Using improved Group 2 and Linguistic Z-numbers combined approach to analyze the causes of railway passenger train derailment accident

In recent years, several passenger train derailment accidents occurred in the world. In order to guarantee the reliability and normal operations of railway passenger trains, an improved Group 2 and Linguistic Z-numbers (IG2-LZs) combined approach is proposed to analyze the causes of railway passenger train derailment accident (RPTDA). The weight and attribute value of each cause of accident are calculated by IG2 and LZs, respectively. In order to relax the three limitations of previous G2 approach, the Fault Tree is applied to figure out the least important cause of accident from all causes of accident; the minimum nonempty intersection sets, which can reflect the opinions of experts, are introduced to improve the interval assignment; left offset rate and right offset rate are introduced to improve the risk attitude factor from the experts. The Puyuma case happened in Taiwan is applied as the case study. The calculation results show that, the failure of Automatic Train Protection (ATP) belongs to the most contributed cause. A comparison of Analytic Hierarchy Process (AHP), IG2-Triangular Fuzzy Number and previous G2-LZs show that the IG2-LZs is more reliable and reasonable than AHP, IG2-TFN or G2-LZs. (c) 2021 Elsevier Inc. All rights reserved.

Automated summary

An improved Group 2 and Linguistic Z-numbers combined approach is proposed to analyze the causes of railway passenger train derailment accidents. Various methods such as Fault Tree, minimum nonempty intersection sets, left offset rate, and right offset rate are applied to improve the analysis of accident causes and risk assessment. The study uses the Puyuma case as a case study and shows that the IG2-LZs approach is more reliable and reasonable compared to other methods such as AHP, IG2-TFN, or G2-LZs.