Previsão em mrp usando a transformação de box-cox através do aplicativo glim, com aplicação

Abstract

The principal objective of this paper is to offer a quantitative methodology for making forecasts to be used as an essential ingredient for MRP. The forecasts are for an individual product and for the very short-term, which is usually one or two weeks before the delivery date. The forecasting equation is estimated through regression analysis, transforming the dependent variable through a Box-Cox transformation and taking into account tendency and several kinds of seasonality, in the week, the month and the year. The resulting equation is shown to be easy to program and place directly within the MRP computational software.