RAM and Safety Analysis of Railway Signalling Systems

Abstract

Demonstrating compliance with RAM and Safety requirements is a prerequisite before railway application products are qualified for their intended application. EN 50126 requires documented evidence which shows how railway application products continue meeting the specified requirements in the event of both systematic and random hardware failures. The standard demands to carry out repetitive risk analysis processes at several stages of the lifecycle to determine the risk associated with the hazards and to establish a process for ongoing risk management. Several formulas work well for RAM and Safety analysis of high demand systems; however it is still very important to describe all the scenarios related to design and operational philosophies, maintenance policies, notion of downtimes, failure detection mechanisms, testing procedures etc and interpret all parameters of the formulas accordingly.For that matter, Markov modeling technique provides rather good understanding of scenarios and results from its simplified formulas are adequately accurate. The author presents a method for RAM and Safety analysis of railway signalling systems and reviews available analytical formulas for quantifying RAMS targets. Event tree is proposed to model the sequential propagation of operational scenarios into their potential outcomes or consequences and the chronological execution of safety functions in response to initiating events. An integrated approach of Markov modeling, reliability block diagram and even tree is used to calculate the hazard rate or probability of dangerous failures per hour (PFH); which is the target measure of loss of safety in high demand systems.

The proposed methodology has been applied to demonstrate compliance of railway signalling systems with top level RAM and Safety requirements set by the Norwegian railway authority.Two safety critical functions namely; signal light and train detection safety functions are chosen for analysis and the result shows that they are compliant with the stated RAM and Safety requirements. Moreover, they fulfill the requirements for SIL 4 certification based on random hardware failures. This master thesis provides rather detailed and comprehensive quantitative safety and reliability analysis on the railway interlocking system in line with the requirements of EN 50126. The proposed approach appears suitable to analyze failure frequencies of high demand systems in general and railway signalling systems in particular as event trees clearly show how development of scenarios spread-out through in depth layers of safety functions constituting the overall safety system. Moreover, this approach enables the user to have a full understanding of the system to be analyzed and helps to identify the strengths and weaknesses of the system.

RAM and Safety Analysis of Railway Signalling Systems