Print Email Facebook Twitter Towards Automated Classification of Aircraft Maintenance Documentation Title Towards Automated Classification of Aircraft Maintenance Documentation Author Scherp, L. Contributor Verhagen, W.J.C. (mentor) Faculty Aerospace Engineering Department Control & Operations Programme Air Transport & Aerospace Operations Date 2016-12-16 Abstract Airline maintenance technicians (AMTs) often still rely on paper-based documentation when performing a task. In the limited time during unscheduled tasks, obtaining the right information from these documents can take up to 30% of the available time. To speed up this process, in this thesis it is explored how an automated system can be created that can pre-classify digital maintenance documents so an AMT can quickly retrieve task-relevant documents. Using a dataset of 14.000 Airworthiness Directives, the main objective is to compare document classification techniques ranging from Boolean to Machine Learning on various document properties (e.g. language), to determine a recommended approach for such a system. Subject maintenance documentationaircraft maintenancemachine learningautomated documentation system To reference this document use: http://resolver.tudelft.nl/uuid:9a3d4bdb-00a4-4ddf-b643-ebc11b2fa3b2 Embargo date 2017-06-01 Part of collection Student theses Document type master thesis Rights (c) 2016 Scherp, L. Files PDF Thesis_LennartScherp_1357 ... 161130.pdf 20.76 MB Close viewer /islandora/object/uuid:9a3d4bdb-00a4-4ddf-b643-ebc11b2fa3b2/datastream/OBJ/view