Worst-case vulnerability assessment and mitigation model of urban utility tunnels

Constructing utility tunnels would bring new threats, among which the terrorist threat has been identified as the most serious one. Because utility tunnels increase geographical interdependencies among urban infrastructure systems (UISs), and attacking a tunnel segment could cause simultaneous failure of all its carried systems. This paper models the terrorist threat as worst-case failure, and formulates a tri-level mathematical model to assess and mitigate the worst-case vulnerability of urban utility tunnels that carry interdependent UISs. Exact solution of the model is obtained by a column-and-constraint generation based decomposition algorithm. Real utility tunnels that carry interdependent power and water systems in Tianjin Eco-city, China, are used to demonstrate the proposed approach. Results show that there exists a threshold value of the defense budget B-D. If B-D is less than that threshold, the utility tunnels greatly benefit the attacker and increase the power and water systems' worst-case vulnerability. If B-D is equal to that threshold, the utility tunnels almost do not affect the worst-case vulnerability. If B-D is larger than that threshold, the utility tunnels benefit the defender and reduce the worst-case vulnerability.