Optimización del inventario en punto de venta de licores mediante modelos predictivos de demanda, periodo febrero 2024 - octubre 2025
Loading...
Date
2026-03-03
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Tecnológica Centroamericana UNITEC
Abstract
La presente investigación analiza la aplicación de técnicas de analítica predictiva para optimizar la gestión de inventarios en Licorería Josway, una microempresa dedicada a la comercialización de bebidas alcohólicas en Honduras. El estudio surge ante las limitaciones de los métodos empíricos tradicionales de control de inventarios, los cuales generan quiebres de stock y sobreinventarios que afectan el nivel de servicio y la eficiencia del capital de trabajo. Se adoptó un enfoque cuantitativo, aplicado y explicativo, utilizando datos históricos de ventas e inventarios correspondientes al período comprendido entre febrero de 2024 y octubre de 2025. A partir de esta información, se desarrollaron y validaron modelos predictivos basados en técnicas de aprendizaje automático, incorporando variables dinámicas como estacionalidad, promociones y eventos locales. La validación se realizó mediante procesos de backtesting y simulación, evaluando el desempeño del modelo a través de métricas de precisión y de indicadores operativos de inventario. Los resultados evidencian una reducción significativa del error de pronóstico, una disminución de los quiebres de stock y una mejora en la rotación de inventarios, demostrando que la analítica predictiva es viable y efectiva incluso en contextos de micro y pequeñas empresas. El estudio aporta evidencia empírica sobre los beneficios operativos y financieros de la toma de decisiones basada en datos en el sector retail hondureño.
This research analyzes the application of predictive analytics techniques to optimize inventory management in Licorería Josway, a small retail business dedicated to the commercialization of alcoholic beverages in Honduras. The study addresses the limitations of traditional empirical inventory control methods, which often generate stockouts and overstock situations, affecting service levels and working capital efficiency. A quantitative, applied and explanatory research approach was adopted, using historical sales and inventory data from February 2024 to October 2025. Predictive models based on machine learning techniques were developed and validated through backtesting and simulation, incorporating dynamic variables such as seasonality, promotions and local events. The performance of the proposed model was evaluated using forecast accuracy metrics and operational indicators related to inventory efficiency. The results demonstrate a significant reduction in forecast error, a decrease in stockouts and an improvement in inventory turnover, confirming the feasibility and effectiveness of predictive analytics in a microenterprise context. This research provides empirical evidence that data-driven inventory management can generate tangible operational and financial benefits, even in organizations with limited technological resources.
This research analyzes the application of predictive analytics techniques to optimize inventory management in Licorería Josway, a small retail business dedicated to the commercialization of alcoholic beverages in Honduras. The study addresses the limitations of traditional empirical inventory control methods, which often generate stockouts and overstock situations, affecting service levels and working capital efficiency. A quantitative, applied and explanatory research approach was adopted, using historical sales and inventory data from February 2024 to October 2025. Predictive models based on machine learning techniques were developed and validated through backtesting and simulation, incorporating dynamic variables such as seasonality, promotions and local events. The performance of the proposed model was evaluated using forecast accuracy metrics and operational indicators related to inventory efficiency. The results demonstrate a significant reduction in forecast error, a decrease in stockouts and an improvement in inventory turnover, confirming the feasibility and effectiveness of predictive analytics in a microenterprise context. This research provides empirical evidence that data-driven inventory management can generate tangible operational and financial benefits, even in organizations with limited technological resources.
Keywords
Analítica Predictiva, Gestión de Inventarios, Aprendizaje Automático, Pronóstico de Demanda, Quiebre de Stock, Microempresa