Familiar things can look strange sometimes, and be hard to handle until you recognize them. Green and sustainable manufacturing were in vogue few years ago, and mostly turned out to be new hats for efficiency. Cybersecurity became more approachable when people equated it with process safety. It’s easier to deal with a new idea or technology when you know more about it, and then cope with any truly new aspects that tag along.
Likewise, all the blogs, LinkedIn, Facebook, Twitter, Instagram and Snapchats are just more websites and webpages. Big data is just more data. The Internet of Things (IoT) and the Industrial IoT (IIoT) are just more Internet, which is just more networking using Internet protocol (IP) over Ethernet and wireless. And virtualized and cloud computing are just bigger and more secure data storage.
“The basics of IIoT have really been in process automation for a long time because we’ve been connecting processes and whole plants for years,” says Will Aja, customer operations VP at Panacea Technologies Inc., a CSIA-member system integrator in Montgomeryville, Pa. “The pharmaceutical companies we serve can’t operate without connected and networked facilities, but today’s smart devices, apps and virtualized computing are making life more interesting. The automation industry is trying to mix in consumer-based Internet technologies, but the industry is still far behind.”
Aja adds this lag isn’t a particularly bad situation because process control applications and users require much different capabilities from Internet technologies. “For us, IIoT didn’t begin when Ethernet kicked off because that doesn’t account for DeviceNet, ControlNet and other older protocols, or the fact that we’ve been able to read data from HART devices for a long time, accomplishing many of the same tasks that IIoT does,” explains Aja. “I think IIoT is less about connecting devices, and really begins with collecting, contextualizing and using data appropriately. For example, Amazon Echo’s voice recognition and intelligent assistant is a nice device, but it doesn’t jive with what we have to do on the plant floor. The consumer side is vastly different than pulling data from a plant with 10,000 I/O points, and we have much bigger operating and safety consequences.”
Alicia Dubay, strategic marketing manager for ABB process automation and control technologies, adds, “Much IIoT technology has been in use in the process industries for years. The gains will start to come when the needed standards emerge, security concerns are addressed and the issue of who owns the data becomes more clear. It is possible now to get process data and diagnostics onto the cloud for analysis, but the question becomes who can access the data to provide value-added applications? Once these questions are answered, the value of IoT in the process industries will become more clear.”
Jason Wright, PlantPAx business manager at Rockwell Automation adds, “Bringing operations technology (OT) and information technology (IT) to develop IIoT can be like trying to bring two positively charged magnets together because they’re different disciplines, but they’re sharing more technology. Still, this is like any technological shift. When we went from proprietary operating systems to commercial off-the-shelf (COTS) in the 1990s, there was a lot of resistance then, too. However, it gets solved when people see the value the new technology can bring, and that’s what going to happen now with IIoT.”
Common language, common ground
Apart from the latest buzzwords and hype, there are a few new and substantive differences between traditional industrial networking and IIoT. Faster, cheaper, more powerful microprocessors can run software anywhere, and are finally making the old dream of distributed control come true, while their Ethernet ports, IP addresses and internal webpages are getting them and their data onto the Internet, up to the enterprise and onto cloud-based services more easily. Also, these functions require more collaboration with IT and its advanced security capabilities. In fact, some users are even asking why plant-floor data needs to stop off at PLCs and DCSs on the way to the enterprise.
Despite these tremors, manufacturing and process control have the same goals as always: do more with less, run lean, and maximize efficiency, production and profit. It’s just that some new Internet-based tools are now available, and more users and applications are waking up to employ them, so some new skills are needed to take advantage of them. One of the most crucial tasks in understanding and adopting IIoT and gaining its benefits is learning the networking concepts, terms and technologies it uses—most of which originated in the IT realm.
Just as most transmitters and other process control devices progressed from point-to-point, 4-20-mA hardwiring to twisted-pair fieldbuses and digital communications in recent years, IIoT is expected to further simplify and accelerate their networks and communications. However, process engineers and operators will need more IT know-how as IIoT moves down to the plant floor and out to the field to interact with more so-called edge devices, which include all the usual sensors, instruments, valves, actuators, transmitters and other equipment.
In the short run, Newton explains, OT and IT can converge using protocol gateways, OPC servers and middleware, but long-term convergence will demand flattened architectures and seamless communication between assets, using open, standards-based protocols and programming languages. As a result, process control assets will need to be designed with built-in web technologies, such as Hypertext Transfer Protocol (HTTP) for interaction, Secure Socket Layer/Transport Layer Security (SSL/TLS) for encryption and authentication for data security, and JSON for data formatting. This approach is presently available and is called Representational State Transfer (REST) architecture. Opto 22 has already added a REST-compliant, or RESTful, application program interface (API) to its industrial programmable automation controllers (PACs).
Newton adds OT assets will need to use this and other APIs, which are sets of routines, protocols and tools for building software applications, and specify how software components should interact. To have OT assets communicate autonomously and directly with each other, he adds they also need to use other Internet protocols, such as:
- MQTT publish/subscribe messaging protocol for constrained devices and low-bandwidth, high-latency or unreliable networks to collect device data and communicate it to servers;
- eXtensible Messaging and Presence Protocol (XMPP) that enables near-real-time exchange of structured yet extensible data between two or more devices on a network;
- Data Distribution Service (DDS) fast bus for integrating intelligent machines; and
- Advanced Message Queuing Protocol (AMQP), an open-standard application layer protocol and queuing system for connecting servers to each other.