Supervisory control and data acquisition (SCADA) systems have an important role on automation projects. Though a lot has changed in 60 years, not many improvements have been made to SCADA systems in the past 20 years. Some important changes have taken place—such as the move from DOS to Windows platforms, as well as the adoption of open protocols and high-performance HMI—but SCADA systems have often lagged behind other types of software.
Nonetheless, SCADA has solidified its place as a necessity on automation systems; you don’t need to explain to buyers why they need a SCADA system. SCADA has also been seen as a great tool to communicate with any device through industrial protocols.
But in the era of the Industrial Internet of Things (IIoT), what’s the role of SCADA systems? Will they still have a long life? How can SCADA evolve for the future?
Several SCADA platform providers are forgoing the large base of protocol interfaces to rely instead on third-part gateways with OPC UA or OPC DA. Tools like the Intel IoT Gateway allow users to create an industrial gateway using free software and affordable hardware, meaning SCADA is no longer required at the collector level. With real-time data being streamed using open-source software, people can also create reliable interfaces to monitor real-time data.
So, is SCADA dying?
There is still a lot that can be done by SCADA that won’t be covered by other tools. One example is alarm management, for which SCADA systems are very reliable. They are also very robust tools for working on premise, where availability is key.
For IIoT, however, SCADA systems will need to evolve (with some providers already on the right path). SCADA systems will need to be able to handle not hundreds or thousands, but hundreds of thousands and even millions of devices and their data points. For that, not only decentralization is required, but also an efficient way to handle the traffic of messages, as it happens with the Apache Kafka distributed streaming platform. Historical databases will no longer be the standard SQL solution (though I’m not saying SQL is dead for time series), but something horizontally scalable.
Perhaps one of the biggest changes has to do with the real-time databases. Today, in many systems, there are variables in random access memory (RAM). In a scalable scenario, they will have to use a database system—in memory or not—that will allow for instant delivery of real-time data to thousands of possible users. It might be weird to think about thousands of users in a SCADA system, but I believe SCADA is entering this stage with IIoT.
In his book “Crossing the Chasm,” Geoffrey A. Moore presents the challenge in sales and adoption of new technology between the early adopters and the early majority. The last tail refers to the laggards—those people who would never buy your product anyway, unless there is no other option or the product is part of another solution. Guess what—that is probably already happening with SCADA. Most people don’t even think about buying a SCADA system for their homes; they don’t even know what it is. But a lot of them are users, monitoring their power generation on solar panels or accessing their thermostats on smartphones. So SCADA is probably now on the late majority, reaching its first laggards. Being at the late majority stage means SCADA is still in its peak. But with IIoT, I believe the laggards tail might be much longer than we think.
Having worked with automation software for my entire career so far, I believe a lot still has to evolve, especially with regard to security, architecture and user experience. But this is already happening in new markets where SCADA does not seem as viable, where innovations such as voice commands and image recognition are being used along with old standard SCADA features. There is a big market for SCADA. It’s just changing.
Mario Gonsales Ishikawa is an advisor at scadaHUB Technology, a member of the Control System Integrators Association (CSIA). For more information about scadaHUB, visit its profile on the Industrial Automation Exchange.
