Gemelos digitales y mantenimiento predictivo en industrias manufactureras de Huancayo

This study aimed to evaluate the impact of implementing digital twins integrated with predictive machine learning models on the optimization of industrial maintenance in textile companies in the city of Huancayo, where mechanical failures occur that result in approximately 12 hours of downtime or inactivity. It employed a mixed-methods approach, using a pre-post intervention design with 72 experts through surveys (72) and interviews (15) with individuals serving as technicians, managers, and supervisors in the textile industry in Huancayo, Peru. The analysis relied primarily on software tools such as SPSS (ANOVA) for quantitative analysis and ATLAS.TI for qualitative analysis, in addition to Power BI to better support the findings identified in the categorization. The results indicate significant reductions in response times to failures, improvements in usability (SUS), and increases in MTBF (p<0.01), demonstrating a positive impact of the digital twin. It is concluded that digital twins with predictive ML can facilitate the transition from corrective to predictive maintenance in Andean textile contexts such as those in Huancayo.