A Multistage Stochastic Optimization Model for Resilient Pharmaceutical Supply Chain in COVID-19 Pandemic Based on Patient Group Priority

In response to challenges brought about by the COVID-19 pandemic in the pharmaceutical supply chain (PSC), this study design the production-inventory-allocation problem to improve the resiliency of PSC during pandemics. It is essential to involve prioritizing patients based on their health status, ensuring those at higher risk of disease have sufficient access to medication. In this paper, we develop a multistage stochastic mixed-integer optimization model that considers various aspects including multiproduct, multi-period, patient group prioritization, human resource capacity, and emergency safety stock which are crucial to the efficient functioning of the PSC during a pandemic. We aim to minimize the total expected costs including purchasing cost of medicine from emergency and primary suppliers, inventory, hiring, transportation, backorder penalty, and ordering costs. This approach considers the demands as stochastic parameters and classifies the patients based on machine learning methods. The results present the optimal decision for allocation, inventory, flow, human resources, and transportation under diverse scenarios. We perform some analysis under diverse scenarios for inventory decisions. The results show that all cost types increase as demand rises except the inventory costs and prioritizing effectively reduces shortages among vulnerable patients.