Abstract: As a natural gas pipeline operates, its failure rate increases over the time. However, the failure characteristics of natural gas pipeline cannot be met when the traditional preventive maintenance with one fixed period is carried out. To improve the safety of natural gas pipeline, an elastic maintenance period model was studied. The reliability of natural gas pipeline and the economy of its maintenance were combined to increase the reliability of the system and control the maintenance costs. According to the change of system failure rate and the theories related with preventive maintenance, a reliability-maintenance cost multi-objective optimization model was established for the system by introducing the age reduction factor. Then, based on the specific engineering cases, the specific calculation examples were solved in this model by applying the multi-objective particle swarm optimization algorithm in Matlab platform. And thus, the elastic preventive maintenance period with reliability and economy as the objectives was calculated. The research results provide the theoretical basis for the maintenance and management of natural gas pipelines.