Even as we stumble into the second half of 2022, you can find those who prattle on about the transformative role that 5G and IoT will play in driving us to the edge of a fourth industrial revolution. But why hasn’t 5G–powered IoT taken off yet?
In the digital 21 st century, these visionaries proclaim smart factories that run on IoT machines and are fueled by the 5G communications boom will deliver automated, connected, and intelligent workplaces imbued with lightning–fast data speeds and mega–low latency. Meanwhile in the real world, everything from baby formula to silicon chips are in short supply thanks to the Covid pandemic.
People at the sharp edge of the IoT world, such Particle COO Dan Jamieson, take a more realistic view of how 5G and IoT interoperate in the industrial realm. Speaking to EE Times in a phone interview this April, Jamieson tells us that 5G in factories will “become a reality over many decades” as companies “build new facilities.”
Tesla gigafactory in Austin, Texas (Source: Shutterstock)
Tesla, Jamieson notes, is the only company that is building gigafactories right now, and even they don’t use 5G yet. To bring in cellular–controlled automation to the shop floor, you need to buy a lot of technology that will work with the 5G signal. This means a lot of expensive machines and systems that need to be integrated to form a cohesive whole.
“You can’t go to Harley Davidson and say, ‘let’s rework your factory and turn it into an intelligent, smart factory based on 5G’, right?” Jamieson contends. “You have to go upstream and start with the OEMs that make the machines…[then] you have to go downstream and be able to layer a system on top of that.”
This is a costly and time–consuming process that most businesses would only deal with if they really had to. “These aren’t aftermarket solutions, they need to be built into the software, the connectivity, the control systems, and built into the machines themselves,” Jamieson said.
Jamieson’s main point is that to grasp the advantages of 5G, “you need to build a new factory.”
IoT today, Jamieson suggests, doesn’t really need 5G. “If what you’re trying to do is prove that I can connect this HVAC system to the internet and collect data from it so that I can better operate the thing, [5G] is not necessary and it’s not worth the risk.”
“The primary reason I see that it’s not taking off is the ecosystem is not ready,” Jamieson said. From modules to software to network deployments, the cellular standard has barely taken off its training wheels yet.
The carrier’s conundrum
IoT hasn’t emerged as the cash cow that cellular operators hoped for five to seven years ago. Selling IoT devices to enterprises is just not the same as flogging the latest iPhone to consumers.
“The carriers haven’t quite figured out how to deploy and sell IoT the way it’s been envisioned because they can’t figure out how to make money off of it,” Jamieson said.
For instance, major mobile network operators still use average revenue per unit (ARPU) as a key metric when reporting quarterly numbers to Wall Street. “The ARPU of an NB–IoT device is an order of magnitude or two orders of magnitude smaller than what their average is today, and so it becomes a problem,” he said.
Enterprises that want to deploy IoT devices run a pilot first, typically rolling out a few thousand units to ensure the system works. The company then scales the IoT system, bringing on tens or hundreds of thousands of devices.
“This can take years,” Jamieson said. “The impact on the P&L [sheet] and on ARPU, the Street’s not gonna like.”
One feature of 5G that could alleviate this problem is the standalone standard that can support 1 million devices deployed in a square kilometer. This is many more units than previous cellular specifications could handle in a small area. Not many operators around the world, however, have deployed standalone 5G yet.
So, it’s still largely a waiting game for 5G.
Jamieson does see a few green shoots on the ground for 5G and IoT in the short term. “Private networks in hospitals, I think, are really interesting because you’ve got HIPAA and the real value there, like data security.”
Providing a hospital administrator with a 3D map of the location and state of use of medical equipment within a facility is “super interesting”, Jamieson emphasized. Obviously, you don’t want dialysis machines sitting around idle when they could be in operation.
Jamieson also thinks that one of the initial ways 5G will be deployed in industrial settings is in passive sensing networks. This is with networks that don’t get inside the firewall of the factory.
“If you come in and say, ‘I can layer in a system in this factory that can measure risk,’… risk can be like accumulations of noxious gases, things that explode, that’s interesting… if you can create value by getting a bunch of passive data.”
So, when people combine the internet of (useful) things with the power of 5G data, then the revolution may start to grow wings.
—Dan Jones is a veteran reporter who has covered many segments of the communications market.
>> This article was originally published on our sister site, EE Times.
- The slow emergence of private 5G networks
- Assessing IoT and the impact of 5G
- Three key applications that benefit from edge computing
- 5G roll-out: a marathon not a sprint
For more Embedded, subscribe to Embedded’s weekly email newsletter.