A bayesian network model with epistemic uncertainty: analysis of a medicine supply chain risk

To effectively manage risk in supply chains, it is important to understand the interrelationships between events that might affect the flow of material, products and information. We present a quantitative modelling process using Bayesian Belief Networks to represent probabilistic dependency relationships between events. A visual modelling process, grounded in Bayesian Network theory and the decision context of supply chain risk management, is developed to capture the knowledge and probability judgements of relevant stakeholders. Building causal maps provides a good basis for translating stakeholder cause-effect knowledge about possible events into a formal graphical probability model. A protocol for eliciting subjective probabilities from relevant stakeholders to quantify the state of knowledge uncertainties about risk events has been developed and applied. The modelling process has been evaluated through an in-depth case for the hospital medicine supply of NHS Greater Glasgow & Clyde. Working in collaboration with relevant stakeholders with expertise in all or part of the medicine supply chain, a model has been developed and their perceptions about the process and results have been analysed. We find that the Bayesian Network model of the medicine supply chain has provided insight into risks not captured by conventional risk management methods and it supported deeper understanding of risk through exploration of modelling scenarios.