Eliminating Cycles in Model Net

Model Net is a graphical model used to represent and analyze the static structure of the model bases. Compared to other models, model composition method based on model net can not only greatly improve the efficiency of model composition, but also find out all the potential composite models for a given decision problem. However, the model net simplification and model composition algorithms may work improperly in practical applications for the reason that a model net constructed based on the concept of generalized data type matching may contain cycles. This paper aims at how to eliminate the cycles existing in a model net. We first introduced a concept of importance factor of an edge in a model net, and then worked out a method to eliminating cycles.