Data mining for prediction of aircraft component replacement.

The operation and maintenance of modern sensor-equipped systems such as passenger aircraft generate vast amounts of numerical and symbolic data. Learning models from this data to predict problems with component may lead to considerable saving, reducing the number of delays, and increasing the overall level of safety. Several data mining techniques exist to learn models from vast amount of data. However, the use of these techniques to infer the desired models from the data obtained during the operation and maintenance of aircraft is extremely challenging. Difficulties that need to be addressed include: data gathering, data labeling, data and model integration, and model evaluation. This paper presents an approach that addresses these issues. We also report results from the application of this approach to build models that predict problems for a variety of aircraft components.L'utilisation et l'entretien de syst\ue8mes sensoriels modernes tels qu'un a\ue9ronef de passagers g\ue9n\ue8rent de grandes quantit\ue9s de donn\ue9es num\ue9riques et symboliques. D\ue9gager des mod\ue8les de ces donn\ue9es en vue de pr\ue9dire les probl\ue8mes li\ue9s aux composants pourrait permettre de r\ue9aliser des \ue9conomies consid\ue9rables, de r\ue9duire les retards et d'am\ue9liorer la s\ue9curit\ue9. Il existe plusieurs techniques d'exploration en profondeur de donn\ue9es qui servent \ue0 d\ue9gager des mod\ue8les \ue0 partir de grandes quantit\ue9s de donn\ue9es. Cependant, l'utilisation de ces techniques pour obtenir les mod\ue8les souhait\ue9s \ue0 partir des donn\ue9es produites pendant l'utilisation et l'entretien de l'avion pose un grand d\ue9fi. Il faut notamment s'attaquer aux difficult\ue9s suivantes : collecte de donn\ue9es, \ue9tiquetage des donn\ue9es, int\ue9gration des donn\ue9es et des mod\ue8les et \ue9valuation des mod\ue8les. Le document d\ue9crit une m\ue9thode qui r\ue8gle ces probl\ue8mes. Nous pr\ue9sentons aussi les r\ue9sultats issus de l'utilisation de cette m\ue9thode pour constituer des mod\ue8les qui permettent de pr\ue9voir les probl\ue8mes li\ue9s \ue0 une gamme de composants d'a\ue9ronef.NRC publication: Ye