Answers are grounded in this paper and cite their sources.
Textile microenterprises present operational losses from production downtime, work-in-process accumulation, and nonconforming products, while their adoption of artificial intelligence and Industry 4.0 technologies remains limited by resource constraints. This study proposes a Lean–TPM 4.0 model that will be enabled by artificial intelligence and integrates YOLOv8n-based computer vision for sewing-defect detection, an IoT–XGBoost architecture for predictive maintenance, Lean tools, and discrete-event simulation to evaluate the overall performance of the system. The case study was conducted in a Peruvian textile microenterprise, where the initial diagnosis identified that 60.69% of the economic losses came from unproductive time and 39.31% from nonconforming products. In the proposed To-Be scenario, the nonconformity rate was projected to decrease from 20.00% to 8.00%, while MTBF increased from 5.70 to 9.41 h and monthly downtime was reduced from 64.50 to 19.87 h. YOLOv8n achieved an overall F1-score of 85.09% and mAP@50 of 87.88% for sewing-defect detection. In 100 Arena replications with 95% confidence, WIP decreased from 46.58 to 2.97 units, while simulated production increased from 299.17 to 386.84 units and operational efficiency from 80.00% to 92.00%. The results demonstrate the potential of integrating AI/ML with Lean–TPM practices to support smart manufacturing and data-driven operational improvement in resource-constrained textile microenterprises.
No takes yet. Share an insight, caveat, or question.
Mallqui-Alzamora et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: