The system presented in this work is focused on prediction of the structure parameters as well as the properties of the fibers – yarns – fabric structural hierarchy. The main target of this system is to obtain optimal fabric design based on virtually created fabric. This system contains original models and procedures for prediction of properties that have influence on the structure and on the textile fabric behavior. Prediction is based on developed theoretical models as well as experimental methods and regression analysis. The regression models are mainly used for estimation of cotton fiber strength measured on common testing instruments. Furthermore, the system deals with prediction of yarn packing density, yarn diameter, hairiness and yarn strength using the basic parameters of fibers, yarns, and technology applied. In the fabric level the basic mechanic, physical and utility properties are predicted. The system is able to predict fabric e.g. cover factor, mass per sq. meter, permeability, surface roughness and many others properties. The system is constructed to visualize yarns, and fabrics in 2D. The main structure of this system will be discussed. The applicability for construction of fabric from prescribed cotton fibers and fiber mixtures will be demonstrated on practical examples.