Using operational research methods to improve decision making in Canada's blood supply chain

Canadian Blood Services (CBS) spends $1B to provide Canadians with a safe supply of blood. As pressures mount to constrain costs, there is a need to improve the efficiency of the blood supply chain. Blood products are perishable and thus subject to wastage. Managing blood inventory is difficult because ordering decisions must consider the age of stock. Historically, the literature on blood supply chain management has focused on Red Blood Cell (RBC) inventory policies for a single supplier or consumer. However, because of proposed changes to the storage limits of red blood cells (RBC), end-to-end supply chain modelling has recently become a topic of interest for researchers and blood agencies around the globe. Currently, we are working on a modelling framework to represent ten regional blood distribution networks in Canada. Our ongoing work has produced an approach that represents a significant step forward in terms of scope, detail, and re-usability of supply chain models for regional blood distribution networks. We propose to extend our regional modelling framework to address four important issues in blood supply chain management. 1) The impact of a shorter shelf-life for RBC: Recent studies have raised questions about the safety of RBC after storage. However, reducing shelf life would likely create shortages or substantially increase product wastage. Thus, there is a need to determine the impact of a reduced shelf life for RBC. We will use our regional modelling framework to evaluate a shorter shelf life for RBC in each of the ten CBS blood distribution networks. By evaluating the impact of a shorter shelf life across a class of problems we can generalize results and frame the ongoing debate regarding the costs and benefits of transfusing newer blood. 2) Non-FIFO issuing policies: It has been well established that optimal policies for perishable blood products follow first-in-first-out (FIFO) rules. However, the spectre of a shorter shelf life for RBC has renewed interest in non-FIFO policies. Recently, several papers describing non-FIFO issuing have appeared. A distinguishing feature of this literature is the unitary scope of analysis, which focuses only on hospital issuing policies; the broader context of how non-FIFO issuing policies would work in a network has not yet been investigated. We will extend our framework to evaluate non-FIFO issuing policies from a network standpoint. 3) Age equitable distribution policies: The age of RBC issued to hospitals in Canada is not uniform. Larger facilities, which tend to turn stock more frequently, are supplied with older stock, while smaller facilities get newer stock. This practice maximizes product availability at distant sites and lowers delivery costs. However, it has been suggested that it leads to more frequent transfusion requirements and poorer outcomes for some patients. There is a need to develop issuing policies that promote equitable age distribution of blood products across an entire network to ensure all Canadians have equitable access to safe and effective blood products. 4) Dynamic inventory re-balancing: While inventory is managed on a regional level in Canada, blood is a national resource and materials flow, either on a planned or ad hoc basis, between regions. Approximately 20% of all RBC collected by CBS are transferred between sites. No study has established a scientific basis for control policies currently used to re-balance blood inventory in Canada. We will employ simulation based optimization (SBO) methods to evaluate policies for inventory re-balancing between CBS distribution sites. This project will help to reduce operational costs at CBS while promoting more reliable access to needed blood products across the country.