It is nearly impossible to ignore the marketing hype behind the Industrial Internet of Things (IIoT) these days. If we believe all that we read, within the next few years, we can scatter prolific, cheap, battery-powered, wireless sensors everywhere, and with assistance from data analytics in the cloud, the process industry will finally fulfill Warren Bennis’ often-quoted prophecy from 20 years ago:
“The factory of the future will have only two employees: a man and a dog. The man will be there to feed the dog. The dog will be there to keep the man from touching the equipment.”
That idealized version of industrial automation is something we’ve been striving toward for decades, not only because of how it represents an idyllic level of efficiency and safety but also because industrial environments are not exactly fun places for people or their dogs. They’re hot, noisy and full of opportunities for human error.
Ideally, we would let the sensors and the cloud do their thing. We humans could let the computers schedule visits from our comfortable, air-conditioned, mountain-view offices to the plants only on beautiful sunny days, and of course just moments before the equipment is set to breakdown, allowing us to extend its lifetime to the max.
However, for anyone who has ever looked under the veneer of the IIoT hype machine at the real products and technologies available to make that a reality, a clear path to that future seems much further out.
This article looks at some surprising and practical ways that leveraging mainstream, ubiquitous wireless technologies like Bluetooth and Wi-Fi can help organizations reach their vision of an IIoT-enabled plant today and be ready for the future. The article discusses:
- Where industrial-grade wireless sensor networks are today
- How savvy reliability teams are coupling consumer wireless technologies with the people and processes they already have to dodge the pitfalls and achieve similar IIoT results
- How intense competition and innovation of these consumer wireless technologies will make them industry-ready in the coming years
The best effort to solve the wireless sensor IIoT problem, so far, comes in the form of WirelessHART and the Industrial Automation Society’s 100.11a wireless technology standard, and with good reason: They were built by and for process automation users and vendors. The benefits include:
- Technology that is extremely reliable and can cover large, unfriendly environments with mesh networking and frequency hopping
- Security that is baked into the way the devices are commissioned and deployed
- Protocols designed to be industry-friendly by being multivendor and open
- Designed-in backward compatible support for hardwired protocols like HART
- Low-power consumption that means a device can operate battery-powered for five years or more in the field before needing attention
- Adding a wireless monitoring network in parallel to the controls network keeps critical operations firewalled from the internet
- Existing since 2007, the technologies are now officially battle-tested
As a result, process industry leaders like Emerson, Honeywell, Yokogawa and ABB have deployed many useful wireless sensor network solutions to help their customers reduce operating costs. One example is Emerson’s case study of helping Denka save 7 percent of their steam costs at a resin plant by monitoring steam traps with WirelessHART sensors and connecting that data to the cloud.
So why can’t businesses simply keep deploying more and more WirelessHART/ISA100.11a networks to bring more automation to factories and take more steps toward Warren Bennis’ vision?
The problem is cost. WirelessHART and ISA100.11a are great technologies worth every penny, but all those pennies and dollars add up pretty fast when these networks are taken to the next level with the number of sensors and devices that are involved in next-generation IIoT initiatives. Sensors that utilize the WirelessHART standard often cost more than $500 per unit and gateways cost several thousand dollars, which means that a single network deployment could cost $100,000 or more. If that number is multiplied by the number of systems that could benefit from this type of remote monitoring, it quickly becomes clear that the scope of IoT deployments that industry analysts are talking about for the coming years simply isn’t feasible cost-wise – at least not if only these existing technologies are used.
This pricing isn’t the result of nefarious business practices – anything but. It’s simply a function of this technology being developed specifically for the process automation world rather than for a mass market the way other wireless standards have been. The investment in the technology has been significant and it shows in how well it works and how it has improved with time, but the process automation world is a niche market in comparison to a true mass market like consumer wireless. It is simple economics that make WirelessHART and ISA100.11a cost what they do, and it is also simple economics that the price means that this technology only makes sense for some IIoT deployments – not all.
The result is that there is a very practical limit to how many wireless sensor networks industrial companies can deploy with WirelessHART/ISA100.11a networks while still having a responsible return on investment (ROI) for their projects. They tend to be truly mission-critical sensor networks aimed at the most important equipment and operational areas. There is so much more potential to these wireless networks, if only the economics made sense.
