Yusuf Secerdin
Resilient Humanitarian Transportation Network Design Against Hurricane Disruptions
Poster Presenter #48
Ph.D. Student in Industrial Engineering
From the logistics network perspective, hurricanes significantly differ from other natural disasters in that they emerge as progressive events sweeping across regions. This distinct aspect calls for new adaptive transportation design procedures across time and space. The significance of a resilient transportation network- either a public or a private network- arises when a transportation network is utilized for humanitarian purposes immediately after the disaster. Delivering relief cargo such as water, food, and fuel as well as critical commodities for private sector firms (mobile generators, communication equipment) is the main goal in the first 24-hours or 48-hours after the storm. In the case of Irma and Maria, an initiative was started to exchange information between companies and organizations to improve logistics operations in the affected areas under the coordination of FEMA along with the American Logistics Aid Network. Thus, any firm having a resilient transportation network may contribute more to any endeavor for humanitarian logistics activities without any delay for coordination. We propose a two-phase solution framework for multi-stage stochastic humanitarian air network design problem for a major carrier which contributes to disaster relief efforts in the post-disaster period in the Caribbean. Due to the possible impact of the hurricane, this air network might be disrupted partially/fully by an airport which is shut down. This affects both the inbound and the outbound delivery made to that airport/city. To overcome this, an operational contingency strategy including rerouting and rescheduling decisions would make the air network resilient against hurricane disruptions. We employ two distinct models that are specified as the outer and inner model in each corresponding phase for solving the overall problem sequentially. In the first phase of this approach, we develop a multi-stage stochastic programming model for the route (path) based network design problem without scheduling decisions. The inner model solves the service network design problem on a time-space network, which corresponds to the routing and scheduling of feeders for each scenario in each stage based on volume data provided by the outer model.