An integration analysis of joint distribution path optimization and cost sharing was carried out, and the bi-level programming mathematical model of joint distribution operation mode under the condition of multi-distribution centers and multi-goods was established. Therein, the upper level model was used to optimize the delivery path, and the lower level model was used to share the cost delivery in the alliance. A genetic algorithm was designed to solve the proposed problem. The initial population was generated based on the sweep algorithm, and the cost sharing mechanism was designed by the Shapley value method. To avoid falling into the local optimal solution, the internal disturbance and the external disturbance process were introduced during the evolution progress. Results show that the total cost of the alliance and the cost of each distribution center are greatly reduced under the condition of joint distribution, which can provide decision reference for joint distribution practice.