Imagine it is time to go home. You leave the office and go out to the parking lot, get in your car, and start your engine. This starts the following sequence:
- Your car tells your house to set the thermostat to a nice comfortable temperature so that by the time you get there it is very pleasant
- Your stove preheats to 450 so that when you arrive you can put your chicken breast in without delay
- Your refrigerator notices you are on your way home and notifies your car that you are low on milk, sending you navigation instructions to a market on the way
- When you get to the market, it knows you are there and sends you a text message that the hazelnut ice cream you crave is on sale
- As you approach the house, your television notices and turns on the channel to the baseball game you want to see
- When you open the door, the lights turn on
This sequence is happening today because the dumb things of the past (cars, appliances, homes, stores) are now smart. They can measure and communicate. The means of communication is a common data communication infrastructure; the Internet. We now have smart things connected by the Internet. This is the Internet of Things (IoT).
For information technology (IT), this is a very new development. They are being hired at a rapid pace to meet the demand of those who want a smarter connected world. When they look at the industrial realm that we work in, they think their new technology can help us the same way. It is easy to see why. In the February issue of InTech magazine, Christopher Logue says
When IT people come into the plant for the first time, they are usually shocked by what they find. Many of the younger technicians have never seen some of the technologies they encounter. "This computer is still running Windows XP." "Where do I find a driver for a dot matrix printer? I do not even know what that is." "What's Modbus?" "Is everything this old?"
Naturally, they believe their newer IoT technology can be helpful to industrial facilities. They call this the industrial Internet of things (IIoT).
To us, this isn't new. We have used smart connected systems for a very long time. We have had sensors in our process send us tons of data rapidly, process it, send signals automatically to valves and motors, share it on a network, and allow us to be smarter through automation.
In prior articles I highlighted how the Internet has made our industry different. How is IIoT different? The problem with that question is that before we can answer we need to know what IIoT is.
In Logue's article he explains how a WirelessHart adapter can be added to a device to put it in a network using wireless Ethernet routers that are familiar to IT technicians. The device is being made smarter. He says this makes IIoT implementations practical and possible. However, the Internet is nowhere to be found in this example. The public communication infrastructure that connects IoT is not used in this case. Does that mean that anything using IP addresses and switches are IIoT?
In that same issue, an article on IIoT remote monitoring explains how a virtual private network can enable Internet connection from a tablet, smart phone, or PC to a programmable logic controller (PLC). This allows someone remotely to see what is happening in the PLC and make logic changes. In this example, there is no device that is being made smarter. The existing devices are being connected by the Internet. This suggests any industrial connection using the Internet is IIoT.
If devices are smarter with no Internet, is that IIoT? If devices are no smarter but connected with Internet, is that IIoT? Which is it?
What is happening is that IT terms are being used to describe industrial process automation technology assuming there is no preceding industrial process automation technology. That leads to a confusing application of terms. Peter Zornio says in the February issue of Control magazine:
For example, an edge gateway used to be called just a gateway, while the Industrial Internet of Things (IIoT) is really just SCADA over the Internet. However, as the IT community discovered IoT and digital transformation technologies, they started coming up with sexier terms for what we'd already been doing in process automation for 30 or 40 years.
Some are defining IIoT as sensors in the cloud. The vision is to have a sensor connect directly to the Internet so the data can be more easily shared without having to go through control system gateways. There are obvious advantages to this approach.
Consider the application of wireless acoustic transmitters to detect failing steam traps. These smarter "things" measure something that were previously undetectable. Their data can be sent through the Internet to cloud computers to analyze and identify failures. Experts offsite can determine what actions to be taken and inform the facility of their recommendations. In this application we have smarter "things" using the Internet for connectivity. This seems to fit the description of IIoT.
However, there are caveats. If there is no demilitarized zone (DMZ) between the process and the Internet, how do we ensure security? What is it we want to put in the cloud? Sensor data? Control? Can we do control with non-deterministic data communication and latency? Is such an approach desirable? Do we want to bypass a control system that provides fast processing local to the process with security in order to use a cool Internet technology? That is why the steam trap application often is a separate system, not integrated into the control system. It is a one-way stream to the cloud with no automatic feedback.
Then we come to an article "At the IIoT crossroads" in the same February issue of Control magazine. The article is about the application of data analytics, such as statistical and neural network models, the danger of doing so without considering first principle process knowledge, use of wireless measurements to help PID loops, and connecting to more data to derive new metrics. Given the title, it sounds like any application of data analytics in industry is IIoT.
In articles such as these, it appears IIoT is anything new; wireless communication, smarter instrumentation, Internet connectivity, data analytics, and more. Any or all of this seems to be what IIoT means. This makes it hard to figure out what IIoT really is.
Consider the following statement from Smart Industry magazine in their Q4 2018 issue: "IoT is propelling us from an era of information scarcity to an age of information abundance ..." First, this statement does not distinguish between IIoT and IoT. I agree that the sequence of events I began this article with greatly increases information flow from cars, homes, and appliances, so that statement is true for IoT. The statement is not true for IIoT; the number of sensors in industrial facilities and data in networks was not in scarcity before the Internet. It is true that more of that data is flowing across the Internet, but the premise of the statement is false. It is problematic when IoT and IIoT are being used interchangeably while IIoT is not defined.
In conclusion, IoT has a clear meaning. It is easy to see how our lives outside of the office are different than they were when we were in our youth. Discerning what IIoT is more mysterious. Over time as more application of these technologies in our industry proliferation, IIoT should become easier to define.
