EFFICIENCY ANALYSIS OF THE USE OF ARTIFICIAL INTELLIGENCE FOR TRANSLATING AND INTERPRETING AERONAUTICAL MAINTENANCE MANUALS

• Grant number: 24/15910-7
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: February 01, 2025
• End date: January 31, 2026
• Field of knowledge: Linguistics, Literature and Arts - Literature - Modern Foreign Languages
• Principal Investigator: Daniela Terenzi
• Grantee: João Pedro Teodoro Batista
• Host Institution: Instituto Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP). Campus São Carlos. São Carlos , SP, Brazil

Abstract

With the advent of the 4th Industrial Revolution, advanced technologies, including Artificial Intelligence (AI), have revolutionized various sectors. In the aeronautics sector, AI has been applied to management and maintenance systems, but it has not yet been used to translate technical manuals, which are crucial to the safety and efficiency of operations. Considering that understanding such documents can be a challenge for maintenance technicians, since all of those in the aeronautics sector are written in English and 80% of technicians are not native speakers, and that these professionals have the possibility of using AI tools on a daily basis as resources to help them understand such documents, it is necessary to understand the results of this process. It is therefore important to compare the efficiency of manual translation and interpretation carried out by AI with that carried out by humans alone. The project proposed here aims to use artificial intelligence, specifically Chat GPT-4 developed by OpenAI, to verify the automatic translation of aircraft maintenance manuals, investigating its characteristics, capabilities and limitations. The justification for the project is based on data showing that communication and document interpretation problems contribute significantly to air accidents and on the possibility of using these tools by professionals who are not proficient in English and need to understand specific documents. The study will involve the selection of a maintenance manual, manual translation assisted by specialized tools, validation of this translation by an experienced professional, automated translation using AI, and comparison of the results of both methods. The aim is to determine whether AI is effective in interpreting these technical documents.