Abstract: In order to study the reliability allocation scheme with the lowest cost under the same reliability of NC grinder, an improved particle swarm optimization algorithm is proposed by combining the objective function and limited optimization conditions. Firstly, the basic theory of improved particle swarm optimization is introduced. Secondly, the reliability cost function model of CNC grinding machine is established by combining the generalized reliability cost function model. Thirdly, the reliability relationship between subsystems of CNC grinding machine is analyzed, and the non-linear programming model of reliability allocation is constructed. Finally, the improved particle swarm optimization method is used to solve the model, which is compared with genetic algorithm. The results show that under the condition of ensuring reliability, the improved particle swarm optimization algorithm is more reasonable. The research results have been fed back to the manufacturer and approved, which is of great significance to improve the reliability level of CNC grinder.