A novel multi-objective endocrine particle swarm optimization algorithm (MOEPSO) based on the regulation of endocrine system is proposed. In the method, the releasing hormone (RH) of endocrine system is encoded as particle swarm and supervised by the corresponding stimulating hormone (SH). For multi-objective problem, the new SH is composed by the Pareto optimal solutions which determined by the feedback of RH and SH of current generation. In each generation, RH is divided into different classes according to SH, the best positions of different classes, the best position of current generation and the best positions that the particles have achieved so far are simultaneously used to generate the new RH. The effectiveness of the method is tested by simulation experiments with some unconstrained and constrained benchmark multi-objective Pareto optimal problems. The results indicate that the designed method is efficient for some multi-objective optimization problems.