The purpose of this work is to increase the effectiveness and the efficiency of the operations carried out during the activities in the aeronautical maintenance and transformation processes. In particular, we examine the Non-Routine Card Resolution Process. An NRC is document issued when a fault, not expected, detected during the maintenance/transformation operations, is found. Typically to describe a defect we need of the aircraft zone (MajorZone, Sub-MajorZone, Zone, etc.) and/or the system (ATA system) and/or the coordinates, the component (sometimes with the sub-component) and the kind of the defect with its attributes. We name this set of information Non-Routine Card.
The costs and the efforts of the NRC management are of the same order of magnitude of the planned activity management (Job Card - JC). The process requires that corrective actions come from applicable technical documentation. There are many kinds of defects and technical documentation does not cover all possible cases. Often, we observed the operators refer to similar NRCs solved in the past.
We found two major issues: the enormous number of NRCs and the lack of a shared and concrete definition of similarity between NRCs. The latter is due to a distinctive feature of the NRCs. We need of many attributes in NRC and even a little change of the values leads to very different corrective actions. The operators find “similar” NRCs using their expertise.
The application we developed relies on the concept of dynamic taxonomy. The operator can find “similar” NRCs, with a little effort, using his own search criteria. The application has a very simple visual user interface and it can use, in very easy way, other search technique too.