Prognostics

Forecast potential equipment failures

Prognostics algorithms allow you to avoid equipment failure by monitoring sensor data from machines to predict when a failure event will happen. Based on those predictions, you can adjust maintenance schedules. These prognostics algorithms provide an alternative to conventional preventive maintenance programs in which the maintenance schedule is determined by a prescribed timeline.

Prognostics algorithms enable customers and equipment manufacturers to:

  • Reduce equipment downtime by identifying issues before failure, thereby extending equipment lifetime
  • Avoid the costs of unnecessary maintenance by scheduling equipment service only when necessary
  • Bring equipment back online faster by determining the root cause of impending failures and faults

Prognostics algorithms are critical to the success of predictive maintenance programs. Temperature, pressure, voltage, noise, or vibration measurements are collected using sensors. This data is processed using various statistical and signal processing techniques to extract features called condition indicators. To monitor the health of the equipment, you can compare these condition indicators with established markers of faulty conditions using data clustering and classification or other machine learning techniques. You can also use the condition indicators as inputs to remaining useful life (RUL) estimation models to train prognostics algorithms. The RUL model used—similarity-based, trend-based, or survival-based—depends on the kind of data available. The end result is a prognostics algorithm that can classify and predict the next failure event and provide a confidence bound on that prediction.

Prognostics algorithm development workflow.

Once validated, prognostics algorithms can be operationalized in an IT environment such as a server or cloud. Alternatively, prognostics algorithms can be implemented in an embedded system directly on the equipment, enabling faster response times and significantly reducing the amount of data sent over the network.

For additional information, see Predictive Maintenance Toolbox™, Statistics and Machine Learning Toolbox™, and Deep Learning Toolbox™.



See also: data analytics, unsupervised learning, predictive modeling, prescriptive analytics, Predictive Maintenance Toolbox, Parallel Computing Toolbox, Signal Processing Toolbox, Image Processing Toolbox, Statistics and Machine Learning Toolbox, Deep Learning Toolbox, MATLAB, Database Toolbox

Three Ways to Estimate Remaining Useful Life: Predictive Maintenance with MATLAB