Most of the current mooring lines are composite lines, and they need to be assembled with clumps. The position and weight of the clump are usually designed based on experience. In order to reduce the amount of calculation and find the optimal parameters, the Particle Swarm Optimization-Back Propagation (PSO-BP) neural network is built, and the sample data of neural network is determined by the orthogonal experiment. The workload of the finite element calculation is reduced by the generalization ability of the neural network as much as possible. The neural network prediction results are applied to the genetic algorithm. The minimum sum of the tension of all mooring lines is taken as the optimization goal to search for the optimal solution within a given range. The sum of the optimized safety factors of 8 mooring lines increases by 36.07, which is higher than that of the randomly selected clump weight scheme. This method can provide ideas for the optimization problems with a large amount of calculation.