Large-Scale Visual Exploration of Radiological and Nuclear Risk-Assessment Methods

• Activity Number: 293
• Type: Invited
• Date/Time: Tuesday, August 11, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Risk Analysis
• Author(s): Landon Sego* and Daniel Fortin and Robert Brigantic and Angela Waterworth and Casey J. Perkins
• Companies: Pacific Northwest National Laboratory and Pacific Northwest National Laboratory and PNNL and PNNL and PNNL
• Keywords: risk assesment ; exploratory data analysis ; big data ; radiological ; nuclear

Abstract

The Radiological and Nuclear Risk Assessment Methods (RNRAM) tool provides holistic risk assessment of radiological and nuclear threats and nuclear detection architectures. It includes a variety of unique capabilities: an extensive tree structure that represents potentially billions of scenarios, an adaptive adversary model that accounts for potential adversary behavior, and a layered architecture with many possible detection and interdiction points. We developed a process for using RNRAM to investigate questions like: What groups of scenarios lead to greatest risk? How do changes in the nuclear detection architecture affect overall risk? What is the relationship among the components of risk (i.e., threat, vulnerability, and consequence)? The stochastic risk simulations produced by RNRAM result in very large datasets that require additional investigation and analysis to be useful. To analyze these data, we used Tessera, an open-source suite of R packages designed to facilitate exploratory data analysis and statistical computing on large datasets.