The development of advanced nuclear facilities is progressing rapidly around the world. Newly designed facilities have differences in structure and operation from existing nuclear facilities, so Safeguards by Design (SBD), which applies safeguards at the design stage, is important. To this end, designers should consider the safeguardability of nuclear facilities when designing the system. Safeguardability represents a measure of the ease of safeguards, and representative evaluation methodologies are Facility Safeguardability Analysis (FSA) and Safeguardability Check-List (SCL). Those two have limitations in the quantification of safeguardability. Accordingly, in this study, the Safeguardability Evaluation Method (SEM), which has clear evaluation criteria based on engineering formulas, was developed. Nuclear Material Accountancy (NMA), a key element of Safeguards, requires the Material Balance Area (MBA) of the target facility and performs Material Balance Evaluation (MBE) based on the quantitative evaluation of nuclear materials entering or leaving the MBA. In this study, about 10 factors related to NMA were developed, including MBA, Key Measurement Point (KMP), Uncertainty of a detector, Radiation signatures, and MUF (Material Unaccounted For). For example, one of the factors, MUF is used in MBA to determine diversion through analysis of unquantified nuclear materials and refers to the difference between Book Inventory and Physical Inventory, as well as errors occurring during the process in bulk facilities, errors in measurement, or intentional use of nuclear materials. This occurs in situations such as attempted diversion, and accurate MUF evaluation is essential for solid Safeguards implementation. MUF can be evaluated using the following formula (MUF=(PB+X-Y)-PE). The IAEA’s Safeguards achievement conditions (MUF < SQ) should be met. Considering this, MUF-related factors were developed as follows. (􀜵􀜧􀜯 = 1 − 􀯆􀯎􀮿 􀯌􀯊 ) In this way, about 10 factors were developed and described in the text. This factors is expected to serve as an important factor in evaluating the safeguardability of NMA, and in the future, safeguardability factors related to Containment & Surveillance (C&S) and Design Information Verification (DIV) will be additionally developed to conduct a comprehensive safeguardability evaluation of the target facility. This methodology can significantly enhance safeguardability during the design stage of nuclear facilities.

Nuclear Material Accountancy (NMA) system quantitatively evaluates whether nuclear material is diverted or not. Material balance is evaluated based on nuclear material measurements based on this system and these processes are based on statistical techniques. Therefore, it is possible to evaluate the performance based on modeling and simulation technique from the development stage. In the performance evaluation, several diversion scenarios are established, nuclear material diversion is attempted in a virtual simulation environment according to these scenarios, and the detection probability is evaluated. Therefore, one of the important things is to derive vulnerable diversion scenario in advance. However, in actual facilities, it is not easy to manually derive weak scenario because there are numerous factors that affect detection performance. In this study, reinforcement learning has been applied to automatically derive vulnerable diversion scenarios from virtual NMA system. Reinforcement learning trains agents to take optimal actions in a virtual environment, and based on this, it is possible to develop an agent that attempt to divert nuclear materials according to optimal weak scenario in the NMA system. A somewhat simple NMA system model has been considered to confirm the applicability of reinforcement learning in this study. The simple model performs 10 consecutive material balance evaluations per year and has the characteristic of increasing MUF uncertainty according to balance period. The expected vulnerable diversion scenario is a case where the amount of diverted nuclear material increases in proportion to the size of the MUF uncertainty, and total amount of diverted nuclear material was assumed to be 8 kg, which corresponds to one significant quantity of plutonium. Virtual NMA system model (environment) and a divertor (agent) attempting to divert nuclear material were modeled to apply reinforcement learning. The agent is designed to receive a negative reward if an action attempting to divert is detected by the NMA system. Reinforcement learning automatically trains the agent to receive the maximum reward, and through this, the weakest diversion scenario can be derived. As a result of the study, it was confirmed that the agent was trained to attempt to divert nuclear material in a direction with a low detection probability in this system model. Through these results, it is found that it was possible to sufficiently derive weak scenarios based on reinforcement learning. This technique considered in this study can suggest methods to derive and supplement weak diversion scenarios in NMA system in advance. However, in order to apply this technology smoothly, there are still issues to be solved, and further research will be needed in the future.