Effective mitigation policies for disaster operations management in Manufacturing Closed Loop Supply Chains: Insights for transition from unknown-unknowns to known-unknowns

The viability of Manufacturing Supply Chains (MSCs) is strongly affected by disaster effects as they have been identified as vulnerable networks to shocks imposed by the external environment. It is remarkable that although the research agenda in disaster operations management of MSCs has received increased attention during the last decade, the studies that have been conducted on disaster-resilient Manufacturing Closed Loop Supply Chains (MCLSCs) are limited. Since disaster for an MCLSC is a risk source of unknown-unknowns, the key issue of this work is to study effective mitigation policies to provide insights for fast transition to pre-disaster operating conditions, through a methodology that combines the well-established Disaster Operations Cycle method with the Rumsfeld Matrix. To this end, the study defines short-/medium-/long-term disaster scenarios with alternative settings regarding their impact on demand and supply chain operations (production rates, shipments), as well as a set of relief policies in pre-disaster, disaster and post-disaster periods. By conducting extensive numerical experimentation in conjunction with a dynamic MCLSC model with remanufacturing and recycling under disaster effects, the system response (transient flows, inventory levels, cost, profit) is compared to that obtained from the disaster-free model by examining the manufacturer's availability of finished goods inventory and profitability. This comparative analysis of results provides managerial insights for a transition from unknown-unknowns to known-unknowns and reveals holding base stocks at the manufacturer level, as a dominant mitigation policy. In terms of manufacturer's serviceable end-product inventory, different inventory level policies before, during, and after disaster periods appear to be the best choices. It is important to note that for small- and medium-scale disaster scenarios, the differences in optimal policies are relatively small, whereas these differences become more significant during large-scale disaster scenarios. In terms of economic impact, the results show the necessity to follow a policy of a constant inventory level before, during and after disaster periods. The insensitivity of the optimal policies to any change in demand in the ±50 % range is a particularly appealing feature of the proposed policies, given the difficulty in obtaining accurate demand forecasts under disaster effects.