Proceedings of the International Telemetry Conference 2008, San Diego, CA (2008)
Predictive maintenance is the use of inspection and data analysis to perform maintenance when the need is indicated by unit performance. The method can provide significant cost savings to operators while preserving a high level of system performance and readiness. The monitoring of complex instrumented systems creates vast collections of observations. Identifying predictors of maintenance requires expert knowledge and the ability to process large data sets. This paper describes a novel application of constraint-based data mining applied to the prediction of exceedence conditions in twin engine aircraft. The approach extends the extract, transformation, and load process with Domain Aggregate Approximation to encode expert knowledge in the transformation step. A data mining workbench enables a human expert to pose hypotheses that constrain a multi-variate data mining process.
For More Information
(Please include your name, address, organization, and the paper reference. Requests without this information will not be honored.)