Bayesian network-based human error reliability assessment of derailments. (May 2020)

• Record Type: Journal Article
• Authors: Dindar, Serdar; Kaewunruen, Sakdirat; An, Min
• Is Part Of: Reliability engineering & system safety. Volume 197(2020)
• Publication Date: 2020-05
• Subjects: Human-errors -- Railway operation -- Derailment -- Bayesian network -- Fuzzy logic; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2020.106825
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: Human errors causing derailments at switches and crossings were identified and classified. A novel methodology dealing with the errors was proposed. A novel DAG (Directed Acyclic Graph) built through Bayesian network was proposed. The risks of errors were identified and analysed using new mathematical expressions. Risk is prioritised by a most-to-least-critical importance ranking of human errors. Abstract: The knowledge acquired in relation to failures associated with components has made significant contributions to the development of components with increased reliability, as well as a reduction in the number of rail incidents caused by certain system defects. These new systems have led to innovative developments in both the operations and technology of rail networks. Hence, rail employees must now function in conditions that have high complexity that are hard to comprehend. The risk of failure caused by human error (such as by dispatchers, train crews and track engineers) has developed into a significant safety problem. This study is the world first to provide novel insights into better understanding human errors, which result in derailments at rail turnouts. A most- to-least-critical importance ranking of these errors is established throughout a novel risk management technique. Moreover, the new findings and recommendations of this research study have a strong potential for industry to improve the reliability of rail operation, and avoid safety concerns regarding train derailments at rail turnouts.