Fluxus AI

Login authentication via Speckle
Import model parameters via Speckle Stream
Amendment of IFC parameters post validator checks
Isolation of selected component in Speckle Viewer

Inspiration

The AEC industry faces significant inefficiencies and has been slow to embrace digital transformation and the ongoing AI revolution. To propel the AEC industry forward, it is essential to automate mundane and repetitive tasks, which are prevalent sources of inefficiency. Computers and AI excel at these menial tasks, outperforming humans in speed and accuracy. This observation inspired the development of Fluxus AI—an automated IDS validator designed to streamline the preparation of BIM/IFC models for submissions to building authorities.

What it does

As an IDS validator, Fluxus extracts information from IFC and BIM models uploaded to Speckle via Speckle Stream. It validates the IFC information against the IFC IDS mapping file provided by the user. After validation, Fluxus offers detailed results, allowing users to make necessary adjustments to model parameters. These adjustments are then synchronized with the connected BIM model via Speckle Stream, ensuring that the Revit model's parameters are updated accordingly. This process significantly reduces the time and effort required to ensure compliance of IFC models with submission standards, particularly within the context of Singapore.

How we built it

Fluxus consists of two main components: a frontend repository and a backend repository, which includes the server for IDS validation and parameter updating within the connected BIM model. For data storage and transmission, we utilize Speckle and Speckle streams.

Challenges we ran into

We encountered difficulties in updating the parameters of the connected BIM model via Speckle. Additionally, mastering the functionalities of Speckle Viewer, such as isolating components by ID and coloring them within the model, proved challenging due to the lack of detailed documentation or guides.

Accomplishments that we're proud of

We are proud to have achieved the fundamental functionalities set out for Fluxus, serving as a successful proof of concept. The application effectively performs data validation and updates, potentially saving countless man-hours traditionally spent manually checking and validating IFC and BIM models to meet submission standards.

What we learned

This hackathon taught us that Speckle is an invaluable tool for AEC digital application development, providing seamless connectivity across diverse AEC platforms. It enables developers to access various model types and information standards through a unified platform. Another key insight is the need for increased collaboration among innovators within the AEC industry to drive the necessary transformations.

What's next for Fluxus AI

In future iterations, we plan to integrate Large Language Models (LLMs) to further enhance Fluxus AI's capabilities. The LLMs will process parameters from models uploaded via Speckle Stream and automatically suggest appropriate parameter values, further reducing the effort required for compliance amendments.