From demand forecasting to inventory ordering decisions for red blood cells through integrating machine learning, statistical modeling, and inventory optimization

Abstract

Background: The demand and supply of blood are highly variable over time.Blood inventory management that relies heavily on experience-based decisions may not be adaptive to real demand, leading to high operational costs, wastage, and shortages. Methods:We combined statistical modeling, machine learning, and optimization methods to develop a data-driven demand forecasting and inventory management strategy for red blood cells (RBCs). We then used the strategy to inform daily blood orders. A secondary semi-weekly (twice per week) ordering strategy was developed to handle the last-mile split delivery problem for blood suppliers, characterized by multi-deliveries to the same location multiple times during a short period of time. Both strategies were evaluated using the TRUST database including all patient data across four hospitals in Hamilton, Ontario. Results: We identified 227,944 RBC transfusions for 40,787 patients in Hamilton, Ontario from 2012 to 2018. The predicted daily demand from the hybrid demand forecasting model was not significantly different from the actual daily demand (paired t-test p-value = 0.163); however, the proposed daily ordering quantity from the model was significantly lower than the actual ordering quantity (p-value <0.001). The proposed daily ordering strategy reduced inventory levels by 38.4% without risk of shortages, leading to an overall cost reduction of 43.0% (95% confidence interval [CI]: 42.3%, 43.7%) compared with the actual cost. The semi-weekly ordering strategy reduced ordering frequency by 62.6% (95% CI: 61.5%, 63.7%). Conclusion:The proposed data-driven ordering strategy combining demand forecasting and inventory optimization can achieve significant cost savings for healthcare systems and blood suppliers.

Better than ‘normal’: Describing the distribution of and outliers in RBC usage

“…The log‐normal model slightly outperformed the normal model as well, although with the caveat that both informed models resulted in higher total RBC inventory which could potentially contribute to wastage. More sophisticated demand forecasting models could potentially reduce ordering frequency while limiting wastage 30 …”

“…More sophisticated demand forecasting models could potentially reduce ordering frequency while limiting wastage. 30 Limitations of this study include the lumping together of all RBC transfusions regardless of ABO or RhD type for a broader analysis. In addition, the measure of individual patient daily RBC usage is less applicable to research centered on total RBC usage during a patient's encounter, although it informs daily hospital RBC usage more directly.…”

Proceedings of the 2022 NHLBI and OASH state of the science in transfusion medicine symposium

“…With the advent of very large‐scale computational power (super computers, massively parallel processing, and cloud‐based analytics) and connected, interoperable healthcare data pipelines, there is the promise of data‐driven in‐silico scientific insights stemming from use of AI/ML in transfusion medicine research. Applications include blood demand forecasting and predicting transfusion requirements 61–63 . Techniques using ensemble and federated methods are relatively new to the field of transfusion medicine and merit focused investigation to define applicability to a range of transfusion research questions.…”

“…Applications include blood demand forecasting and predicting transfusion requirements. [61][62][63] Techniques using ensemble and federated methods are relatively new to the field of transfusion medicine and merit focused investigation to define applicability to a range of transfusion research questions.…”

A data‐driven approach to determine daily platelet order quantities at hospitals

“…Inventory management of blood products has received considerable attention from both the Operations Research [2][3][4][5][6][7][8] and Transfusion Medicine [9][10][11] literatures. Despite the large literature, the use of these methods in practice has been limited.…”