AI helps reduce waste and improve the quality of dyed fabrics

A new study found that machine learning can help reduce textile manufacturing waste by mapping more accurately how colors change during the dyeing process.

Fabrics are usually dyed while wet and change color as they dry. This can make it difficult to know what the fabric looks like in its finished state, said Warren Jasper, a professor at Wilson University's Textile University and author of a paper on research published in the journal. fiber.

“The fabric is dyed while it's wet, but the target shade is dry and wearable. This means that if there is an error in the color, it won't be visible until the fabric is dry,” he said. “More fabrics are dyed all the way while waiting for that drying to occur. That leads to a lot of waste because you can't catch the error until later in the process.”

The amount of color change from wet to dry is not uniform between different colors. This nonlinear relationship means that the amount of color change between wetting and drying is unique to each color, and data from a sample of one color cannot be easily transferred to another color.

To tackle this problem, Jasper has developed five machine learning models, including neural networks specifically designed to map this type of nonlinear relationship. He then trained the model by entering visual data from 763 fabric samples in various colors, both wet and dry. Jasper said it took several hours for each dye to be completed, which made collecting data a key business.

All of these models outperformed non-machine learning models in terms of accuracy, but neural networks stood out much more accurately than any other option. The neural network used Ciede2000, a standardized color difference equation, to show an error of as low as 0.01 and a median error of as 0.7. Other machine learning models show the CIEDE2000 error range between 1.1 and 1.6, with the baseline at 13.8. In the textile industry, Ciede2000 values ​​above 0.8-1.0 are generally considered outside the tolerance limits.

This neural network can significantly reduce waste caused by color errors, allowing fabric manufacturers to better predict the final outcome of the dyeing process before large amounts of fabrics are accidentally dyed. Jasper said he hopes similar machine learning tools will be adapted more widely in the textile industry.

“We're a little behind the curve of textiles. The industry has begun to move further towards machine learning models, but it's very slow,” he said. “These types of models can provide powerful tools to reduce waste and increase continuous dyeing productivity, which accounts for more than 60% of dyed fabrics.”