Blood components are perishable, scarce, and irreplaceable resources requiring strict governance. This study aims to develop and validate datadriven inventory management frameworks for regional transfusion networks in Northern Italy, addressing the trade-off between service continuity and waste reduction across multiple blood components and ABO/Rh blood groups. Two complementary analyses were conducted using real operational data from the transfusion departments of Vicenza Province (January 2023-May 2025, covering RBCs, platelets, and plasma across a hub-and-spoke network) and Brescia Province (2022-2024, focused on whole blood by blood group). In Vicenza, a weekly (s,S) inventory policy was calibrated using a 5-day lead time, MA5 demand forecasting, and a 95% target service level. Application of the policy reduced systematic bias in the two main hubs: for O+ at Vicenza, mean error decreased from approximately +43.8 to +4.1 units, with RMSE nearly halving (from ~50.1 to ~24.2). A Low/Medium/High severity framework enabled transparent, site-comparable operational monitoring. In Brescia, four forecasting models (Brown, Holt, Holt-Winters, Markov) were systematically compared. Holt-Winters outperformed alternatives for high-volume groups (O+, A+) due to superior handling of seasonality, while Markov transition matrices provided an allocative framework governing inter-group compatibility flows and protecting critical donors such as O-. Forecast-guided (s,S) thresholds achieved lower safety stocks at equivalent service levels for common groups, while higher z-quantiles were retained for rare and negative groups. Both studies demonstrate that operationally grounded inventory policies, anchored to observed network rhythms, component-specific lead times, and data-derived demand variability, can substantially improve transfusion stock management. The integration of time-series forecasting, (s,S) policies, and compatibility-aware allocation rules provides a transferable framework applicable to comparable provincial transfusion networks, consistent with Patient Blood Management principles and national self-sufficiency objectives.

Blood Supply Chain planning and optimization in Italian provincial transfusion networks

Lucio Zavanella;Beatrice Marchi;Anna Savio;Simone Zanoni
2026-01-01

Abstract

Blood components are perishable, scarce, and irreplaceable resources requiring strict governance. This study aims to develop and validate datadriven inventory management frameworks for regional transfusion networks in Northern Italy, addressing the trade-off between service continuity and waste reduction across multiple blood components and ABO/Rh blood groups. Two complementary analyses were conducted using real operational data from the transfusion departments of Vicenza Province (January 2023-May 2025, covering RBCs, platelets, and plasma across a hub-and-spoke network) and Brescia Province (2022-2024, focused on whole blood by blood group). In Vicenza, a weekly (s,S) inventory policy was calibrated using a 5-day lead time, MA5 demand forecasting, and a 95% target service level. Application of the policy reduced systematic bias in the two main hubs: for O+ at Vicenza, mean error decreased from approximately +43.8 to +4.1 units, with RMSE nearly halving (from ~50.1 to ~24.2). A Low/Medium/High severity framework enabled transparent, site-comparable operational monitoring. In Brescia, four forecasting models (Brown, Holt, Holt-Winters, Markov) were systematically compared. Holt-Winters outperformed alternatives for high-volume groups (O+, A+) due to superior handling of seasonality, while Markov transition matrices provided an allocative framework governing inter-group compatibility flows and protecting critical donors such as O-. Forecast-guided (s,S) thresholds achieved lower safety stocks at equivalent service levels for common groups, while higher z-quantiles were retained for rare and negative groups. Both studies demonstrate that operationally grounded inventory policies, anchored to observed network rhythms, component-specific lead times, and data-derived demand variability, can substantially improve transfusion stock management. The integration of time-series forecasting, (s,S) policies, and compatibility-aware allocation rules provides a transferable framework applicable to comparable provincial transfusion networks, consistent with Patient Blood Management principles and national self-sufficiency objectives.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11379/655186
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact