Data-Driven Geographic Diversification for Pandemic Resilient Supply Chain Network

Manufacturing and service industries across the world are discovering the vulnerabilities in their supply chain network design and strategic and operations policies during the global COVID-19 pandemic. Both industries are struggling to manage unforeseen fluctuation in both demand and supply. This unpreparedness is attributable to the insufficient consideration for building resiliency while designing supply chains and developing strategic and operations policies. Historical data suggests that a pandemic with the severity and scale like COVID-19, SARS, and MERS starts from the one epicenter, which can be replaced by another epicenter or appended by multiple epicenters across the world as infected people travel to and from different locations and act as carriers. Evidence shows that there is a time lag in the range of months before a shift or a new appearance of epicenter occurs. Moreover, all regions do not experience the same level of severity or the same length of time to recover. Therefore, incorporating geographic diversification in supplier selection models is necessary for making supply chains resilient against events like COVID-19 pandemic. The state-of-the-art models for supplier selection or strategic-level supply chain network design often include geographic diversity into consideration to a limited degree, e.g. as expert ratings or as supplier-plant distance metrics. However, inclusion of such form of geo-diversification measures are not sufficient to develop supply chains decision in achieving resilience capacity to withstand disruptions during events like COVID-19 pandemic. In this research, therefore, we intend to design a data-driven robust decision framework for developing a resilient supply chain configuration considering the uncertainty in the trajectory of pandemic spread and the recovery pattern