A general equilibrium approach to model the infrastructure interdependencies
Extreme events over the past decade in the USA, ranging from the 9/11 terror attacks to the 2003 Northeast power blackout to the 2005 hurricanes, have highlighted the urgent need to understand the security, engineering and economic implications of the interdependencies among civil infrastructure systems. The need is further highlighted by the challenges arising from the capacity requirements of rapid urbanization and the renewal of aging infrastructure. This motivates a new generation of models that can incorporate multiple infrastructure systems in a single framework, capture their interdependencies, and conduct decision-making analysis for more efficient, effective, robust, and resilient infrastructure systems.^ This dissertation proposes a generalized modeling framework that combines a multilayer network concept with market-based economic approach to capture the interdependencies among the various civil infrastructure systems with disparate physical and operational characteristics. Thereby, the systems are modeled as individual networks connected through flows representing market interactions, and the different types of infrastructure interdependencies are captured through supply-demand mechanisms. ^ Based on a multilayer infrastructure network (MIN) concept, the modeling framework uses the computable general equilibrium (CGE) theory and its spatial extension (SCGE) to formulate the equilibrium and disequilibrium problems. Fundamental introduction and comprehensive literature review are conducted on the related topics. The conceptual framework for the mathematical formulation is discussed, followed by the description of static equilibrium models. The equivalent variational inequality (VI) formulation is also proposed to analyze the mathematical properties such as the existence and uniqueness of solution. The dynamics issues are then discussed, followed by the description of a multi-period dynamic model, which enables the disequilibrium analysis. Numerical experiments are conducted to illustrate the characteristics of the proposed models and their capability to capture the various types of interdependencies. A case study involving the transportation (airline) and energy (fuel) systems is presented to provide insights on the implications of the infrastructure interdependencies to individual systems in the real-world problem context.^
Srinivas Peeta, Purdue University.
Engineering, Civil|Transportation|Urban and Regional Planning