A Model for Forecasting Material Demand with Its Previous Data

Abstract

This paper uses Bayesian model to propose two different conditions in which the ordering quantity and ordering time are two previous and relative data. Based on cost balance view of point, productivity is the prime goal in material manufacturing process. Thus, this research calculates the full model of productivity with considering these two different conditions cases. The empirical data is used to estimate parameters of these two proposed models and make the comparisons of model calibration between these two conditions. The results show good fitness between empirical data and simulation data when ordering time is conditioned by ordering quantity. Finally the conclusion is made for practical application in operations and production management. 