Editor's Pick (1 - 4 of 8)
The Evolution of Manufacturing Processes
Industrial Scientific Blazes a Trail in IoT
IoT Platforms: What are they and do you need one?
Preparing IT for the Internet of Things
Can We Survive The IoT Data Tsunami?
By Cyril Perducat, Executive VP, IoT & Digital Transformation, Schneider Electric
Think about this: an average-sized refinery has about 50,000 sensors teeming with data. A large one? 500,000. Once the cost of computing and sensors plummeted, even aging industrial facilities could claim a seat at the IoT table. In a smart manufacturing facility, thousands of sensors spewdata related to conditions such as temperature, pump speed, vibration, power conditions — and the list goes on.
But is innovation’s industrial IoT hitting us with a massive tsunami of data? In the last few years, the rush to stay ahead was all about infrastructure. From wifi to wired machines, connectivity was the golden bridge of getting from old-school manufacturing to industry 4.0. Modernization efforts are no small feat, and many, many companies are still in the thick of it. Some are only just beginning their digital transformation.
No matter where you are, there is a big pot of gold waiting for you on the other side. And it’s not (big) data. Instead, it’s better business.
We all face the weighty question: How do I leverage data to solve operational problems? We know that data can offer valuable insights into the overall “health” of a system. Since data isn’t actionable without context, the challenge is to aggregate, contextualize, and analyze all that data in a meaningful, dynamic way. Well beyond traditional static graphics and data charts. The goal is to be able to act quickly.
Here is where artificial intelligence (AI)will lead the way. AI will accelerate the value of IoT. In a nutshell, a combined set of AI technical approaches collectively enables more intelligent systems to drive better decision-making — not in a descriptive way but in a more predictive and prescriptive one. This approach ultimately creates new and unique experiences for our customers.
For example, connected buildings generate massive amounts of data. By enabling us to build better energy efficiency models, data from hundreds of smart building scan transform what we know about any facility’s energy efficiency opportunities. Some estimate that 82 percent of buildings’ energy efficiency potential has yet to be tapped (). AI is the key to unlock this potential: extract data then transform it into value. That’s the gold mine of the 21st century.
Can we create data-based models in a classical way? No way! We therefore are turning to AI, machine learning (an AI technology that gives computers the ability to learn and adjust based on patterns), and deep learning (technology that mimics the way the human brain operates).
At Schneider Electric, we’ve innovated a way to survive the data tsunami
These innovations turn data into insights and, ultimately, into action to improve any building’s energy efficiency.
How many buildings does your enterprise own or manage? See how the landscape of bottom-line savings from energy alone is becoming much more attractive? This is just one example of what going beyond the sensors can do.
A more dynamic IoT world
Now envision any industrial machine. There is a lot of maintenance required under the hood. The traditional approach has been to monitor the machine and then fix the problem in a reactive way. Connected machines, by contrast, enable more proactive maintenance. Thanks to predictive analytics, a customer service representative can know whether a part is about to fail and ship that part before the operator even receives an alert. Schneider Electric’s Avantis PRiSM solution, which is based on machine learning, can diagnose equipment issues and provide early warning notification – days, weeks, or months before failure. We can even classify risks based on their potential financial or operational impact.
But what does that operator do when the part arrives? Coupled with advancements such as smart bot in mobility, artificial intelligence can transform the operator’s experience where the action is. After having identified a machine with a QR code, a beacon, or even automatic image recognition, an operator can connect with the digital twin of that machine. That’s augmented reality. Now imagine access to all documentations, customized online support, diagnostics, etc., all at once. That’s what I call situational operational maintenance on AI steroids! If the operator sees infrared using mixed reality-enabled applications, he or she not only could clearly visualize that there’s an abnormally hot part that needs immediate attention based on contextual analysis and proactive diagnosis. The operator also could see the automatic analysis of the potential root cause, all processed by AI using operational data combined together. Thanks to insights gleaned from machine learning and AI, this information is readily transformed into action. That future is just around the corner
Virtual reality (VR) goes even deeper. In addition to advancing proactive maintenance, VR can be used for training applications. Here is where the quality of the experience is much more relevant and important than the technology itself. VR is especially helpful for closing the gaps from retiring experienced workforces and advanced skill shortages. And in a variety of process industries, VR has huge potential for improving safety. Specifically, VR lets you run simulated, “what if?” scenarios without affecting safety and uptime. You can know consequences and change course accordingly. Those VR experiences are made even closer to reality by using data models generated by actual running real-life processes. To me, that’s beyond exciting.
Our IoT lifeboat
At Schneider Electric, we’ve innovated a way to survive the data tsunami. EcoStructure is our open IoT architecture. It integrates connected products; edge control; and apps, analytics, and services, and it works with third-party hardware and systems so users can turn data into valuable information. We have solutions for industries, buildings, data centers, and energy infrastructure.
There’s no doubt that IoT and AI are fast-moving and changing constantly. To get a handle here, I highly recommend a pragmatic approach. Learn by doing. Look at where transforming data into business impact is going to have the most value, and then race ahead accordingly. This approach will take trying, perhaps failing, and then trying again to discover what works for your business. At Schneider Electric, we can partner with you on that journey. I promise you that in the long run, your digital transformation will be well worth it.