Data-Driven Platelet Inventory Management Under Uncertainty in the Remaining Shelf-Life of Units

Abstract

Platelets are perishable (3-5 day shelf-life) blood products required for a variety of clinical treatments. In North America, hospitals typically procure platelet units from a central supplier. As such, the remaining shelf-life of the delivered units could be subject to high uncertainty. Our work focuses on developing new models that leverage the increasingly available data from hospital information systems to prescribe ordering decisions in the presence of this uncertainty. Specifically, we consider a periodic-review, perishable inventory system with zero lead-time and uncertainty in demand and remaining shelf-life of orders, operating under an Oldest-Unit, First-Out (OUFO) allocation policy. We adopt an Empirical Risk Minimization (ERM) approach to obtain linear decision rules using historical data on demand and features correlated with demand. The decision rules map the observed features to a daily ordering quantity and are obtained by minimizing an approximate measure of the in-sample empirical cost comprised of a weighted sum of shortage and expiry costs. The fixed initial age model is obtained assuming a constant remaining shelf-life for all units. The model is obtained assuming that an adversary selects the remaining shelf-life of units subject to an uncertainty budget determined through an endogenous uncertainty set. We investigate the out-of-sample performance of the proposed models in a case study using data from two Canadian hospitals. Both models achieve significant improvements over the historical expiry and shortage rates observed in the data. For instance, when shortage is assumed to be twice as costly as expiry, the fixed initial age model achieves a 86% and 93% reduction in expiry rate, and a 82% and 99% reduction in shortage rate for the two hospitals, respectively. The model achieves better out-of-sample generalizability and demonstrates a more robust performance under counterfactual remaining age distributions.