Research Uncertainties Probabilistic Graphical Models Based On Supply Chain

With the development of science and technology, the trend of economic globalization and the size of the enterprise supply chain are growing and the structure is increasing complexity. Now in the field of enterprise supply chain and inventory, the uncertain problem becomes more and more serious. Probabilistic graphical model has become the hotspot because it’s most efficient theory model for solving uncertain problems and expressing uncertain information.According to the uncertain problems in the enterprise supply chain and inventory, this paper has a deep research in the principles and process of decision-making in the supply chain, the state and node parameters in inventory. The main jobs are followed:The theory of probabilistic graphical model and common probabilistic graphical model, the structure learning and inferring of Bayesian networks in uncertain problems are systematically summarized. Analysis the manifestation pattern of the uncertain problem in the enterprise supply chain. This paper has researched the causes of the uncertain problem, the decision-making principles and decision-making process in enterprise supply chain. According to the enterprise supply chain, the paper discuss the uncertain problems in the business inventories, research the state and node parameters of Bayesian network in uncertain problem. The modeling and decision-making process of the supply chain in Bayesian network also discussed.An uncertain problem system in inventory based on Bayesian network is set up, and the system is carried out with Delphi, including network modeling and decision-making. Through the experimentation and analysis, this method gets the best solution of inventory in the supply chain.