Machine Learning para mejorar la toma de decisiones en una farmacia hospitalaria de Lima, 2025

Fuentes: histórico de OJS (PKP, por objeto) + tracker en la revista + 3 vista(s) en este índice. Citas: se consultan en Crossref para publicaciones con DOI.

Resumen

The main objective of this research was to determine how the use of machine learning contributes to improving decision-making processes in a hospital pharmacy in Lima during the year 2025. The study was classified as applied, with a quantitative and descriptive approach, employing a pre-experimental single-group design with pretest–posttest measurements, which allowed the evaluation of the impact of the analytical proposal. The population consisted of historical medication sales data, considering a representative sample for analysis through time series techniques. Regarding the procedure, ARIMA models were applied after evaluating stationarity, autocorrelation, and seasonality criteria, and their validation was carried out using error metrics such as RMSE and MAE. The results showed a significant improvement in the accuracy of medication demand forecasting, as well as in inventory control, reducing stockouts and overstock levels. Likewise, hypothesis testing showed significance values of p < 0.05, with a 95% confidence level, confirming statistically significant differences between the pretest and posttest. It is concluded that the implementation of machine learning models based on time series is effective in optimizing hospital pharmaceutical management, improving operational efficiency and the quality of decision-making, provided that adequate data management and institutional support are ensured.

Lecturas por mes y tendencia

Barras: vistas del resumen y lecturas de archivos (tracker + índice). Línea punteada: tendencia lineal y predicción de 3 meses (estimación estadística, no un dato).

Palabras clave

Machine LearningARIMADemand ForecastingMake decisionsHospital PharmacyPredicción de demandaToma de decisionesFarmacia hospitalaria