| Summary: | A computational tool to assess the most likely path a state proliferator would take in making a nuclear weapon was created in a Bayesian network. The purpose of this work was to create a tool to facilitate analysts and policymakers in learning about state proliferation. In carrying out this work, a previous Bayesian network based on nuclear weapon proliferation was expanded to include dual-use export controlled technologies. The constant nodes in the network quantifying technical capability, international networking, and available infrastructure were developed to be based on pertinent characteristics that were appropriately weighted. To verify the network, nine historical cases of state proliferation were tested over time, and the enrichment and weapon pathways were graphed. The network sufficiently modeled the cases, so it was concluded that, while one can never truly being able to sufficiently validate a network of this type, sufficient verification was achieved. The tool was used to gain knowledge and insight concerning technology transfers with four countries in hypothetical cases. This exercise proved that the network can in fact be used to learn about state proliferation under different policies and conditions.
|
|---|
