A hybrid pareto–fishbone and IoT-based monitoring framework for reducing DTY yarn defects

Quality Control (QC) challenges in the textile industry increasingly require data-driven and real-time solutions to reduce critical production defects. This research aims to develop a hybrid Pareto Fishbone analysis integrated with an IoT-based monitoring framework to reduce the incidence of dominant defects in Draw
Textured Yarn (DTY) yarns (X-stitch and Broken Filament). Defect data collected in 2024 (n=2,396) and early 2025 (n=1,177) were analyzed using Pareto charts, which identified X-stitch (40.15%) and Broken Filament (37.15%) as contributing 77.3% of total defects in 2024. Fishbone diagrams traced root causes to machine vibration and yarn tension anomalies. An IoT prototype was designed using ADXL345 vibration sensors (200 Hz sampling), tension monitoring, and MQTT communication to a Node-RED dashboard to enable real-time alerts. Preliminary testing achieved
95% MQTT transmission success and detected vibration anomalies correlating with 85% of X-stitch incidents. The proposed hybrid framework combines the diagnostic strength of Pareto–Fishbone
analysis with the preventive capability of IoT monitoring, offering a scalable Industry 4.0-oriented solution for textile QC and predictive maintenance.