Since there is no analysis of the resources for the current regional comparative advantage analysis system and its lack of the ability in performing the cropping structure programming, this paper focuses on how to build a more useful collaborative analysis system of the cropping comparative advantages by system science, artificial intelligence, software engineering and model design. The system takes the advantage of the agent technology, uses the collaborative analysis method, implements several reusable agent subsystems, and provides an interface to integrating the related comparative advantage systems, in order to show a new idea for studying traditional cropping comparative advantages.