Beartooth: Nuclear Material Processing with MBSE and Digital Twin

effort and lead to uncontrollable risk during project execution, resulting in
significant delays and cost overruns.
Furthermore, these projects, when completed, are not designed for the rapid digitization that allows for
advanced analytics, such as artificial intelligence (AI), machine learning (ML), or digital twinning. These
digital advancements provide great opportunities to reduce cost and schedule. The path to these
opportunities is embracing digital engineering. Digital engineering can break down these siloes and
ensure a project is ready to be part of the digital future from its onset. The wide variety of tools, vendors,
and processes creates the need for a standard toolset to normalize and integrate these datasets. To that
end, Idaho National Laboratory (INL) has developed DeepLynx as a key tool in solving this problem.
DeepLynx brings those siloed efforts into an integrated platform that operates over the course of a
projects lifecycle and integrates to widely used enterprise scale software.
Beartooth is a cutting-edge research and development testbed for processing novel special nuclear
material (SNM) feedstocks. In addition to its processing capabilities, the testbed will incorporate digital
engineering concepts, such as AI/ML and digital twins, to bring new insights into material processing that
will assist in alleviating problems that have plagued the nuclear industry. This testbed is physically
reconfigurable and can be used for novel chemical processes, which creates a challenge to ensure that
sensor systems are adequately configured to enable a digital twin. Additionally, the digital twin and its
data system will need to be equally adaptable and configurable.
This paper will go into the benefits, challenges, and lessons learned in using digital engineering on a
complex system utilizing model-based systems engineering (MBSE) and digital twin as core components
of the engineering effort. The paper will breakdown how MBSE was used in discovering early issues with
the configuration as initially laid out using a legacy document-based process. The paper will also discuss
the opportunities brought on by using a digital thread and twin to create interactive collaboration in design
and the future possibilities for these technologies during operation of the fuel cycle testbed. Finally, this
paper will go over the use and planned use of mixed reality and metaverse technologies for use in design,
training, and operation.