ML-Based IFC File Interpretation Automation
Budget: $750 – $1,500 USD
Project Description: Automated Interpretation of IFC Files Using Machine Learning
Background and Objective
The project is based on the ambition to streamline and automate the process of extracting relevant building information from IFC models. Today’s workflows are manual and time-consuming, particularly when it comes to setting up material specifications, cost estimates, and quantity take-offs. Our goal is to combine machine learning with IFC analysis to automatically identify rooms and building components.
Machine Learning-Based Interpretation of IFC Files
The model will be trained to recognize rooms and components based on objects within the IFC file.
Examples:
Window → Exterior wall
Toilet/Shower → Bathroom
Kitchen units → Kitchen
The model will define which walls, floors, and ceilings belong to the recognized room.
Objective for the First Phase
The first step of the project is to develop a prototype that can:
Read and analyze an IFC file
Automatically identify bathrooms, based on objects such as showers, toilets, sinks, etc.
Link the correct walls, floor, and ceiling to the identified room
Extract surface areas (m²) for each component
Technical Requirements and Methodology
The program will be developed in Python, using IfcOpenShell for reading and interpreting IFC files.
The model will initially use rule-based identification (e.g., object type = "Toilet").
In later phases, machine learning will be implemented to improve accuracy and adaptability.
The extracted data should be prepared for display in a web-based platform or exported to other systems for further use (e.g., Excel, PDF, or web integration).
Background and Objective
The project is based on the ambition to streamline and automate the process of extracting relevant building information from IFC models. Today’s workflows are manual and time-consuming, particularly when it comes to setting up material specifications, cost estimates, and quantity take-offs. Our goal is to combine machine learning with IFC analysis to automatically identify rooms and building components.
Machine Learning-Based Interpretation of IFC Files
The model will be trained to recognize rooms and components based on objects within the IFC file.
Examples:
Window → Exterior wall
Toilet/Shower → Bathroom
Kitchen units → Kitchen
The model will define which walls, floors, and ceilings belong to the recognized room.
Objective for the First Phase
The first step of the project is to develop a prototype that can:
Read and analyze an IFC file
Automatically identify bathrooms, based on objects such as showers, toilets, sinks, etc.
Link the correct walls, floor, and ceiling to the identified room
Extract surface areas (m²) for each component
Technical Requirements and Methodology
The program will be developed in Python, using IfcOpenShell for reading and interpreting IFC files.
The model will initially use rule-based identification (e.g., object type = "Toilet").
In later phases, machine learning will be implemented to improve accuracy and adaptability.
The extracted data should be prepared for display in a web-based platform or exported to other systems for further use (e.g., Excel, PDF, or web integration).