Data Fusion at Scale: Strengthening Safeguards Conclusions Through Applied Analytics

The timely collection and analysis of all safeguards-relevant information is the key to drawing andmaintaining sound safeguards conclusions. The volume, velocity, and diversity of data available toState Evaluation Groups (SEGs)1 are growing exponentially, increasing the demands on SEGs to stayabreast of activity in relevant areas of the nuclear fuel cycle. In parallel, technologies have beendeveloped to efficiently process, store, and extract information from this data at scale. In responseto the opportunities and challenges this presents, the IAEA has developed the Collaborative AnalysisPlatform (CAP) to apply structured analytic techniques in an all-source environment in order totranslate this mass data into actionable information for faster, more informed decision making.The CAP is a socio-technical system that takes a holistic approach to the problem of analysis,including business process reengineering and the corresponding information technology to processhistoric and current safeguards relevant information. Data is fused from all relevant Safeguardsinformation systems and new systems have been developed to increase the collection capabilitiesacross the organization. Machine learning is combined with human-driven tools to augment theSEGs’ analytical activities. Through the use of standard work products, multi-level ontologies, taskorientedinterfaces, and clear mission directives and prioritization, the CAP enables the IAEA tosignificantly improve the comprehensiveness of the iterative cycle of collection, analysis,implementation, and evaluation.