Air Liquide can produce an extra 500 tons of output per year thanks to DataHub bridging and a DataHub script written by Software Toolbox.
A recent article in Food Engineering collected insights and opinions from executives at a number of industrial automation and control companies regarding how IIoT, Industrie 4.0 and digitization of the plant floor will play out in the area of food and beverage manufacturing.
The shared view was that these new initiatives are worth pursuing in food processing, as long as they are balanced with caution and good judgement. The article said, “While no one interviewed would suggest that a food or beverage manufacturer convert all its manufacturing software systems to an industrial internet of things (IIoT) platform in one fell swoop, many would suggest that to remain competitive in a fast-changing consumer product environment, it wouldn’t be a bad idea—for those that haven’t already done so—to embark on the ‘digitalization journey’ now.”
Skkynet’s contribution to the conversation focused on the value of connecting real-time production data to management, and what that might look like in the context of food processing. We also shared our thoughts on a number of other topics, such as the value of secure-by-design data communications, but these were outside the scope of the final published article.
In the late 1980s I was working at company in the USA that sold natural gas through direct purchase. Always on the lookout for new opportunities in the energy field, at one point they gave me a special assignment to research solar energy. We needed to know when solar would be cost-effective, competitive with coal, oil, gas, or nuclear.
In those pre-Internet days, I had to head over to the nearby university library and pore through scientific journals and economic publications to come up with some predictions. From all I could gather, the experts seemed to agree that solar would take at least 5 years before it would be worthwhile to invest in. So the project went on hold, but I was eagerly looking forward to a clean, renewable source of energy to become widely available in the early ’90s.
Well, it has actually taken closer to 30 years, but it appears that the tide is finally turning. A recent report from the International Renewable Energy Agency (IRENA) says that renewable power, including solar and wind, were more affordable in 2017 than ever before, and are now strongly competitive in many locations and applications.
A few IRENA report findings
“Renewable power generation costs continue to fall and are already very competitive to meet needs for new capacity.”
“The levelized cost of electricity (LCOE) from solar photovoltaics (PV) decreased by 69% between 2010 and 2016 – coming well into the cost range of fossil fuels.”
“Onshore wind, whose costs fell 18% in the same period, provides very competitive electricity, with projects routinely commissioned nowadays at USD 0.04/kWh.”
“As installation accelerates, the cost equation for renewables just gets better and better. With every doubling of cumulative installed capacity for onshore wind, investment costs drop by 9% while the resulting electricity becomes 15% cheaper.”
I find these statistics very encouraging, because now more than ever, the world needs abundant sources of energy at affordable prices. This is the energy that will power industry, commerce, and households for decades and centuries to come.
Data connectivity plays a role
I’m also pleased that Skkynet is responsible, in a quiet but important way, for some of these cost reductions. With increasingly more efficient and secure data connectivity, our customers who install and monitor wind turbines or solar panels are improving their quality of service and cutting their costs. Ultimately these savings are passed on to the consumers and industrial users of the energy. As the costs continue to drop, investment in renewables will increase.
Of course, conventional power plants, pipelines, and offshore oil platforms are also cutting their costs through secure remote monitoring and supervisory control. Improved access to production data benefits everyone across the board. So, I expect that the changeover to renewables will continue in the same gradual, steady way into the foreseeable future. Let’s see what the next 30 years have in store.
After years of riding high on the Gartner Hype Cycle, Industrial IoT (IIoT) is beginning to take shape in various ways. Early adopters tend to be large companies who have identified specific applications in which IoT connectivity provides an immediate advantage. The Internet of Things Institute recently named Top 20 Industrial IoT Applications, giving an overview of the best of what is happening. All of these are interesting, and we’d like share our thoughts on a few that you may not have heard of elsewhere.
Compressed Air as a Service
The Kaeser Kompressoren company in Germany has been manufacturing and selling air compressors for almost 100 years. Lately they have adopted an IIoT perspective, and have changed their business model. Now they provide compressed air as a service. Instead of selling their equipment, they install it at a customer site and sell its ability to compress air.
