Predictive maintenance of equipment based on artificial intelligence models holds great promise. An asset management system could alert a plant or facilities manager that a piece of equipment is about to fail. However, predictions are only as good as the data they are based on. A false alarm triggered by a lack of data or inaccurate data can lead to incorrect actions and additional costs. Bottom line: predictive maintenance will only be able to predict those problems for which it already has accurate data.

The U.S. Navy is planning to use artificial intelligence-based predictive maintenance. The technology aims to eliminate unplanned downtime for critical shipboard systems powered by pumps and motors. Failure of such devices in a land-based plant or facility is disruptive enough. At sea, and in combat situations, breakdowns can endanger the crew. The initiative, which is headed by the Defense Innovation Unit, builds on previous Department of Defense AI-based predictive maintenance programs for the Army and Air Force.

Facilities are still sorting out the current value of predictive analytics in regards to maintenance, repair and operations (MRO). It turns out the key to success is having current asset data to complement historical data.