PNNL

Abstract

Process monitoring traditionally relies on facility operator measurements. New concepts for future inspection regimes include inspectorate-owned systems that enable unattended monitoring (UM) of all declared material flow \cite{smith_potential}. If such systems can be established as viable alternatives, they may provide an important new capability: a true inspector's estimate of MUF that can achieve the efficiency and effectiveness goals outlined by the IAEA. A rigorous statistical framework for comparing traditional and new methodologies is needed to evaluate this potential and enable inspectorates to optimally balance UM and sampling in the context of timely detection and cost constraints. Recognizing that these are only two among a wide range of safeguards activities that support mass balance evaluation to detect missing material, this paper compares UM to traditional safeguards sampling in terms of detection probabilities as a function of time based on simplistic material loss scenarios, and shows how the detection probabilities can be rather easily combined. Using this approach can help enable tradeoff analysis and optimal resource allocation.