Aviation security remains a priority amid growing passenger traffic. Billions of people undergo screening annually, facing risks ranging from contraband to terrorism. Traditional methods, including manual checks and scanners, overload personnel and cause delays, reducing efficiency. The author proposes the implementation of artificial intelligence (AI) for predictive threat analysis based on big data. Models based on recurrent neural networks (RNN) and deep learning analyze historical incidents, traffic, weather, and behavior, allowing for resource optimization and focusing on high-risk areas. FAA reports (2022) indicate that 70% of incidents are due to human error. Russia has a Federal Aviation Security System (2019), integrating ICAO standards with local measures against terrorism and illegal movement. The key screening principles are a risk-based approach with measures adjusted according to threat levels. Methods include visual inspection, metal detectors, X-rays, body scanners, and manual screening of passengers and baggage. For ships and infrastructure, these include dog handlers, sensors, and patrols. AI complements these measures by predicting risks. Research in aviation demonstrates up to 95% accuracy in recognizing anomalies. In Russia, the focus is on reducing false alarms by 30% and integrating with existing systems. Currently, there are a number of challenges, including data shortages and privacy concerns. Adaptation to Russian airports, taking into account geopolitics, testing, and collaboration with regulators are recommended. The purpose of this article is to develop and test an AI model for predictive risk analysis in airport screening procedures.
artificial intelligence, predictive analytics, aviation security, screening procedures, machine learning
1. Mezhdunarodnaya organizaciya grazhdanskoy aviacii. URL: https://www.un.org/ru/ecosoc/icao/. (data obrascheniya: 21.12.2025 g.).
2. Godovoy otchet NextGen. URL: NextGen Annual Report Fiscal Year 2022. (data obrascheniya: 21.12.2025 g.).
3. «Federal'naya sistema obespecheniya aviacionnoy bezopasnosti (Nacional'naya programma aviacionnoy bezopasnosti)» (odobreno Mezhvedomstvennoy komissiey po aviacionnoy bezopasnosti, bezopasnosti poletov grazhdanskoy aviacii i uproscheniyu formal'nostey 04.04.2019). URL: https://www.consultant.ru/document/cons_doc_LAW_328563/. (data obrascheniya: 21.12.2025 g.).
4. Federal'noe agentstvo vozdushnogo transporta Rosaviaciya. URL: https://favt.gov.ru/. (data obrascheniya: 21.12.2025 g.).
5. Al'gamdi M.I. Opredelenie vliyaniya osvedomlennosti o kiberbezopasnosti na povedenie sotrudnikov: primer Saudovskoy Aravii. Materialy segodnya: Materialy konferencii, 2022. – S. 122.
6. Bolyuta E.A. Pavlov D.Yu., Samysheva O.A. Preimuschestva i nedostatki primeneniya sovremennyh tehnologiy v oblasti aviacionnoy bezopasnosti // Aktual'nye voprosy sovremennoy nauki: teoriya, tehnologiya, metodologiya i praktika: Sbornik nauchnyh statey po materialam X Mezhdunarodnoy nauchno-prakticheskoy konferencii, Ufa, 27 dekabrya 2022 goda. Tom Chast' 1. – Ufa: Obschestvo s ogranichennoy otvetstvennost'yu «Nauchno-izdatel'skiy centr «Vestnik nauki», 2022. – S. 110-115.
7. Kolbasina A.A., Sevryukova E.M., Burcev D.S. Sravnitel'nyy analiz avtomatizirovannyh sistem predpoletnogo dosmotra passazhirov v aeroportu // Ekonomika i biznes: teoriya i praktika, 2025. – № 3 (121).
8. Sostoyanie bezopasnosti poletov v grazhdanskoy aviacii gosudarstv-uchastnikov soglasheniya o grazhdanskoy aviacii i ob ispol'zovanii vozdushnogo prostranstva v 2024 g. URL: https://mak-iac.org/upload/iblock/54e/gh74inwm1ez2bj2o89jqqwthgf8nii5k/bp-24.pdf. (data obrascheniya: 21.12.2025 g.).
