The risk assessment of manufacturing supply chains based on Bayesian networks with uncertainty of demand

Abstract

The great changes in the external environment of the manufacturing supply chain make its demand more complex and difficult to control. This paper takes China as an example. According to questionnaire survey and principal component analysis, the risk indicators caused by uncertain demand are screened and classified to construct evaluation system and complete risk identification. The Bayesian network integrating fuzzy set theory and left and right fuzzy ranking is used to explore the relationship between risk indicators and supply chain to achieve risk evaluation. In view of the highest risk factors, an incentive mechanism model based on information sharing is put forward to prove theoretically that information sharing is an important strategy to reduce risk. The results are as follows: The uncertain demand will lead to a high level of risk in China’s manufacturing supply chain, in which the level of information technology is the biggest cause. Only when manufacturing enterprises are willing to share information and other node enterprises join the information sharing team, can demand uncertainty be fundamentally reduced. The proposed risk assessment model realizes the method innovation and theoretical innovation. It can practical and effectively help relevant enterprises to determine and control risks.

Optimizing Production Supply Chain With Markov Jump System for Logistics Collaboration

“…Although the use of Markov jump systems in the collaboration of production and supply chain logistics information is a popular research topic (Meng et al, 2022), this method relies on the estimation of historical data and state transition probabilities for modeling the system state, which may be affected by incomplete or inaccurate information, leading to challenges in the accuracy of the model . In addition, for complex supply chain network structures and dynamic environments, there are also problems such as difficulty in fully capturing the complexity and variability of the actual supply chain, as well as insufficient recognition and collaborative consideration of important decision-making nodes (Tang et al, 2023).…”