To solve the split delivery vehicle routing problem, a hybrid heuristic algorithm combining genetic algorithm and ant colony algorithm is adopted. This algorithm takes the genetic algorithm as the main structure and uses the ant colony algorithm to construct the population and improve the crossover operation. In the algorithm, the total number of vehicles used to complete the transportation task and the total distance traveled by all vehicles are taken as the fitness function. The ant colony algorithm designs the path construction principle with the vehicle capacity, the maximum one-way driving distance, and the number of empty containers at the replenishment nodes as the constraints. Combined with the genetic algorithm, the crossover and mutation methods are improved to enhance the convergence speed and ensure the generalization ability of the algorithm. By randomly generating split delivery transportation problems of different scales, this algorithm is compared with three other heuristic algorithms. The accuracy of the algorithm is verified through small-scale problems, and the hybrid heuristic algorithm performs better in solving large-scale problems. The proposed hybrid heuristic algorithm can effectively solve the split delivery transportation problem.