Among other things, this requires a mental shift when calculating where their revenues come from. Previously, when the customer owned the machinery, Kaeser could make money on repair services. Now that Kaeser owns the equipment, repairs have become a cost center, and it is in their interest to keep those costs as low as possible. Since they they started working under this business model, the company has been using IIoT technologies to sustain a healthy predictive maintenance (PdM) program. The cost savings revert directly to Kaeser.
This ability to adapt, to transform business models and capitalize on the Industrial IoT applications will set the leaders apart from the followers in the next few years as the IIoT moves from hype to reality.
Keeping Track of Tools
How many screws does it take to build a commercial airliner? How tightly must each one be turned? What’s the right tool for the job, and how should it be calibrated? A joint IIoT project between Bosch, Cisco, National Instruments, and Tech Mahindra coordinated through the Industrial Internet Consortium is demonstrating the value of the IIoT in answering those questions.
At a testbed location that simulates aircraft assembly, Bosch cordless screwdrivers are connected wirelessly via National Instruments technology and send identification data about themselves, as well as torque data about the screw they are tightening, to a central database. Their exact physical location is calculated using a triangulation technology from Cisco. An integration program from a Tech Mahindra program uses the screwdriver’s location data to look up the amount of torque specified for that screw at that location, and configures the screwdriver accordingly. When the operator moves to a different location on the aircraft body to drive other types of screws, the screwdriver gets reconfigured automatically and precisely.
These four companies working together highlight the value of cooperation in developing Industrial IoT applications, especially at the beginning stages. Many successful early adopters have emphasized the value of partnerships and collaboration. Those who take a do-it-yourself approach often find the IoT more complicated to implement than expected.
Automated Mining and Haulage Systems
The largest private railroad in Australia with over 1,700 kilometres of track is owned and operated by the Rio Tinto mining company. Using IIoT technologies, the company is now running unmanned, autonomous trains along this line, hauling iron ore from mines in the Pilbara region to ports along the north coast. The pilot project will be expanded to full service next year, as the world’s first fully-autonomous heavy haul, long distance railway system.
This initiative is just one of several IIoT-related initiatives that Rio Tinto is developing. They are also pioneering in the operation of autonomous trucks and drilling systems for their mines, and are even looking at self-navigating ships to cut the cost of delivering their products worldwide.
Not every company is in Rio Tinto’s position, but their broad vision, wide range of IoT applications, and obvious success can be an inspiration for all of us. The message is clear: Industrial IoT is not only possible, it is profitable. Learning from these examples, anyone venturing into this space needs to consider the opportunities and challenges unique to their industry and company, learn how and when to work with others, and then start taking action to gain the maximum benefit from Industrial IoT.
In our previous blog we looked at how the Industrial IoT (IIoT) is transforming the field of Asset Performance Management (APM). One key aspect of APM is the ability to maintain physical assets like hardware and machinery. As the IIoT bolsters APM, we can expect it to also impact plant maintenance systems in a positive way. In fact, experts in the field are suggesting that the IIoT is going to usher in a new era, an era of wide-scale Predictive Maintenance (PdM) for industrial systems.
Until now, there were essentially two approaches to equipment maintenance: run-to-failure, or fix it even if it “ain’t broke.” These two approaches were necessary because it was often difficult or expensive to figure out when a machine would break down. Both approaches were costly, though, because they meant either wasting time and materials repairing or replacing parts that still had plenty of life left, or shutting down the plant when something broke―or possibly both.
An Alternative – PdM
The alternative is Predictive Maintenance (PdM), which means regularly monitoring machines, and repairing them just before they break. Regular testing can be done by having staff periodically walk around the plant and take readings on portable instruments. This has value, but more effective is continual monitoring―installing sensors directly onto machines to detect changes in vibration, temperature, and sound on the machine, and connecting them by wire to APM software. Until recently this was prohibitively costly for all but the most expensive machinery. But the advent of tiny, low-cost, wireless sensors, cloud computing, and new AI (Artificial Intelligence) solutions―in other words, IIoT technologies―have put the cost of efficient PdM within reach of far more companies, to be used on far more equipment, than ever before.
