딥러닝 기반 항공안전관리시스템 예측모델 개발

Development of Prediction Model for Aviation Safety Management System Based on Deep Learning

This study was conducted with the aim of improving the safety performance of the aviation safety management system. The aviation safety management system is developing into a system that preemptively prevents risks through future predictions. In order to perform safety performance efficiently, data-based on D3M decision-making and Forecasting methodology that can predict the future must be applied. This study reflects the future-oriented paradigm of the aviation safety management system and intends to contribute to maintaining and developing aviation safety. The basis for the current SPI calculation method is not clear, and the SPI system by frequency analysis is limited to be used in aviation safety indicators with complex causal relationships and needs improvement. The reason is that the paradigm of the approach is shifting to a predictive method, and the development of an advanced form of SPI system is required.