Optimización de la proyección de pesos en pollos Broiler de la granja San Bernardo del cantón Cumandá mediante un análisis comparativo de modelos matemáticos utilizando Matlab y Excel.

Abstract

The objective was to optimize the projection of the weight in broiler chickens from the San Bernardo farm in Cumandá Canton, through a comparative analysis of mathematical models using Matlab and Excel. Field data was collected over the weight of the chickens from production lots 36, 37, 38, 39, and 40 of the six chicken coops found at the farm. In total, 6,440 stats were collected, with which an average weight was calculated for each corresponding chicken coop in each lot. After gathering this data with the help of the polynomial interpolation or extrapolation method, missing stats were projected for the purpose of obtaining an analysis interval that runs from day 0 to day 43. Next, a measurement of central trend and data bias analysis was realized to develop a table of past average weight for each weight statistic from each chicken coop on the farm, which helped in the realization of the projection programs for weight carried out on the Excel and Matlad softwares, where the mathematical basics were polynomial regression and polynomial extrapolation respectively. To determine the informatic tool that will optimize the projection of weight, a statistical analysis was carried out between related and normal samples and the errored samples that were obtained with the difference of the data between the final weight of production lot 40, having the report data of the weight production from production lot 40 and the data projected in the Excel and Matlab softwares. Upon finalizing said statistical analysis, it was determined that the Excel software has a better projection of the weight of the chickens with 90% reliability. Finally, implementation of the excel software was proposed for use at the farm.