Note, this article continues from a previous instalment: ‘Connectivity is 5% of Industry 4.0 spend – we’d rather talk apps and services’. Go here to read this article.
So, what about those IoT use cases, which Orange Business Services (OBS) reckons are proven (see above link for part one of this article)? “Some of the more, let’s say, ‘unsexy’ use cases have the best success,” explains Werner Reuss, in charge of Industry 4.0 at OBS, picking up where we left off.
OBS is focused, in the main part, on just two use cases: smart tracking (mostly indoors) and condition monitoring (mostly in retrofit ‘brownfield’ environments). It is the same conclusion Vodafone has come to, and the market at large appears to have settled on; that most IoT applications can be boiled down into these twin offers.
The first, indoor tracking, is a gateway application for site automation, and a means for enterprises to scope out the practicalities of digital transformation. It brings quick wins in terms of time, that most valuable of enterprise commodities. “The thing that costs time – and you may be surprised – is just workers searching for items in storage,” says Reuss.
“Most processes are not highly automated, regardless of the industry, and it is the same in all of them: a warehouse brings together components –- boxes, pallets, whatever – for a certain step in the manufacturing. Those items are required directly, at a specific location, and it takes time and effort to find those parts.
“In some cases it works well; in others, it doesn’t. In a facility of 10,000 square meters, say, you’re going to have five people, more or less full time, searching for stuff – and they’re not juniors, but skilled technicians. If you can reduce that, and bring down buffer times, you can have a very direct impact on productivity. It’s just process optimization, and a very straightforward thing.”
It is the same with condition monitoring, he says; the practicalities of it are more to do with breathing new life into old machinery, by retrofitting new sensing and sense-making tools, than it is about starting over, and the promise of high-fidelity lights-out automation.
He explains: “Most industrial environments are, plainly speaking, brownfields, designed and built 10 or 20 years ago, with equipment installed at various stages over those time periods. Before you do predictive maintenance and artificial intelligence – and all those pretty things – you have to go the first step, which is about connecting [old machines] and starting to collect and understand their data. That’s where most customers are at.”
Reuss gestures to the Orange stand behind (we are at IoT Solutions World Congress, way back in October), where OBS is showcasing its industrial IoT partnership with UK based Ubisense, which is offering a modular SmartSpace factory and warehouse platform and digital twin of “untethered production activity”.
The pair are joined on the stand by Finnish IoT tracking company (and Nokia spinoff) Quuppa, Dutch smart lighting firm Signify (formerly Philips Lighting), and German manufacturer Advanced Realtime Tracking (ART), which produces optical tracking, industrial measurement, and motion capture systems for virtual and augmented reality (VR/AR) and other industrial outlets.
The group is partnering variously with OBS on the sale of connectivity, hardware, and software for industrial tracking and monitoring. But Ubisense is running pin-point three-dimensional location tracking in ultra wideband (UWB) frequencies, with tags operating at 1Hz. Quuppa uses Bluetooth Low Energy (BLE), along with its own angle-of-arrival (AoA) signal processing and proprietary algorithms.
What about good-old cellular? Rightly, OBS is not interested in selling connectivity, for its own sake; the question of how to connect industrial equipment only comes after the solution has been scoped out, and made affordable. This, as discussed, is a core realisation for operators seeking to reinvent themselves as more than utility providers.
“We discuss the solution in the context of the type of factory, the kind of application, and the budget as well,” comments Reuss. “There are three components, basically: the sensors and antennas, the engineering and localization, and then the business logic on top of that. And the fourth is the connectivity in the whole process. The business case, at the end, is about performance and cost.”
OBS is effectively curating a solution, taking certain elements from partners, plugging in its own connectivity and services, as appropriate. On that last ingredient, cyber-security is the clearest beyond-airtime proposition from OBS. But it is another conversation, says Reuss, which runs in parallel with the immediate problem solving.
“The solution needs a certain level of security and safety. But the bigger question is how it is handled within the context of the factory – what are the steps to make it secure? Again, it has to start with assessment – what’s the current status, what’s already on the network, which ports are open, are the antennas wired or no? From there, you can work to harden the environment.”
He explains OBS’s curatory position, pulling together third-party hardware and software. It is, increasingly, the role major operators are pursuing in the chaos of IoT, which is a cluster-shambles of sensors and platforms. To wit, Orange’s enterprise IoT platform is based on Siemens’ Mindsphere IoT platform, which is based on Software AG’s Cumulocity IoT platform, which is running on Azure and AWS.
The IoT market is a mess of technologies, marketed via smoke and mirrors; there is an urgent need for solution providers in the space to create order from the chaos. AT&T is calling itself a “master integrator”; it is systems integration by any other name. Reuss puts it more plainly.
“We are the prime contractor, bringing in partners, with tracking solutions – with trackers, antennas, platforms. We bring them in for the hardware and software; we plan and manage the deployment, and customize and integrate the platform so it works in the environment. And if we need to bring in specific knowhow for certain steps, then we will also do that,” he says.
But the question of cellular connectivity niggles, still. What, then, about hyped industrial versions of LTE and 5G, whether hived off public networks or ring-fenced in private campus setups? What place do these technologies have in the factory, if tracking and monitoring are adequately served by other radio solutions?
“Industrial 5G could be very interesting,” says Reuss. “The question is when it will happen, and how it is set up. Because it looks like a very powerful tool, if you want to really standardize your infrastructure, and you don’t want to stop at the factory walls. It has the most meaning when you think of it holistically, as a full campus network.”
But industrial 5G is in the future still; it is not really in front of us, on the table, says Reuss. “The reality is we’re pretty much at the beginning with 5G; it is being evaluated for new deployments of factories, campuses, whatever. But do we see the demand for 5G in retrofit environments? I’m not so sure. Does the business case work? We will see.”
Interestingly, and to close, Reuss briefly considers, and rather dismisses, the two primary use cases for industrial 5G in factories: automated guided vehicles (AGVs) for ferrying items and augmented reality (AR) for maintenance support. “If you are in a retrofit environment, AGVs are not so easy to implement – just because you need space to put them to work,” he explains.
Meanwhile, AR hardware and software for maintenance and support are novel still, and require behavioural change.
He says: “AR is an interesting case, but it’s early days – and there’s a question mark over the acceptance [in working environments]. That’s the thing that comes up with customers. I’m not saying it won’t change, because it probably will. But gesture control will take some getting used to, and the latest AR devices are heavy, especially if they are worn for six hours in an industrial environment. The feedback we are getting is mixed.”
