OPTIMIZING LAST MILE DELIVERY OF DISASTER RELIEF SUPPLIES FOR OAHU, HAWAII

Abstract

The state of Hawaii and island of Oahu are vulnerable to supply chain disruptions, such that any major disaster will require emergency food distribution to local populations. However, organizations across federal, state, and local government have proposed different distribution concepts that rely on different points of distribution (PODs), where no concept on their own may be sufficient to feed Oahu communities. In this work, we develop a data set and series of models to test these concepts for populations on Windward Oahu. We develop two models that select optimal POD locations for either a pickup concept, where populations drive to receive food, or a delivery concept, where food is brought to communities. We further study hybrid concepts that prefer either pickup or delivery. Our results show that ideal plans for Windward Oahu will prefer delivery PODs and utilize 17 hybrid PODs that serve both pickup and delivery purposes. Moreover, we identify four POD locations that would be used no matter which distribution concept is implemented. We recommend developing a hybrid distribution concept centered on delivery that can be tested and implemented at these four locations.