This paper presents a preliminary study on integrating Model Context Protocol (MCP) into Computer-Aided Engineering (CAE) workflows. We developed a proof-of-concept system that combines FreeCAD for 3D modeling, Gmsh for mesh generation, and ADVENTURE for finite element analysis, orchestrated through MCP-enabled AI. The system demonstrates automated design iteration, mesh quality optimization, and boundary condition generation through natural language interactions. Results show that MCP can effectively bridge the gap between AI reasoning capabilities and specialized engineering tools, enabling more intuitive and efficient CAE workflows. Performance evaluation indicates successful automation of the complete design-to-analysis pipeline with minimal human intervention.
MIYOSHI et al. (Wed,) studied this question.