For the collaborative optimization of multi-depot vehicle routing problem with split pickup and cross-dock door assignment, considering joint distribution of the multiple cross-dock center, split pickup and cross-docking operation, an optimization model is established to minimize the sum of vehicle dispatching cost, vehicle fuel cost, fixed and transportation cost of forklift in the cross-dock center and time window penalty cost. According to the characteristics of the problem, the hybrid genetic algorithm is designed through adding the directed optimization method of the particle swarm algorithm, the evolutionary reversal and elitist retention strategies are also used to improve the solution quality. The validity of the model and algorithm is verified by multiple examples, the influence of distribution mode and vehicle type on the distribution scheme formulation are also analyzed. The results show that the proposed model can effectively reduce the cross-dock centers’ operation cost. The research results expand and enrich the research of the vehicle routing problem with cross-docking, and provide theoretical basis for logistics enterprises with multiple cross-centers to make use of resources efficiently and formulate the decision distribution scheme scientifically.