Abstract: Quantitative forecasting of the inventory for key products can help to reduce the amount of inventory obsolescence and prevent production delays due to raw material stock-outs. Predicting productive inventory is beneficial to promote the sustainability of production management. In this work, a prediction model is constructed that predicts the pass rate of products and the processing path of unqualified products and simultaneously calculates the quantity, time, and probability of each path. Using the Graphical Evaluation and Review Technique (GERT), the manufacturing process of a square tube can be transformed into a stochastic network. Then, grey parameters are introduced into the GERT network to solve uncertainty in manufacturing. Finally, a numerical example is given to obtain a productive inventory prediction for beam square tubes using grey GERT(G-GERT). The main contribution of this work is the integration of inventory quantity, time, and probability. These three results can be predicted simultaneously, and the algorithm can be extended to any product production network.

Key words: G-GERT, Forecasting, Stochastic Networks, Inventory, Grey Parameters