Selecting forecasting model parameters in Material Requirement Planning systems

Abstract

This paper investigates how the choice of parameters for forecasting models influences the performance of MRP systems. The results of the study show that the error measures, which are used to estimate forecasting parameters, have a significant effect on the system performance. Minimising Mean Absolute Deviation (MAD) and Mean Square Error (MSE) in choosing the forecasting model parameters will result in total cost that is much closer to the minimum cost than minimising the mean error (Bias). While operating parameters such as Freezing Proportion (FP) and Cost Structure (CS) do significantly influence the relationship between total cost and the error measures that are used to estimate forecasting model parameters, both MAD and MSE are better than Bias under all conditions. The use of Safety Stock (SS) does not influence the conclusion.