Supply Chain Viability: Risk-Averse Decision-Making

In this chapter, stochastic optimization of CVaR (Conditional Value-at-Risk) is applied to maintain risk-averse viability and improve resilience of a supply chain under propagated disruptions. In order to establish the risk-averse boundaries on supply chain viability space, two stochastic optimization models are developed with the two conflicting objectives: minimization of Conditional Cost-at-Risk and maximization of Conditional Service-at-Risk. Then, the risk-averse viable production trajectory at between the two boundaries is selected using stochastic quadratic MIP model. The proposed approach is applied to maintain the supply chain viability in the smartphone manufacturing, and the results of computational experiments are provided. The findings indicate that when the decision-making is more risk-aversive, the size of the viability space between the two boundaries is greater. As a result, more room is available for selecting viable production trajectories under severe disruptions. Moreover, the larger is viability space, the higher is both worst-case and average resilience of the supply chain. Risk-neutral, single-objective decision-making may reduce the supply chain viability. A single objective supply chain optimization which moves production to the corresponding boundary of the viability space should not be applied under severe disruption risks to avoid greater losses. The major decision-making insights are summarized at the end of this chapter.