Modeling Inventory Policies in Multi Echelon Supply Chain by Beysian Networks

Determining the optimal policy of inventory have been always one of the main challenges in inventory management area in which several research have been conducted to address this issue. A vast majority of proposed approaches are very simple models to the extent that they simplify the real-world conditions and fail to consider the real uncertainties. To determine the optimal inventory policy of supply chains, on the other hand, there exist many influential uncertainties. In this thesis, an integrative probabilistic model is developed to model the uncertainty of the optimal inventory policy of multi-echolon supply chains using Bayesian networks (BNs) a state-of-the-art technology in modeling uncertainty. BNs provide a framework for presenting cause and effect relationships and enable probabilistic inference among a set of variables. The new approach explicitly quantifies uncertainty in qualitative and quantitative uncertain variables in customer, retailer, manufacturure, and supplier levels and provides an appropriate method for modeling complex relationships for process capability analysis, such as common causal factors, formal use of experts' judgments, and learning from data to update previous beliefs and probabilities. The capabilities of the proposed approach are emplemented in Agenarisk software by a real case study.