A Bayesian simulation approach for supply chain synchronization

While simulation has been used extensively to model supply chain processes, the use of a Bayesian approach has been limited. However, Bayesian modeling brings key advantages, especially in cases of uncertainty. In this paper, we develop a data informatics model that could be used to realize a digital synchronized supply chain. To realize this model, we develop a hybrid model that combines Bayesian modeling with discrete-event simulation and apply it to the supply chain process at a Proctor & Gamble (P&G) manufacturing plant. Moreover, we use approximately one year of transactional data, including information on customer orders, production, raw materials, inventory, and shipments. A driving force for creating this model is to better understand Total Shareholder Return expressed in terms of cash, profit, and service.