Improving the resilience of social infrastructure systems to an extreme event

Abstract

Presented in this thesis are several models to assist in the decision making process regarding the mitigation and restoration of social and civil infrastructure systems. These models utilize various techniques such as mixed-integer optimization, stochastic optimization, and Monte-Carlo simulation to identify vulnerabilities and interdependencies in these systems, to provide guidance in determining the best practices for the pre-event allocation of resources, and to support decisions regarding the restoration of damaged infrastructure components. The models in this thesis are presented along with an application using a comprehensive artificial community dataset called CLARC county. In addition, a detailed explanation of the translation of this research into decision support technology for the emergency management community is presented.

This thesis presents research conducted in the recovery of social infrastructure systems following an extreme event. A social infrastructure is a system that is primarily operated by people and which uses physical components only for support. Examples of social infrastructure systems include healthcare, emergency response, education, fuel distribution, etc. Individuals and organizations rely on the critical services that these social systems provide both during normal and disaster conditions. One important consideration in these social infrastructure systems are the relationships or interdependencies that exist among these systems as well as with and among the civil infrastructure systems such as power and communications.