Abstract: Regarding the defects of a posteriori and experience-depended of traditional method, the research proposed use BP neural network to predict the length of garment marking. In the course of achievement, it starts from the order information analysis directly and picks out which factors are concerning with marking length instead of the process of pattern making & marking, and uses these factors as input parameters to establish a BP artificial neural network model for predicting marking length. The experiments show that the model has good prediction. If provide the actual value of factory production to the BP model in practice，the prediction will be more close to the reality，and may be the construction for factory to estimate the fabric consumption and design optimized cutting plan.