From a standing start in Texas – the story of Dow Chemical’s multi-site private LTE push
US system integrator Kyndryl has published a nice-looking blog post, complete with videos, on its headline work, announced at the start of the year, to deploy private LTE for Michigan-headquartered Dow Chemical Company, manufacturer of plastics, chemicals, and agricultural products. The post puts focus on the initial scope of the project to modernise plant maintenance at its Freeport manufacturing site in Texas.
But the Freeport site is no ordinary factory; it covers 40 different production plants, each with “thousands of mechanical components”, across more than 50 square kilometres of Brazoria County. And the exploratory work across this industrial sprawl, carried out with Kyndryl, and eventually with Nokia as well, set the blueprint for the trio to explode their single-site testbed into a 200-site deployment.
Indeed, along with publicly-referenced multi-site, multi-market deployments by Airbus (in process, with Ericsson) and Schneider Electric (on paper, with NTT), and perhaps some others, the Dow Chemical project with Kyndryl and Nokia represents a seminal case study for how to scale private cellular in the complex Industry 4.0 space. The narrative of the blog post – as told by Dow Chemical, as reported by Kyndryl – shows how far, and how quickly, the group came.
Because the start-point in Freeport, for Dow Chemical and Kyndryl, was just to digitalise its old industrial processes – across the largest integrated chemical manufacturing complex in the Western Hemisphere. Everything was manual, as the blog tells it; procedures for production and maintenance were recorded, largely, only in the heads of its workers. The first job was to make digital copies of these process flows and actions.
Jason Jackson, chief technology officer for industrial sector clients at Kyndryl, writes: “To execute a maintenance work order, an engineer from the control room often had to go out into the field to inspect the components, pull the relevant process documents from the paper filing system, write a safety analysis report, and deliver it to a field operator. Operators then carried the printed documents back into the field to perform… tasks.”
He goes on: “And because field operators needed to coordinate with the control room on certain steps, there was often much walking back-and-forth to communicate face-to-face, sometimes doubling the time needed to complete an operation. The printed process documents also posed a safety risk as operators worked to keep up with the paper while inspecting, and sometimes climbing, components and manually checking off all the steps in order.”
It is a classic case of baby steps, and a cautionary tale about how far even the biggest industrial mega firms have to go on the road to Industry 4.0. NTT, which is working with Schneider Electric on another rapid-scale private networks project, has said the same in these pages; that a vendor crew often turns up to consult on a ‘digital factory’ process only to find the instructions for it “on a two-by-four yellow card on the wall”. Kind of thing.
But the message from Kyndryl is valuable, as well, because it makes clear how fast this road can be travelled, so long as the starter-case to stand-up the project is well defined, and the parties are well aligned. “Despite great care, the potential for human error remained,” writes Jackson, revealing another motivation, besides work efficiency and worker safety, for Dow Chemical to make the leap.
The article quotes Clark Dressen, director of the Dow Chemical’s global IT innovation centre. Dressen says: “One of the primary problems we were trying to solve was getting information into the field to change the way we do operations and maintenance tasks. We wanted to make that information available at the fingertips of the front-line professionals, impacting how they work, how they interact, how they collaborate, how they solve problems.”
Melanie Kalmar, chief information and digital officer at Dow Chemical, was on board for the project, alongside “business leaders”, writes Jackson. Kyndryl and Nokia were both at the table, too. Together, they devised a “digital manufacturing initiative… [with] technology partners both in and outside of Dow. In the end, after relevant processes were digitally memorialised, the brief was ultimately just to connect – at least to start with.
Jackson writes: “The answer was to digitalise and modernise operations, including communications… The effort would digitalise all process documents, set up remote communications through Microsoft Teams, provide rugged devices for operators to use in the field, and implement augmented reality applications to aid control room operators in working with multiple documents related to different operations happening simultaneously.”
The blog tells it like a cliffhanger; how to connect a large industrial site, amid all the “steel and concrete [to] make building a radio network difficult”? Except, of course, we know how it ends – with a private LTE network, specified by Dow, made by Nokia, built by Kyndryl (or devised in co-creation by all three, at every stage). “But there was one major challenge,” writes Jackson, before concluding: “It was time for (cost-effective) mobile connectivity.”
Dressen is quoted: “We looked at other solutions like Wi-Fi, but it’s hard to get that type of service extended over such a large geographic area. The plant in Freeport is made up of four distinct areas where we have plants, operating units, and people working. To provide that type of coverage, it was going to be very expensive and very difficult to maintain and operate.”
Jackson writes that the Freeport project evolved from a proof-of-concept to a fully scaled production environment in about 12 months. Dressen comments: “We were introducing a lot of new tools to our operations and maintenance teams in terms of changing the way we do our daily work. Kyndryl was a great partner for us, working with our internal teams from a security architecture point of view to make sure they would fit the Dow environment.”
Enterprise IoT Insights wants to write that ‘the rest is history’, in line with the well-deployed story cliches in the blog; that the Freeport project has multiplied and mutated across the Dow Chemical footprint. But it isn’t history, yet, of course; private networks are brand new, and the Dow Chemical expansion is mostly on LTE, and mostly just for connectivity. It is IoT-for-industry, really, rather than industrial IoT, as it were.
And it will probably only be judged as history when we discuss the industrial 6G, in the context of the successes and failures of industrial 5G, as a platform for industrial revolution.