Citas bibligráficas
Cayo, M., Cruz, L., Peche, M., Palomino, H. (2023). Reducción de la subjetividad en el proceso de clasificación de color de teñido de lotes de producción textil mediante machine learning [Tesis, Universidad Peruana de Ciencias Aplicadas (UPC)]. http://hdl.handle.net/10757/669755
Cayo, M., Cruz, L., Peche, M., Palomino, H. Reducción de la subjetividad en el proceso de clasificación de color de teñido de lotes de producción textil mediante machine learning [Tesis]. PE: Universidad Peruana de Ciencias Aplicadas (UPC); 2023. http://hdl.handle.net/10757/669755
@mastersthesis{renati/526611,
title = "Reducción de la subjetividad en el proceso de clasificación de color de teñido de lotes de producción textil mediante machine learning",
author = "Palomino Lopez, Hugo Miguel",
publisher = "Universidad Peruana de Ciencias Aplicadas (UPC)",
year = "2023"
}
One of the main challenges in textile production is to reproduce as best as possible the color tone in the fabric, a tone that is obtained from a sample given by the client. The color evaluation procedure normally has a lot of subjectivity due to the visual appreciation that the human quality analyst makes when evaluating a dye lot. When a batch of dyeing is rejected and the color difference is not very evident, there are delays in the final decision, this is the problem at the EcoTextil company. Workflows including supervised learning algorithms for the classification of dye batches according to hue were investigated, reducing human subjectivity in the evaluation of color hue. For the design of the classification workflows, the Fundamental Methodology for Data Science was used. The proposed automated workflows were classified into high, regular and low performance flows, high performance flows have an average F1 value of 0.92, which is higher than the F1 value of the current workflow, human evaluation, which is 0.82. The use of the proposed automated workflows means a saving of 30,000 soles per year due to the reduction in man hours, about 60,000 soles per year due to the reduction of unnecessary reprocessing, 144,000 soles per year due to the reduction of downtime and about 30,000 soles per year for reduction of production balances.
Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons