As companies shift to digital systems on the factory floor the first step is often preventive maintenance. In previous years, highly experienced factory hands would go around and listen to the motors and drives. They could hear the health of the machines. They could smell it. Those folks are retiring, and they’re being replaced with sensors and alarms tied into IoT systems.
This is changing the nature of the factory floor, said Saar Yoskovitz, CEO of Augury, a predictive maintenance company. He offers three predictions for manufacturing facilities during the coming year. For one, he sees a big future in fog computing as floods of data choke cloud storage and processing. He sees growth in the awareness of security and risk. Finally, he expects facility managers will make smarter decisions as they process greater data from their machines.
One of the changes Yoskovitz has recently observed in the manufacturing market is that smaller companies are now implementing predictive maintenance. “Predictive maintenance has been done for more than 30 years, but it’s been in the high-end market,” Yoskovitz told Design News. “We bring predictive maintenance to all markets. We attach sensors to chillers, auxiliary equipment, pumps, compression and send it to the cloud for analysis. Then we tell the user when something is wrong and what needs to be done to fix the machine.”
1.) Fog and Cloud Computing Will Bring Light to Manufacturing Facilities
According to Yoskovitz, the future of the industrial IoT will be a combination of cloud and fog computing: a hybrid solution where some computing happens on the edge – on the devices in the field (or fog computing) – and some computing happens in the cloud. Sensors will collect data from machines and monitor it for system health. The data will then be analyzed to determine options for improvement as well as predicting potential problems – before they happen.
The idea behind fog computing is that not all digital decisions and alarms need to go up to the cloud for analysis. “When you keep the algorithm at the machine level, you only send data to the cloud for updates. You don’t send all of it to the cloud,” said Yoskovitz. “You want the machine to respond quickly. Also, with fog computing, you can still work, since you don’t need the cloud to keep on working. How do the two systems – cloud and fog – work together? The analogy is that you bring the cloud down to the asset.”
2.) Awareness of Security and Risk Will Grow
With greater connectivity, you increase the risk, since you’re adding additional points of contact. Yet the IoT doesn’t necessarily present more risk, according to Yoskovitz. Instead, it presents a different kind of risk. “Facilities that implement devices connected to the IoT need to think about communication and the security protocols between devices: sensor-to-sensor communication, sensor-to-gateway communication,” said Yoskovitz. “This involves updating and maintaining all on-premise equipment to better secure the data.”
Yoskovitz noted that security is continually improving, which is reducing the risk of connectivity. “Having the right guards in place lets us use the system without putting ourselves at risk. There has been a lot of security innovation in the past five years,” said Yoskovitz. “You can protect the sensors with encryption chips on the hardware. You can use a communication protocol to make sure all the data that goes to the cloud is secure. You can use artificial intelligence to analyze the data that goes from the sensor to the cloud. You analyze it to detect anomalous behavior, so you can detect intrusion as it happens.”
3.) Facility Managers Will Make Smarter Decisions
Yoskovitz believes facilities managers will make more informed decisions about manufacturing operations during 2018. “As more facilities implement the necessary elements of connectivity, people will make smarter decisions,” said Yoskovitz. “First, facility managers will adopt more SaaS technologies and solutions to begin this process. They will then leverage computers and mobile phones for machine-to-machine communication via sensors to remotely run a critical facility.”
The promise of predictive maintenance is the ability to determine when to order spare parts and whether you need to shut down immediately. This is a shift from scheduled maintenance to condition-based maintenance. When you replace something, it’s not just because it’s scheduled. You know at any given movement what the condition of the equipment is. This changes how machines are run and operated.
To implement this, you add sensors to the equipment and analyze the data to determine condition. You need specific data – not necessarily tons of data. “The first wave of IoT was connecting as many devices as possible. Facility managers were overwhelmed by the onslaught of raw data. Connecting for connectivity sake is a flawed idea,” said Yoskovitz. “It’s about making better decisions based on actionable insight. For that, you need just the right data.”
ESC Silicon Valley is Back!
The Embedded Systems Conference (ESC) is back in Silicon Valley and it’s bigger than ever. Over three days, Dec. 5-7, 2017, receive in-depth education geared to drive a year’s worth of work. Uncover software design innovation, hardware breakthroughs, fresh IoT trends, product demos, and more that will change how you spend time and money on your next project.
Rob Spiegel has covered automation and control for 17 years, 15 of them for Design News. Other topics he has covered include supply chain technology, alternative energy, and cyber security. For 10 years, he was owner and publisher of the food magazine Chile Pepper.
Photo courtesy of Augury.