Pronóstico de la demanda intermitente de repuestos aeronáuticos mediante métodos clásicos estadísticos y modelos lineales dinámicos.

Abstract

Military aviation is essential for the fulfillment of the Constitutional Order and the purposes of the State, which is why efforts must be centralized to guarantee the availability, safety and reliability of aircraft, for this reason, this study sought to implement statistical methodologies to through classical forecasting methods, bootstrapping simulations and linear dynamic models to generate a forecast in the demand for aeronautical spare parts with characteristics of intermittence and variability and thus contribute to the continuous improvement in the logistics processes of the aeronautical administration. Within the classical forecasting methods, the exponential smoothing family (Croston, Syntetos-Boylan Approximation), bootstrapping simulations and first-order polynomial dynamic linear model were used, the mean square error (RMSE) and mean absolute error (MAE) were used for comparison and selection of the best model, showing as a result that the dynamic linear model has a better performance compared to the others.