Deploying Frontier Agentic Technology in MOOSEnger, a Multiphysics-Capable AI Assistant

2026-08-16

AI Summary

This study addresses the high entry barriers and lack of intelligent assistance in MOOSE multiphysics simulations by proposing MOOSEnger-GPT-5.2, a localized AI agent. The system integrates retrieval-augmented generation, environment interaction verification, and persistent memory modules to establish an automated architecture featuring a complete execution chain and an experience accumulation closed loop. Experimental evaluations across eight simulation task categories demonstrate a 90% success rate, significantly outperforming baseline models. By effectively lowering the expertise threshold for domain specialists and enhancing end-to-end workflow efficiency, this work provides a reliable intelligent solution for complex scientific computing, thereby facilitating broader adoption of advanced multiphysics simulation frameworks.

Abstract

The Multiphysics Object-Oriented Simulation Environment (MOOSE) is an open-source finite-element framework for building multiphysics simulation applications. Using a multiphysics environment effectively demands specialized expertise, creating a barrier for many domain scientists and engineers. MOOSEnger, developed at Idaho National Laboratory (INL), is a domain-specific, tool-enabled AI agent built for the MOOSE Framework. This work extends MOOSEnger with a harness focused on locally-hosted models. The harness gives the agent a full pipeline: it retrieves contextual knowledge from the MOOSE repository, validates and diagnoses the resulting input through interaction with the simulation executable environment, and extracts and stores lessons in a persistent memory. The resulting framework is demonstrated on an engineering problem from the National Reactor Innovation Center Virtual Test Bed (VTB), illustrating its potential to support realistic multiphysics simulation workflows. Additionally, the agent performance is evaluated on different categories including diffusion, Navier--Stokes, phase field, plasticity, porous media flow, solid mechanics, transient heat transfer, and reactor mesh generation. Each category consists of 25 prompts/cases. We compare MOOSEnger-Gemma4 against MOOSEnger-GPT-5.2, alongside baseline Gemma4 and GPT-5.2 without agentic capabilities. MOOSEnger-GPT-5.2 shows a slight edge, achieving a 90\% success rate versus 76.5\% for MOOSEnger-Gemma4. The baseline models perform far worse, at just 5\% (GPT-5.2) and 0\% (Gemma4), underscoring the impact of the agentic harness.

Problem

• Multiphysics Simulation
• MOOSE Framework
• Domain Expertise Barrier
• AI Agent

Innovation

• Agentic AI
• Multiphysics Simulation
• Local Model Harness
• Persistent Memory
• Executable Environment Validation