An Affinely Adjustable Robust Optimization approach to emergency logistics distribution under uncertain demands

After disaster, effective distribution of relief commodities to the affected areas is vital to minimize the loss. Generally speaking, the exact demand data are hard to obtain immediately after the disaster, which will cause difficulties to the decision-making process. In this paper, we present a prediction method of the relief demands after an earthquake. We also propose a distribution model considering the satisfaction rate of the relief demands and distribution cost. The uncertain demands are addressed by the Affinely Adjustable Robust Counterpart (AARC) method when the model is solved. Finally, a numerical experiment is given to demonstrate the computational efficiency of the proposed model.