On the use of Natural Language Processing for improving automatic maintenance report labelling in the aviation field

The use of Natural Language Processing (NLP) is spreading more and more everyday and in almost every sector; it is applied for example to automatize activities, improve efficiency, and support data-driven decisions. As with other sectors, in view of the digital revolution brought by Industry 4.0, maintenance in aviation should be supported by NLP in many different activities; precisely, this paper aims to demonstrate the importance of it in the improvement of data collection and analysis. In fact, the host Maintenance, Repair, and Overhaul (MRO) company counts with an enormous quantity of maintenance data available that, at the moment, are not suitable for data analysis due to the presence of many errors in classification; the analyzed models have the objective to support MRO companies in classifying data into the correct physical zone of the aircraft.