multiple Access) technology allows multiple devices to be served simultaneously, reducing latency and increasing capacity. This is crucial for environments with high signal density, where numerous sensors and machines need to communicate in real time.
5G
While Wi-Fi makes sense in scenarios where machines and workers are contained on the scale of a single facility, 5G technology is a game-changer for more remote or geographically distributed operations that require high bandwidth and ultra-low latency across longer distances.
Unlike its predecessors, 5G offers URLLC, massive machine type communications (mMTC) and enhanced mobile broadband (eMBB). These capabilities make it ideal for supporting critical industrial applications that require real-time decision-making, such as autonomous robotics, remote control operations and predictive maintenance. Like Wi-Fi 6, the ability of 5G to support a high density of connected devices with minimal delays (as low as 1 millisecond) and its reliability underpin the digital transformation in manufacturing.
Bluetooth and advanced variants
Bluetooth, particularly its advanced variants like Bluetooth 5, has seen increased adoption in industrial settings for applications requiring short-range communication.
Bluetooth, with its low power consumption (and the resulting long battery life) is ideal for asset tracking or connecting sensors, wearables and small devices within a manufacturing environment. Bluetooth may be lower bandwidth than other options, but it’s perfect for transmitting sensor data and audio. And while it’s also a short range technology, its mesh networking capabilities allow for the creation of large-scale device networks, facilitating complex automation and control systems that can adapt and respond in real time.
Other wireless technologies
While low latency is paramount for many critical tasks, a wholesale upgrade to new standards across every segment of your IIoT is probably not in the cards, and in many cases, not required.
For one thing, some older short-range wireless standards remain perfectly adequate for their current use case. Beyond Wi-Fi, 5G and Bluetooth, several other wireless technologies are perfectly well-suited to a role in low latency manufacturing environments. For instance, Zigbee and Z-Wave are used for creating low-power, secure mesh networks for sensor data collection and actuator control.
Low power wide area network (LPWAN) technologies like LoRaWAN are employed in IIoT applications that require long-range communication and low power consumption, although they typically offer higher latency, not every connection demands speeds near real-time. Even so-called big data collection that happens outside of direct control-feedback loops can be offloaded from URLLC networks, to fast-enough legacy networks.
Making low latency lower
Control-feedback bottlenecks can arise from any component in the chain of communication. It’s not all down to network latency. Other components are a factor too. Additional considerations for performance improvement range from shifting more compute closer to the edge, to optimizing processing times and minimizing the number of “hops” between distinct networks.
But the evolution of wireless technologies stands apart as a key driver in the transformation of manufacturing environments to the next level of efficient, flexible and responsive operations.
The bottom line
Wi-Fi 5/6, 5G, Bluetooth and other wireless solutions are at the forefront of enabling real-time communications in industrial settings, reducing latency to levels where machines, sensors and people can interact seamlessly and instantaneously. Development of the next Wi-Fi standard, Wi-Fi 7 (802.11be), is well underway, although availability of its next generation performance remains a few years away.
As these technologies continue to evolve and converge, the potential for innovation in manufacturing processes is boundless, promising a future where smart factories are not just a concept but a global standard.
About the author
David writes on a diverse range of science and technology topics. David previously directed enterprise software projects for government, healthcare and gaming. He graduated from Cal Poly University with a degree in aerospace engineering and is based in Los Angeles.
