Title

Agent-based Modeling for Off-Normal Modeling of Fuel Cycles: Disruptions and Diversions

Abstract

Nuclear fuel cycle modeling - the tracking of nuclear material as it passes among many facilities - allows us to develop insight into socio-political outcomes of changes from the status quo, whether related to resource utilization, waste management, economic performance, or risk of nuclear weapons proliferation.  Agent-based modeling is an approach that creates a virtual environment for different facilities to interact in complex ways according to a limited set of behaviors.  The Cyclus fuel cycle simulator was developed to provide flexibility in an open source tool to analyze transitions from one technology base to another, the impact of disruptions in those transitions, and the opportunities for diversion of material for clandestine military use.  This seminar will outline the initial motivations and design choices for Cyclus, describe some innovative analysis of disruptions to fuel cycle transitions, and outline ways to adapt Cyclus for analyses relevant to nuclear non-proliferation.  Graduate student research opportunities are available for those wanting to join the Computational Nuclear Engineering Research Group.

Bio

Paul Wilson is the Grainger Professor of Nuclear Engineering and current department chair of the University of Wisconsin-Madison‘s Department of Engineering Physics. His research interests focus on developing improved tools for computational modeling of complex nuclear energy systems, with applications in radiation shielding, nuclear waste management, nuclear non-proliferation and energy policy. Wilson joined the University of Wisconsin-Madison as an assistant professor in August 2001 as part of the Energy Systems and Policy Hiring Initiative.

A reception with light refreshments will be held immediately following the colloquium.