When the Industrial Internet Consortium (IIC) started up a little more than three and a half years ago, it introduced the concept of the Industrial Internet of Things (IIoT) to differentiate connected technologies from consumer-related IoT. Now almost every major manufacturing company is pursuing some kind of IIoT project—at least that’s what it feels like if you attend vendors’ user group meetings or IIoT-focused events, or follow any of the IIoT-focused websites.
The data coming from LNS Research paints a somewhat different story. Our research shows that only 40 percent of companies have actually started an IIoT project and only 24 percent more plan to do so within the next year. That leaves more than a third (36 percent) of the market with no IIoT projects or plans. The same survey also reveals that the top three challenges associated with an IIoT initiative are funding, building a business and understanding how IIoT applies to the specific business. Despite the hype and all the proof-of-concept projects apparently happening, progress has been slow.
In an effort to kick-start IIoT spending, there has been a lot written about how to get started. One popular approach that seems to be emerging as a supplier community favorite is the idea that companies need “IIoT momentum projects.” The thought process is to get industrial companies to do something—anything, really—just to get the momentum going, get some experience with IIoT and demonstrate its value. We think there is value with this approach, and even published an e-book on the topic: “Think Like MacGyver to Drive Better Asset Performance.” We agree that it isn’t the right approach to wait for a long-term strategic solution that requires a lot of time to bear fruit. If a problem exists today and there is technology that can cost-effectively address it, then it makes sense to use the technology. However, we do challenge the idea that simply pursuing momentum projects will lead to successful digital transformation led by IIoT.
The manufacturing industry has been notorious for operating its businesses in silos. For more than 40 years, these “islands of automation” defined the way most plants were designed and operated. Each machine center or process train was controlled and operated by its own bespoke automation solution, and information accumulated in one center was thrown away as the product was handed off to the next machine center or cell. At that point, a new set of measurements, scans and the generation of current-state information was initiated, because previous knowledge and information was no longer available.
With the advent of the IIoT—without an architecture of IT/OT integration—the next frontier will be “islands of IIoT platforms.” A proliferation of IIoT momentum projects will take root and the problem will repeat, although likely at the next layer of abstraction.
To avoid this scenario, companies need a framework to define how all the plant floor systems do and should communicate. LNS refers to this as operational architecture, which is the discipline of defining for operational technology (OT) systems and how they will use technology to satisfy the business needs. This is similar to how the enterprise architecture methodology in IT defines business needs to establish IT system linkages.
An operational architecture also provides the mechanism to define how IT will link to OT and create a top-to-bottom plan for technology deployment. When companies have an operational architecture in place, they are in a much better position to pursue those IIoT momentum projects because the investments then apply toward the business’s long-term goals and are more than just technology tinkering.
>>Dan Miklovic is a research fellow with LNS Research, focused primarily on industrial operations, asset performance management and energy management, with collaborative coverage across IIoT, paper and packaging, mining, metals processing and other industry verticals.