Rogue actors need certain radioactive materials to construct nuclear weapons. In an effort to keep these materials out of their hands, states develop methods for identifying, quantifying, and locating these materials. This is especially important for high-throughput areas such as border crossings and ports. From a mathematical or algorithmic perspective, the problem of determining nuclear material identity, quantity, and location based on limited detector measurements can be categorized as an inverse problem. With the maturation of parallel simulation tools for neutral-particle radiation transport, solving these discrete inverse problems now is within our grasp, even in three dimensions. While my dissertation research is the development of a method to characterize a general radiation source, we focus here on a similar, but simpler, problem. A recurring problem in the literature, we estimate the thicknesses of the materials in a shielded uranium system by using probabilistic methods.
