PREVISÃO DE DEMANDA NO MERCADO DE VAREJO: UM ESTUDO DE CASO EM UM COMÉRCIO DE BATERIAS AUTOMOTIVAS

Abstract

The demand forecasting method is used to support decision making through quantitative analysis. However, its use has been mainly in production planning than retail planning. The objective of this study was to perform demand forecasting simulations, based on the history of a retail company, through the forecast calculation models to study and determine which method represented better accuracy for the retail industry. To develop this work was based on historical data for evidence of a retail company and was adjusted the data to propose a demand forecasting model that best matched the reality of the retail market of the studied company. The results showed that the demand forecasting methods that resulted in outcomes that best accompany the demand behavior are the simple moving average, weighted moving average, moving average and exponential smoothing method for linear regression, the simple exponential moving average the best it has lower absolute error.