Research and prototype tools for analyzing BIM/CDE parameters, engineering model data and structured information from 3D BIM models.
This project focuses on the analysis of BIM model parameters and engineering data exported from CAD/BIM environments.
The main goal is to explore how structured model data can be processed, checked, grouped and prepared for further analysis using automated tools and AI-assisted workflows.
Large BIM models often contain many objects, systems and parameters. Manual review of this data is slow and difficult.
Typical issues include:
- inconsistent parameter naming;
- missing or empty parameter values;
- duplicated or similar parameters;
- different parameter structures across model objects;
- difficulty in grouping and analyzing engineering data;
- lack of automated tools for preparing BIM data for AI-based analysis.
The project explores workflows for extracting, organizing and analyzing BIM/CDE parameter data.
The focus is on practical data preparation for engineering analysis, quality control and future AI-assisted workflows.
- BIM/CDE parameter analysis
- Detection of missing or empty values
- Grouping of parameters and model objects
- Preparation of structured datasets
- Prototype scripts for engineering data processing
- Research workflow for AI-assisted BIM analysis
- CADLib model data
- CDE parameter exports
- Excel tables
- CSV files
- BIM object parameter tables
- Engineering system data
- Python
- Pandas
- Excel / CSV processing
- BIM data analysis
- CAD/BIM parameter structures
- AI-assisted data processing workflows
Research / prototype.
The project is connected with practical BIM data analysis and academic research on the integration of 3D BIM models and artificial intelligence.
- Parameter quality checking
- Automatic grouping of model objects
- Data cleaning workflows
- Export reports for engineering review
- AI-assisted analysis of BIM parameter structures
- Documentation with example datasets
The public repository contains project documentation and portfolio materials. Source code and datasets may be published later after cleanup and anonymization.