Relief distribution is a focus of attention of emergency  response after a public emergency. In this paper, a bi-level programming  model is developed to describe the transfer facility location and joint   transportation problem in a two-echelon relief distribution system.  The upper level is to minimize the delivery time of relief materials,  and the lower level is to maximize the fairness of relief material  distribution. Then a hybrid genetic algorithm with heuristic rules
  and a two-phase decoded operation, i.e., transfer facility locating   in the first phase and vehicle routing in the second phase,  is proposed to solve the model. Finally, the validity of the  model and algorithm is demonstrated by a numerical example based  on Wenchuan earthquake relief distribution. The results show that  the proposed genetic algorithm has good performance and is suitable for the joint location-transportation problem in relief distribution.