As the internet of things gains traction, manufacturers are starting to use the technologies to transform critical processes -- particularly in areas such as product and plant floor maintenance, quality management, energy efficiency and product development.
IoT is being cast in a starring role in everything from smart connected cities to wearable systems and autonomous cars, but it is expected to have a significant effect in the industrial sector as a means for manufacturers to reduce costs, garner efficiencies and drive additional revenue streams through new business models. According to IC Insights, revenue from industrial internet of things (IIoT) spending surged from $6.4 billion in 2012 to $12.4 billion in 2015. Cisco is projecting that manufacturing will account for 27% of the projected $14.4 trillion IoT market between 2013 and 2022, with smart factories contributing $1.95 trillion of the total value at stake in the same timeframe.
While manufacturers have long had access to data collected on the plant floor, it's typically been locked away in proprietary manufacturing software silos, restricting their ability to leverage it for decision making, according to Matt Wells, product general manager for automation software at GE Digital, based in San Ramon, Calif. That changes with IoT, which makes it far easier to collect and manage large amounts of manufacturing data not just in a single factory, but across multiple production sites through the cloud, he said. When paired with analytics, companies will gain better insights, allowing them to optimize plant operations, reduce quality defects and perform preventative maintenance, according to Wells.
"The ability for companies to make better business decisions is what will drive the explosive use of IoT and the industrial cloud," he said.
IoT: The ears and eyes of the plant floor
Already, GE customers are using IIoT and analytics for asset performance management -- the process of applying analytics to predict and prevent catastrophic failure on large-scale plant floor equipment, like boilers or compressors. Typically, manufacturers have processes in place to divert production in the case of failed equipment, but there is still associated downtime and unexpected costs, Wells said. "What if you could identify that a particular asset would fail up to a month in advance?" he asked. "Customers could more effectively schedule maintenance and plan for outages, ordering parts and scheduling people, so there would be zero impact on capacity and production."
GE customers are also beginning to use IIoT analytics and machine learning capabilities for energy-monitoring applications, he said. In this case, sensors measure conditions like input flow and pressure settings, and when combined with machine learning, manufacturers can determine whether a piece of equipment is running at optimal performance. "By measuring where it is on the energy curve, you can ensure that piece of equipment consumes less energy, which is impactful to manufacturers who are keen on lowering costs," he explained.
Siemens PLM Software sees the role of IoT in the plant as the means for finding out what's going on, according to Alastair Orchard, vice president of the company's Digital Enterprise project. "One of the factors that has stopped us in the past from delivering digital goodness to the customer is the relatively laborious task of equipping factories with all the necessary sensors and getting data organized and contextualized so it is available for simulation and analytics," he explained. "IoT allows us to sprinkle devices around the factory quickly and cost-effectively to find out what's going on without going through the multiyear process of reorganizing the shop floor."
New opportunities through industrial IoT
Eric van Gemerenvice president of R&D, Flowserve Corp.
For Flowserve Corp., an Irving, Texas, maker of large-scale industrial pumps, valves and seals, IoT is opening up the door to new business models, allowing it to sell preventative maintenance services, which, in turn, foster tighter relationships with customers, said Eric van Gemeren, the company's vice president of research and development. Through a combination of sensored equipment and machine learning and analytics capabilities, Flowserve is able to accurately pinpoint the root cause of problems, helping customers avoid unnecessary plant floor shutdowns for bogus alerts, he said.
"In the old world with raw sensor data, you didn't have the information to determine if and when you needed to shut down," he said. "For an operator in an oil refinery to shut down to find out what's going on, it could cost as much as $1 million an hour."
Machine learning provides answers to those questions, enabling plant operators to keep equipment running longer; reduce unnecessary maintenance costs, because you're not fixing what's not broken; improve operator safety; and reduce energy consumption, according to van Gemeren. "IoT provides the end-user customer and the operator of the machinery the ability to make decisions they weren't able to before," he explained.
How to create your supply chain traceability strategy
Manufacturing gets a digital transformation
IoT vs. IIoT: How to tell them apart