In this paper, by considering the benefits of logistics planners and retailers, a bi-level programming model is proposed to describe the location inventory problem. In the bi-level programming model, the upper level is used to determine the optimal location by minimizing the location cost of logistics planners and the lower gives a demand allocation by minimizing inventory and transportation costs of retailers with stochastic demands. A new heuristics algorithm combining genetic algorithm and simulated annealing algorithm with memory container and historical solution set is proposed to solve t...