Bayesian methodology for supply chain risk analysis: Concept and case studies

Supply chain risk becomes increasingly important as manufacturers attempt to harness the benefits of global procurement and distribution practices. Commonly available tools used for supply chain risk management (SCRM) attempt to prepare the supply chain for the occurrence of negative events. Often, however, these tools overlook the key concept of risk interdependency. To begin consideration of such risk relationships, a standard procedure is required for risk identification and risk network specification. A promising method for probabilistic, quantitative SCRM is modeling with Bayesian Belief Networks (BBN). Root cause analysis and sensitivity analysis regarding risk interdependency are examples of the useful results of BBN. Utilizing BBN, this paper presents a method for addressing supply chain risk that was developed and applied in two cases. The first case addresses the issue of supplier evaluation and serves as a proof of concept for the implementation of a BBN SCRM technique. The second case applies the same concept to examine internal risk exposure at a manufacturing company due to supply chain relationships. Both case studies illustrate the unique features available via risk modeling with BBN. The method provides valuable insight into risk interactions and probabilities that could help inform the decision making process of supply chain managers.