Predictive maintenance sits between reactive repair, which waits for breakdowns, and preventive maintenance, which replaces parts on a schedule whether they need it or not. Sensors capture vibration, temperature, current and acoustic signals; condition monitoring tracks their trends; and models flag the early signatures of bearing wear, imbalance or clogging. Maintenance is then planned for a convenient window.
The approach pays off most on expensive or critical assets where unplanned downtime is costly, such as compressors or wind turbine gearboxes. It is harder on equipment with few historical failures, because there is little data to learn from, which is where anomaly detection against a healthy baseline helps. See Predictive maintenance with IoT and AI for a practical walkthrough.
Part of the IoT Glossary. Spotted an error or a missing term? Tell us.