“In today’s competitive industrial world, predictive maintenance (PdM) is no longer a nice-to-have; it has become a necessity,” says Abhinav Khushraj of Petasense. “Traditional PdM methods have several limitations. However advancements in wireless, cloud and AI technology are disrupting the way PdM has been done in recent decades.”
Valuable as this new implementation of PdM may be, it’s possible that it’s just the tip of the iceberg. Low-cost, IIoT-based PdM could cause a shift in thinking about preventative maintenance that could affect the whole enterprise, according to Eitan Vesely of Presenso. He identifies a number of trends, such as PdM becoming more holistic in scope, covering multiple assets, and becoming a source of top-line growth, all thanks to Industrie 4.0 and IIoT.
In a recent blog Vesely said, “With Industry 4.0, executives are starting to consider the impact on top-line revenue from their big-data investments. With the shift from Industry 3.0 to Industry 4.0, metrics such as improved uptime and higher-production yield rates are replacing downtime as the driving force for investments in this technology category.”
Infrastructure is Needed
All of these initiatives require infrastructure. The IIoT data that powers this far-reaching PdM must be transmitted and received securely, robustly, and quickly. The wide variety of sensors with their multiple data protocols need to connect within the plant and, via gateway or directly, to the cloud. Analytical engines rely on seamless connections to real-time streaming data. Every step is needed, every step adds value. As the new vision of IIoT-powered PdM begins take shape, Skkynet is there, helping to make it happen.
How well is our equipment performing? Are we getting the best mileage from it? When will machine/pump/motor X fail? What can we do better? How can we even find out? These are the kinds of questions that keep production managers awake at night, and that may eventually lead them to consider implementing Asset Performance Management.
The term “Asset Performance Management” or APM, has been broadly defined by MESA International as “an approach to managing the optimal deployment of assets to maximize profitability and predictability in product supply.” In other words, APM means making sure your equipment works, and works well.
The need for APM is clear. Asset failure costs millions of dollars in lost production each year. But that’s not all. Workers who lack of critical data about the state of their equipment can make expensive mistakes. Broken or poorly-functioning machinery can cause accidents. Failure to manage assets properly will eventually lead to higher insurance costs.
A New Approach to APM
Until recently, APM has been used mainly for high-cost, mission-critical machinery. Data collection was based on scheduled manual readings of sensors mounted on machines, a costly, time-consuming effort. Performance stats for less expensive assets were typically derived from plant walk-throughs, operator experience, and educated guesses. In some cases it was more cost effective to simply allow equipment to run until failure, and then replace it.
The advent of the IIoT is changing all of that. As the cost of sensors drop, as wireless technologies improve, and as Big Data systems come online, it is now becoming cost-effective to monitor more and smaller assets, and do it in real time. Doing APM in real time allows managers to run advanced analytics on the data, and do predictive maintenance. Now, instead of shutting down a process to replace a burnt-out fan motor, or guessing when the motor might need to be replaced, an engineer can run the motor until just before failure, and then switch it out at the optimal time.
Commenting about APM, Rich Carpenter, the Chief Technology Strategist for GE Intelligence Platforms, said, “The purpose of running advanced analytics is to have early detection of problems, so that we can prevent in real time, rather than react in real time.” GE follows a simple, four-step process for IoT-based APM:
- Connect to sensors on the plant equipment or for the control system.
- Send the sensor data via DMZ or other protected environment to the cloud.
- Organize the data, according to type of customer, site, machine, etc.
- Run advanced analytics to manage the assets.
Hartford Steam Boiler (HSB), an industrial insurance company, has a vision that IoT-powered APM may transform their business into becoming an industrial service provider. Instead of simply insuring against mechanical failure, by putting IoT-connected sensors on their customers equipment, the company can check performance, find the risk of breakdown, and recommend timely replacement or repair. “The Internet of Things is the next industrial revolution and we have to position ourselves,” said Greg Barats, CEO of HSB. “IoT start-ups are a fantastic way of jump-starting your thinking.”
What visionaries like GE and HSB have in common is an understanding of the potential of IIoT to be a game-changer for APM. We share that vision. In fact, we see APM as a logical application for IIoT technology, and supply the necessary software and services to securely access the necessary data in real time, to make it happen.