A stochastic program to evaluate disruption mitigation investments in the supply chain

Supply chain risk management is becoming increasingly important due to a variety of natural and man-made uncertainties. We develop a methodology to evaluate the costs of disruptions and the value of supply chain network mitigation options based on a two-stage stochastic program. To solve the model, we rely on a solution scheme based on sample average approximation. We explicitly differentiate between disruption periods and business as usual periods to decrease the model size and computational requirements by approximately 85% and 95%, respectively. Furthermore, the decrease in model complexity allows us to include the conditional value at risk in the objective function to incorporate the risk aversion of decisions makers. Based on a case study of a chemical supply chain, this study shows the trade-off between long-term expected costs minimization and short term risk minimization, where the latter leads to a more aggressive investment policy.