Comparative analysis of demand forecasting methods for products of a stationery store in Minas Gerais
Keywords:
Demand forecasting, Retail, Quantitative analysisAbstract
This paper presents demand methods applied in a retail trading company in order to compare the performance of the models and propose the best demand forecasting model. The data used for the application of the models correspond to the back-to-school period, a period that shows an increase in sales of certain products in the company. Currently, the company uses only qualitative methods to make purchases and the results are not always satisfactory, whether they are large inventories or lack of products for the great demand. For the present work, the demand forecasting models of simple moving average, weighted moving average and exponential smoothing were analyzed. The performance of the methods was compared by means of the errors obtained, which were the Mean Absolute Deviation (MAD), the Mean Absolute Percentage Error (MAPE) and the Mean Square Error (MSE). The results obtained were not satisfactory for any of the forecasting models to be used by the company, because they presented very large errors, and the forecasting model with the smallest error was the weighted moving average.
