Open Access
Choi, Sung Hee
Graduate Program:
Industrial Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
March 05, 2015
Committee Members:
  • Soundar Rajan Tirupatikumara, Dissertation Advisor
  • David Arthur Nembhard, Committee Member
  • Tao Yao, Committee Member
  • Kyusun Choi, Committee Member
  • Blood Supply Chain
  • Agent-Based Simulation modeling
Blood is a critical resource for hospitals and other care giving institutions, and shelf life of blood units tends to be very short. Moreover, the demand for the blood units generally outruns the supply. So, it is important to use and manage blood units efficiently. A community blood bank normally struggles to distribute the blood units based on the demand at each hospital under its geographical reach in fair a manner, keeping shortage as low as possible. The community blood bank should assign blood units through a balanced assignment strategy. Hospitals strive to secure blood units through an efficient ordering policy through which they can fulfill orders, minimize wastage and shortage. Each of these institutions - hospitals and blood banks have their own goals, are autonomous in decision making and exemplify a multi agent system. Given the conflicting yet collaborative nature of the objectives, this research addresses the issue of generating optimal blood unit assignment policies for the blood bank and optimal ordering policies for hospitals working in a collaborative manner. We develop methods for communication between hospitals and blood banks. We formulate and solve for optimal ordering policies for hospitals and assignment policies for community blood banks. From this understanding we suggest two ordering policies to fulfill the shortfall to test whether the current ordering policies are appropriate or not. To compare the effectiveness of optimal solutions, we build a multi-agent based simulation model in NetLogo and study different scenarios and the effectiveness of optimal solutions generated. We use the metrics of wastage and shortfall in each hospital to study the effectiveness of optimal solutions.