Abstract: Virtual manufacturing resource selection can be seen as a selforganization evolution process of a complex adaptive system. With this viewpoint, the selforganization evolution process is reflected by the interactions between resource agents and the task agents, and between these agents and the environment. Aiming at obtaining a most satisfactory solution from the resource selection route set, a selforganization evolution model is constructed. Based on the model, information interaction mechanism between agents is designed. Then, by using Swarm platform, simulation is carried out with evolutionary genetic algorithm embedded. In this way, an optimal solution can be obtained. An example is used to test the feasibility and validity of the proposed method. Results show that the solution obtained is very useful for decisionmaker so as to make a quick and good selection of virtual manufacturing resource.