The maintenance and service records collected and maintained by the aerospace companies are a useful resource to the in-service engineers in providing their ongoing support of their aircrafts. Such records are typically semi-structured and contain useful information such as a description of the issue and references to correspondences and documentation generated during its resolution. The information in the database is frequently retrieved to aid resolution of newly reported issues. At present, engineers may rely on a keyword search in conjunction with a number field filters to retrieve relevant records from the database. It is believed that further values can be realised from the collection of these records for indicating recurrent and systemic issues which may not have been apparent previously. A faceted classification approach was implemented to enhance the retrieval and knowledge discovery from extensive aerospace in-service records. The retrieval mechanism afforded by faceted classification can expedite responses to urgent in-service issues as well as enable knowledge discovery that could potentially lead to root-cause findings and continuous improvement. The approach can be described as a structured text mining involving records preparation, construction of the classification schemes and data mining.
|Title of host publication||Foundations of Computational Intelligence Volume 6|
|Subtitle of host publication||Data Mining|
|Number of pages||19|
|Publication status||Published - 2009|
|Name||Studies in Computational Intelligence|