Deep Learning-Based Predictive Models For Rail Signaling And Control Systems: Improving Operational Efficiency And Safety

Abstract

The main objective of this study is to understand the current operational structure and features of Turkey's rail signaling and control systems. From this subtle information, potential deep learning techniques that improve both operational efficiency and transportation security are proposed. To this end, current signaling and control structures are detailed, while the current migration priorities and the accessibility of deep learning are emphasized. As a result of this assessment, the possibilities created by deep learning algorithms in the signaling and control systems of the signaling territories are highlighted. The limitations and opportunities identified at the end of the study are based on the data set. So far, there is no deep-learning model for the prototypical data set. However, the data set that will emerge at the end of the application and pre-configuration, such as variable extraction, will make deep learning models in this field more probable.