Computational Technology Resources - CCC

T. Itagaki, T. Takata, T. Fukuda, H. Mochizuki, S. Takahashi, H. Nakamura, "Applicability of Generative AI to Railway Signalling Control Based on the Evolution of Fail-Safe Design", in J. Pombo, (Editor), "Proceedings of the Seventh International Conference on Railway Technology: Research, Development and Maintenance ", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 15, Paper 24.2, 2026, doi:10.4203/ccc.15.24.2

Keywords: railway signalling, fail-safe, interlocking, block system, fail-safe logic, generative AI, intrinsic control.

Abstract

Safety design in railway signalling systems has evolved along two major paths: the incorporation of safety-related non-functional requirements into systems from the outset and the prevention of the recurrence of past accidents. Historically, many such accidents have been attributed to human error. Japanese railway safety systems have achieved a high level of sophistication by reducing human involvement while continuously integrating technological advancements. As technology has advanced, the concept of safety design has also evolved significantly. Recently, attention has been directed toward constructing more rational systems based on the concept of intrinsic control, in which safety is ensured through consistency verification at system interfaces. Furthermore, recent technological developments have prompted interest in the application of Generative AI to railway signalling control systems. However, due to challenges such as limited explainability and unpredictability, its adoption remains cautious. This paper proposes an alternative approach in which Generative AI is not used for control decisions but for evaluating the consistency of data sequences, thereby enhancing safety without compromising system reliability.

download the full-text of this paper (PDF, 11 pages, 400 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCC CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Conferences ISSN 2753-3239 CCC: 15 PROCEEDINGS OF THE SEVENTH INTERNATIONAL CONFERENCE ON RAILWAY TECHNOLOGY: RESEARCH, DEVELOPMENT AND MAINTENANCE Edited by: J. Pombo Paper 24.2 Applicability of Generative AI to Railway Signalling Control Based on the Evolution of Fail-Safe Design T. Itagaki¹, T. Takata¹, T. Fukuda², H. Mochizuki², S. Takahashi² and H. Nakamura² ¹, Kyosan Electric Manufacturing Co., Ltd., Yokohama, Japan ²College of Science and Technology, Nihon University, Japan doi:10.4203/ccc.15.24.2 Full Bibliographic Reference for this paper T. Itagaki, T. Takata, T. Fukuda, H. Mochizuki, S. Takahashi, H. Nakamura, "Applicability of Generative AI to Railway Signalling Control Based on the Evolution of Fail-Safe Design", in J. Pombo, (Editor), "Proceedings of the Seventh International Conference on Railway Technology: Research, Development and Maintenance ", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 15, Paper 24.2, 2026, doi:10.4203/ccc.15.24.2 Keywords: railway signalling, fail-safe, interlocking, block system, fail-safe logic, generative AI, intrinsic control. Abstract Safety design in railway signalling systems has evolved along two major paths: the incorporation of safety-related non-functional requirements into systems from the outset and the prevention of the recurrence of past accidents. Historically, many such accidents have been attributed to human error. Japanese railway safety systems have achieved a high level of sophistication by reducing human involvement while continuously integrating technological advancements. As technology has advanced, the concept of safety design has also evolved significantly. Recently, attention has been directed toward constructing more rational systems based on the concept of intrinsic control, in which safety is ensured through consistency verification at system interfaces. Furthermore, recent technological developments have prompted interest in the application of Generative AI to railway signalling control systems. However, due to challenges such as limited explainability and unpredictability, its adoption remains cautious. This paper proposes an alternative approach in which Generative AI is not used for control decisions but for evaluating the consistency of data sequences, thereby enhancing safety without compromising system reliability. download the full-text of this paper (PDF, 11 pages, 400 Kb) go to the previous paper go to the next paper return to the table of contents return to the volume description
Back to top	©Civil-Comp Limited 2023 - terms & conditions