Recovery strategies from major supply disruptions in single and multiple sourcing networks

The increase in supply network disruptions has shown that many companies have been caught by surprise and were not able to quickly, if at all, recover from these disruptions. In this paper we propose models for optimal recovery from major unpredictable disruptions in a supply network. We consider a supply network that is comprised of customers and production facilities where disruption may occur at any of the facilities and can cause partial failure or complete shutdown of the supply facilities. The novelty of our model is that it incorporates dynamic pricing as a lever to manage demand during such disruptions. In addition to pricing we also incorporate recovery strategies that are based on inventory, transshipment and outsourcing. We allow for the recovery duration and the disrupted capacity to be uncertain and use pricing to reflect temporary impacts of disruption on demand. Demand is price sensitive and accounts for uncertainty in the customers’ willingness to pay during the recovery period. We investigate the cases of multi-sourcing and single-sourcing. The multi-sourcing model is a convex programming model which is easy to solve using commercial solvers. An accelerated Benders decomposition method with valid inequalities is proposed and tested for solving the more complex single-sourcing model. Using a US case study we find that a dynamic pricing recovery strategy can improve profits during recovery from major supply disruptions. Furthermore, we find that a dynamic pricing recovery strategy is more efficient in single-sourcing networks than multi-sourcing networks.