Routing decision is an essential element to the efficiency of a distribution network as its configuration determines the flow of different products via various transition points. Traditionally, factors, such as cost and distance, will usually be considered. However, different handling reliability and the ability of recycling wastes, packaging materials for example, in some transition points could affect the performance of the network. The objective of this paper is to propose a genetic algorithm to determine the flow and route of products in a distribution network. It simultaneously considers various factors such as cost, distance, the handling reliability, and recycling ability in different transition point. A numerical example is presented to illustrate the conflicts between the different factors to the performance of the distribution network under study, and thus how the proposed algorithm can handle the trade-off in such a multi-criterion decision making problem.