The impact of transshipment on an integrated platelet supply chain: A multi-stage stochastic programming approach

This paper presents a multi-stage stochastic optimization model to minimize total cost under centralized control for a multi-echelon platelet supply chain which takes into account transshipment, demand uncertainty, stock age information, two types of supply, and three types of patients. The objective is to study the impact of trans-shipment between hospitals on performance improvement using a multi-stage stochastic framework to capture dynamic demand so as to explore approaches to improve an integrated platelet supply chain. Applicability of the model is tested through a real-world platelet supply chain with one blood center, three collection facilities, and five hospitals. The proposed multi-stage stochastic model is shown to perform better than a two-stage stochastic program in handling random daily demand and it also shows transshipment helps with alleviating shortage and wastage and reallocating stock at hospitals to improve system cost efficiency. Further, sensitivity analysis in-dicates that recruiting more apheresis platelet donors at hospitals and maintaining an initial inventory level that is close to the daily demand of hospitals are efficient strategies to improve performance of the platelet supply chain. In addition, we observed that imposing a FIFO issuing policy is not necessary for achieving optimal performance. These findings suggest that administrators and decision makers can improve cost efficiency in the platelet supply chain by implementing transshipment between hospitals, recruiting more donors, maintaining a suitable inventory level, and making collection and production plans in a shorter stage length.