Pronósticos de ventas para la empresa Fabrinorte Cia. Ltda para los años 2020-2026

Abstract

The objective of this study was to reduce uncertainty in the company FABRINORTE CIA.LTD, through deterministic forecast models, regarding its sales for correct decision making, as well as an analysis of the company to observe its behavior in sales. For the study, an analysis of atypical data was performed using the statistical software R, then the most representative variables of the entire matrix were analyzed and an analysis of the qualitative variables through easy-to-interpret graphs, for the quantitative variables a descriptive analysis was done, as histograms for the description of the behavior of each of the variables. For the elaboration of the forecasts, the total variable was used, which represents the sales that the company made over time, the process of decomposition of the series was used to observe their behavior and their relationship with each other. The models used to make the forecasts were the Holt Winters and Seasonal Auto-regressive Integrated Moving Average (SARIMA) models, which were chosen through the components presented by the time series. The results through the decomposition process of the series permitted observing that our series had a trend and seasonality, therefore, through this analysis, the model that provided the best efficiency results was the SARIMA (1 1 2) (1 1 2) model, said model fulfilled all the assumptions of its remainders. It is concluded that the company's sales will have growth over time, in addition to the fact that it was observed that during the months of June to November the company shows its highest sales points in the course of the year.