So, where do we go from here?
The best wireless technology today – A reliability engineer
Despite the IIoT hype, there are real people who work hard to help reduce operating costs by doing predictive maintenance and data analytics – and they’ve been doing this for decades. They accomplish it in large part by walking around these hot and noisy environments by choice. They know that all it takes is one piece of equipment to fail to cause real trouble, so until everything can literally be monitored via sensors, they’ll be out there checking on things themselves.
Their problem: Employers are asking them to keep driving up efficiency with fewer people.
This creates a scenario with two important realities:
- Reliability teams recognize they can and should benefit from remote intelligence offered with IIoT solutions.
- Since we have not yet arrived at the zero-person facility, maintenance routines still bring people in proximity to equipment on a regular basis.
How does one take advantage of the internet-required remote experts and data centers while simultaneously being disconnected from wireless infrastructure? Simple – bring a smartphone on the route.
The reality is that many of the most common equipment issues don’t go from healthy to broken in the time it takes to complete a route and get back to the office. By connecting to sensors and equipment in the field via a Bluetooth connection or a machine’s local Wi-Fi hotspot, the smartphone can store critical machine information in areas with no connectivity. When the operator is back in the office, that data can be forwarded to the cloud and benefit from the analytics in time to address an issue before it becomes critical.
Innovative industrial product companies are recognizing this reality and embracing it. An example of what is becoming possible as Bluetooth and Wi-Fi enter the industrial world is the work that ITT Goulds Pumps has done with its i-ALERT2 machine health monitor. This monitor uses Bluetooth Low Energy (BLE) as the foundation for a next-generation machine health monitor that allows engineers to identify potential problems with machinery before they become costly failures. It tracks vibrations, temperature, run-time hours and other key diagnostics – and it provides that information wirelessly to engineers in real-time via a mobile app that makes it all accessible and actionable on a smartphone or tablet.
Dan Kernan, the global product director for Aftermarket Solutions at ITT Goulds Pumps offered these insights about why his company is embracing Bluetooth:
“We recognized many years ago that we could differentiate ourselves and our pumps by embracing digital technologies, and making them standard on every pump. There are a lot of early warning signs to pump failure that can announce themselves as subtle changes in how the pump vibrates. Our customers recognized this as a valuable predictive tool, but many were struggling to turn it into something actionable. Identifying problems requires analyzing vibration data and interpreting what they mean. It’s like being a radiologist looking at CT scans. You have to have some experience looking at these things to identify the signal from the noise. Most facilities either have only a small number of people with this expertise, or they outsource the job to a third party. In either case, with hundreds or thousands of pumps in a plant and limited resources, they can only concentrate on the most critical, most expensive assets. We knew we could take advantage of digital technologies to help our customers monitor more of their pumps and avoid safety incidents and costly downtime.
“A majority of our customers were hitting ROI dead ends trying to install wireless sensor infrastructure to capture the vibration data. Portable tools that manually capture these data could do the job, but still required extra training to know how to use properly and avoid bad data. They already had people on the ground due to initiatives like operator-driven reliability, those people just weren’t always vibration experts. They needed an inexpensive, easy-to-use solution anyone on their team could use to capture the information and get it to their vibration analysts.
“We quickly recognized that the best way to do that was to use the high-tech, standard-issue connectivity tool everyone already had in their pocket — their smartphone. The i-ALERT2 monitor allows anyone doing their routes to capture the vibration data over a Bluetooth connection then share that data using the companion app connected to the cloud. We added even more value by building in some basic analytics intelligence to the sensor itself. The i-ALERT2 monitor can identify critical conditions and send alarm push notifications to smartphones in proximity. This way even a non-expert can know something needs to be done right away and can get in touch with someone who can help. And because consumer-grade products are driving high volumes for things like MEMS [microelectromechanical] sensors and Bluetooth connectivity modules, we can keep our hardware costs down.”
While Bluetooth and local Wi-Fi hotspots on machines can offer a practical way to wirelessly capture machine data and get it to the cloud by way of an operations team’s smartphones, is there a way to get any closer to Bennis’ IIoT vision without the added cost?
