A New Method for Assessing the Resiliency of Large, Complex Networks

Designing resilient and reliable networks is a principle concern of planners and private firms. Traffic congestion whether recurring or as the result of some aperiodic event is extremely costly. This paper describes an alternative process and a model for analyzing the resiliency of networks that address some of the shortcomings of more traditional approaches – e.g., the four-step modeling process used in transportation planning. It should be noted that the authors do not view this as a replacement to current approaches but rather as a complementary tool designed to augment analysis capabilities. The process that is described in this paper for analyzing the resiliency of a network involves at least three steps: 1. assessment or identification of important nodes and links according to different criteria 2. verification of critical nodes and links based on failure simulations and 3. consequence. Raster analysis, graph-theory principles and GIS are used to develop a model for carrying out each of these steps. The methods are demonstrated using two, large interdependent networks for a metropolitan area in the United States.