Pattern Recognition Options to Combine Process ...

Nuclear material accounting (NMA) is a component of nuclear safeguards, which are designed to deter and detect illicit diversion of special nuclear material (SNM) from the peaceful fuel cycle to a weapons program. NMA consists of periodically, but at relatively low frequency, comparing measured SNM inputs to measured SNM outputs, and adjusting for measured changes in inventory. Process monitoring (PM) is a relatively recent component of safeguards that consists of data more frequently collected than NMA data. PM data are often only an indirect measurement of the SNM and is typically used as a qualitative measure to supplement NMA, or to support indirect estimation of difficult-to-measure inventory for NMA. This paper introduces quantitative diversion detection options for NMA and PM data, which can be regarded as time series of residuals. Unique statistical challenges in combining NMA and PM residual time series include: PM and NMA data are collected at different frequencies; PM residuals often have a probability distribution that cannot be adequately modeled by a Gaussian distribution, not all PM and NMA data streams are independent, and the monitoring scheme must have reasonably high detection probability for both abrupt and protracted diversion.