Human-Enhanced Loop Modeling (HELM): Agent-Based Finite Element Modeling of Concrete Bridge Barriers

The paper introduces the Human-Enhanced Loop Modeling (HELM) framework, which automates the finite element modeling process for concrete bridge barriers by integrating human oversight into a structured agent-based protocol. It demonstrates a significant improvement in modeling success rates, increasing from 20% to 75% across a 20-case matrix under MASH TL-4 and TL-5 loading conditions, utilizing ANSYS and LS-PrePost software. This framework highlights the importance of human-in-the-loop interventions in addressing spatial reasoning and algebraic logic limitations, providing open-source agent design code to facilitate further research and application in safety-critical infrastructure modeling.