Abstract: The grey modeling method has been developed according to the theory of the grey system and is used for correlation analysis of experimental data. Compared with the method of data regression, the grey model analysis (GMA) enjoys an obvious advantage. It can enhance the reliability of prediction in particular, in the testing of samples of small size. In this paper, the grey GM (1,2) model and grey optimum GM^* (1,2) model for the testing and analysis of the fineness of ramie fiber are described. Correlation analysis is then performed to predict the relationship between the data Tex (sample Y) and the data projection breadth (sample X). The error is evaluated and proves acceptable. The model for testing the fineness of ramie fiber established on it provides a concise and scientific method for technical control of production, quality inspection and supervision.

Key words: grey optimum model, ramie fiber, Tex, projection breadth