The fuzzy rule-based Genetic Algorithm (Fuzzy-MOGA) is proposed in the paper to solve the problem of multi-objective optimization. The original chromosomes are generated with heuristic information. The Pareto optimal solutions are built by the arena’s principle on the partial order set, so that it is easy to search the non-convex solution and it dos not need to determine weight value any more, which encountered by the aggregation function approach. Fuzzy-MOGA is used to solve the production plan optimization in SCM. Because the fuzzy-rule for production distribution facilitates can easy express explicit knowledge, the limitation of greedy algorithm can be avoided. Therefore, it not only displays evidently efficiency of the algorithm and but also can find the complete Pareto front.