Vulnerability assessment in urban metro systems based on an improved cloud model and a Bayesian network

Hongyu Chen,Qiping Shen,Zongbao Feng,Yang Liu

DOI: https://doi.org/10.1016/j.scs.2023.104823

IF: 11.7

2023-11-01

Sustainable Cities and Society

Abstract:To effectively evaluate the vulnerability of urban rail transit operations, a hybrid method that combines an improved cloud model and Bayesian network is proposed. This research consists of four parts: evaluation system development, model establishment, result aggregation, and analysis. The developed improved cloud-Bayesian network model is composed of 13 root factors and 4 s-level factors. The Wuhan Metro is adopted as a case study to provide instructions and verify the proposed method. The results indicate the following: (1) Line 2 is the most vulnerable line in the case; (2) The equipment-related factor (D2) is the most significant second-level variable in the vulnerability management of urban rail transit operation; (3) The employee professional level (X2), equipment anti-interference ability (X6), and urban rail transit line density (X11) display high correlations with the corresponding second-level factors; and (4) Peak duration rate (X3), platform passenger density (X4), urban rail transit line density (X11) and disaster seriousness level (X12), especially X4, are considered the key factors when urban rail transit reaches a high vulnerability level. Accordingly, corresponding countermeasures are proposed. The results show that the research conclusion is consistent with the actual situation, and the proposed method can provide a reference for other similar circuits.

energy & fuels,construction & building technology,green & sustainable science & technology