The influence of fibre, yarn, and fabric properties on the bagging behaviour of woven fabrics is studied by developing rheological models and comparing them with the results of statistical analysis on experimental measurements. Three rheological models are developed to describe the bagging-deformation features at different stages of bagging processes, from which three types of deformation energy are identified: elastic and viscoelastic energies of the fibres, and the frictional-deformation energy resulting from the frictional slippage between fibres and yarns. To evaluate the theoretical analysis, 24 fabrics were tested to determine their bagging behaviour, the mechanical properties of constituent fibres and yarns, and fabric structural features. Statistical regression analysis is applied to derive the parameters describing bagging fatigue and their relationships with the fibre and yarn mechanical properties and fabric structural features. A comparison of the experimental results with the theoretical analysis indicates that a decrease in the work of loading with repeated bagging deformation resulted from two basic causes: the stress relaxation of the fibres due to the fibres' viscoelastic behaviour and the friction between fibres and between yarns due to frictional restraints inherent in the fabric structures. The three parameters (Q o, U σ) describing the fabric-fatigue process are shown to be dependent on fibre and yarn mechanical properties and fabric structural parameters, such as fabric thickness, weight, cover factor, and interlacing points. Using the regression equations for the three parameters (Q o, U σ), we show that the bagging-fatigue behaviour and fabric-residual-bagging height are predictable from the fundamental fibre and yarn mechanical properties and fabric structural parameters, indicating that a subjective perception of fabric-bagging is also predictable from the parameters.