Predictive Maintenance in Textile Manufacturing Using Artificial Intelligence

Chambi N.; Sanga C.; Sanga A.; Sanga P. · Proceedings - IEEE CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies, ChileCon (2025) · DOI: 10.1109/chilecon66915.2025.11476013

Resumen

This article addresses predictive maintenance (PdM) applied to textile machinery using artificial intelligence, a key approach to improving efficiency and reducing operating costs. The objective is to optimize equipment lifespan through early fault detection, avoiding costly repairs and unplanned downtime. The challenge lies in intensive operating conditions and high production demand, which make failure prediction difficult with traditional methods. The methodology includes continuous monitoring of critical parameters (vibrations, temperature, humidity, yarn tension, and component wear) using sensors and real-time data analysis. These data are processed with artificial intelligence and machine learning techniques to identify patterns that precede failures. The results demonstrate that PdM can reduce maintenance costs by 8% and increase equipment availability by 10%, achieving greater productivity and quality in textile manufacturing.

Instituciones: Universidad Nacional de San Agustín de Arequipa