IIoT Progress Report: The Barriers and the Best Practices

IIoT Progress Report: The Barriers and the Best Practices

By Jason Andersen, Vice President Business Line Management, Stratus Technologies

For many in the industrial automation world, it may seem like the industrial Internet of things (IIoT) is like a distant mountain peak—perpetually on the horizon but never getting closer. In reality, there is a growing number of enterprises taking tentative steps to climb the IIoT peak.

Among these early adopters, we’re seeing a clear trend of IIoT implementations as limited pilots focused on a specific use case or objective. Improving the efficiency of a particular area of production, for example. The goal is usually to demonstrate the business value of IIoT in solving real problems. But these projects are often not scalable to other parts of the business. What’s often missing is a broader, unifying principle that places IIoT at the heart of the automation value chain.

There are some good reasons for this. In fact, we see three key challenges that are acting as barriers to more widespread, enterprise-wide IIoT implementations:

1. Lack of a clear IIoT blueprint.

There is currently no clear “template” for what an IIoT architecture should look like. There’s a lack of standards. And there’s a common misunderstanding that “IIoT” doesn’t necessarily mean “pure cloud”—that a hybrid cloud/data center model may be what’s needed, at least at this juncture.

Many industrial enterprises have adopted the Purdue architectural model, implementing enterprise functions in multiple layers and multiple stages. Whether or not this model, first developed in the early 90s, can adapt to a service-oriented, IIoT infrastructure remains an open question. Meanwhile, there remains a need for an entirely new, IIoT-ready architecture.

2. Legacy mindset and skills.

If enterprises change slowly, it’s because people change slowly. Faced with a need to update infrastructure, people in the OT space will tend to rely on established vendor relationships and technologies with which they are comfortable. That’s just human nature. Meanwhile, IT often will try to force IT-centric solutions, which may not be what’s needed. That’s because the hub of an IIoT implementation will not necessarily be a server; it may be an intelligent node out at the edge, closer to where the work is performed.

Breaking out of the legacy mindset and embracing new and different technology approaches and solutions is a key barrier enterprises will need to overcome.

3. Uncertain funding.

Who will pay for IIoT and how will the case for those investments be made? To date, most IIoT projects have been an efficiency play, with payback in terms of incremental efficiency gains. But this offers limited potential; how much efficiency can you ultimately wring out of the operation?

To spark broader IIoT implementation, what’s needed is a more compelling value proposition—an IIoT “killer app” that delivers significant, long-term value to the enterprise. What will such a “killer app” look like? I believe real-time analytics will be at the heart of the next generation of high-value IIoT applications.

Instead of simply driving greater efficiency, this higher order of analytics enables a whole range of capabilities that can literally transform the way businesses run. For example, driving “intelligent” supply chains that automatically optimize production in real time, based on a granular, up-to-the-second understanding of each link in the chain, from materials pricing to equipment dynamics to market conditions. Imagine a plant that automatically determines to manufacture product A today rather than product B due to a sudden change in raw materials pricing or availability, labor dynamics, or a spike in market demand—or even identifies a market opportunity for a new product variant.

Building a New Foundation for the IIoT Enterprise

How can industrial enterprises overcome these barriers and begin building the foundation for the IIoT enterprise they envision? A good place to start is by rebuilding their outdated industrial automation infrastructure. In many industrial environments, this is long overdue. Walk through most plants and you’ll see a patchwork of old desktop hardware and no-longer-supported software controlling processes. Or proprietary, “home grown” systems that have been patched and bandaged for years. If it ain’t broke…

So it’s time to update the environment. But with the IIoT future still coming into focus, how can industrial enterprises retool, while having confidence they are making the right investments? Based on our experience in the field, we see some best practices that will keep you on solid ground today and well into the IIoT future.

1. Embrace secure connectivity.

Legacy industrial control environments often consist of isolated system “islands.” This is due not only to the piecemeal manner in which they were built, but also with an eye to protecting critical production systems from the perceived threat posed by open connectivity. But data feeds are the life blood of the IIoT, making connectivity a core requirement for operational technology. Connecting the enterprise—from sensors and actuators to basic control and supervisory control systems to business planning and logistics systems—is essential to reaping the advantages of the IIoT.

The key is to implement systems that support secure connectivity between industrial control systems and IT resources, including data warehouses, analytics engines, and ERP systems.

2. Leverage virtualization.

Any discussion on modernizing operational technology would not be complete without a nod to virtualization. Deploying multiple applications on a single, virtualized server running on off-the-shelf, standards-based hardware significantly reduces the number of platforms that must be purchased, configured and maintained, minimizing CAPEX, OPEX and the staff required to manage it all.

The idea of hosting diverse application workloads on a single system is old news in the data center. Now, the operational technology (OT) side of the enterprise needs to discover the efficiency advantages of virtualization. 

3. Make it reliable and available.

When replacing aging systems, the last thing you want to do is replace them with something that can disrupt the smooth operation of your business. That means investing in systems built to last and built to stay up and running, day in and day out.

A key factor in achieving this goal is leveraging automation systems built from the ground up to provide the degree of availability your applications require—from basic fault tolerance to high availability and disaster recovery. Making reliability and availability core requirements of your infrastructure rather than an afterthought will pay dividends not only in system uptime but also in the longevity of those systems. That’s key to achieving your projected payback time horizon.

4. Prioritize simplicity.

As you consider your available options, exercise a bias for simplicity. Will it be easy to maintain? If there is a minor glitch, is it easy to service? A solution intended to improve efficiency and reduce costs will do neither if it ends up requiring a lot of time-consuming care and feeding. In many cases, such as remote processing plants or pumping stations, it could be days or even weeks before personnel can be freed up to go fix something that breaks.

Reducing complexity by simplifying your architecture with streamlined technologies is a good first step. Just as important is choosing solutions designed from first principles for easy service and maintenance, with a solid support model behind them. For example, can the system vendor resolve many issues remotely, without requiring any intervention of your staff? Asking questions like this up front can make a big difference in your total cost of ownership and business risk over the long term.

Putting the IIoT Principles into Practice

Some forward-looking organizations have already put these principles into practice. We’ve seen a growing number of industrial enterprises that have revamped their architectures for IIoT using a hybrid model. They have updated their infrastructures for connectivity, reliability and simplicity, and layered on cloud-based services. For example, we’ve seen organizations in the natural gas industry that have massively simplified the control systems at their remote compression stations with fault-tolerant servers running localized SCADA and HMI applications, but they also interoperate with cloud-based analytics engines. This provides them with mission-critical, real-time command and control, together with the ability to generate new insights for improving their business based on up-to-the-minute data.

That’s a big step forward. And well worth the effort to build a foundation for the IIoT-enabled industrial enterprise. For the innovators, that reality isn’t looming in the distance; it’s here now.

About the Author

Jason Andersen is Vice President of Business Line Management at Stratus Technologies and is responsible for setting the product roadmaps and go to market strategies for Stratus Products and Services. Jason has a deep understanding of both on-premise and cloud based infrastructure for the Industrial Internet of Things (IIoT) and has been responsible for the successful market delivery of products and services for almost 20 years. Prior to joining Stratus in 2013, Jason was Director of Product Line Management at Red Hat. In this role, he was responsible for the go to market strategy, product introductions and launches, as well as product marketing for the JBoss Application Products. Jason also previously held Product Management positions at Red Hat and IBM Software Group.