Abstract
The International Union of Railway provides an annually safety report highlighting that human factor is one of the main causes of railway accidents every year. Consequently, the study of human reliability is fundamental, and it must be included within a complete reliability assessment for every railway-related system. However, currently RARA (Railway Action Reliability Assessment) is the only approach available in literature that considers human task specifically customized for railway applications. The main disadvantages of RARA are the impact of expert’s subjectivity and the difficulty of a numerical assessment for the model parameters in absence of an exhaustive error and accident database. This manuscript introduces an innovative fuzzy method for the assessment of human factor in safety-critical systems for railway applications to address the problems highlighted above. Fuzzy logic allows to simplify the assessment of the model parameters by means of linguistic variables more resemblant to human cognitive process. Moreover, it deals with uncertain and incomplete data much better than classical deterministic approach and it minimizes the subjectivity of the analyst evaluation. The output of the proposed algorithm is the result of a fuzzy interval arithmetic, $\alpha $ -cut theory and centroid defuzzification procedure. The proposed method has been applied to the human operations carried out on a railway signaling system. Four human tasks and two scenarios have been simulated to analyze the performance of the proposed algorithm. Finally, the results of the method are compared with the classical RARA procedure underline compliant results obtain with a simpler, less complex and more intuitive approach.
Highlights
Railway engineering is a complex field in which many aspects of work throughout the complete system life cycle are performed by human operators
Human Error Probability (HEP) evaluation since it severely contributes to the overall RAMS (Reliability, Availability, Maintainability, Safety) parameters of the system under study
The nominal HEP of each task has been selected within the range of the admissible value provided by Railway Action Reliability Assessment (RARA) following the guidelines of the company that manage operation and maintenance of the Automatic Train Protection (ATP) under analysis
Summary
Railway engineering is a complex field in which many aspects of work throughout the complete system life cycle are performed by human operators. The quality of the estimation is extremely related to the experience of the analyst performing the assessment Trying to solve these needs, this research aims at finding an innovative HRA method developed for railway engineering applications which must be able to solve the major drawbacks of the RARA method available in literature, namely the impact of analyst subjectivity and the complexity of a numerical assessment of the influence factors. The proposed approach simplifies the assessment of the HEP by means of linguistic variables to describe both the probability of error and the affect level of each internal and external factor In this way, it is possible to develop a simple and effective tool for HEP assessment which minimizes the subjectivity and the impact of analyst experience. The rest of the paper is organized as follows: section II presents a detailed literature review regarding HRA, Section III presents the proposed fuzzy-based approach along with the developed tool, in section IV the proposed method is applied to a real case study and Section V presents the VOLUME 9, 2021 results validation through comparison with an existing HRA method
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
