Modelling supply chain disruption analytics under insufficient data: A decision support system based on Bayesian hierarchical approach

Abstract

Recent disruptions in global and local supply chains (SCs) due to manmade and natural disasters have drawn a lot of attention to academics and practitioners. Such disruptions are often characterized by a lack of insufficient data. To tackle such a data scarce supply chain ecosystem, this article examines the potential disruption risks in SCs under insufficient input data. For this, a decision support system (DSS) based on the Bayesian hierarchical approach and value at risk (VaR) reduction analysis is proposed to assess a supplier's disruption risk events probability as well as the supplier revenue impact on a company of interest. Empirical data are used to examine the DSS. The findings show that the DSS is effective in generating the suppliers’ risk profiles. The proposed DSS can be utilized by supply chain managers and practitioners to manage SC disruption risks in a more efficient manner.

Keywords

• Bayesian hierarchical model
• Decision support system
• Disruption analytics
• Supply chain disruptions
• Value at Risk