SIMULATION OF A BLOOD NETWORK WITH COLLECTIONS AND DEMAND NORMALIZED TO CANADIAN POPULATION MEANS

Red blood cells (RBCs) can be classified into 8 types: 4 blood groups (A, B, O) and an Rh factor (+ or -). When a transfusion is required, the best outcomes are achieved when patients and donors share the same ABO/Rh. However, blood can be transfused from compatible donors to compatible patients; fo...

Full description

Bibliographic Details
Main Author: Hashemi, Neda
Other Authors: Department of Industrial Engineering, Master of Applied Science, Dr. Leslie Anne Campbell, Dr. Alireza Ghasemi, Dr. Corinne MacDonald, Dr. John T.Blake, Not Applicable, Yes
Format: Thesis
Language:English
Published: 2016
Subjects:
Online Access:http://hdl.handle.net/10222/72078
Description
Summary:Red blood cells (RBCs) can be classified into 8 types: 4 blood groups (A, B, O) and an Rh factor (+ or -). When a transfusion is required, the best outcomes are achieved when patients and donors share the same ABO/Rh. However, blood can be transfused from compatible donors to compatible patients; for instance, any patient can be transfused with O- blood. Because blood is also a perishable product, many hospitals, especially small facilities in remote locations, over stock with O type or A type blood. This requires suppliers to collect more of these types than would be expected, especially O- blood. In this study we adapt a simulation framework of a regional blood network to determine the impact of changing collection and transfusion practices in which blood is collected and used at rates equivalent to the ABO/Rh distribution of the underlying population. The model was developed to represent the blood supply chain in Newfoundland (a small region) and ported to Ottawa (a medium centre) and BC and Yukon (a large centre). The results of the study show that normalizing demand and collections to match the distribution of blood types within the Canadian population would lead to healthier inventory levels for most blood types, at the cost of additional hospital wastage. Results also show that O- shortage and emergency orders increase when collections and demand precisely match the distribution of blood types in the population. Therefore, we undertake an additional set of runs to determine the minimum level of O- over collection necessary to achieve nominal shortages for all blood types. Our results suggest that if hospitals request blood types at a rate equivalent to their distribution in the underlying population, collections of O- blood can be reduced from 5.8% above the population distribution to 1-2% above the population distribution without creating product shortages. Implementing a collection and ordering policy reflecting the distribution of blood types in the population would, however, increase hospital wastage to between 1.12% and 4.79% of average daily demand.