A Use Case for Developing Meta Architectures with Artificial Intelligence and Agent Based Simulation in the Kidney Transplant Complex System of Systems

The gap between the demand for kidneys and the supply of kidneys keeps growing by approximately six percent per year. There is an opportunity to utilize discarded kidneys, with roughly twenty percent of kidneys discarded each year. A real-time generation of an optimized architecture is desired to assist kidney allocation. Current practice for allocating high-risk deceased donor kidneys requires organ procurement organizations (OPOs) to engage in a prolonged manual process, accruing cold ischemic time and exacerbating kidney quality. An interactive digital simulation tool is developed to identify high-risk kidneys that are hard to place and justify starting the accelerated placement process sooner. The simulation tool will estimate the increased likelihood of placement and improve the current allocation model by providing a systematic basis for changing the current process and demonstrating how simulation can be leveraged to evaluate a policy change before implementation. The simulation meta-architecture and AI were all integrated on the Anylogic platform to conduct real-time optimization and prediction of the kidney transplant system of systems. This integrated simulation allows UNOS to generate optimized policies for the kidney transplant systems of systems. Future work includes validating and verifying the platform with additional stakeholders and testing different policies with UNOS.