Desarrollo de una herramienta estadística para el pronóstico de la demanda en una empresa de balanceado acuícola

Abstract

CONDITION FOR PUBLICATION OF PROJECT. This project focuses on the development of a statistical tool for demand forecasting in an aquaculture feed company, with the objective of increasing forecast accuracy by at least 65% in order to optimize production planning and the use of raw materials. The hypothesis states that the incorporation of robust statistical models and the technical classification of demand patterns can reduce estimation errors and improve operational efficiency. For its implementation, historical sales data were collected and verified from the company’s ERP system, which were then cleaned and analyzed through time series techniques and the evaluation of different forecasting models. The results demonstrated a significant improvement, achieving an accuracy level of 71% when comparing the projected values with the actual demand of July 2025, thus surpassing the initial objective. In addition, economic, social, and environmental benefits were identified as a result of more reliable and efficient planning. It is concluded that the proposed tool increases forecast reliability, strengthens strategic decision-making, and contributes to the sustainability of the company’s operations. Keywords: demand forecasting, accuracy, planning, time series.