Simulation-Optimization of Emergency Supply Pre-Positioning for Flood Response in South-South Nigeria

Pre-positioning of emergency supplies before predictable flood events can dramatically reduce response times and improve humanitarian outcomes in flood-prone regions, but optimal pre-positioning locations and quantities depend on uncertain flood impact scenarios. This study develops a simulation-optimization framework for pre-positioning emergency relief supplies in preparation for flood events in Nigeria's South South geopolitical zone, which experiences annual flooding affecting multiple states. A scenario-based stochastic optimization model determines pre-positioning quantities and depot locations to minimize expected unmet demand across flood scenarios, subject to budget, storage capacity, and perishability constraints. Flood extent and affected population scenarios are generated from a hydrological simulation model calibrated using 15 years of flood event records from the Nigeria Hydrological Services Agency. Emergency supply demand functions are derived from historical relief operation data. The optimization model is solved using a sample average approximation approach combined with a Lagrangian relaxation decomposition for computational tractability. Results from 500 sampled flood scenarios demonstrate that the optimal pre-positioning plan reduces expected unmet demand by 53 percent compared to reactive post-flood procurement, at 22 percent higher pre-event cost. The optimal depot configuration identifies three strategic locations, two in Rivers State and one in Delta State, that minimize expected response travel time across the range of simulated flood extents. Recommendations include embedding the simulation-optimization model in the National Emergency Management Agency's annual flood preparedness cycle. Keywords: simulation-optimization, emergency pre-positioning, flood response, South South Nigeria, humanitarian logistics