Abstract: The relief preparation of adequate facilities and resources in areas frequently struck by natural disasters is a cornerstone that guarantees the effective execution of relief operations. Within a limited budget, a two-stage stochastic programming model is developed with an aim to minimizing the expected number of casualties. The first stage of the model is focused on pre-disaster strategic decisions including the expansion of hospitals and shelters, the site selection of warehouses, the allocation of emergency supplies and relief vehicles, and the second stage is concerned with operational decisions on emergency logistics plans, including taking the wounded from disaster areas to hospitals, transferring the victims from disaster areas to shelters, sending the medical supplies from warehouses to hospitals and shelters, and dispatching the living materials from warehouses to resettlement areas. The computational results of an illustrative case prove the validity of the proposed stochastic optimization model.