Emergency Supplies Distribution For Early Disaster Response Operations Under Demand Information Asymmetry

Quick response to the urgent demands in the affected areas after a disaster through a timely and effective distributing emergency supplies is of great importance in reducing disaster impact.In this study,we consider emergency supplies distribution for early disaster response operations under uncertainty,and propose a single-commodity,two-stage robust model that determines the number of supplies to be distributed from relief facilities to affected areas in a multi-sourcing disaster relief logistics network.In the early response stage after a disaster,the providers of relief-demand information take communities as the statistical standard,but the actual demanders include not only the community population but also the floating population.The resulting demand information asymmetry is the main reason for an uncertain demand,which is the uncertain parameter in our model.Meanwhile,the two stages are defined with respect to demand information asymmetry,and we use the upper bounds,the lower bounds,and the most likely values of uncertain demands to define an uncertainty set.The objective is to minimize the sum of the first-stage cost and the worst-case second-stage cost among all possible realizations of uncertain demands in the uncertainty set.In order to illustrate the advantage of the model,we use a case study about the 2010 Yushu earthquake in P.R.China to compare the robust model with the deterministic and two-stage stochastic models.The computational results demonstrate that the robust model outperforms the other two models for the same problem and the MIP equivalence of the robust model can solve larger scales problems more efficiently than the LP equivalence of the robust model.