This paper studies the relief-allocation problem for short-supply relief goods in humanitarian logistics. Considering the conflict between ever-increasing demand and short supply due to typhoon evolution, the Markov decision approach is employed to demonstrate the ever-varying demand, and a dynamic relief-allocation model is proposed. Then the model is solved using the binary particle swarm optimization algorithm. Finally, the proposed model is applied to a case study of a post-typhoon rescue. Computational results show the potential advantages of the Markov decision approach for its dynamic relief-allocation adjustments, resulting in gradual increase and minimum quantity of total demand.