The location problem is closely related to people's lives and property, and the harmonious development of society. Based on the discussion of location principle, to meet the demand of emergency rescue point as large, the mathematical model for location problem of emergency service facilities is established. And then the Genetic-Simulated Annealing Algorithm (GSAA) for solving this model is designed, which enriches searching behaviour in optimisation process, and has strong abilities of exploration and exploitation in large real search spaces. In this algorithm, genetic algorithm controls the search direction, and simulated annealing algorithm plays an important role in local optimum convergence. Case study shows that the algorithm is efficient, and it not only gets optimal location results quickly and accurately, but also clearly gets the afflicted points that are covered by the selected facilities. So the model and the algorithm have strong practicability.