There has been a lot of debate recently over the increasingly competitive and overlapping roles of operational technologists (OT) and information technologists (IT) within industrial automation. While traditionally very separate, the increasingly pervasive use of IT technologies at the operational level – specifically in areas of IIoT and cybersecurity –is making it clear that OT/IT convergence is here to stay. In a recent webinar with LNS Research, analyst Matthew Littlefield explored the idea of how IT can play an important role in advancing the adoption of IIoT and analytics.

No one can deny that IT has critical skills around managing data and using cloud technologies, networking and cybersecurity. These are all an integral part of IT’s role. Certainly, most of these areas, except for cloud, have some overlap with OT. What sets IT apart is the scale of operations that they must deal with in these areas. While Historians are a critical part of an automation implementation, the amount of data they generally collect, relative to IT networks, is often quite small. When we talk about IIoT and analytics, the scale of the data collected is often magnitudes greater than current automation systems. This is a result of increased rates of data collection, additional sources of data (i.e., sensor instrumentation), or a combination of the two.

IT’s experience in dealing with large volumes of data and moving that data to secure, long-term repositories, such as the cloud, can make them an important partner or advisor in an industrial automation digitalization and analytics implementation. And let’s not forget that they bring additional expertise in cybersecurity, particularly important if that data is stored outside the plant or in a public cloud. IT may also bring networking expertise, enabling segmentation and adding another dimension to security through their enterprise responsibilities in this area.

With all that IT brings to the table, it is important to recognize where they may have limitations and where OT has a better understanding of requirements. This crossover point is often where OT and IT clash and where conflict can arise. The question is, why does this happen? It all comes down to how they approach technology needs, based on what they are used to.

For example, OT is accustomed to dealing with a variety of equipment, from mechanical valves, pumps, and motors, to sensors, PLCs/PACs, compute platforms and various automation applications, SCADA, HMIs, Historians etc. And specific analytics applications incorporating sophisticated algorithms and machine learning are clearly the next wave. To OT, edge platforms running these applications are just a tool, and like everything in the OT world, eliminating unplanned downtime is an important KPI, particularly at the edge, where support resources may be scarce or unavailable. Automation systems are often highly deterministic and real-time in nature; specific events must happen at specific times, in a specific order and within tight tolerances.

The IT world is somewhat different. While eliminating downtime is important to IT, any IT professional will tell you that security is their number one priority. Typical IT clustering or high-availability virtualization techniques enable IT to standardize and scale their operations. But the inherent complexity of these implementations does not translate well into remote edge deployments, such as control rooms, or control cabinet environments. Additionally, the IT world does not deal in deterministic, real-time events, and variability of a few seconds, or even minutes does not really matter. Think email or web page loads.

From an analytics perspective, however, the automation industry is in general agreement that there is the need for both cloud-based capabilities (IT) and edge-based analytics (OT). Cloud-based analytics are ideal for addressing broad requirements where data sharing across an enterprise can be beneficial, such as profit optimization, asset performance benchmarking or digital twinning. Edge-based analytics are critical for improving operations in areas such as advanced process control, real-time quality inspections and identifying device failure. This is the domain of OT. These locations require simple, easy to manage platforms, with integrated redundancy and self-monitoring capabilities that enable OT to plan maintenance or replacement, instead of having to react when something fails. Certainly, these systems must inter-operate with the rest of the network, up to and including ensuring data can reliably be stored until it can be transmitted to the cloud or a higher-level Historian.

What can we conclude from all of this?

IT has a critical role to play in modern automation implementations; handling large volumes of data, architecting networks, providing cloud connectivity expertise as well as advising and possibly implementing layers of cybersecurity. However, as the implementation extends deeper into the edge environment, OT skills and requirements become paramount and IT’s role becomes much more advisory. Implementations can be successful if one of the following approaches is followed:

  • IT and OT collaborating, yet each being open to understanding its own limitations and where the other should take the lead
  • Developing a hybrid OT capability, where people within the OT organization who have the requisite IT skills and understand OT requirements, implement IT like requirements and liaise with their IT counterparts at critical interface points

Whatever approach is taken, no one can ignore that the cloud and the edge represent new frontiers for industrial automation.