ImpactU Versión 3.11.2 Última actualización: Interfaz de Usuario: 16/10/2025 Base de Datos: 29/08/2025 Hecho en Colombia
Modelos estadísticos para el pronósticos de los parámetros fisicoquímicos (porcentaje de acidez y densidad) y composicionales (porcentaje de grasa y proteína) de leche de búfala recibida en Colanta planta Planeta Rica
Decree 616 of 2006, refers to the requirements that milk must meet for human consumption in Colombia, detailing the characteristics of raw milk from cattle, excluding buffalo milk, which leads to industries having no normative basis for the acceptance / rejection of this raw material, it is added that the physicochemical, compositional and organoleptic characteristics of this milk are different from cow's milk. It is considered that through the statistics of time series a model can be adjusted to each parameter studied of buffalo milk, making it possible to forecast them correctly. For the adjustment of the models, the buffalo milk reception database is used in the Colanta plant in Planeta Rica from 2008 to December 2018, with the help of the RStudio Software and using the methodology proposed by Box and Jenkins. adjust the ARIMA model (1, 0,0) x (0, 1, 1) 12 for the acidity parameter, the ARIMA model (0, 1,0) x (0, 1, 1) 12 for the density, ARIMA (1, 0,0) x (0, 1, 1) 12 to predict the percentage of fat and the percentage of protein is achieved by modeling with the adjustment ARIMA (3, 0,2) x (1, 0, 1) 12 . The results of the out-of-sample forecasts show concordant results with percentage absolute percentage errors (ASM) of 1.75% for the percentage of acidity, 0.03417% for the density, 0.85% for the percentage of fat and 2, 75% for the buffalo milk protein, the Theil's U statistic of the 4 parameters studied