A novel mathematical model for the classic facility location problem is presented in this study. This paper tries to allocate the customers to the facilities in a way that the total costs are reduced and the welfare level of the customers is increased. As in the real world, the demands of the customers are uncertain, we have considered the customers’ demand probabilistic. Then, in order to prepare a robust solution for the problem, we introduced some scenarios and formulated the two-stage robust model of the problem. Finally, the two well-known genetic and simulated annealing algorithms are employed to solve the proposed model.