Design and evaluation of disaster response organizational structure for resource demand and supply satisfaction

Abstract

Disaster response operations are critical, challenging tasks that decrease the devastating impacts of disasters in casualties and property damages. Even such a simple operation as providing resources often calls cooperation between delivery routers, suppliers and demanders. This cooperation has to be dynamic and complex in a disaster, adapting robustly to changing contexts, always efficient. This requires a solid organizational structure which is represented by network. This work computationally designs and evaluates alternative organizational structures as network for disaster response to resolve disconnects between resource demand and supply. Particularly, this research consists of 1) organizational structure designs with two distinct optimization techniques, 2) agent-based simulations that virtually replicate disaster response contexts, and 3) social network analysis that observes and interprets the relation between the structure and the performance under network perspectives. Simulations of Hurricane Katrina suggest optimized organizational structures improve resource deliveries. For instance, an organizational structure optimized to solve the multi-commodity network flow problem delivers resources more accurately by emphasizing the information broker organizations. Another organizational structure optimized to solve the resource delivery shortest path problem delivers more resources by strengthening the links of the leadership and edge organizations.