Private 5G is outpacing Wi-Fi for its early growth, and its trajectory is diverging as well. Just as Nokia checks-out, Ericsson argues that a rising tide of high-value industrial use cases is making the case for private 5G very clear – as evidenced in its rapid ROI, and new propensity to global scale.
In sum – what to know:
Parallel narrative – private 5G is Wi-Fi adjacent, says Ericsson – it follows a similar, steeper, early growth curve, but serves more-critical business needs, with associated ROI.
Industrial demand – remote operations, AI video analytics, AGVs/AMRs are all uniquely served by 5G, finally taking cellular into indoor environments alongside Wi-Fi.
Tangible returns – enterprises typically start with a handful of business-critical use cases, and, within a year, expand their deployments – to thousands of devices and global rollouts.
Note: this article is continued from a previous post, found here.
So the question hangs: if private 5G is following the early Wi-Fi playbook (and tearing it up), does it need its own catalytic breakthrough – its iPhone moment, circa 2007 – to tip it into the enterprise-mainstream? Manish Tiwari, head of enterprise 5G at Ericsson, pushes back; the comparison only holds so far because private 5G is a different technology, serving a different market, with different requirements and expectations. Which is where our story picks up again. Because for all the parallels with the salad days of Wi-Fi, the real shift with private 5G is more about a widening field of high-value industrial applications – which are already multiplying fast, and reshaping demand.
Tiwari explains: “The thing with 4G is that it solved the problem of wide-area connectivity for enterprises. So most deployments were outdoors where the transmit power and coverage area made it valuable for certain verticals – like oil and gas, offshore drilling, mining environments, wind farms, outdoor logistics yards. But it was mostly for handheld connectivity and so on – outdoors. A lot of 4G projects have transitioned to 5G over the past three years. That’s the thing that has changed – 5G. And it has enabled new use cases (in these environments), mostly driven by remote operations, automation, and AI. But it has made the case for private networks indoors, as well.”
He goes on: “So we have started to see these enterprises deploy remotely-operated vehicles for worker safety in hazardous environments; AI-enabled video cameras, whether fixed or mobile, for remote inspection of errors and faults, and for health-and-safety compliance; AGVs and AMRs (to ferry assets around plants and yards). All of these use cases – which are more 5G centric, as opposed to just 4G-based – are becoming more prevalent, and every year we see more and more. So the number keeps increasing. AGVs, say, are a pretty big deal, and they’re time-sensitive, with an upper bound on latency; and 5G has unique capabilities so they can operate smoothly.
“Many customers have struggled for many years to run AGVs on Wi-Fi. When they see the difference with 5G – well, that’s what leads to this growth with private 5G.” Outside of these core Industry 4.0 disciplines, he also mentions remote surgery (“using, for example, Microsoft HoloLens”), plus “robotics and sensors” for asset management, patient safety, and digital operations in healthcare. “So more sensors, more automation. That’s the theme across many verticals, and it is uniquely enabled by 5G,” he says. The sense is that the iPhone moment for this technology will not come with a single product drop or app release, but in rolling waves of high-value industrial use cases.
Which befits the user market, of course – in charge of critical systems and processes, where disruptions are hugely costly and upgrades are enormously expensive; which operates according to its own industrial rhythm, defined by long investment horizons, and justifiable conservatism and a touch of paranoia. Point is, this private 5G swell has already started, reckons Tiwari. He says: “Most deployments start with three to five use cases, and they tend to be business critical. And if the enterprise has thought through its use cases, and if they are successful a year later, then the deployment scales quickly.” Again, this is the difference with 5G, and 5G-enabled use cases, he suggests.
Three years ago, when enterprises were still kicking the tyres on 5G, the ROI calculation looked complex – especially indoors, where it got skewed because of legacy investments in Wi-Fi infrastructure. “So outdoor-cellular solves certain problems, but there was a question about indoors, where enterprises have invested in Wi-Fi. Do I need another network? But this ROI question has started to be addressed, so long as stakeholders are focused on the right use cases. Because you don’t need to boil the ocean; it is wrong to think 5G will replace Wi-Fi, especially given the low price points for Wi-Fi chipsets and the long tail of Wi-Fi devices. But both have their place.”
Tiwari adds: “With 5G, it is all about those three-to-five use cases. And in almost all cases, when we look back with customers a year or two later, there are many more than three-to-five cases on their networks. Once it is running, the numbers of applications and devices quickly scale. (And from there) the number of networks scale, as well. Many enterprises that started on this journey two years ago, say, now have hundreds of thousands of devices on their network, and are deploying globally – not just in a plant or city, but across countries around the world.” As above (see here), certain of Ericsson’s biggest global clients, rolling out across regions, are well known.
He adds: “The technology has real ROI, and it has a place, and every year it gets clearer just what a difference it will make in Industry 4.0.”
This indoor discussion is interesting. Tiwari suggests the old mix-and-match models from the 4G/LTE era, where integrators pair core and radio networks from different vendors, will not work with private 5G, especially for the kinds of deterministic high-end use cases that 5G is being deployed for – which have gotten private cellular a foot in the factory door, in indoor industrial environments, alongside familiar best-effort WLAN systems. “You see it a lot in Europe – where the enterprise takes a core from this vendor and a radio from this vendor, and an integrator puts a dashboard around to manage the whole system,” explains Tiwari.
“That kind of architecture worked in the 4G era because the use case was mostly wide-area outdoor connectivity. But when you come indoors, with 5G, when you’re doing time-sensitive networking, time-critical comms, URLLC, time-bounded latency – it does not work at scale. It might work in POC, but when you go to production with hundreds of thousands of devices, and every minute of outage costs hundreds of thousands of dollars, then it does not work.” Which gives way to the Ericsson pitch, of course – because it is one of the few that does it all; and even Nokia, the erstwhile leader, looks like it is about to separate its seminal NDAC core network product from the mothership.
Tiwari says: “An integrated stack gives access to rich data across every layer of the network – whether it is the RF layer, the RAN software layer, the different components of the core, the voice and video applications, the security and policy layer, the management layer, the APIs. All of these have to work very closely together for this product to support those business-critical cases deployed with 5G. And it is an important distinction to make because – and I would agree there has been a lot of noise and hype about private 5G over the past few years, but – when you start deploying in production and scaling with real deployments, you learn what works well and what does not work well.”
To be continued (again)…
