Evaluation of Bending Rigidity of Dressing Materials using Artificial Neural Networks

Meenakshisundaram Chenganmal


In garment industries, the formability is one of the factors to construct the garment. Like the cuff and collar of a garment, the fabric is turned over on itself then the inner layer of fabric has to conform to smaller radius of curvature than the outer layer. In this regard, the outer layer has to stretch and inner layer has to contract. So, the fabric is unable to accommodate this change in length, the inner layer will get pucker. This ability to deform is known as “Formability”. The measurement of formability is derived from the bending stiffness of the fabric and its modulus of compression. Hence, the bending rigidity is another important factor for the garment industries. It is necessary for the garment industries, to predict the flexural rigidities of 15 different varieties of Polyester with micro denier, polyester/viscose, and polyester/cotton plain woven dressing materials were used for designing the back-propagation network and a statistical regression analysis approaches. The optimum construction of neural network was investigated through the change of layer and neuron number. The results showed that the back-propagation network could predict the bending values of the above fabrics with a meaningful difference. In this study, neural network has been successfully applied to predict the bending properties of plain woven dressing materials with good reliability. .
Key-Words: Garment fitting, Back- propagation, Neural Network, neuron number, pure bending tester, Formability.

Full Text: PDF