The objectives of this Grant Opportunity for Academic Liaison with Industry (GOALI) award are: (1) to develop and analyze mathematical and computational models of key elements of the blood products collection, production, and inventory system at a large regional level in order to improve the overall system performance and (2) to advance the state of knowledge in the analysis of these models. The methods that will be used and extended include: completely observed and partially observed Markov decision processes (MDPs) and Markov games, chance constrained MDPs, artificial intelligence-based heuristic search algorithms, mathematical programming, exact and approximate dynamic programming, and stochastic programming. In collaboration with the American Red Cross (ARC), a comprehensive decision support system (DSS) based on proposed models and algorithms will be built. The outcomes of this DSS will be tested for validation in real collection, production, and inventory environments at the ARC production facility, and at hospitals and blood banks in the region.
This research is intended to directly impact the quality of delivery in 120 regional hospitals and health facilities and hence can potentially affect millions of people in the Southeast United States. If successful, this research will lead to a more cost-effective blood products collection, production, and inventory system with fewer blood product unit stock outs and with a reduced number of blood product units exceeding their lifetime. Furthermore, the models and decision support tools developed as part of these projects may ultimately be adapted by other service regions in the United States.
