Prediction of tear strength of bed sheet fabric using machine learning based artificial neural network

Fabric tear strength is an important parameter in textile materials since it is linked to the fabric's life. When it comes to bed sheets, a consumer's main concerns are mechanical comfort and durability. The fabric's tear strength has a direct impact on its durability. In this study, a novel algorithm was developed using machine learning based neural network approach taking fabric parameters as input and warp and weft tear strength as output. Different state-of-the-art algorithms were used among which the XGboost model (Extreme Gradient Boosting) gave the best results. The training accuracy of the machine learning model was approximately 99.99% and mean absolute error was 0.61 for the weft model and 1.78 for the warp model. Correlation between various fabric parameters and tear strength was determined. The model has the potential to be beneficial to textile industries since it can reduce the number of production efforts for a high-quality bed sheet fabric, saving both time and money. Artificial neural networks are used to forecast fabric performance in textile manufacturing. Tests show that these artificial neural networks are good at predicting future difficulties.

Automated summary

A novel algorithm based on machine learning was developed to predict the tear strength of bed sheet fabrics, showing high accuracy and reliability in forecasting fabric performance.