Among flows in every supply chain (finance, information, and material), material flow according to its part in product cost is very important. Many studies in the field of material flow showed that material flows are often single level. This study offers a comprehensive model for appropriate order assignment in different levels of supply chain according to minimizing related cost to with genetic algorithm method. Results based on genetic algorithm is compared to other common pattern search methods such as Latin Hypercub and Mead-Nelder. The evidence shows that genetic algorithm method is the best method among other methods.