Mark Twain famously quipped that everyone complains about the weather, but nobody does anything about it. Similarly, everyone these days seems to be talking about the industrial internet of things, but what are they actually doing about it?
For the most part, early adopters are dipping their toes in the IIoT waters with pilot projects for very specific use cases. Most of these are focused on improving efficiency of a production process or replacing manual processes. These are effectively proof-of-concept projects intended to establish the value of IIoT, potentially paving the way for broader implementation.
This phenomenon is driving an interesting shift in the way industrial automation decision-makers are viewing their operational technology (OT) infrastructures. Traditionally, technology acquisitions were a “relationship buy;” organizations bought entire end-to-end control systems from a single, trusted vendor with whom they had worked for years or even decades. The emergence of IIoT point solutions is changing this paradigm. Increasingly, OT decision makers are considering the best-of- breed approach to assembling control systems.
This presents a challenge. The key to a best-of-breed approach is the establishment of clear, robust industry standards to ensure interoperability. Disparate standards exist, but there is not yet a comprehensive, standards-based reference architecture for industrial IoT. To be sure, there are a number of groups working on next-generation standards, including the Open Group and Industry 4.0, but we are still early in the development curve.
Other barriers exist, including a skills gap, inertia caused by legacy OT mindset and, of course, questions around how the transformation to industrial IoT will be funded. Overcoming all of these barriers will require a better understanding of the true value of IIoT. That means going beyond pilot projects with a simple efficiency play and tapping the real value potential of the technology: harnessing big-data analytics to drive new capabilities that transform the way industrial businesses operate and engage with their customers and vendors. We’re talking things like “smart” supply chains that optimize production automatically, in real time, based on up-to-the-second intelligence about everything from raw materials pricing and availability to equipment dynamics to market demand.
That’s an exciting and achievable vision. But how will industrial enterprises get from where they are today to that future? A good place to start is by taking steps now to retool their OT infrastructures. Based on my experience in the field, here are some best practices to building a foundation that will deliver advantages today while preparing for the IIoT-driven enterprise on the horizon.
1. Get connected. Traditionally, industrial control systems have been deployed as isolated “islands” with little or no connectivity to other OT systems. Unlocking the power of industrial IoT, however, requires bidirectional data feeds across the enterprise. That means connecting everything from sensors and actuators to supervisory control systems to ERP and business planning systems. The key is implementing an infrastructure that enables secure connectivity, enabling the data exchange that is the lifeblood of IIoT while protecting the integrity of business-critical systems and data.
2. Go virtual. Any plan to modernize infrastructure must include virtualization. Hosting multiple applications on virtual machines residing on shared physical servers running on standards-based commodity hardware is crucial for optimizing efficiency, reducing capital costs and simplifying system maintenance and management. IT made the shift a long time ago; now it’s OT’s turn.
3. Build in availability. Continuous availability of critical production systems is non-negotiable. So minimizing the potential for business interruption is a key priority for any OT infrastructure. Reliability and availability should be core requirements for all system decisions, rather than an afterthought. Leveraging solutions specifically designed to support the levels of fault tolerance or high availability required throughout the infrastructure not only helps ensure uptime, but also the longevity of those systems, shortening the payback time horizon.
4. Keep it simple. Back in the data center, where technical skillsets are at a high level, complexity is acceptable. OT is different; technical skills are often less developed and more narrowly focused and staffs may be smaller. Seeking out solutions that are simple to manage, maintain, and service is crucial. For control systems in remote locations, remote maintenance is key. But also take a close look at the vendor’s support model and track record.
No matter where your enterprise is on the path to IIoT, taking these steps to modernize your OT infrastructure will pay dividends immediately in terms of efficiency, reliability and maintainability. Even if you’re just in the first tentative steps toward a smarter industrial enterprise, those advantages alone should make for a compelling business case right now.