A Heuristic Method for Identifying Near-Optimal Defending Strategies for a Road Network Subject to Traffic Congestion

In this study we investigate the vulnerability of road networks to interdictions. We consider that an intentional attacker wants to maximize the congestion level on the network by interdicting some links of the network. The drivers are assumed to be selfish and always choose the route that minimizes the individual travel cost. In this case, network traffic flow follows user equilibrium (UE) traffic assignment, which however is affected by the interdiction initiated by the attacker. In this problem, the role of the defender is to minimize the damage that can be caused by an attack. A heuristic method is developed to solve for near-optimal attack (interdiction) and defense strategies, the effectiveness of which is demonstrated by simulation results. Numerical experiments are conducted to examine factors that influence the application of this method. Specifically, we compare the efficiency of this method when applied to different Cartesian grid-like networks. Managerial insights into the vulnerability and defense of road networks are drawn from the analysis.