Prepositioning emergency supplies to support disaster relief: a case study using stochastic programming

This paper studies the strategic problem of designing emergency supply networks to support disaster relief over a planning horizon. The problem addresses decisions on the location and number of distribution centres needed, their capacity, and the quantity of each emergency item to keep in stock. It builds on a case study inspired by real-world data obtained from the North Carolina Emergency Management Division (NCEM) and the Federal Emergency Management Agency (FEMA). To tackle the problem, a scenariobased approach is proposed involving three phases: disaster scenario generation, design generation and design evaluation. Disasters are modelled as stochastic processes and a Monte Carlo procedure is derived to generate plausible catastrophic scenarios. Based on this detailed representation of disasters, a multi-phase modelling framework is proposed to design the emergency supply network. The two-stage stochastic programming model proposed is solved using a sample average approximation method. This scenario-based solution approach is applied to the case study to generate plausible scenarios, to produce alternative designs and to evaluate them on a set of performance measures in order to select the best design.