Uma comparação dos Modelos SARIMA e Holt-Winters para a previsão do desvio do custo relativo de manutenção dos veículos da frota de uma empresa de distribuição de energia elétrica brasileira.

Abstract

In times of great economic uncertainty, performing good predictions can help the financial planning of companies going forward. Therefore, the present study compared the Seasonal Autoregressive Integrated Moving Avergage SARIMA and Holt-Winters models to predict the deviation of the relative maintenance cost, measured in R$ per Km, of the fleet of a brazilian company in the electricity distribution sector. The Holt-Winters results, for a proposed time series, performed better, according to the criteria of Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). Other model models will be incorporated and compared in future works.