Improving Situational Awareness in Construction Projects
BIM RAG
A brief about the project
What is BIM?
Used in the construction industry to store data and metadata about a project along with its timelines and other factors.
Why is it used?
Reduces costs and also improves project workflow making it more effective.
What is the problem?
Due to the requirement of advanced skillsets, there is a big knowledge gap in the industry and it is difficult for people to collaborate efficiently.
What was the proposed solution?
Using AI RAG which will analyze the data and metadata of the projects and will have a simple text-to-text interface where the user can ask questions without even knowing anything about BIM or the project.
How does RAG work?
Any LLM model which generates an output, uses a much larger dataset of data to mine information. But RAG or Retrieval Augmented Generation limits this data input. Adding onto this, advanced prompt engineering will help get better results.
This part was the main issue as the BIM file (also called ifc files) was too large and too complex to understand, so I wanted to use an AI model to allow the user to ask questions like “What is the height of the building” and they would get the answer directly from the LLM instead of needing to sit and understand BIM files and navigate through them to get the data.
An example of how it would work
Without the tool
Manual Search: The manager spends time navigating through complex BIM files to find basic information.
With the tool
Instant Query: The manager types a question like "What is the building height?" and gets an immediate response.
Presented work to over 500+ visitors during the Open day on 23rd Feb 2034
Met Robert Metcalfe during his visit to iisc and discussed about the future of connectivity and user experience
Indian Institute of Science, Bangalore. 2023-2024
The Department of Design at IISc is a globally renowned hub for interdisciplinary research, blending technology with human-centered design to address complex, real-world challenges. Known for its work in Human-Computer Interaction, product design, and sustainability, the department excels in AI-driven, data-informed design.
As a Research Assistant in the Impact for Impact Lab, I contributed to projects focused on creating impactful, user-centered solutions, particularly in public health and sustainable development. My research combined data analytics with UX insights, driving innovations that address pressing societal needs.
