Fire and Smoke Digital Twin – A computational framework for modeling fire incident outcomes

Jufeng Jiao: University of Texas at Austin, Austin, Texas, USA jjiao@austin.utexas.edu

Ryan Hardesty Lewis University of Texas at Austin, Austin, Texas, USA rhl@austin.utexas.edu

Kijin Seong University of Texas at Austin, Austin, Texas, USA kijin.seong@austin.utexas.edu

Arya Farahi University of Texas at Austin, Austin, Texas, USA arya.farahi@austin.utexas.edu

Dev Niyogi University of Texas at Austin, Austin, Texas, USA dev.niyogi@jsg.utexas.edu

Paul Navratil Texas Advanced Computing Center Austin, Texas, USA

Nate Casebeer Austin Fire Department, Austin, Texas, USA

Team

https://github.com/UrbanInfoLab/FireIncidentBackend. This repository contains strictly the back-end files required torun our model. The base back-end finds itself in a curated list ofexecutable events to fetch, clean, and synthesize data on an hourlybasis from chosen fire departments. Our custom implementation of the VSmoke algorithm is also included, alongside our MantaFlow algorithm, with proper usage examples.

FireIncident Backend

https://github.com/UrbanInfoLab/FireIncidentData. This repository contains all accumulated data from running ourmodel for the last year. Multiple cities like Los Angeles, Houston,Seattle, and others are included in addition to Austin, and all datais collected and updated on a nightly basis to the repository. Wealso provide access to an API integrating with our live server incase the data is needed on a real-time basis.

The material on this website can be used freely in any publication provided that:
1. It is duly credited as a project by the UT Austin Urban Info Lab 2. A PDF copy of the publication is sent to jjiao@austin.utexas.edu