Predicting Risks for Supply Chain Management Networks with Machine Learning Algorithm

Abstract

Predicting risk in supply chain management networks has been interested by many researchers because supply chain management can be seen as a core factor of businesses activities. Using machine learning algorithm, especially with Bayesian networks, to predict risk in supply chain management network can help to control and monitor the supply chain process in particular in the second step of identification. If the supply chain risk is effectively evaluated, it can support supply chain partners to assess, identify, monitor, and mitigate risks in order to increase robustness and resilience, reduce supply chain vulnerabilities, ensuring continuity and profitability. The Bayesian network have advantage of optimization in explosion dataset based on treating the weights and outputs which find their marginal distributions that best fit the data. The contribution of the paper focuses on summarize the applying of machine learning algorithms in predicting in supply chain management field, proposing supply chain management risk framework in which machine learning algorithms are applied, and demonstrate the case study to show the advantage of using machine learning algorithm in particular of Bayesian networks in risk prediction. The experimental case study shows the good results with the risk model. This implicates the performance of using machine learning in predicting risk in support supply chain management networks.