How car & home audio systems may help connect plants in the future
Familiarizing one’s self with Bluetooth and Wi-Fi may be beneficial in the long-term as well. The abundance of smartphones in people’s hands has nearly every industry – consumer products, medical, automotive, home automation and others – looking for ways to get on the bandwagon.
The result of this all-encompassing, all-industry race toward IIoT solutions is a huge investment in Bluetooth and Wi-Fi as technologies. The investment is far beyond anything imaginable just within process automation, let alone all industrial automation combined. The price of hardware is rapidly falling, and the capabilities of this cheap hardware are rapidly skyrocketing.
Several interesting technological developments are coming out of this race that are starting to directly address the challenges WirelessHART and ISA100.11a set out to solve in the process world a decade ago.
Briefly revisited, these challenges are: secure, reliable in difficult environments, deterministic and able to work over large areas.
When WirelessHART and ISA100.11a were introduced, comparing them to Bluetooth and Wi-Fi was apples to oranges. Bluetooth did have some of the desired robustness features like multichannel frequency hopping and reliable performance in crowded radio frequency space, but it was a personal area network, meant to pair one’s headset to his phone sitting on the desk next to him, not scale to an entire plant and deliver mission-critical data. Wi-Fi offered good range and excellent encryption and security (as it was relied on to protect patient data in hospitals), but as anyone trying to stream a movie while sitting in a Starbucks can attest – it cannot be relied on to always get data where it needs to go.
Fast forward to sometime around 2020, and you will see a number of now (in 2017) nascent capabilities of these networks in full bloom:
- Bluetooth 4.2 introduced huge security improvements in how devices pair and identify as a way to stymie advertisers trying to track consumer behavior in stores.
- Bluetooth 5.0, released in 2017, boasts longer range, better throughput and denser sensors to grow its abilities to service the home automation industry.
- Bluetooth Mesh, to be released in the coming months, creates an interoperable standard for mesh communications, makes anyone’s Bluetooth-capable product another agent able to improve the reliability of the overall network, another home automation improvement.
- Widespread adoption of 802.11ax, which greatly improves reliability and speed for areas with a high density of devices using improvements like MU-MIMO, developed for public Wi-Fi spaces like hospitals, stadiums, airports and cafes.
- Time Sensitive Networks (TSN), originally developed to synchronize microphones, monitors and speakers wirelessly at Beyoncé concerts then added on by the automotive industry to synchronize communications (as well as audio) between different vehicle systems, gets introduced to the industrial world to add deterministic communications over Wi-Fi.
- Wi-Fi Mesh becomes more popular as consumers want their Amazon Alexa in one room to be able to know what is going on and coordinate with the Amazon Alexa at the other end of the house.
- 802.11ah, known as HaLow, will extend Wi-Fi in to the 900 MHz band for low-power, long-range applications, another entrant designed for the smart home.
- You can see that these developments did not come from the industrial world, but the problems they are addressing are directly relevant. Furthermore, the pace of major advancements is dizzying compared to new releases and adoption of new specifications and technology within the industrial world.
- And perhaps the best part? Because of the low-margin, high-volume nature of the markets where these products do come from, more functionality does not come at a higher cost – in fact, each generation is usually cheaper than the last.
- Extrapolating that trend out, it will not be long before the capabilities of Bluetooth and Wi-Fi are knocking at the door of WirelessHART and ISA100.11a, but at a much lower price point.
The work that ITT Goulds Pumps is doing, and the technology examples of things to come are just some of many ways that show the potential for BLE and Wi-Fi for process automation. This is just the beginning. Maybe we’re not as far off from Warren Bennis’ vision of a factory, a person and a dog, as it may seem. With that said, it seems appropriate to end with one more quote from Mr. Bennis:
“In life, change is inevitable. In business, change is vital.”
Paul Elvikis is the strategic business development director for Industrial Automation at Laird, which provides a full range of modules and engineering services that simplify the process of using wireless technology, including embedded wireless solutions for industrial IoT. Elvikis oversees Laird’s development of solutions for the industry, bringing the next generation of wireless sensors and controls to market that meet the needs of challenging industrial environments. Prior to joining Laird, Elvikis worked for HMS Industrial Networks and Orion. He earned his bachelor’s and master’s of science in Mechanical/Industrial Engineering at the University of Illinois. Elvikis may be reached at firstname.lastname@example.org.