Modeling network vulnerability of urban rail transit under cascading failures: A Coupled Map Lattices approach

In terms of urban rail transit network vulnerability, most studies have focused on the changes of network topology and functional characteristics under disruptive events while ignoring the impacts of cascading failures on network performance. In this study, an improved Coupled Map Lattices model is proposed to address the evolution process of cascading failures on rail transit network. Particularly, the model quantifies important factors of cascading failures integrating quantitative failure spreading mode, anti-risk ability of stations, and perturbation index. The rail transit network vulnerability is then modeled addressing numbers of failed stations, degree of affected passengers, and loss of passenger travel time as a result of cascading failures. The methodology is applied to the metro network of Shanghai, China. Results show that the propagation of network cascading failure depends on the station's anti-risk ability, geographical location, and type of adjacent stations. Stations located in the urban center area are more susceptible to failures of adjacent stations. Furthermore, higher affected passenger volume and station coupling strength would lead to more serious and rapid spreading of failures, and thus more vulnerable of the metro system. However, the anti-risk ability of stations could effectively resist the spread of failures in the early stage, and stations with higher topology redundancy demonstrate stronger anti-risk ability.