Automating BIM (IFC) Data Analysis Using LangGraph
Bibliographic record
Abstract
This study introduces a novel methodology for automating the analysis of Building Information Modeling (BIM) data using LangGraph and integrating Google's Gemini Large Language Model (LLM) with IfcOpenShell.BIM, and specifically Industry Foundation Class (IFC) files, are widely used in the construction industry for representing and managing building data.However, analyzing this data effectively remains a significant challenge due to its volume and complexity.Additionally, analyzing BIM data typically requires knowledge of different BIM software depending on the application.This research addresses this challenge by creating a workflow that utilizes LangGraph's ability to develop different AI agents designed to handle tasks like extracting element data, analyzing spatial relationships, and categorizing risks based on predefined criteria, without the need for any BIM software at all.The integration of Gemini LLM provides advanced language-based reasoning and decision-making capabilities that allow the system to process complex queries, in human language, and provide valuable insights from the BIM data.As a proof-of-concept, four applications of the LangGraph methodology were created, providing significant insights regarding the strengths and limitations of this framework.The models were validated through hypothetical case studies and real-world applications, and responses were evaluated based on their accuracy, validity, and completeness, demonstrating the framework's effectiveness in analyzing BIM data in construction projects.However, the results also revealed limitations that can affect the system's performance in large-scale real-world applications.These findings suggest that while the proposed system shows great potential, further optimization is needed to enhance its usability and reliability in more complex and large-scale scenarios.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.004 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".