To make a reasonable node location in the urban underground logistics system and improve the service level of the urban underground logistics， the characteristics of urban underground logistics node are analyzed， the problem of the location of the underground lo-g istics node could be abstracted as Location Allocation Problems.With further separation a-nalysis， the paper made Allocation Problem inserted in the Location part.The upper layer planning model from the point of view of the decisionmaker and the lower layer planning model from the customer's point of view are constructed.A combination of genetical go-rithm was used to solve the bi-level programming model.The results show that the bi一level programming model considering the cost of the logistics department and the customer 's benefits can effectively solve the urban underground logistics node location problem， and the algorithm of the model is more efficient.