The rapid and automatic solution of ship stowage scheme is of great practical significance for the efficient delivery of marine equipment. Based on the characteristics of ro-ro loading of emergency rescue vehicles, the ship stowage problem of multi-layer and multi-type rescue vehicles under different emergency tasks is abstracted as a multi-objective optimization model. The algorithm of ship stowage is designed based on particle swarm optimization, and the optimization of stowage is divided into two stages. In the first stage of ship stowage, adaptive mutation and local search idea represented by Tabu search are integrated into PSO algorithm and the above algorithm is discretized; in the second stage, the adaptive mutation and heuristic rules are integrated into the PSO o optimize the allocation of vehicle parking positions. As the determination of stowage scheme needs to be dynamically changed according to the changes of the disaster situation, different rescue tasks will correspond to different target weights. The effectiveness and feasibility of the multi-objective ship stowage algorithm are verified by a case study.