Predictive maintenance can grow far beyond traditional condition monitoring when the data from equipment is gathered through the Industrial Internet of Things (IIoT) and then stored and processed through Big Data analysis systems, such as IBM’s Watson.

“When you gather the regular maintenance data, you can build a history of the data. Then you use algorithms to detect anomalous behavior in the historical data. In time, you learn that when you see this anomaly, you know—based on the history—that this component is likely to fail in the next 10 to 17 days,” Tom Craven, VP of product strategy at RRAMAC Connected Systems, told Design News.

Source: designnews.com
Author: Rob Spiegel